978 3 642 71552 5

Ecophysiology
of Spiders
Edited by Wolfgang Nentwig
With Contributions by
C. W Aitchison . I-C. Bonaric . I E. Carrel
I L. Cloudsley-Thompson . K.-G. Collatz . R. E. Coville
I E. Dalingwater . A. E. Decae . S. Heimer' A. Holl
WE Humphreys' R. R. Jackson' W Kirchner' I Kovoor
A. Lopez' A. M. Macnab' Z. Maretit . K. Nakamura
W Nentwig . S. E Perry· H. M. Peters' G. O. Poinar, Jr.
S. D. Pollard' R. Pulz . M. Schaefer' E. I. Schlinger
K.-A. Seitz' E Strazny . E. K. Tillinghast· M. Townley
E Vollrath
With 133 Figures
Springer-Verlag
Berlin Heidelberg New York
London Paris Tokyo
Dr. VVOLFGANGNE~G
Institut fUr Zoologie
Universitat Regensburg
UniversitatsstraBe 31
D-8400 Regensburg, FRG
Cover illustration:
Mastophora, a bola spider (Araneidae)
Redrawn from GERTSCH, W.J. (1947) Spiders that lasso their prey.
(Nat. Rist. 56:152-158)
ISBN-13:978-3-642-71554-9 e-ISBN-13:978-3-642-71552-5
DOl: 10.1007/978-3-642-71552-5
Library of Congress Cataloging-in-Publication Data. Ecophysiology of spiders. Bibliography: p.

Includes index. 1. Spiders - Physiology. 2. Spiders - Ecology. 3. Arachnida - Physiology.
4. Arachnida - Ecology. I. Nentwig, Wolfgang, 1953- . II. Aitchison, C.W. QL458.4.E26
1987 595.4'4041 86-22077.
This work is subject to copyright. All rights are reserved, whether the whole or part of the material
is concerned, specifically those of translation, reprinting, re-use of illustrations, broadcasting,
reproduction by photocopying machine or similar means, and storage in data banks. Under § 54 of
the German Copyright Law, where copies are made for other than private use, a fee is payable to
"Verwertungsgesellschaft Wort", Munich.
© Springer-Verlag Berlin Heidelberg 1987
Softcover reprint of the hardcover 1st edition 1987
The use of registered names, trademarks, etc. in this publication does not imply, even in the absence
of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
2131/3130-543210
Preface
Recently another book on insect physiology was published. It was restricted

to a few focal points as are many of these new insect physiology books, but
there was considerable depth in its specialized point of view. We were dis-
cussing the structure of this book and of insect physiology books, in general,
when Prof. Remmert asked me " ... and what about books on spider physio-
logy?" Silence. Then I started to explain "oh yes, there is a congress pro-
ceedings volume on this topic and there is a group with excellent publica-
tions on another topic ... ", but I felt that this answer was weak. One can no
longer buy the proceedings volume in a bookshop and to read a series of
publications on a given topic one must search in a library for a dozen journals.
Why is there not a single book on spider physiology comparable with the
many books on insect physiology? Are spiders a scientific ivory tower, far
from public interest and commercial importance? I do not think so, although
spiders are one of the many "forgotten" animal groups which always grew
in the shadow of the insects. There are research groups working on spider
physiology, there are fascinating phenomena in this animal group and there
are plenty of exciting results. Spiders may have been always underresearch-
ed, but research is progressing. In the last few years, new books have been
published, e.g. field guides on spiders and books on behaviour and neuro-
biology of spiders. Moreover, a volume on spider ecology will soon appear.
Though spider taxonomy is still disastrous, physiological research has in-
creased considerably in the last decade. Many spiders have proved to be
"good laboratory animals" and investigations of the invertebrate silk com-
position or the haemocyanin structure are inseparably linked with some spe-
cies of spiders. From an ecological point of view more and more attention is
being paid to these ubiquitous predators and in the field of integrated insect
pest control spiders are perhaps a key group. Therefore, I think that this is a
good time to gather together our knowledge of the physiology of spiders.
A book on the physiology of spiders perhaps promises too much. Mole-
cular biochemistry in spiders is just beginning and there is more "ocean"
than "safe land". Ecophysiology, on the other hand, circumscribes a wide
range of physiological facts which determine the ecology of the species.
Therefore, this term provides more a framework for the contents of this book
rather than an exact description. I prefer to use "ecophysiology" in a broad
sense, which includes different factors such as thermal relations, gland physio-
VI Preface
logy, silk biology, feeding ecology, parasites or biorhythms. There are other
topics which I would have gladly included in this book, but nothing is known
about the immune system of spiders, special metabolic pathways, or the bio-
chemistry of pheromones (to mention only some examples). There are, how-
ever, a few other gaps in the completeness of this book, mainly due to tech-
nical reasons.
When I decided to edit a book which covers such a wide range I realized
soon that its success would depend mainly on the cooperation of the con-
tributors. I want to thank the many co-workers who made this work pos-
sible. None of them really had the time to write a book chapter, but I could
convince them all to do so. Most of them critically read other chapters of
this book or gave general advice. Special thanks are due to Prof. Dr. H. Rem-
mert for his initial impetus and Dr. D. Czeschlik and the Springer Verlag
for their eager interest. Further thanks go to many people who helped me
in various ways: H. Hallmer, J. Kien, M. Kredler, R. Loftus, R. Novack,
B. Tschirner, C. Warcup. Special thanks go to my family who had to do
without all that time I spent on this book.
Regensburg, Autumn 1986 Wolfgang Nentwig

Contents
A Cuticle, Temperature and Respiration
I Chelicerate Cuticle Structure

J.E. Dalingwater (With 5 Figures) 3
II Coloration and Chromes
A. Holl (With 4 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . .. 16
III Thermal and Water Relations
R. Pulz (With 6 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26
IV Behavioural Temperature Regulation
W.F. Humphreys (With 4 Figures) . . . . . . . . . . . . . . . . . . . . . 56
V Behavioural and Physiological Adaptations to Cold
W. Kirchner (With 1 Figure) . . . . . . . . . . . . . . . . . . . . . . . . . 66
VI Respiratory System: Structure and Function
F. Strazny and S.F. Perry (With 5 Figures) ............... 78
VII Heart Rate and Physiological Ecology
J .E. Carrel (With 7 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 95
VIII Moulting Hormones
J.-C. Bonaric (With 3 Figures) . . . . . . . . . . . . . . . . . . . . . . . . 111
B Glands, Silk and Webs
I Glandular Aspects of Sexual Biology

A. Lopez (With 2 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . 121
II Communication with Chemicals: Pheromones and Spiders
S.D. Pollard, A.M. Macnab, and R.R. Jackson
(With 3 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
III Spider Venoms and Their Effect
z. Maretic (With 7 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . . 142
VIII Contents
IV Comparative Structure and Histochemistry of Silk-Producing

Organs in Arachnids
J. Kovoor (With 12 Figures) . . . . . . . . . . . . . . . . . . . . . . . .. 160
V Fine Structure and Function of Capture Threads
H.M. Peters (With 9 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 187
VI Chemistry, Physical Properties, and Synthesis of Araneidae
Orb Webs
E.K. Tillinghast and M. Townley (With 3 Figures) . . . . . . . . . . . 203
VII Ecological Aspects of Spider Webs
W. Nentwig and S. Heimer (With 6 Figures) . . . . . . . . . . . . . . . 211
C Feeding Ecology
I Structure and Function of the Digestive Tract

K.-G. Collatz (With 2 Figures) . . . . . . . . . . . . . . . . . . . . . . . . 229
II Excretory Organs
K.-A. Seitz (With 5 Figures) . . . . . . . . . . . : . . . . . . . . . . . . . 239
III The Prey of Spiders
W. Nentwig (With 6 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 249
IV Feeding Ecology of Winter-Active Spiders
C.w. Aitchison (With 5 Figures) . . . . . . . . . . . . . . . . . . . . . . 264
V Kleptobiosis in Spiders
F. Vollrath (With 6 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 274
VI Hunger and Starvation
K. Nakamura (With 6 Figures) . . . . . . . . . . . . . . . . . . . . . . . . 287
D Parasites
Nematode Parasites of Spiders

G.O. Poinar, Jr. (With 2 Figures) . . . . . . . . . . . . . . . . . . . . . . 299
II Spider-Hunting Sphecid Wasps
R.E. Coville (With 2 Figures) . . . . . . . . . . . . . . . . . . . . . . . . 309
III The Biology of Acroceridae (Diptera): True Endoparasitoids of
Spiders
E.I. Schlinger (With 3 Figures) . . . . . . . . . . . . . . . . . . . . . . . 319
Contents IX
E Environment and Life History
I Life Cycles and Diapause

M. Schaefer (With 6 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 331
II Dispersal: Ballooning and Other Mechanisms
A.E. Decae (With 3 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 348
III Growth, Foraging and Reproductive Success
F. Vollrath (With 7 Figures) . . . . . . . . . . . . . . . . . . . . . . . . . 357
IV The Biorhythms of Spiders
J.L. Cloudsley-Thompson (With 3 Figures) ............... 371
Appendix .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 380
References ............................... ..... 389
Subject Index .......................... ........ 437

Contributors
You will find the addresses at the beginning of the respective contributions
Aitchison, Cassie W. 264 Macnab, Aynsley M. 133

Bonaric,lean-<::laude III Maretic, Zvonimir 142
Carrel, James E. 95 Nakamura, Kazuo 287
Cloudsley-Thompson, Nentwig, Wolfgang 211, 249
John L. 371 Perry, Steven F. 78
Collatz, Klaus-Gtinter 229 Peters, Hans M. 187
Coville, Rollin E. 309 Poinar, George 0., Jr. 299
Dalingwater, John E. 3 Pollard, Simon D. 133
Decae, Arthur E. 348 Pulz, Ralf 26
Heimer, Stefan 211 Schaefer, Matthias 331
Holl, Arthur 16 Schlinger, Evert I. 319
Humphreys, William F. 56 Seitz, Karl-August 239
Jackson, Robert R. 133 Strazny, Franke 78
Kirchner, Walter 66 Tillinghast, Edward K. 203
Kovoor , Jacqueline 160 Townley, Mark 203
Lopez, Andre 121 Vollrath, Fritz 274,357
A Cuticle, Temperature and Respiration
I Chelicerate Cuticle Structure
JOHN E. DALINGWATER 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Structure of the Arthropod Cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.1 Basic Subdivisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 4
2.2 Epicuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 4
2.3 Pro cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 5
2.4 Microstructural Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.5 Cuticular Sensillae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 8
3 Structure and Ultrastructure of the Cuticles of the Chelicerate Groups . . . . . . . . . . .. 8
3.1 Eurypterid Cuticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2 Xiphosuran Cuticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9
3.3 Acarine Cuticles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.4 Cuticles of Minor Chelicerate Groups . . . . . . . . . . . . . . . . . . . . . . . . . . " 10
3.5 Scorpion Cuticle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 10
3.6 Spider CutiCle. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12
4 Conclusions.............................................. 15
1 Introduction
It is almost impossible to exaggerate the significance of the cuticle in the life of an

arthropod, for the cuticle and the epidermal cells which secrete it together comprise
the tissue system which forms the exoskeleton. The ability of the cuticle to become
hardened in some regions, yet to remain soft in immediately adjacent areas, allows body
segments to move against adjacent segments and limb podomeres to operate against
adjacent podomeres. The strength and developmental plasticity of the cuticle has al-
lowed a tremendous variety of body shapes and limb types to be developed. Even one
of the supposed disadvantages of possessing a rigid exoskeleton, the necessity for peri-
odic moulting to allow further growth, may be viewed as conferring the advantage of
allowing gradual remodelling during development or even drastic reshaping during
metamorphoses.
Cuticle covers all external surfaces of the body and may also line tracheae, book
lungs and parts of the fore- and hind-gut. It can, therefore, act as a protective barrier
between the arthropod's tissues and the environment but equally, in appropriate areas,
can allow selective passage of gases or fluids between the animal's tissues and the ex-
1 Department of Zoology, The University, Manchester M13 9PL, England

4 J.E. Dalingwater
ternal medium. Arrays of sensillae arise from the cuticle and are capable of detecting
chemical, tactile and visual stimuli; ducts through the cuticle can play a part in its
development and can transport materials into it or onto its surfaces. So the cuticle,
though non-cellular, is not an inert covering but a complex and vibrant frontier be-
tween the arthropod and its environment. In short, the great success of the arthropod an
level of organisation rests firmly on the exploitation of the potential contained in cuticle.
2 Structure of the Arthropod Cuticle
2.1 Basic Subdivisions
The cuticle is secreted by epidermal cells; collectively, the basement membrane of the
epidermal cells, the epidermal cells themselves and the cuticle are termed the integu-
ment. The first major subdivision of cuticle is into a thin outer zone, often with several
distinct layers, termed the epicuticle and a thicker inner region, the procuticle, also
frequently further subdivided. By convention, the epicuticle is defined as an outer zone
containing no chitin, contrasting with the pro cuticle which does contain chitin. Chitin,
poly-N-acetylglucosamine, is the characteristic molecule of cuticle and can contribute
up to 50% of its dry weight. Proteins are the other major cuticular component: together
with chitin they account for more than 90% of the organic content of cuticles. Poly-
saccharide-protein complexes, enzymes, lipids, phenols and pigments are other signifi-
cant organic constituents. Inorganic material usually contributes less than 1% of dry
weight except in the calcified cuticles of crustaceans, diplopods and the larval and
pupal stages of a few insect species (R.H. Hackman 1984).
2.2 Epicuticle
Epicuticle is rarely more than a few micrometers thick, often considerably less. It
consists of a number of layers, often four or five. Correlation of the various layers
described from light and electron microscope studies can be difficult and discrepancies
in nomenclature, even between authors of various electron microscope studies, can also
cause confusion. Many authors (e.g. Neville 1975) recognise four layers. From outside
to inside these are: a thin, outermost cement layer; a wax or lipid layer; a cuticulin
layer, which is often the first-formed layer of the epicuticle and acts as a selective sieve
during resorption of the old endocuticle prior to ecdysis; and an inner epicuticul&r
layer which often makes up the bulk of the epicuticle. However, Hadley (pers. commun.
1985) has commented that a terminology for epicuticular layers should perhaps be
based on location and appearance rather than on supposed chemical composition, be-
cause of present uncertainty about the constituents of most epicuticular layers.
Although the whole cuticle may be involved in reducing and sometimes even con-
trolling water loss, the epicuticle, by virtue of its position and the composition of some
of its layers, is of great importance as a waterproofing barrier.
Chelicerate Cuticle Structure 5
2.3 Procuticle
Neville (1975) distinguished lateral and horizontal divisions of the procuticle. Lateral
divisions include the solid cuticle of sclerites, rubberlike cuticle and the soft cuticle of
arthrodial membranes between adjacent segments and podomeres. The major horizontal
divisions are exocuticle and endocuticle. Newly-formed cuticle consists entirely of un-
tanned endocuticle, but in sclerites part or nearly all of its outer region may subse-
quently be tanned to form exocuticle. Only regions of endocuticle formed before ec-
dysis become tanned to form exocuticle. The tanning process is complex, but in many
groups essentially involves impregnation with proteins and lipids and cross-linking by
quinone tanning. Exocuticle is often brown: the result ofthe tanning process and the
chemically-related, but distinct, process of melanisation. The intermediate, impregnated
but untanned, condition sometimes persists in areas of fully-formed cuticles as meso-
cuticle, identified by its staining reactions with Mallory's triple stain (see Neville 1975).
2.4 Microstructural Features
Fine vertical canals and horizontal laminations are commonly encountered microstruc-
tural features of procuticle. Towards the outer surface and within the epicuticle, minute
(ca. 7 nm) wax canal channels may be observed; several of these may fuse into a pore
canal which extends inwards to the epidermis. Pore canals are extremely abundant in
most cuticles and are considered to be cytoplasmic extensions of the epidermal cells;
they may transport impregnating materials into the cuticle, but their full and precise
range of functions has yet to be documented. Pore canals follow a pathway through
the cuticle that has been likened to that of a twisted ribbon, although in some organisms
their course is truly helical.
Explanations of the laminate nature of the procuticle have aroused considerable
controversy over the last 20 years. Under the light microscope this lamination appears
as narrow dark bands (laminae sensu Dennell 1976) separated by wider light bands
(inter-laminae); one dark ancj. one light band together comprise a lamina unit. In many
tangential sections of laminate cuticle, a series of bow-shaped arcs appear to pass across
the inter-laminae, thus linking successive laminae. In 1965, the French biologist and
mathematician Y. Bouligand suggested that the bow-shaped arcs were an artefact
resulting from sectioning sheets of parallel fibrils: each sheet is also parallel to the
cuticle surface and the orientation of the fibrils changes by a small angle between suc-
cessive sheets (Fig. 1). The dark bands observed under the light microscope thus re-
present regions where the orientation of fibrils is normal to the plane of the section.
The elegant simplicity of this model, the ease with which such a system can be imagined
to work in three dimensions, its compatibility with suggestions of the cuticle as a self-
assembling system and its linking in with an explanation of the twisted ribbonlike path-
way of the pore canals (Neville et al. 1969) quickly led to its widespread acceptance as
the universal and underlying explanation of lamination in all arthropod cuticles. Later
observations that fibrils might curve in arcs in the plane parallel to the cuticle surface
were readily incorporated into the model by its main proponents, Bouligand and Neville
(see Neville 1975). But acceptance of such 'fmgerprint patterns' for the fibril sheets
6 I.E. Dalingwater
~
:.--
- -- --
----:-------:------------
--
--- ----
---- ----- ---- --- b
Fig. la,b. BouJigand helicoidal model. (Redrawn after Bouligand 1965). a Truncated pyramid of
cuticle viewed from above. Flat sheets of fibres are arranged in stacks, and fibre orientation changes
from one sheet to the next in a counter clockwise direction. b Simplified, 'exploded' view of the
system: apparent laminae (I) result when sheets of fibres are sectioned parallel to fibre orientation,
i.e. in the plane of the page
presents a serious difficulty: the regular C-shaped arcs seen in angled light and electron
microscope sections can be explained as artefacts within the basic Bouligand model,
but with a fingerprint pattern regular arcs are most unlikely to be formed. Dennell
(1976, 1978) has maintained his view that laminae are real structures and that fibres
form true , not artefactual, arcs between successive laminae. His strongest evidence for
the reality of laminae is the presence of lamina membranes which persist even after
cuticles have been subjected to violent treatment in potash and acetic acid. Further
evidence supporting the reality of laminae is provided by examining right-angled pieces
of cuticle or specially shaped pyramids (Dalingwater 1975). According to the basic
Bouligand model, there should be a discontinuous shift in the position of artefactual
laminae around the right angle; in fact, laminae seem to be continuous.
Some procuticles contain areas of non-laminate cuticle where fibres are unidirection-
ally orientated, e.g. some areas of mesocuticle in Limulus. The two distinct types of
cuticle, laminated cuticle explained by the Bouligand model and cuticle with uni-
directional fibres, form the basis of Neville and Luke's (1969) two-system model of
cuticle architecture. There are also intermediate conditions between the two main
types; Barth (1973) has described some of these (Fig. 2).
Some details of the nature of the cuticular fibrils should now be discussed. Under
the transmission electron microscope (TEM) at a near-molecular level of organisation,
microfibrillar structures which appear to consist of lucent rods embedded in an opaque
matrix have been described from the cuticles of several arthropodan groups. The lucent
rods were interpreted as chitin crystallites embedded in a protein matrix, and this per-
mitted Neville (1970) to produce an analogy of cuticle with fibreglass, with the chitin
A c o
Fig. 2A-E. Variations in pattern of microfibre architecture (reproduced from Barth 1973 with per-
mission of the author). A Unidirectional type. B Bidirectional reinforcement with fibre orientation
changing abruptly by 90° between zones of unidirectional reinforcement. C Unidirectional cuticle
alternating with layers continuously rotating for a total of 90°. D Unidirectional cuticle alternating
with layers whose fibre directions rotate continuously for a total of 1800 • E Multidirectional type
with fibre direction rotating regularly and continuously. Arrows indicate the dimensions of a 'lamel-
la' as defined by a rotation of 1800
crystallites equivalent to glass fibres and the protein matrix comparable to the epoxy
resin filler. Furthermore, this analogy tied in with Jensen and Weis-Fogh's (1962) con-
clusions from mechanical experiments that cuticles behave as two-phase systems.
Giraud-Guille (1984) gave a number of reasons for questioning that the lucent rods are
chitin and instead suggested a rather different two-phase system: one formed by the
hydrophobic (dark) and hydrophilic (clear) parts of cuticular proteins, chitin linking
with the polar amino acids of the dark phase . In the same paper, which is largely con-
cerned with crab cuticle, Giraud-Guille showed how microfibrils are associated in dif-
ferent modes in different cuticular regions; these associations form macrofibrils. In-
cidentally, Bouligand's (1965) original helicoidal model was based on the interpre-
tation of macrofibrillar patterns, though Neville and Luke (1969) extended his con-
cepts to the microfibrillar level for insect cuticles and Barth (1969, 1970) to that
level for spiders. Associations of microfibres in macrofibres (Giraud-Guille's second
level of organisation) have not been described from arthropods other than crustaceans.
Her third level of organisation is the helicoidal model level discussed in detail above.
The intellectual appeal of the two-system model, which seems to explain the under-
lying organisation of all arthropod cuticles, is strong and a considerable body of evidence
can be assembled to support it. Yet many observations cannot readily be fitted into its
framework and the model is neither universally accepted nor of universal validity. So,
its truth should not be taken for granted, nor should it hinder investigations which
might produce results which cannot be interpreted within the context of this model.
8 I.E. Dalingwater
2.5 Cuticular SensiUae
A great variety of sense organs are intimately associated with arthropod cuticles. One
has only to examine the surface of almost any region of any arthropod cuticle to be
astounded by the variety and complexity of sensillae. As Steinbrecht (1984) has re-
marked, relatively few receptors have been examined both for fine structure and for
function even from insects, and the discrepancy between descriptive and experimental
work is even more pronounced for non-insectan arthropods. Steinbrecht (1984) has
provided an excellent review of arthropodan chemo-, hygro- and thermoreceptors,
which is complemented by Barth and Blickham's (1984) article on mechanoreception
in the same volume.
3 Structure and Ultrastructure of the Cuticles of the Chelicerate Groups
3.1 Eurypterid Cuticles
Fossil chelicerate cuticles have not been extensively studied, but the remarkable preserva-
tion of some eurypterid cuticles has encouraged their examination (Dalingwater 1975).
Eurypterid cuticles were almost certainly entirely organic, but impregnation by silica
has preserved many fine structural details. Most eurypterid cuticle remains are exuviae,
and in these only laminate regions of cuticle are present: curved fibrous patterns in
angled breaks can be resolved with the scanning electron microscope (SEM) as shown
in Fig. 3a. More rarely, complete cuticles are preserved and from these the presence of
Fig. 3a,b. Eurypterid cuticle. a Scanning electron micrograph of an angled break of Dunsopterus
cuticle. Preparation RSM GY 1887.25.1074.D from the Namurian of Scotland. Bar = llLm. blight
micrograph of gnathobasic tooth cuticle of the fourth appendage of Baltoeurypterus tetragonoph-
thalmus. Specimen BMNH I 3406 (12) from the Upper Silurian of Estonia. Bar = 10 ILm
preferred orientation zones of cuticle can be established. Pore canals and tegumental
ducts are also preserved and, in the remarkable Silurian Baltoeurypterus specimens, setal
sockets (Fig. 3b) and even remains of the setae themselves can be seen under the light
microscope. Tobien (1937) and Eisenack (1956) have described a whole range ofsensil-
lae from this material. Thus, characteristic features of cuticle, such as lamination and
its underlying fibrillar structure, pore canals and cuticular sensillae, were present in
chelicerate cuticles in the Palaeozoic era. Although the most striking material so far
described dates from the Silurian and Carboniferous periods, these features have also
been observed in Ordovician eurypterid cuticles by the present author (unpubl. work).
3.2 Xiphosuran Cuticles
Despite studies of various aspects of Limulus cuticle, its basic structure remains in-
completely analysed. Most of the structural studies have employed light or scanning
electron microscopy and so little is known about the detailed structure of the epicuticle.
Below the epicuticleis a thin,yellow, non-staining zone similar to the hyaline exocuticle
of scorpions (see below). The rest of the procuticle, in most regions of the body, can be
divided into mesocuticle and endocuticle. This division, based on staining reactions in
Mallory'S triple stain, was established by Krishnakumuran (1962); Dalingwater (1975)
made similar observations, but followed Neville et a1.'s (1969) terminology for the zones.
Karlson et al. (1969) noted that the outer fraction of the pro cuticle stained in both
acid fuchsin and orange G and, because insect meso cuticle stains with only one or
other of these stains during the Mallory procedure, concluded that: 'something is added
to the brown cuticle of Limulus carapaces to permit this staining'; they termed this
region 'exocuticle 2'. In addition, Karlson et al. (1969) named the outermost fraction
below the epicuticle 'exocuticle 3' since it contains black pigment. However, following
Neville's (1975) comments on the action of Mallory's triple stain, it is clear that the
outer part of the pro cuticle is not quinone-tanned and should, therefore, be termed
mesocuticle for the present. Both meso cuticle and en do cuticle are laminated, but in
some regions of the cuticle (e.g. the edge of the carapace, the telson) a preferred orienta-
tion zone of cuticle is intercalated between the two major laminated zones (Dalingwater
1975, 1980). From this brief account, it is clear that a full and detailed TEM examina-
tion of Limulus cuticle is needed.
3.3 Acarine Cuticles
Hackman (1982) has produced a detailed, yet concise, review of tick cuticle. The
major layers of epicuticle, exocuticle and endocuticle are generally present and meso-
cuticular regions have often been detected by light microscopic examination of stained
cuticles. The importance of the epicuticle in waterproofing the cuticle is emphasised,
though the role of other cuticular layers and of the epidermis in water conservation is
also recognised. Hackman's conclusion on transition temperatures is that there is no
definitive evidence from the study of tick cuticles to support this concept nor for
changes in the physical structure of the lipid layer of the epicuticle. References to the
relatively few papers on mite cuticles are given in Neville (1975).
10 ] .E. Dalingwater
3.4 Cuticles of Minor Chelicerate Groups
The only detailed examination of ricinuleid cuticles was made by Kennaugh (1968)
using the light microscope. He remarked on some peculiar and distinctive aspects of
their cuticles: the lack of endocuticle over most regions of the body; the presence
(especially in nymphal stages) of considerable thicknesses of impregnated but un-
tanned mesocuticle; the tubercular nature of the cuticle, especially in Ricinoides; but,
particularly on the distribution and form of the pore canals which are tightly coiled, but
which also follow large diameter spiral pathways through the cuticle. The possibility of
a hyaline exocuticle (see next section) being present was discussed; few conclusions
were reached on the nature of the epicuticle apart from the possible presence of a cuti-
culin layer.
A brief, light microscope study of two opilionid cuticles was reported by Grainge
and Pearson (1966). The cuticle structure of the pedipalpid Thelyphonus sp. was out-
lined by Krishnakumuran (1962): a double-Iayeredepicuticle, an outer amber-coloured
exocuticle, a laminated mesocuticle and two endocuticular layers. In the same paper,
the cuticle of the solifugid Galeodes sp. was briefly described: over most regions of the
body three epicuticular layers (including a cement layer), laminated mesocuticle and
endocuticle are present with amber-coloured exocuticle present only at the cheliceral
tips.
35 Scorpion Cuticle
Scorpion cuticle has been the most extensively studied of all chelicerate cuticles. light
microscope studies in the 1950's centred around the supposed presence of chitin in
epicuticular layers (despite the convention that layers containing chitin cannot be
considered part of the epicuticle). Kennaugh (1959) resolved the dilemma by establish-
ing that scorpion cuticles possess a zone below the true epicuticle which is colourless
and refractory, but which contains chitin: this he termed the hyaline exocuticle, show-
ing it to be two-layered in sclerite regions. Above the hyaline exocuticle lies an epi-
cuticle with paraffin and cuticulin layers and below it a quinone-tanned exocuticle,
mesocuticle (Kennaugh's 'colourless exocuticle') and endocuticle. Mutvei (1977) dem-
onstrated the presence of a preferred orientation zone of cuticle ('vertically lamellar
layer') interposed between 'hOrizontally lamellate layers' in some specialised regions of
cuticle (e.g. pedipalps); Dalingwater (1980) defmed the position of this intercalation in
Heterometrus/ulvipes: below the first five laminar units of the mesocuticle.
Filshie and Hadley (1979) and Hadley and Filshie's (1979) work thus rested on a
reasonable base of light and scanning electron microscope studies: they have consider-
ably refined and extended our knowledge of scorpion cuticle by their TEM studies of
the cuticle of Hadrurus arizonensis. In the dorsal and ventral sclerites Filshie and Hadley
(1979) reported the presence of four main zones: epicuticle, hyaline exocuticle, inner
exocuticle and endocuticle (see Fig. 4 for details). Intersegmental membrane cuticle
differs from sclerite cuticle in that hyaline and inner exocuticles are absent. The dense
homogeneous layer of the epicuticle is considerably thicker than in sclerite cuticle and
this may account for Kennaugh's (1959) description of a single layer of supposed hyaline
ep
, -
hx
ix
en
a b
Fig.4a,b. Cuticle of the scorpion Hadrurus arizonensis: diagrams based on transmission electron
micrographs in Filshie and Hadley (1979) and Hadley and Filshie (1979). a Transverse section of
sclerotized dorsal plate cuticle. Abbreviations: ep epicuticle; hx hyaline exocuticle with two layers:
laminate and non-laminate; ix laminate inner exocuticle; en endocuticle, which is laminated ; how-
ever, the lamina units seem to be composed of broad bands of unidirectional microfibrils; vertically-
oriented columns of fibrous cuticle contain the pore canals. Bar =5 ",m. b Wax canal (we) and epi-
cuticular layers of sclerite cuticle. "Tubular' regions within the wax canal only become evident after
heat treatment. Abbreviations: m outer membrane; 0 outer epicuticle; e cuticulin layer; d dense
homogeneous layer (inner epicuticle); hx outer part of the laminate zone of hyaline exocuticle. Bar
= 0.1 ",m
exocuticle from arthrodial membrane cuticle. Further, there are no pore canals and the
wax canals are distinctly different from those in sternite cuticle.
Hadley and Filshie (1979) gave a more detailed analysis of the epicuticle of Hadrurus
especially in connection with the location of lipids. This scorpion had been shown to
have extremely low cuticular permeability by Hadley (1970b) and its lipids had been
qualitatively and quantitatively analysed by Toolson and Hadley (1977). Lipids were
extracted from the epicuticle using hexane and chloroform/methanol; Hadley and
12 J .E. Dalingwater
Filshie used these same solvents prior to SEM and TEM examination and compared the
results with untreated controls. Under the SEM, sclerite cuticle treated with hexane
showed clearer surface detail, while cold chloroform/methanol treatment removed
greater amounts of surface wax and revealed indentations corresponding to wax canal
openings. However, neither solvent system significantly altered the appearance as com-
pared with untreated sclerite cuticle when examined with the TEM. But even untreated
cuticle has to be prepared for TEM examination, by dehydration and embedding, so
lipids may be lost from the surface or removed from layers in which other components
remain to maintain structural integrity. Hadley and Filshie (1979), therefore, considered
that their attempts to establish the location of lipids in the epicuticle were only partial-
ly successful. They did, however, provide further information on the sublayers of
Hadrurus epicuticle. The outer membrane, corresponding in position to Wigglesworth's
(1947) cement layer (formed by secretions from dermal glands and protecting the wax
layer), may contain mucopolysaccharides. The wax layer thought to be so important
in waterproofing the cuticle is termed outer epicuticle since it is resistant to organic
solvents and if it contains lipids must certainly also contain non-lipids. The cuticulin
layer and the dense homogeneous layer (= inner epicuticle or protein epicuticle) both
probably contain abundant lipid and protein stabilised by quinone tanning.
These coordinated studies of scorpion cuticle involving experimental physiology,
biochemical analyses, SEM and TEM examination serve as a model for the approach
necessary to provide a reasonable understanding of any arthropod cuticle.
3.6 Spider Cuticle
Spider cuticle has been rather infrequently studied and it is only relatively recently
that the work of Barth (1969, 1970, 1973) and of Hadley (1978, 1981) has provided
a reasonably full account of the cuticle structure of certain spider species. The cuticle
of most regions of the body is very thin; even the cephalothorax cuticle of larger species
rarely exceeds 30 p.m in thickness, so TEM work is almost essential to provide detail
of microstructure and ultrastructure. Bonaric (this volume) reviews recent work on
spider moulting honnones.
Browning (1942) described the moult cycle of fifth instar Tegenarwatrica and gave
a brief account of its cuticle structure with comments on the surface appearance of this
and a few other species and notes on the relative thicknesses of the procuticular zones
(cf. Fig. 33). He decided that there was no epicuticle in the species he studied. How-
ever, Sewell (1955) not only established that there was an epicuticle in Tegenaria do-
mestica, but also described its ridgelike thickenings and, by a series of histochemical
and staining techniques, was able to suggest that the epicuticle was four-layered: cement,
wax, paraffm and cuticulin layers. Later light microscope studies added a few further
details: Krishnakumuran (1962) and Barth (1969) reported a mesocuticular layer in
Argiope catenulata and Cupiennius salei respectively. Barth's (1969) TEM examination
of the epicuticle of Cupiennius confirmed subdivisions as cement, oriented lipid, cu-
ticulin and electron-dense layers. The epicuticle is less than 0.5 p.m thick with a ribbed
surface: wax canal channels reach the surface or near the surface between epicuticular
ribs and join to form pore canals (see Fig. 5c). Barth (1970) was essentially concerned
with microfibril arrangement and its relationship with pore canal shape, a theme he
II!
III~-ep
-=-=== ex d
l
-
ex
--- I
--------
i
<-.--? "'"
~_V',--_~~... en
--========: ---
Fig. Sa-c. Spider cuticles: diagrams based on electron micrographs in Hadley (1981) for Latrodec-
tus hesperus and in Barth (1969) for Cupiennius salei. a Transverse section of dorsal cephalothorax
cuticle of Latrodectus hesperus. Abbreviations: ep epicuticle; ex exocuticle ; en endocuticle; ed epi-
dermis. Bar = 2 I'm. b Pore canal/wax canal system and epicuticular layers of walking leg cuticle of
Latrodectus. Abbreviations : m outer mem brane ; 0 outer epicuticle ; d dense homogeneous layer (in-
ner epicuticle) ; ex exocuticle ; pc pore canal, which links with wax canal channels running within
the epicuticle. Bar =0.1 I'm . c Pore canal/wax canal system and epicuticular layers of walking leg
cuticle of Cupiennius salei. Abbreviations as in b , but cuticulin layer (c) additionally present. Bar =
0.1 J.lm
continued in 1973 with a detailed analysis of fibril arrangements. Five main patterns
of microfibril arrangement were recognised, ranging from unidirectional to helicoidal
(see Fig. 2). The widths of lamina units in various regions of cuticle were examined: in
all exocuticles lamina width increases regularly and continuously from outside to inside,
with inner lamina units 6-15 times wider than the outermost units ; in the opisthosomal
cuticle lamina units are relatively narrow and their width does not change systematical-
ly ; the lamina unit width does change systematically in dorsal prosomal meso- and endo-
cuticle, narrowing from outside to inside; lamina unit widths are irregular in the meso-
and endocuticle of metatarsus, femur and prosomal sternum ; fmally, lamina unit width
increases abruptly at the borderline between exo- and mesocuticle, especially in the
metatarsus and tarsus. Barth analysed the factors which might influence lamina unit
width, basing his ideas on interpretations of laminae according to the Bouligand-Neville
model. His favoured explanation was that lamina unit width is determined by the num-
ber of microfibril layers and consequently by the amplitude of directional change be-
14 I.E. Dalingwater
tween successive layers; this is supported by evidence provided by densitometry and

electron microscopy. Dalingwater (1985) discussed Barth's analysis and suggested an
alternative explanation: that if one considers laminae as real structures with lamina
membranes present (see Dennell 1976), then laminae might act as crack stoppers; this
is supported by the presence of very narrow lamina units in the outer zones of many
hardened cuticles. Barth (1973) concluded his paper with an analysis of the mechanics
of the various types of microfibrillar orientation seen in Cupiennius cuticle: unidirec-
tional fibre orientation is strongest in resisting tension exerted parallel to fibre direction,
whilst isotropic fibre direction gives the most efficient reinforcement in relation to any
stress direction experienced in the plane of the fibres.
Hadley (1978, 1981) has provided an integrated analytical study of the cuticle of
the black widow, Latrodectus hesperus. Water loss rates for adult female spiders at 0%
relative humidity increased by only a factor of two between 25°C and 35 °c (Hadley
1978). These low transpiration rates are similar to those recorded for xeric-adapted
arachnids and indicated the presence of effective barriers to water loss in the black
widow cuticle. Cuticular lipids were analysed using thin-layer and gas chromatography:
hydrocarbons and wax esters were the main lipids extracted from the cuticle, with
traces of triglycerides, free fatty acids, alcohols and sterols. The presence of saturated
long-chain hydrocarbons was considered to improve the effectiveness of the epicuticle
as a barrier to water loss; wax esters were also considered as being in some way involved
in waterproofing.
Hadley (1981) followed up this analytical approach by an SEM and TEM examina-
tion of Latrodectus cuticle, concentrating on locating the surface lipids, but incidentally
also providing details of cuticle ultrastructure, especially epicuticular ultrastructure.
TEM examination of dorsal prosomal cuticle (see Fig. Sa) showed it to be divisible
into: an extremely thin (ca. 0.1 ~m) epicuticle, an exocuticle (ca. 4.5 ~m thick) and an
endocuticle (ca. 7 ~ thick). Lamina units in the prosomal cuticle show a similar pat-
tern to that described by Barth for Cupiennius: in the exocuticle an increase in thick-
ness from outside inwards and a gradual decrease in the inner region of the endocuticle.
The outer region of the endocuticle may be equivalent to the meso cuticle described by
light microscopy from other species, but cannot be so distinguished under the TEM.
Wax canal channels (ca. 9 nm in diameter) appear to penetrate to the outer edge of the
epicuticle and at their inner ends connect with pore canals at the epi-exocuticle junc-
tion (see Fig. 5b). Pore canals are sparse in prosomal cuticle, only regularly revealed in
glancing sections in clumps of three or four, but are more numerous in leg cuticle. The
epicuticle is also most clearly seen in leg cuticle where it is up to 0~2 ~m thick, but it
does not show the ribbing seen in the epicuticle of Cupiennius. It has three distinct sub-
layers: an extremely thin (2 nm) outer membrane, equivalent to the cement layer of
Barth (1969); an electron lucent outer epicuticle (12 nm), equivalent to Barth's oriented
lipid layer; and a dense homogeneous layer, 116 nm thick. No cuticulin layer was
reported which seems surprising since this layer was considered to be a constant feature
of arthropod epicuticles. Hadley (1981) explains his use of the term outer epicuticle in
place of oriented lipid layer: although this layer may initially have contained free lipids,
other constituents remain after preparation for transmission electron microscopy which
maintain its structural identity.
SEM comparisons of untreated and uncoated material with material washed in
chloroform :methanol and gold -coated showed that large amorphous droplets, particular-
ly abundant at setal bases, were largely removed by treatment which also exposed pos-
sible dermal gland duct openings. At higher magnifications untreated, but coated,
material showed that the droplets and entire cuticular surface also appeared to be
covered by a fine film; both droplets and film seemed to be removed by washing in
chloroform: methanol and thus surface detail resolution was enhanced. Hadley consider-
ed that the surface droplets might be produced by glands associated with setae (analo-
gous to the sebum of mammalian hair follicles) and may be lipid and protein in com-
position. Their role in waterproofing might be to contribute to the lipid film which is
spread over the cuticle surface.
4 Conclusions
Although the cuticles of many individual chelicerate species possess some unique aspects
of structure, it is difficult to pinpoint any criteria which can be used positively to dis-
tinguish all chelicerate cuticles from those of the other major arthropodan taxa. The
hyaline exocuticle is a particular feature in some chelicerate groups (scorpions, xipho-
surans and possibly eurypterids), but does not seem to be present in the cuticles of
others. Mesocuticle has been identified in many chelicerate cuticles, and preferred
orientation zones of cuticle are possibly more widespread than our present state of
knowledge indicates, but both types of cuticle are also found in insects. At the moment,
it is easier to recognise basic differences between the cuticles of scorpions and spiders
than between those ofspiders and insects. Future work, including biochemical analyses
of cuticular components and detailed examination of microfibrillar patterns, may
eventually allow us to recognise features of all chelicerate cuticles which set them apart
from those of the other major arthropod groups.
II Coloration and Chromes
ARTHUR HOLL 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2 C h r o m e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1 Ommochromes......................................... 17
2.2 Bilins.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 18
2.3 Purines: Guanine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 20
2.4 Questionable or Unidentified Chromes . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
3 Structural Colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 21
4 Variability of Coloration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
5 Color Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22
5.1 Morphological Color Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22
5.2 Ontogenetic Color Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3 Physiological Color Change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 23
6 Functions of Colors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 24
1 Introduction
Compared to the richness of integumental colors of numerous insects, coloration of

chelicerates appears to be of little diversity, except for spiders and certain mites. Par-
ticularly in spiders a wide range of coloration is presented (Millot 1926; Kaston 1948;
Crome 1955; Levi and Levi 1968; Stern and Kullmann 1975), based upon integumental
chromes and also specific cuticular structures. Furthermore, color patterns may be
supplemented by white guanine inclusions of intestinal storage cells (Millot 1926;
Blanke 1972; Seitz 1972).
Spiders are white, green, yellow, orange, red, brown and black. Coloration may be
distributed uniformly or in graded shades over the entire body (prosoma, opisthosoma,
legs) and may lack any pattern as seen in various species of the suborder Orthognatha
and also in some labidognath families (Levi and Levi 1968). Most species, however,
exhibit particular color patterns (mainly on the dorsal opisthosoma) which contrast
more or less with the basic coloration (Kaston 1948; Crome 1955; Levi and levi 1968).
Iridescent or metallic colors mainly occur in jumping spiders (Salticidae) and in species
ofa few other families (Kaston 1948; Locket and Millidge 1951).
Although a number of arachnologists have been engaged in the description of spider
coloration, rather few studies pertain to the biochemistry of chromes that contribute
1 Institut flir Allgemeine und Spezielle Zoologie der Justus Liebig-Universitat, D-6300 GieBen, FRG
Coloration and Cluomes 17
to color patterns or function as eye pigments (Weigel 1941; Linzen 1967; Vuillaume
1969; Seligy 1969, 1972; Holl and RUdiger 1975; HollI986).
Apart from phenotypic and genotypic color variations of particular species (Oxford
1976; Blanke 1982), integumental and also guanine related coloration in spiderlings
and subadults generally diverges from that of their adults (Moles 1916). Furthermore,
in a number of species males and females not only differ noticeably in size and ap-
pearance (sexual dimorphism), but also in coloration (Gerhardt 1925; Nct>rgaard 1941;
Locket and Millidge 1951; Bristowe 1958; Stern and Kullmann 1975).
The relations of zoochromes to excretory functions have been extensively elucidated
(Linzen 1967; Needham 1974,1978), but nevertheless, the actually evolved roles of
chromes towards the colorative outfit of animals are more exceptional. Taking those
functional aspects into consideration, spider coloration with respect to underlying
chromes appears to be related - similar to insects - to temperature regulation, protec-
tion against radiation (UV), phenomena of color change and to the complex roles of
crypsis (camouflage) and semasis (advertising).
2 Chromes
Whereas in other arthropods (especially in insects) compounds of most of the known

classes of zoochromes are reported to be present (Needham 1974, 1978), in spiders
only ommochromes, bilins and guanine (a common purine compound) have been
identified as being related to coloration (Millot 1926; Seligy 1969, 1972; Seitz 1972;
Holl and RUdiger 1975; Holl 1986). Except for guanine storing intestinal cells, specific
subhypodermal 'pigment' cells are absent. The cuticle, in particular that of the opistho-
soma, is at most faintly colored, thus transmitting the subcuticular color patterns (Cutler
and Richards 1972). The brown or black appearance of the prosoma in certain species
is supposedly due to substances (perhaps benzoquinones) involved in sclerotisation of
the endocuticula.
2.1 Ommochromes
Ommochromes, as reported by Stamm-Menendez and Galarza Basanta (1961), Linzen

(1967), Seligy (1969) and Vuillaume (1969) for spiders are the predominant chromes
in this arachnid order. Ommochromes, which are considered to be phenoxazines
(Fig. 6a), may be subdivided into ommatins, ommidins and ommins, according to their
structural and chemical properties (Linzen 1967). Apart from their slight solubility in
neutral organic solvents, a peculiar feature of the ommochromes is that the reduced
forms are more bathochromic (red) than the oxidized ones (tending to yellow) (Fig. 6a).
Compounds are mostly bound to protein and are stored intracellularly as membrane-
shielded granules. Based upon conventional techniques, ommochromes were found to
be present in all of the 24 species (belonging to 14 labidognath families) investigated
by Seligy (1972). In particular, orange, red and light-brown hypodermal color patterns
could be related to an ommatin, almost identical to xanthommatin; dark brown or
black patterns to both xanthommatin and an ommin (unidentified), the latter being
the more predominant, whereas yellow coloration resulted from compositions of
18 A.Holl
O~
:(2
1;12
,
"=
0:
COOH
asp
N NH
6 5 I 10" 4")
~ ~ 2
o9 0
nz: Di-H-xanthommatin IT Xanthommatin
(Butenandt's numbering)
A
I Phenoxazine m Ommin A
x y
I I
C=O C=O
B
o o Fig. 6A,B. Structural formulas of

representative ommochromes (A) and
of micromatabilin (B);// oxidized
a X=OCH 3 , Y= sugar or peptide xanthommatin (yellow); /V reduced
or Y = OCH J • X = sugar or peptide form (red); a monoconjugate, b di-
conjugate of micromatabilin. (A after
b X and Y : sugar or peptide Needham 1974)
kynurenine and 3-hydroxykynurenine. All compounds were found to be granular in-

clusions of hypodermal cells (Fig. 7). Surprisingly, no further chromes like melanins,
carotenoids and pterins were detected in hypodermal patterns of those species.
This is in agreement with findings in mature males of Micromata virescens, which have
an opisthosoma with a striking red~n-yellow marking. In this case the red hypodermal
pattern is due to reduced xanthommatin and to ommins (unidentified). A yellow pat-
tern, however, results from oxidized xanthommatin, kynurenine and 3-hydroxykynu-
renine as well (HoIl1986) (Fig. 7).
2.2 Bilins
The bilins are pyrrolic zoochromes, denoted as linear tetrapyrroles because of their
open-chain structure. In contrast to porphyrins, reactivity of bilins is remarkable,
particularly, as many reactions are reversible. In general, all of the structurally known
bile compounds of invertebrates may be derived from protoporphyrin-IX, thus being
Coloration and Chromes 19
Fig. 7a,b. Hypodermal ommochrome deposition and intestinal guanine storage (section of dorsal
opisthosoma). a Micromata virescens (X 520) (untreated freeze section); b Araneus spec. (X 750).
Abbreviations: cu cuticle without pigments; hy hypodermis; om ommochrome granules in hypo-
dermal cells; he heart; div intestinal diverticulum with heavy guanine deposits appearing dark
degradation products of the latter (Rudiger 1970). Compounds of the biliverdin IX

a-structure appear to be green or blue. Most bilins are protein-bound and some are in
liquid phase.
Though common in a number of insects and crustaceans (as well as in other in-
vertebrate groups) bile compounds had not been reported for chelicerates before 1975.
Identification of two green bilins characterized as a mixture of two biliverdin conjugates
(named "micromatabilin") was performed in the overall green-colored females of
20 A.Holl
Micromata virescens (Holl and Riidiger 1975). Differences in the non-hydrolytic degrada-
tion and in the polarity of compounds led to the conclusion that one bilin was a mono-
conjugate and the other a diconjugate of biliverdin (Fig. 6b). Such mixtures have been
described only for bilirubin in the bile fluid of man. Micromatabilin is present in the
hypodermis (predominantly) and in subhypodermal tissues, but also in the hemolymph
and in yolk masses. Astriking feature of micromatabilin contained in the yolk (oocytes
and larvae) is a temperature-dependent, rapidly reversible color change from a pale
green (below 29°C) to bluish-green (above 29°C), which may be due to a specific
yolk protein bound to the biliverdin chromophore (Holl 1982).
2.3 Purines: Guanine
Some purines are regarded as biochromes because in a number of animals they confer
structural whiteness, which is produced by the reflection from crystalline compound
deposits. With regards to their chromatology, strong absorption in the UV has been
detected. In spiders guanine is stored by specific cells of their intestinal diverticula
(Millot 1926; Seitz 1972; Seitz, this volume; Fig. 80). Guanocytes of these diverticula
are closely arranged beneath the integument. In several species (e.g. araneid spiders)
the white matter of guanine crystals is transmitted by integumental areas (preferably
of the opisthosoma) which have little or no coloration of their own (Blanke 1972).
The best known example is the garden spider Araneus diadematus, in which the dark
coloration of the dorsal opisthosoma is supplemented by a white ornamentation
(Fig. 8), resulting from subhypodermal guanine masses (Seitz 1972; Foelix 1979).
Even overall, uniform whiteness may exclUSively be due to intestinal guanine deposits
as in individuals of the crab spider Misumena vatia, when hypodermal chromes are
absent (Weigel 1941). Besides, integumental coloration may be brightened by the
Fig. 8. a Opisthosoma of Araneusdiadematus (dorsal view) (X 5). Guanine storage is visible through
the transparent cuticle and appears as a white cross. b Section of dorsal opisthosoma of Araneus
cornutus with guanine storage cells (G) at the periphery of the intestinal diverticula (D) ; C cuticle,
H hypodermis (x 350). (From Foelix 1979)
Coloration and Cluomes 21
translucent white matter of intestinal guanocytes (Blanke 1972). In overall dark brown
or black species, however, guanine obviously does not affect integumental coloration.
2.4 Questionable or Unidentified Chromes
Considering a few cases of questionable chromes, mention will be made fIrst to the
crab spiders Misumena vatia and Thomisus onustus, which are known for their color
change. According to Weigel (1941), the yellow color of female Misumena is due to a
hypodermal liquid carotenoid, deposited when white or pink individuals change their
appearance to yellow. By Seligy (1972), however, the yellow coloration of Misumena
was proven to be related to hypodermal granular storage of kynurenine and 3-hydroxy-
kynurenine. No examination has been performed on the integumental chromes of
Thomisus onustus, the coloration of which is white, green-yellow or pink corresponding
to substrate colors (Heckel 1891 ; Levy 1970).
Little is known of the nature of chromes producing the specillc cuticular colors of
some spiders. For example in Diaea dorsata and in Heriaeus hirtus (Thomisidae) green
coloration is mainly based upon greenish toning by an unidentifIed chrome of cuticular
depOSition. Concerning the conspicuous coloration in mature males of Eresus niger
(N!/>rgaard 1941) and Philaeus chrysops (Bonnet 1933), the opisthosomal red results
from orange to reddish chromes of unknown nature accumulated by the cuticular
hairs. Dark brown or black colors, which are common in a number of spiders, appear
to result from diffusely brown exocuticle alone (Plectreurys) or (e.g. in the genera
Gasteracantha, Lycosa and Zelotes) from a combination of exocuticular browning and
hypodermal chromes (Cutler and Richards 1972). Granular-type, dark chromes reported
for several spiders and supposed to be melanins (Millot 1926, 1949; Weigel 1941) may
at least, in part, be ommochromes. Yet, the occurrence of eumelanins in spiders should
not be excluded. In all those cases in which unidentified chromes affect color, the
compounds probably belong to one of the known classes of biochromes.
3 Structural Colors
Iridescent or metallic colors (greenish, bronze, bluish, violet) appearing in many species
(Salticidae, Clubionidae and a few other families) are due to the effect of superimposed
planes created by modifIed setae of the cuticle, rather than to diffraction gratings by
cuticular ridges (Cutler and Richards 1972). True integumental colors generated on the
basis of chromes may be modifIed by iridescent patterns. Besides, some species are
equipped with "hairs" that are white due to the total light reflection (e.g. the prosoma
of Argiope). Also, whiteness by scattering reflection and refraction, which is due to
intestinal guanine crystals, has to be considered in this context (Seitz, this volume; see
Figs. 8 and 80).
4 Variability of Coloration
The depth and pattern of coloration in spiders are often quite varied between speci-
mens of the same species, but usually all conform to a fundamental design (Locket
22 A.Holl
and Millidge 1951 ;Crome 1955). Variability of coloration may be due to either pheno-
typic or genotypic differences (Oxford 1976; Blanke 1982), which have been analysed
only in a few instances.
Color variation with respect to the dark form Araneus diadematus stellatus of
A. diadematus has been proven to be an ecophenotype, which is the adaptive result of
varying environmental brightness (Blanke 1982). In Pityohyphantes p/vygianus (Uny-
phiidae) variable coloration was recorded for a population of both sub adults and adults
varying from a pale phenotype, through intermediate varieties, to a melanic phenotype
(Gunnarsson 1985). Melanic variation is probably maintained by balancing selective
factors. Another species of considerable, suggestively phenotypic color variation, even
in individuals of one clutch, is Anelosimus jucundus (Theridiidae) (Nentwig and
Christenson 1986).
Polymorphic color variation appears to be best known in Enoplognatha ovata (Theri-
diidae) from which several morphs have been described (Locket and Millidge 1953;
Hippa and Oksala 1979; Oxford 1976). Results of breeding experiments suggest that
polymorphism in Enoplognatha ovata is genetically con trolled by a one-locus mechanism
(Oxford 1983). Considerable variation of color pattern is also recorded for the black
widow Latrodectus mactans (Fig. 41a) (Kaston 1948; Maretic 1965), whereby some
individuals almost or entirely lack the orange to red ornamentation. But all those
specimens occur syntopically and are not morphs.
5 Color Change
Often, striking alterations in color throughout lifetime may be related to either environ-
mental conditions or to metamorphosis, and hence, they are subject to internal mecha-
nisms of the animal (Biickmann 1974). In several spiders phenomena of color change
occur, being either of "morphological", "ontogenetic" or "physiological" type.
5.1 Morphological Color Change
Misumena vatia has attracted much attention because adult females (but no juveniles
and males) can change their color reversibly (packard 1905; Gadeau de Kerville 1970;
Rabaud 1923; Gabritschevsky 1927; Weigel 1941). Most of the females are white or
yellow and are found on flowers with corresponding colors. White individuals, placed
on yellow flowers (or a yellow background) turn yellow within a couple of days, but
will regain their original color when placed on white flowers. Yellow coloration is
produced by hypodermal accumulation of chromes (see ommochromes) which disap-
pear when a yellow background is replaced by white matter, leaving an overall body
whiteness due only to intestinal guanine. Blinded individuals do not react to alteration
of environmental colors (Weigel 1941). Color change of the crab spider Thomisus
onustus is similar to that in Misumena (Heckel 1891 ; Bristowe 1958) and there may be
more crab spiders capable of that type of color change.
Apart from crab spiders, morphological color change has been reported for the
araneid Cyrtophora citricola (Kullmann 1959c; Blanke 1982). Their subadult females
and males can reversibly change the degree of brightness of the opisthosomal coloration
in correspondence to the varying in tensity of environmental colors. Dark mature females,

however, do not generally respond to alteration of background colors (Blanke 1972).
"Slow" color change, related to the alteration of substrate colors, was observed in
the green lynx spider Peucetia viridis (Neck 1977) and in sub adults of the heteropodid
Micromata virescens (Holl and RUdiger 1975). In both cases reversibility of color change
has yet to be proven experimentally.
5.2 Ontogenetic Color Change
Ontogenetic color change is developmental in Significance and the resulting coloration

is irreversible. In arthropods it appears to be associated with metamorphosis or moult-
ing, respectively. Quite a number of spiders display ontogenetic color change. Particu-
larly in males of those species the final moulting is accompanied or followed by this
event, whereas females mostly preserve their nymphal coloration (Millot 1949).
For instance, in the lace web spider Eresus niger premature males resemble the
females in color (almost black) until their final moulting during which color change oc-
curs. As a result, the dorsal opisthosoma of the moulted mature male is vivid scarlet
supplemented by two pairs of black spots that are surrounded by white hairs (Nct>rgaard
1941; Bristowe 1958). Also in males of the jumping spider Philaeus chrysops, along
with their final moulting, a striking color change occurs resulting in an overall red
opisthosoma with dorsomedian black ornamentation (Bonnet 1933).
In contrast, males of Micromata virescens sustain their nymphal coloration (either
hay-colored or greenish) until the last moulting has been achieved. A splendid red-on-
yellow opisthosomal pattern is completed only after a few days (Homann 1946; Holl
1986). Premature green females keep their color in the mature stage, but they change
color after the final moulting, if they are hay-colored as prematures.
In all those cases, color change is assumed to be due to physiological processes in
the course of final moulting and ( or) definite maturation of the gonads. Internal mecha-
nisms in charge of controlling the event - possibly by hormonal operation - have not
as yet been studied in spiders.
5.3 Physiological Color Change
Based upon responses to environmental conditions, physiological color change is a rapid

and reversible process by either migration of chromatic inclusions within cells or altera-
tion of chromatocyte morphology. This type of color change in spiders was first ob-
served for the Australian araneid Phonognatha wagneri (Roberts 1936). When disturbed
in its web, Phonognatha drops to the bottom and (associated with a cataleptic state)
the creamy mottling of the opisthosoma changes instantly to a dingy brown. Recovery
of the original color scheme requires several minutes after the spider has climbed back
to its web.
Rapid, reversible color changes occurring in the same manner as in Phonognatha
have been reported also for several web-building spiders such as Leucauge subgemma
and Chrysso (=Argyria)venusta (Uyemura 1957), Floronia bucculenta (Bristowe 1958),
Gea heptagon (Sabath 1969) and Cyrtophora cicatrosa (Blanke 1975b). In all those
24 A.Holl
Fig. 9. Instantaneous color change of Floronia bucculento (dorsal view). Left: Unaffected white
pattern (due to guanine). Right: Modified pattern after color change. (After Bristowe 1958)
cases color change results in a considerable darkening of the opisthosomal coloration

(Fig. 9).
According to Blanke (1975b), color change in Cyrtophora cicatrosa (and in the
other species as well) is produced by a sudden decrease or retraction of the intestinal
areas of the guanocytes, leaving the greyish brown of the gut system and the dark
hypodermal pattern. Thus, the basic hypodermal coloration is not affected during the
process of color change . With regard to the rapidity of the "reaction", it may be as-
sumed that in spiders responses are mediated by integumental mechanoreceptors and
are neurally con trolled.
6 Functions of Colors
As in insects and crustaceans, coloration of spiders not only plays some role in the
visual effect functions, but may also be related to thermoregulatory and antiradiant
strategies (Humphreys, this volume) . In this chapter attention will be drawn only to
the roles of cry psis and semasis, which in a number of spiders seem to be of consider-
able significance for purposes of camouflage and advertising.
With regards to crypsis, spiders of various families have colors and patterns that well
resemble background schemes (mimesis). Matching a background may be either relevant
to escape detection by predators or to avoid detection by animals preyed upon. In
most cases of cryptic coloration in spiders, however , definite biological roles have been
merely discussed rather than experimentally elucidated.
Striking examples of background matching are Drapetisca socialis and other spiders
living on bark (e.g. hersiliids). Drapetisca with a grey, mottled opisthosoma is almost
undetectable when sitting motionless on tree trunks unless it is made to move (Bristowe
1958). Deceptive, unconspicuous coloration is often associated with specific postures
or ease positions, thus resembling elements of the inhabited plant substrate. For in-
stance, Hyptiotes paradoxus almost exactly resembles dry buds of spruce twigs (Reu-
kauf 1931; Stern and Kullmann 1975). Striking mimetic resemblance (including color)
of dry leaves or twigs is also reported for Arachnura scorpion ides (Stern and Kullmann
1975) and for Dinopis spinosus (zaW 1971). Furthermore, several tetragnathid and
philodromid spiders are supposed to be protected against avian predators by their
coloration similar to dry grass and by stretching their legs along grass stems (Steininger
1934).
Though conspicuous, the gorgeous coloration of Heterodictyna waeckenaeri, Lys-
somanes bradyspi/lus, Peucetia viridis, Micromata virescens, Aranie/la cucurbitina (all be-
ing green) and the coloration of a number of crab spiders (white, green, yellow or pink)
seem to be well suited for deceptive functions (including both aggressive and protective
advantages) when specimens are located on the substrates of their natural habitat: green
leaves, grass, flowers (Heckel 1891; Weigel 1941; Crane 1949; Bristowe 1958; Levy
1970; Neck 1977). According to Hinton (1976), individuals of white or yellow Misu-
mena vatia with red patches when waiting for their prey on identically colored flowers
are cryptic for (red-blind) insects in both the visible and ultraviolet parts of the spectrum.
Red patches, however, may function as warning signals for predatory birds which
perceive the red color.
With respect to its role in advertising and signalling, semasis provides conspicuous
or contrasting coloration specifically directed towards warning or attracting, and thus,
being of adaptive significance. Bright warning colors, such as yellow or red, occur in a
few spiders with highly effective venoms, for instance in Latrodectus mactans (Theridi-
idae) (Fig. 41a). Color patterns of various Argiope species (Fig. 73e) similar to those of
wasps (yellow/black bands), however, supposedly 'serve disintegration of the body
relative to the structures of habitats (somatolysis) rather than warning functions (Stern
and Kullmann 1975). The gorgeous coloration of mature males of a number of jump-
ing spiders is of intraspecific attractive quality in courtship or agonistic behavior (Peck-
ham 1909; L.M. Forster 1982; Jackson 1982a).
The first detailed information on mimicry in spiders involving their color patterns
was presented by Peckham and Peckham (1892) and Pocock (1909). Striking examples
are spiders that resemble snails (e.g. Cyrtarachne conica), coccinellid beetles (Araneus
coccine/la, Parap/ectana thorntoni and P. walleri), ants or mutillids (almost 100, pre-
dominantly clubionid, gnaphosid and salticid species; Nentwig 1985f). As to the con-
spicuous coccinellid-like pattern in mature males of Eresus niger a mimicry function,
which was suggested by Gerhardt (1928a), has yet to be proven. The possible role of
coloration and patterns in ant-mimicking spiders is not considered here.
III Thermal and Water Relations
RALF PULZ 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 26
2 Background: Comparison of Spiders and Insects. . . . . . . . . . . . . . . . . . . . . . . .. 27
3 Thermal Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... . . . . . .. 28
3.1 The Body Temperature in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 28
3.2 Effects of Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 30
3.3 Evaporative Cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 35
3.4 Control of Temperature Responses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4 Water Relations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 39
4.1 Water Content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2 Water Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 42
4.3 Water Loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
4.4 Osmotic and Ionic Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5 Autecological Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 51
5.1 Hygrothermal Preferenda and Resistance to Heat and Desiccation. . . . . . . . . . .. 51
5.2 Comparison of Developmental Stages . . . . . . . . . . . . . . . . . . . . . . . . . . .. 54
6 Conclusions.............................................. 55
1 Introduction
The study of the thennal and water relations of spiders began with observations on the
thennal relations of growth (Wagner 1888), with experiments on temperature sensing
(Gaubert 1892; Mclndoo 1911) and with the description of interspecific differences in
the drinking behaviour of spiders (Gerhardt 1923, 1928b). Millot and Fontaine (1937)
measured the water content of ecologically different species, introducing both a com-
parative approach and experimental methods into the analysis of spider water relations.
Autecological aspects were introduced by Weese (1924), who studied the response of
various spiders to humidity gradients, and by Savory (1930), who related the habitat
choice of two araneids, Zygiella x-notata and Z. atrica, to positive and negative hygro-
taxis, respectively. Palmgren (1939) combined both field and laboratory studies and
demonstrated that access to water is crucial in the habitat selection ofa pisaurid,Do-
lomedes fimbrilltus.
Subsequent research has largely focussed on detennining autecological parameters,
e.g. water loss rates, thennal preferendaand resistance to heat and drought, in a variety
1 Zoologisches Institut der Universitiit Kiel, Zoophysiologie, Biologiezentrum, Olshausenstr. 40-60,

0-2300 Kiel, FRG
Thermal and Water Relations 27
of species from different habitats (e.g. Holm 1950; Nq,rgaard 1951; Cloudsley-Thomp-
son 1957; Engelhardt 1964; Vollmer and McMahon 1974; Baehr and Eisenbeis 1985).
Moreover, temperature-related behaviour in the field has been frequently studied (re-
views: Pulz 1986b; Humphreys this volume). Main (1982) and Cloudsley-Thompson
(1983) reviewed adaptations of spiders to arid habitats, summarizing various aspects
of spider hygrothermal physiology for the first time.
Basic physiological problems, e.g. excretion and control mechanisms underlying
water loss and thermal responses, have not attracted much attention so far (see Edney
1977 for water relations). The respective literature is scattered rather heterogeneously
on different subjects. Recent progress in some fields, e.g. osmotic and ionic regulation
(Burton 1984, Tillinghast et al. 1984; Riddle 1985), contrasts with a lack of detailed
or even of first investigations in many other fields.
This chapter reviews the state-of-the-art of spider hygrothermal physiology and will
hopefully point to some interesting lines offuture research. The first two sections will
primarily be concerned with general physiological aspects, whereas the third section is
dedicated to autecological parameters: hygrothermal preferenda, heat and desiccation
resistance. With regards to the thermal relations, I will only deal with the 'upper' part
of the temperature range. Problems of overwintering and of cold adaptation are dis-
cussed by Kirchner (this volume) and Aitchison (this volume). Before turning to the
subject, one final remark: the separate discussion of thermal and water relations is a
matter of convenience only and does not reflect the true situation. Rather, complex
interactions exist that have been thoroughly worked out by Willrner (1982) for insects.
2 Background: Comparison of Spiders and Insects
GeneralAspects. Spiders,like all terrestrial arthropods, must keep fluctuations in body

temperature (Tb) and in body hydration within tolerable limits. Otherwise, the animals
will be physiologically hurt or at least restricted in their normal activity. Three major
problems, however, conilict with the maintenance of thermal and hygric balance in
terrestrial arthropods: (1) Land habitats may show large fluctuations in temperature
and humidity, in the supply of drinking water and wind speed, and in many other
parameters relevant to hygrothermal balance. (2) Water activity within the body (i.e.
the concentration of 'effective' water molecules in the body fluid) is normally much
higher than water activity in the surrounding air (i.e. the partial pressure of water
vapour) (see Edney 1977 for details). Arthropods, like any land-living organism, will,
therefore, tend to continuously lose water along this gradient. (3) Insects and spiders
are relatively small animals reaching a body length of a few millimeters or centimeters
and a body mass of a few milligrams or grams. The low mass of the body, along with
the high surface to volume ratio, generally facilitates heat and gas exchange with the
surroundings. This implies the danger of first, rapid fluctuations in Tb , which may even
reach dangerously high levels, and secondly, a severe depletion of body water stores
due to rapid water loss via the integument, the respiratory organs and other avenues.
Comparative Aspects. We may predict that in both spiders and insects similar mecha-
nisms and adaptations should have developed to withstand these difficulties. This is, in
28 R. Pulz
fact, the case: for instance, either arthropod groups possess a largely impermeable waxy
epicuticle keeping cuticular water loss low (Hadley 1984; Dalingwater, this volume).
The respiratory organs and their openings, the spiracles, are adapted to restrict and
control respiratory water loss, respectively (Levi 1967b; Kestler 1985; Wharton 1985;
Strazny and Perry, this volume).
On the other hand, spiders and insects differ in some important aspects with respect
to hygrothermal physiology: unlike many insects and ticks, spiders are obviously not
capable of actively taking up water vapour from the atmosphere (Machin 1979; Kntille
and Rudolph 1982). The principal nitrogenous excretory product in spiders is guanine
(J.F. Anderson 1966), which is still more insoluble in water than uric acid, the chief
nitrogenous excretory product in insects. No evidence of endogenous heat production
exists in spiders which, again unlike many insects (Kammer 1981; May 1985) must,
therefore, be regarded as strictly ectothermic animals.
Finally, a characteristic aspect of spider biology should be remembered: many
spiders, above all adults of those species building burrows or permanent webs, live
more or less sedentary. They cannot shuttle between distant microhabitats as readily
as, for instance, the flying insects, but will rather stay at the once chosen site even
when microlimatic conditions become suboptimal. Otherwise, they would expend
valuable energy in chOOSing new microhabitats and reconstructing burrows or webs.
It follows that the selection of appropriate microhabitats is most crucialin these spiders:
thermal and hygric conditions, along with other abiotic factors (e.g. soil property,
sites for web attachment, exposition to wind) and prey availability must rather precisely
fit the physiological needs of the respective species. The solution of this problem has
been extensively studied in a desert-inhabiting, funnel-web spider, Agelenopsis aperta
(Riechert and Tracy 1975; Riechert 1985). However, control mechanisms underlying
the selection of microhabitats , like the sensory reception of temperature and humidity,
have so far largely remained obscure (see below).
3 Thermal Relations
The profound effects of temperature on organisms have been extensively studied both
at the molecular, cellular and systemic level, as well as at the level of the whole organ-
ism. General reviews in Precht et al. (1973), Wieser (1973a) and Hochachka and Somero
(1980), together with special reviews on the temperature relations of insects (Heinrich
1981; K.H. Hoffmann 1985a; May 1985), provide an excellent background for the
present discussion.
3.1 The Body Temperature in Spiders
Methodical Aspects. Indirect and direct methods have been used to predict and mea-
sure T b , respectively: measuring air temperature (Ta) close to spiders (N¢rgaard 1956;
Pointing 1965), measuring the temperature of spider models (Riechert and Tracy 1975;
Tolbert 1979), infrared thermometers (Suter 1981), thermistors attached to be body
surface (Krakauer 1972; J.F. Anderson and Prestwich 1985) and inserted (Carrel 1978)
or permanently implanted thermocouples. The latter technique has proved to be most
useful both in the laboratory (Humphreys 1978a; Pulz 1985b, 1986a) and in the field
(Humphreys 1974, 1978a; this volume). It is advantageous in allowing the measure-
ment of Tb in the same individual several times at exactly the same body site. So far,
it has not been observed that spiders are hindered in their normal behaviour when
carrying fine thermocouples.
General Aspects. The Tb of an organism is determined by its metabolic heat produc-

tion and by evaporative heat loss (physiological factors) as well as by radiative, con-
ductive and convective heat exchange with the surroundings (physical factors, see
Church 1960a,b;May 1985;D.A.Parry 1951;Digby 1955).
In spiders, Tb is normally controlled by behaviour and by appropriate colouration
or colour changes of the body surface. These mechanisms modulate the physical fac-
tors of heat exchange and allow for both raising Tb at low Ta and for keeping Tb low
when the animals are exposed to locally high temperatures, to high Ta or to direct
insolation (Humphreys, this volume).
With regards to physiological factors, spiders lower and stabilize Tb by evaporative
cooling (see below), but do not raise Tb by metabolic heat production. J.F. Anderson
and Prestwich (1985) and Pulz (1986a) measured Tb in large theraphosid spiders (weigh-
ing up to more than 20 g) during rest and forced activity. Brachypeima smithi was
forced to continuously run on a treadmill at different speeds for 10 min, however, no
sign of endothermy was observed (J.F. Anderson and Prestwich 1985). Rather, Tb
decreased below Ta , possibly indicating evaporative heat loss (see below). PUlz (1986a)
studied behavioural responses to temperature in Brachypelma smithi and Phormicto-
pus sp. No increases in Tb were measured during 10-30 s bouts of vigorous struggling
at low (25°C) and high (40 DC) Ta.
These results may be predicted from the comparatively low resting metabolic rates
(J.F. Anderson 1970) and factorial aerobic scopes (the ratio of maximal to resting
metabolism) in spiders: different spiders, including theraphosids, showed an ap-
proximately two- to ninefold increase in oxygen consumption when forced into activity
(Seymour and Vinegar 1973; Prestwich 1982; J.F. Anderson and Prestwich 1985;
Carrel, this volume). This contrasts with scopes of more than 100 or even 200 in large
flying insects (Bartholomew 1981). In these animals, however, endothermy is not only
a by-product of high metabolic rates during sustained activity of the wing muscles, but
may also be employed in preparing for activity: warming-up is then achieved by mecha-
nisms of shivering and non-shivering thermogenesis (Kammer 1981; May 1985). Similar
thermogenetic mechanisms have not been detected in spiders so far, nor do we know
whether spiders can produce fever. Scorpions showed 'behavioural' fever when in-
jected with prostaglandin E 1 ; the animals then preferred higher temperatures in a
temperature gradient (Cabanac and LeGuelte 1980). The ability to produce fever might
help to defeat pathogenetic infections (May 1985).
Changes in Body Temperature. To summarize, Tb in spiders will always tend to ap-

proximate Ta, provided the body does not heat above Ta due to external heat sources.
Temperatures at different sites within the body will then be homogeneous: in live
theraphosids exposed to homogeneous Ta no differences in Tb were measured within
the body. Rather, both tagmata heated and cooled at about the same rates and showed
30 R. Pulz
20
Fig. 10. Cooling curves in life Brachy-
pelma smithi. Prior to cooling the animals
10 were heated with implanted high-fre-
quency electrodes in the opisthosoma;
Tb was measured in the prosoma. The
5
spiders were exposed to a slowly moving
airstream (70 cm/min) during recordings.
Individuals weighed 21.0 g (1) and 11.6 &
(2)_ Note that the cooling curves change
from linear to curvilinear in the lower
range. According to Bakken (1976) this
PROSOMA indicates that Tb would be lower than
Ta in the steady state. However, in the
0.5
present experiments Tb exactly attained
the level of Ta
0.1
o 16 32 48 64
Time (min)
similar cooling curves closely following Newton's law of cooling (Buatois and Croze
1978; Bartholomew 1981). Figure 10 shows cooling curves for the prosoma in live
Brachype/ma (Pulz 1986a)_
Heating and cooling rates, however, differed considerably between the body core
and the body surface: cuticula temperatures (T c) changed far more rapidly at the be-
ginning of temperature changes, but reached steady temperatures at approximately the
same time as core temperatures (Pulz 1986a). Therefore, large differences between Tb
and Tc can occur in spiders being rapidly exposed to thermal gradients. Moreover,
Humphreys (this volume) found that when either soma was separately heated, heat
flow from the opisthosoma to the prosoma was more rapid than in the opposite direc-
tion. This result may be explained as follows: blood flow in spiders is primarily con-
trolled by the heart and only modulated by body and leg muscles (Carrel, this volume;
Stewart and Martin 1974; Blickhan and Barth 1985). The heart responds to direct
thermal stimulation by an increase in heart rate (pulz 1986a); convective heat transfer
within the body will, therefore, rapidly increase when the opisthosoma (containing the
heart), but not when the prosoma is heated.
3.2 Effects of Temperature
According to the Arrhenius-van't Hoff rules, rates of chemical reactions and hence of
biological processes increase with rising and decrease with falling temperature. At an
increase of 10°C, rates of reaction normally increase by a factor (QIO) of 2-3. The
QIO is hardly constant over the whole physiological temperature range of an organism,
but often increases at low and decreases at high temperatures. Therefore, exponential
relationships between the rates of reaction (R) and temperature (T) may only apply to
a limited temperature range (K.H . Hoffmann 1985b). In general, temperature-dependent
physiological processes in spiders are consistent with these principles, but some modi-
fications of the typical pattern also exist and will be considered below.
3.2.1 Metabolism
General Aspects. The parameters most frequently used to determine metabolic activities
are the rates of oxygen consumption and of carbon dioxide release, respectively ('met-
abolic rates'). In spiders, these may show the predicted exponential increase with in-
creasing temperature (K. Miyashita 1969b; Hagstrum 1970b; Moulder and Reichle
1972; Steigen 1976), but may also linearly increase in a temperature range as wide as
10°-40°C(Humphreys 1975b). Decreasing Ql0 values in the upper temperature range
occur frequently and may generally be expected in lycosids (Humphreys 1977a). In a
North American theraphosid,Aphonopelma sp. B, the QlO dropped from 3.35 (10° to
20°C) to 1.11 (30°-40°C), the high Ql0 in the lower temperature range matched
values from other species at the same temperature (J.F. Anderson 1970).
Between 30° and 40°C a flattening (sigmoid) R-T curve was observed in Aphono-
pelma (Seymour and Vinegar 1973). In other spiders, however, abrupt decreases in
oxygen consumption occurred at high temperatures, probably indicating severe thermal
stress or even heat injury (Moeur and Eriksen 1972; Moulder and Reichle 1972).
Absolute levels and thermal sensitivity of metabolic rates in spiders not only vary with
temperature, but also with feeding conditions, activity and body weight (Humphreys
1977a). Moreover, metabolic responses to temperature differ significantly when com-
pared at constant and cyclic temperatures. In Geolycosa godeffroyi, respiration rates
under cycling temperatures (10°-30 °C) were 1.35 -1.46 times higher than at the mean
temperature of the cycle (Humphreys 1975b, 1977b). In Lycosa carolinensis, however,
metabolic rates were elevated at constantly high temperature (43°C) relative to rates
at cycling temperature (35°-40°C) (Moeur and Eriksen 1972). Similar results were
obtained in Pardosa palustris (Steigen 1976).
Modifications. Metabolism in certain species does not strictly follow changes in Ta and
Tb, respectively, but exhibits some degree of thermal independence. Like decreasing
Ql0 values at high temperatures, such modifications may be biologically advantageous
in spiders experiencing large thermal fluctuations in their natural habitats.
Spiders initially increased oxygen consumption when transferred from lower to
higher acclimation temperatures ('overshoot'), but during the following days gradually
reduced metabolic rates to new constant levels. This pattern of thermal adaptation
(partial compensation) was observed in different species, including lycosids (J.F. An-
derson 1970; Moeur and Eriksen 1972; Humphreys 1975b), a salticid and theridiid
(J.F. Anderson 1970) and spiderlings of an araneid, Argiope aurantia (Riddle and
Markezich 1981). On the other hand, exposure to temperatures below acclimation
temperature was followed by an initial drop and within some days by a compen-
satory rise in metabolic rates (Riddle and Markezich 1981). The reverse was observed
in Filistata hib erna lis , showing gradually declining metabolic rates when transferred
from high to low temperatures (J .F. Anderson 1970).
Long-term modifications occur in the desert-inhabiting lycosid, Lycosa carolinensis,
and in diapausing eggs of a temperate-zone linyphiid, Floronia bucculenta. Lycosa
32 R. Pulz
exhibits seasonal acclimatization as indicated by lower metabolic rates in June relative

to January when compared at 29°-45°C; consequently, the R-T curve is shifted
laterally to higher temperatures (translation, Precht et al. 1973; Wieser 1973b). More-
over, lower Ql0 values in June (29°-39°C) indicate a decrease in absolute thermal
sensitivity (Moeur and Eriksen 1972).
In Floronia, overwintering eggs show a diapause phase (September- January). During
this period, metabolic sensitivity to temperature (5°C vs 25°C) is Significantly de-
pressed. In the postdiapause phase, however, a sharp increase in oxygen consumption
occurred at 25°C, but not at 5°C, indicating that the normal metabolic sensitivity to
temperature had been restored (Schaefer 1976b). These phenomena are closely related
to embryonic development (see below). Similarly, eggs of Argiope aurantia (this species
overwinters in the spiderling stage) exhibited an increasing temperature dependence of
metabolism towards the end of development with QlO = 2.59 (5 0 -15°C) and 2.61
(15°-25 DC) (Riddle and Markezich 1981).
3.2.2 Development, Growth and Reproduction
Development. Temperature affects all characteristic phases and processes during the
life of spiders, whether embryonic development, growth rate, food consumption, adult
period, egg production Or mortality. Embryonic development is considered as the phase
between deposition of eggs, hatching of spiderlings and their emergence from the egg
sac (Schaefer 1976b; van Praet and Kindt 1979; Schaefer, this volume). Since the
growth of spiders is attached to successive moults, growth rate is indicated by the
moulting frequency (number of moults per time unit) or by the length of intermoult
periods (ins tar length).
Developmental processes are basically coupled to metabolism. It can be expected,
therefore, that increasing temperatures normally increase developmental and growth
rates and shorten developmental phases, respectively. This was clearly shown in post-
diapause eggs of Floronia bucculenta (Schaefer 1976b) and in eggs of Erigone atra and
Oedothorax fUscus (van Praet and Kindt 1979): the time of embryonic development
decreased exponentially with temperature (Fig. 11), reflecting the characteristic pat-
tern of metabolic responses to temperature (see above; Ratte 1985). In diapausing
eggs, however, increasing temperatures may first inhibit embryonic development before
a period of chilling removes the diapause mechanism (Schaefer 1976b, this volume).
Growth. Thermal influences on subsequent developmental stages are far more difficult
to assess since temperature effects may strongly interfere with other factors, including
humidity, illumination,spider age and food (Browning 1941 ;S.E. Jones 1941; Hagstrum
1970a; Workman 1978). Moreover, the need to regenerate lost limbs can induce a
shortening of the intermoult period (Stewart and Martin 1982).
With regards to temperature, moulting seems to depend on certain threshold values.
Tarentula kochi needed 15 OCto moult, irrespective of the fact that this was only the
upper temperature of a temperature cycle (6°-15 DC). At the constant mean of the
cycle (10 DC) moulting did not occur (Hagstrum 1970a). Moreover, moulting can be
stimulated by temperature changes: Dugesiella hentzi moulted out of the normal moult-
Fig. 11. Relation between the length of em-

bryonic development (days) and temperature in
Erigone atra . The graph was constructed after 50
Er igone atra
data from van Praet and Kindt (1979). The
regression line was fitted visually
~30
o
o
10
10 20 30
Temperature (OC)
ing season (September) when transferred to higher temperatures during the winter
(Stewart and Martin 1982).
Instar length generally decreases with increasing temperature , in Age/ena naevia with
an average QI0 of 2.2 (17°-27 °C)(Browning 1941; S.E. Jones 1941; Hagstrum 1970a;
Workman 1978; Palanichamy 1985). As predicted , the decrease in instar duration fol-
lows the logarithmic pattern (Workman 1978). The situation becomes more complicat-
ed, however, when cycling instead of constant temperatures are considered. In Age/ena
naevia moderate temperature fluctuations (25°- 3l.5 °C) produced slight increases in
growth rate relative to the constant mean temperature of the cycle. Gross fluctuations
(22.5°-35 °C), on the contrary, reduced growth rates, possibly because peak tempera-
tures were unfavourably high (S.E . Jones 1941). In Trochosa terricoia , duration of the
first, second and fifth instars was also shorter at fluctuating (10°-25 °C) than at con-
stant temperatures (17.5 °C). Instar length showed considerable individual variation
(which may decrease with temperature, Browning 1941), and the absolute time until
maturity differed only slightly between the two thermal regimes. The effect of moderate
thermal fluctuations on growth, at least in Trochosa, was inferior to effects of age,
food availability and probably also food quality (Workman 1978).
It is interesting to note that the structure of egg sacs potentially endangered by
direct insolation provides for Significant buffering of thermal fluctuations within the
egg mass (Hieber 1985). Additionally, females may position egg sacs in such a way
that thermal loads impairing the developmental process are restricted to a minimum
(Nct>rgaard 1956 ; Humphreys, this volume).
Temperature appears to influence only growth rates, but not growth increment
(S.E. Jones 1941; Workman 1978). Individuals belonging to the same instar will, there-
fore , show equal linear dimensions, irrespective of developmental speed.
Reproduction. Riechert and Tracy (1975) have elaborated that reproductive success
in a thermally , instable environment may depend more on thermal balance than on
prey availability. The crucial parameter for offspring production being improved by
34 R. Pulz
thermal balance was total food consumption: individuals of Agelenopsis aperta keep-
ing Tb within the optimal activity range were expected to catch more prey, develop
faster, mature earlier and produce offspring under more favourable seasonal condi-
tions than individuals exhibiting poor or no thermoregulatory abilities. Their offspring
will, therefore, have a greater chance of surviving and of contributing to the next
generation.
Some related aspects still deserve attention. Whereas food intake generally increases
with temperature, the efficiency of food utilization apparently does not (Humphreys
1977b; Workman 1978; Palanichamy 1985). The same holds for the efficiency of egg
production (ratio: energetic equivalent of produced eggs/energy consumed by the
female), which remained fairly constant within the range of 22°-37°C in Cyrtophora
cicatrosa (Palanichamy 1985). However, the total number of egg sacs and eggs produced
decreased with temperature (22°-37°C), resulting from a shorter lifetime at high
temperatures: the life span fell with increasing temperature with QI0 = 2.1 in females
and WithQI0 =2.9inmales(22°-32 0C); the adult period offemales fell with QI0 = 1.5
within the same temperature range. These results indicate that temperature per se, feed-
ing conditions being optimal, may significantly influence the reproductive success in
spiders by limiting the time available for offspring production. This effect is apparent-
ly not balanced by a higher rate of egg production per day (Palanichamy 1985).
3.2.3 Behavioural Activity
It is common experience that most arthropods are sluggish at low temperatures, but
become increasingly more active at rising temperatures. Temperature may then stimulate
and modify different behavioural patterns, e.g. locomotion, hunting, feeding and mat-
ing. Surprisingly, however, many araneid spiders reconstruct their webs at dawn when
temperatures are very low. We do not yet know how these spiders overcome the ob-
vious thermal disadvantage.
Stimulating Effects. Surface-dwelling lycosids, like Pardosa pullata, become most

active on warm days (N¢rgaard 1951), but restrict activity maxima to the most favour-
able thermal conditions during the day (Hag strum 1970a). On the other hand, spi-
ders capable of behavioural thermoregulation (Humphreys, this volume) may over-
come this restriction: the desert-inhabiting agelenid, Agelenopsis aperta, maintains T b
in the optimal range (21°-35°C) and, hence, extends its activity period on the web
sheet (Riechert and Tracy 1975, see above).
The threshold temperature for activity on the ground surface was 10°C inP. pul-
lata and 12.6 °c in P. chelata (Hallander 1970). On the other hand, Lycosa carolinensis
showed normal activity only at a minimal temperature of 30 ° C (Moeur and Eriksen
1972). Courtship display started at 22°-24°C in P. chelata and at 20°-26°C inP. pul-
lata; it ceased immediately when temperatures fell below these temperature thresholds
(Hallander 1967). In an araneid, Cyrtophora citricola, courtship occurred at 16°-23°C
and stopped when both the upper and lower limits of this range were exceeded (Blanke
1972).
Feeding did not occur below 5 °c in various forest floor species, but was performed
optimally at 10°-15°C. The whole temperature range for activity was 5 0 -20°C in
these spiders (Moulder and Reichle 1972). Geolycosa godeffryoi feeds regularly only
above 15°C; this species normally regulates Tb at very high levels (30°-36°C) in the
field (Humphreys 1977b, this volume). like the respiration rates, the number of prey
items killed under cycling temperatures (11.7°-29.3 °C) was higher than the corres-
ponding value at constant temperature (20°C).
Some spiders may respond to temperature just in the reverse direction, i.e. stimu-
lated by falling and inhibited by rising temperatures. Carrel (1980) suggested that
cooling of the soil could trigger the nocturnal movement of Lycosa ceratiola out of
its burrow toward the surface. Theraphosids were stimulated by low (Minch 1978) or
falling temperatures (Cloudsley-Thompson 1981 ;Pulz et al. 1984; Pulz 1986a,b) quite
similarly. Whereas Cloudsley-Thompson observed increased activity at the moment
when the temperature (30°C) began to fall, I noted bursts of locomotoric activity
only when the animals (Phormictopus sp., Brachypelma smithi) were re-cooled from
high constant temperatures (> 35°C): activity then began at a certain threshold Tb
(see below). Theraphosids were spontaneously active at low T a , however, they were
immediately inhibited in activity when Ta rose. These responses suggest neuronal
mechanisms possibly modulating the activity patterns of nocturnal spiders. Interesting-
ly, slugs and various arthropods reacted quite similarly to temperature changes (Dainton
1954; Dainton and Wright 1985; Kerkut and Taylor 1958; Gardefors 1964).
Modifications. In Lycosa lenta and Filistata hibernalis, forced locomotion between

15° and 33°C was only weakly influenced by temperature. QIO values for both maxi-
mal speed of locomotion and maximal distance traveled in 2 min were between 1.2
and 2.1. On the other hand, recovery from forced activity (lactate removal) was most
rapid at the acclimation temperature (25°C), but was reduced at 15° and 33°C
(Prestwich 1982). The speed oflocomotion (prey approach) in Salticus scenicus signif-
icantly increased with experimental temperature (QIO = 2; 18°-28°C). Exhaustion of
preying jumps was not temperature dependent, but the recovery rate of the prey-
catching reaction increased with temperature (Dalwigk 1973). In Lycosa malitiosa, the
copulation duration was inversely related to temperature (14°-33°C). The frequency
of single actions (palpal insertions, copulatory side changes) apparently remained con-
stant throughout the whole temperature range (Costa and Sotelo 1984).
3.3 Evaporative Cooling
General Aspects. Evaporative cooling may not be expected to be significant in spiders,

mainly because most spiders simply avoid overheating by behavioural strategies (e.g.
burrowing, hiding in retreats or in the shade), by nocturnal patterns of activity and, in
general, by the selection of favourable microhabitats (Main 1982; Cloudsley-Thompson
1983; Humphreys, this volume). For physiological reasons, evaporative cooling may
not be used because water stores are too limited to provide for efficient long-term cool-
ing, but possibly also because high rates of water loss will raise the concentration of
body fluids, which again involves osmotic problems (J .F . Anderson 1974; Edney 1977).
On the other hand, considering evaporative cooling as a long-term strategy, lasting
for hours, is probably wrong. Rather, situations are conceivable in which short-term
cooling ('emergency cooling'), e.g. for 1 h, is of high adaptive value: •
36 R. Pulz
1. Web spiders may be directly exposed to the sun when waiting for prey. Provided
that changes in body posture alone do not sufficiently reduce or stabilize T b below
critical values, evaporative cooling could be additionally used to extend the activity
in the web. Otherwise, the animals would have to abandon activity and move into
the shade.
2. Spiders may be exposed to high Ta in their retreats, e.g. under stones or in flat bur-
rows, when these are directly heated by the sun. This situation must be expected to
occur frequently in young, vagrant individuals of species living sedentary as adults,
e.g. in theraphosids (Seymour and Vinegar 1973; Gertsch 1979; Pulz and Schulz
1986). Evaporative cooling could allow them to remain in the shelter instead of leav-
ing it and thus running the risk of encountering still higher temperatures or a predator
outside.
Mechanisms. How then would spiders cool? The water loss via the body surface and
respiratory organs draws heat continuously from the body: at 20°C about 2453 J
(586 cal) and at 40°C about 2406 J (575 cal) per gram water. Passive transpiration,
however, will not suffice to effectively cool spiders since transpiration rates, even when
increasing with rising Tb, are generally too low to balance heat input. Spiders must
therefore actively augment total water loss, whether by increasing transpiration or by
releasing fluids on the body surface.
J.F. Anderson and Prestwich (1985) measured decreases in Tb of 0.3°-1.0 °c in
continuously active Brachype/ma smithi at Ta = 23°C. They assumed that evaporative
cooling occurred as a by-product of forced book-lung ventilation during activity. It re-
mains to be studied, however, whether increases in respiratory water loss can actually
lower T b , and whether this mechanism could also be employed in resting animals and
at high Ta. In insects, analogous mechanisms have long been known (Edney 1977).
First evidence of a specialized evaporative cooling mechanism in spiders came from
Krakauer (1972): at Tb = 36.6 °c, large orb-weaving spiders,Nephila clavipes, released
a drop of fluid from the mouth and manipulated it with the chelicerae. The effect of
this behaviour on Tb was not analyzed. However, Krakauer suggested that it might
represent physiological thermoregulation, analogous to evaporative cooling mechanisms
in insects (Adams and Heath 1964; Kammer 1981; Cooper et al. 1985; May 1985).
I have confirmed Krakauer's hypothesis in Brachype/ma smithi andPhormictopus sp.
(Pulz 1985b, 1986a, in press). Both species release fluids at Tb = 33°-35°C mainly
from the gnathocoxal glands (also called 'salivary' glands, which should not be con-
fused with the 'sexual' glands; Lopez et al. 1980; this volume), but possibly also
from the mouth. Fluid secretion was regularly accompanied by cheliceral and pedi-
palpal movements in Brachypelma, but only rarely in Phormictopus. Both species (1)
reduced Tb, (2) stabilized T b at lower levels than Ta and (3) slowed down increases in
Tb by these behaviours. At falling Tb , extruded fluids (which normally accumulate
between the gnathocoxae) were often sucked back through the mouth.
Origin and Efficiency. There is considerable evidence that these behavioural patterns
are closely related to the normalleg-cleaning behaviour in spiders: in either behaviour
the gnathocoxal glands are activated (Sittertz-Bhatkar 1980), cheliceral and pedipalpal
movements occur and cyclic patterns of water loss are observed (see below). Moreover,
"0
To =2".5 °c
.
T opist hosoma
oJ-O
-O-O----'-0--0
-0-0--0-0-0--0-0-_0
.2 35
Q)
E!Q) 33
a.
E
~
30
o 30 60 90 120
Time (min)
Fig. 12. Internal heating of the opisthosoma (35°-37.5 °C) elicited a short jerk, cheliceral move-
ments (chewing) and leg cleaning in Brachypelma smithi (body weight : 14.0 g) freely behaving in a
terrarium (lleft walking leg; r right walking leg; 1-4 position of walking legs from front to back).
Note the decrease in temperature of the unheated prosoma, which is due to the evaporation of
cle~\ning fluid at the leg surface. The animal did not show any movements after cleaning, but as-
sumed a resting 'posture with largely extended legs
thermal stimulation within the body or at the body surface elicited normal leg clean-
ing in freely behaving spiders (Pulz 1985b, 1986a). This activity resulted in a marked,
but relatively slow reduction of Tb (Fig. 12). It resembles a similar behaviour in
honey bees, which under heat load extrude a fluid from the mouth and smear it over
the thorax (Esch 1976). Clearly, distributing fluids over a greater body surface will
facilitate cooling, in particular in still air. Optimal cooling, however, may be expected
in spiders being normally exposed to moving air, namely in the orb weavers. Field ob-
servations are now needed to support these findings.
3.4 Control of Temperature Responses
Complex thermal responses, like locomotoric activity (see above) , evaporative cooling
or shuttling between the sun and shade (Humphreys, this volume), raise questions
about the underlying steering mechanisms. Problems of the sensory reception of environ-
mental stimuli, the central processing of this information and the release of behaviour
patterns are involved here. At present, these questions must remain largely unanswered
in spiders : there is still more to ask than to review at this stage. Some basic ideas have
already appeared, however, and will be outlined below. They were derived from my
experiments in Phormictopus and Brachypelma (Pulz et al. 1984, 1986a, b).
38 R.Pulz
3.4.1 Temperature Perception
The term 'temperature perception' refers to all steps of central nervous activity be-
tween the reception of a temperature stimulus and the fmal release of autonomic or
behavioural responses, e.g. an increase in heart rate or the withdrawal of a leg. In
properly designed experiments, these responses can provide information on the thermal
sensitivity of an animal.
Perception of Ambient Temperature. In spiders, behavioural responses to local thermal

stimuli have long been demonstrated (Gaubert 1892, McIndoo 1911, Schlott 1931,
Blumenthal 1935). However, the structures and mechanisms involved in thermorecep-
tion still await discovery (Altner and Prillinger 1980). A certain hair sensillum occur-
ring on the distal leg segments of Sericope/ma rubronitens (Theraphosidae) was assumed
to be a thermoreceptor (Den Otter 1974). This hypothesis could not be confirmed in
later studies (Pulz 1986a; Kloppenburg and Pulz,in press). Rather, electrophysiological
experiments indicate that not a single, but various receptor types might contribute to
temperature sensing in spiders. From electrophysiological studies it appeared also that
theraphosids are particularly sensitive to temperature changes at high initial tempera-
tures (Ta > 35°C). These results were confIrmed in behavioural studies: sensitivity to
changes in Ta increased exponentially with initial temperature. At 35°-40°Cincreases
in Ta of only 0.3°-1.0 ° C elicited behavioural responses. With regards to the localiza-
tion of thermosensitive structures at the body surface, I found that the distal parts of
the tarsi with the claw tufts, the spinnerets and the tarsal and metatarsal scopulae were
most sensitive to thermal stimuli.
Perception of Body Temperature. Spiders may not only respond to peripheral tempera-
ture changes, but also to changes in Tb . Strong evidence came from those thermal
responses which are released at certain threshold body temperatures (Pulz 1986b;
Humphreys, this volume). In theraphosids, behavioural activity at both rising and fal-
ling Tb was elicited at the same threshold body temperature (32 0e), suggesting that
Tb may, at least partially, be measured by neurones speCifIcally sensitive to this tem-
perature, irrespective of the direction of temperature changes. However, the animals
responded also to fIne increases and decreases in Tb both below and above the threshold
value (Ta was constant in these experiments; the animals were internally heated by
high-frequency current). High sensitivity to these thermal stimuli was similar to peri-
pheral temperature changes.
Thermal sensitivity within the body is obviOUsly not restricted to one soma: rather,
my findings indicate that temperature changes are independently sensed both in the
prosoma and opisthosoma. In the opisthosoma, the heart ganglion together with as-
sociated nerve fibres (Gonzalez-Fernandez and Sherman 1984; Carrel, this volume)
deserves further attention as a possible candidate for mediating temperature sensations
within the body (Pulz 1986a; Pulz and Hinz in prep.). Separate temperature sensors
are also expected in the tagmata of insects (May 1985).
3.4.2 Interactions
Thermal cues may interfere with a variety of environmental factors in controlling

thermal responses in the field, including humidity (N<t>rgaard 1951; Hallander 1970;
Coyle 1971), rainfall (Pointing 1965), light intensity (N<t>rgaard 1956; Hallander 1970;
Blanke 1972), habitat structure (Riechert and Tracy 1975), substrate (Hallander 1970;
Coyle 1971) and wind speed (Suter 1981). The exact contribution of these parameters
to the control of temperature responses is not known. Therefore, we will consider here
only thermal interactions. In view of the normally homogeneous Tb in spiders (see
above), temperature sensing in both tagmata may be regarded as significant only in
relatively large species with behaviours potentially producing gradients in Tb , e.g. in
spiders sunning their egg sac in a head-down posture (Humphreys, this volume). On the
other hand, the separate sensing of T b and peripheral temperatures seems to be essen-
tial in all species. Quantitative behavioural analysis in theraphosids revealed that thermal
responses (locomotoric activity, secretion) depend both on Tb and Ta. These tempera-
tures are obviously compared centrally and do not only influence the pattern, but also
the intensity of a behaviour: for instance, given a certain Tb , the intensity of locomo-
tion (speed, frequency of single actions), will increase with T a' On the other hand,
peripheral temperature stimuli can elicit phobic or kinetic responses before any
changes in Tb occur. Moreover, activity patterns in the field could be modulated at
night (Carrel 1980) and during the day (Riechert and Tracy 1975; Minch 1978) by Ta
alone.
4 Water Relations
Research on the water relations of arthropods has largely dealt with insects (Wharton
1985), mites and ticks (Kaufman and Sauer 1982; Kniille and Rudolph 1982), but
relatively seldom with other arachnids (Edney 1977). Among these, spiders and scor-
pions have certainly received the greatest attention, whereas in other groups only few
studies exist (Uropygi: Ahearn 1970; Crawford and Cloudsley-Thompson 1971; Soli-
fugae: Cloudsley-Thompson 1961b). This section reviews the water relations of spiders,
proceeding from the body water content to mechanisms of water gain and loss. Mecha-
nisms of osmotic and ionic regulation will be dealt with in the last part. However, many
questions related to this field still remain open or may have not yet even been asked.
4.1 Water Content
Water is the major constituent of the body fluids of an organism. It serves a variety of
physiological needs, whether as a transport medium for gases, chemical compounds or
cells, as a solvent, as a means for maintaining thermal balance or as a chemical reagent.
In spiders, body fluids are additionally involved in the generation of hydrostatic pres-
sure for locomotion, a function which is so important that losses of hemolymph are
soon replaced by drinking (Stewart and Martin 1970; J.F. Anderson 1974). However,
this mechanism may be seriously hampered by the necessity to maintain osmotic balance.
Table 1. Selected data on water content and lethal water loss of spiders. [See Millot and Fontaine (1937) and Vollmer and McMahon (1974) for more data on 0
""'"
water content]
Species Stage a Water content Remarks c Reference

(% body weight)b
Dugesiella hentzi A 73 2.5 -4 months before ecdysis Stewart and Martin (1982)
77 1-2 months before ecdysis Stewart and Martin (1982)
72 10 days after ecdysis Stewart and Martin (1982)
Pardo sa pullara Juv. 72-84 First instar after hatching from egg sac Pulz (unpubl. work)
Pardo sa /ugubris Cocoon 68-72 Baehr and Eisenbeis (1985)
A9 74 Mesic habitats Baehr and Eisenbeis (1985)
Pirata uliginosus Cocoon 64-65 Baehr and Eisenbeis (1985)
A9 72-73 Hygric habitats Baehr and Eisenbeis (1985)
Pirata Iatitans Cocoon 7074 Baehr and Eisenbeis (1985)
A9 72-74 Hygric habitats Baehr and Eisenbeis (1985)
Pirata hygrophilus Cocoon 74--76 Baehr and Eisenbeis (1985)
A9 74--75 Hygric habitats Baehr and Eisenbeis (1985)
Meta menardi Egg 58--60 Dresco-Derouet (1960)
A9 76 Before egg laying Dresco-Derouet (1960)
Nephila senegaiensis Egg 56-65 Dresco-Derouet (1960)
A9 75-78 Before egg laying Dresco-Derouet (1960)
65-70 After egg laying Dresco-Derouet (1960)
Tegenaria saeva Egg 68-71 Dresco-Derouet (1960)
A9 75-76 Before egg laying Dresco-Derouet (1960)
Achaearanea tepidariorum A 85 Dry habitats Vollmer and McMahon (1974)
Lithyphantes paykullianus A 61 Millot and Fontaine (1937)
Filistata insidiatrix A 67 No drinking observed Millot and Fontaine (1937)
Dictyna viridissima A 73-87 Collected on Opuntia ficus-indica Pulz (unpubl. work)
12 different species Egg 68 J.F. Anderson (1978) ::c
."
N
'"
>-3
::r'
Lethal water loss 'S...."
(% body weight) b
.,eo.
;:l
Dolomedes fimbriatus A 22-31 At 20° -23 ° C/50 and 90% r.h. Palmgren (1939) Po
Geolycosa godeffroyi A 16-23 At 5°-45 °C/30 and 80% r.h. Humphreys (1975a) .,....::E
Trochosa spp. (3 species) A >25 At 28°-30 °C/75 and 92% r.h. Engelhardt (1964) ....
'":;0
Tarentula barbipes, Lycosa A 20 At 20° and 35 °C/O% r.h. D.A. Parry (1954)
radiata '"
E:
Pardosa pullata A 23 At 19°-20 °C/8-1O% r.h. Pulz (unpubl. work) o·
;:l
'"
Amaurobius spp. A 20-25 At 20° ±1 °C/O and 50% r.h. Cloudsley-Thompson (1957)
Dictyna viridissima A 18 At 19°-20 °C/8-1O%r.h. Pulz (unpubl. work)
Gnathonarium dentata A 26-31 At 18°-20°C < 10% r.h. Palmen and Soumalainen (1945)
Linyphia montana A 26 At 19°-21 °C/O% r.h. Lagerspetz and Jaynas (1959)
a A = adults (females, males or both sexes); Juv. = juveniles.

b Rounded mean values.
C r.h. = Relative humidity of the air.
*'"
-
42 R. Pulz
The body water content of spiders (Table 1) lies roughly in a range of 60- 85% and
agrees well with data obtained in other arthropods (Hochrainer 1942; Edney 1977;
Table 1). Whereas body water content shows high interspecific and intraspecific vari-
ability, eggs appear to be generally characterized by a low water content. This can
be explained by high amounts of lipids contained in spider eggs as an energy store
(J.F. Anderson 1970; Rakotovao and Rahandraha 1973).
Principles underlying the variability in water content between species and individuals
are briefly summarized as follows:
1. Unlike water loss rates or resistance to desiccation, water content does not reflect
habitat characteristics, but is species specific (Millot and Fontaine 1937; Vollmer
and McMahon 1974).
2. Individual water content depends on numerous parameters, including developmental
stage, lipid content (which increases in gravid females, see above), the physiological
state (e .g. state of feeding and moulting), age, humidity of the air and temperature
(Millot and Fontaine 1937; Hochrainer 1942; Dresco-Derouet 1960; Rakotovao and
Rahandraha 1973; Stewart and Martin 1982; Pulz, unpub!. work).
3. There is no consistent trend of differences in water content between the sexes
(Hochrainer 1942; Baehr and Eisenbeis 1985).
We may conclude from the high variability that water content in spiders is not an
appropriate parameter for interspecific comparisons or ecological interpretations.
Rather, it is interesting by itself in reflecting individual physiological variability in
response to various biotic and abiotic factors. Moreover, the state of hydration deter-
mines the actual water loss rates and the water reserves that can be further lost without
physiological damage.
4.2 Water Gain
Arthropods may (1) actively absorb water from subsaturated atmospheres, (2) drink
free or soil capillary water, (3) gain metabolic water and may (4) extract water from
the food (Edney 1977). Since data is lacking, we shall consider here only the first two
possibili ties.
Absorption vs Adsorption. Active water absorption has not been demonstrated in

spiders, rather, previously desiccated animals continued to lose water even in nearly
saturated atmospheres (Palmgren 1939; Nemenz 1954; Aspey et a1. 1972; Humphreys
1975a; Punzo and Jellies 1983). Sometimes, however, slight gains in body mass were
observed in fasting spiders, exposed to high r.h. at low temperatures (Hadley et al. 1981;
Pulz, unpub!' work). These weight gains certainly reflect the accumulation of metabolic
water rather than active water uptake.
On the other hand, the cuticle may passively adsorb water in subsaturated atmo-
spheres (Eskafi 1974; Constantinou and Cloudsley-Thompson 1983). Water adsorption
was demonstrated in theraphosids exposed to an airstream when changing from low
(2-3%) to higher r.h. (40-50%): parallel to the increase in r.h., cuticular temperatures
showed a transient rise up to 1 0 C and dropped to lower, but still elevated levels.
This effect is due to the formation of adsorption heat at the cuticular surface (Pulz
1986a).
Spiders in nearly saturated atmospheres will profit from slight decreases in tempera-
ture, since water may condensate on the cuticle, the droplets being readily available
for ingestion (Humphreys 1975a).
Drinking. Spiders show a striking variability with respect to drinking behaviour: from
species that apparently never drink or even die(?) when exposed to moisture (Filistata
hibernalis; Millot and Fontaine 1937; Prestwich 1982) to species that require free li-
quid water in their habitats to survive (Dolomedes fimbriatus; Palmgren 1939). Erigone
longipalpis, a species common in salt marshes, even drinks salt water with a salinity of
up to 30%0 to compensate for water losses (Bethge 1973).
Provided no free water is available, Lycosa radiata and Tarentula barbipes may
imbibe soil capillary water against suction pressures of up to 20 kPa (150 mmHg) and
53.2 kPa (400 mmHg), respectively (D.A. Parry 1954), corresponding to the minimal
suction pressure of 17.3 kPa (130 mmHg) developed by the pharynx of a scorpion,
Paruroctonus mesaensis (Yokota 1984). Geolycosa godeffroyi obtained water from
soil containing more than 11% water (Humphreys 1975a); a similar threshold (10%)
was measured in Loxosceles ree/usa (Eskafi 1974).
Drinking from a free water surface lasted from several minutes to nearly 1 h (Engel-
hardt 1964) and allowed the almost complete restoration of previously lost water
(Nemenz 1954; Engelhardt 1964). Drinking rates (mg water/min) were correlated with
body weight both in spiders (D.A. Parry 1954) and scorpions (Hadley 1971), but not
with the extent of desiccation (Hadley 1971). The rates observed so far range from
0.2 mg/min in juveniles of Pardosa pullata (Pulz, unpubl. work) to more than 33 mg/
min in Lycosa radiata (D.A. Parry 1954).
Dondale (1965) made the striking observation that spiders drink their moulting
fluid: a few minutes after ecdysis, Philodromus rufus started ingesting excess fluid
from its old exuvia. It is not clear to me how the spider managed to move so soon
after ecdysis, since the new cuticule is not yet fully hardened.
The question whether or not drinking behaviour in spiders is related to certain species
or even families has been a matter of discussion (Gerhardt 1923, 1928b; Millot and
Fontaine 1937; Nemenz 1954). Apparently, filistatids and thomisids seldom drink, at
least in the laboratory, whereas other species, e.g. theraphosids, lycosids, agelenids and
ctenids drink frequently. The tendency to drink may correlate with the normal body
water content, i.e. species with a water content> 70% drink most readily (Millot and
Fontaine 1937). However, spiders certainly always drink when starved (J.F. Anderson
1970), when body fluids have been lost (Stewart and Martin 1970), prior to moulting
(Stewart and Martin 1982) and, more generally, when the body water content has suf-
ficiently decreased (Nemenz 1954).
4.3 Water Loss
Methodical Aspects. Until recently, gravimetry has been the dominant method in study-
ing the water relations of spiders (Toft 1980; Hadley et al. 1981; Stewart and Martin
1982; Punzo and Jellies 1983; Opell 1984; Baehr and Eisenbeis 1985). This technique
44 R.PuIz
originally allowed to make only punctual measurements, but has been improved to
continuously register water loss in arthropods even as small as ticks (Kniille and Rudolph
1982) and collembola (Eisenbeis 1982). It has now become possible to study both
absolute rates (usually expressed as weight loss per hour, related to the body weight
or surface) and the course of water loss with fairly high time resolution. Thus, one
can correlate water loss with actual behavioural patterns and analyze it with respect to
its origin and control.
Water loss has been continuously measured with two alternative methods, i.e . tracer
techniques (W.W. King and Hadley 1979; Low 1983; Nicolson et al. 1984; Vannini
et al. 1985) and electronic humidity sensors in combination with flow-through systems
(Loveridge 1980; Quinlan and Hadley 1982; Pulz 1985a,b, 1986a,in press). The former
technique implies difficulties, e.g. in the assessment of quantitative data (W.W. King and
Hadley 1979), and involves a comparatively low time resolution. These problems have
been largely overcome by the latter method. However, basic difficulties in the study of
whole, intact or experimentally pretreated animals still remain unsolved (Edney 1977;
Loveridge 1980;Pulz 1986a).
Water Loss in Spiders. Measurements of total water loss in arthropods are generally as-
sumed to reflect the sum of respiratory and cuticular transpiration, other sources being
negligible . At low temperatures this assumption is certainly valid, however, difficulties
may arise when applied to water loss in live and intact spiders at temperatures above
35°C (Stewart and Martin 1970; Humphreys 1975a; Hadley 1978): under these condi-
tions, secretory or excretory activity significantly contributed to total water loss in
theraphosids as well as in a ctenid, Cupiennius salei (Pulz 1985b, 1986a, in press; Pulz
and Sperber in prep.). Various additional sources of water loss (Fig. 13) must therefore
be taken into consideration at high, but also at normal temperatures, the gnathocoxal
glands being particularly important (see below). It remains a challenge for future re-
search to determine the specific role of single avenues of water loss in the overall water
WATER LOSS
LI QUID! VAPOUR VAPOUR
1
2
3
8
~ t.
~5
Fig. 13. Possible origin of water loss in spiders. The
~ 5
diagram shows a female theraphosid (ventral view).
g~
1 poison gland; 2 gnathocoxal ('salivary') gland ;
3 rostral gland ; 4 mouth; 5 coxal gland; 6 sexual
g~
opening; 7 anus; 8 cu ticle; 9 book lung; 10 spin-
8 neret. At sites 1-7 water is normally lost in liquid
form. It is uncertain to which extent water vapour
-/---- 10 ~ may be released. Asterisks denote outlets or out-
-----~a{J/
leading ducts which are actively controlled by
----- 10 ~
muscles
balance of spiders. First steps in this direction were the separate measurements of
respiratory and cuticular transpiration in Eurypelma californicum (Low 1983; see
Cooper 1983; Nicolson et al. 1984 for insects) and the study of water extraction by
the major ampullate duct in Argiope aurantia (Tillinghast et al. 1984) (see below).
At this stage two conclusions can be made: (1) It is necessary to continuously ob-
serve the experimental animal, to inspect it for possible fluid extrusions and to note its
behaviours during recordings of water loss. (2) It should be strictly differentiated be-
tween transpiratory and evaporative water loss instead of using both terms synonymous-
ly. Transpiration should refer only to primary vapourous water loss in live animals,
such as water loss through the integument, the respiratory organs or other body open·
ings. Evaporation should refer to primary non-vapourous water loss, as it occurs when
fluids evaporate at the body surface. Moreover, evaporation may also refer to water
loss in dead animals.
4.3.1 Cuticular Water Loss
Quantitative Aspects. Early experiments of Palmgren (1939) suggested that the integu-
ment is the main route of water loss in spiders. This was confirmed in later studies,
cuticular water loss being estimated to contribute from 88% (Low 1983) to 98-99%
(Nemenz 1954) to total water loss. The former value was obtained by direct and sepa-
rate measurements of cuticular and respiratory transpiration, a method which avoids
several problems involved in an alternative technique, sealing experiments (Edney 1977;
Loveridge 1980).
At a given temperature, cuticular water loss may decrease with decreasing body
water content (Edney 1977; Yokota 1979; Hadley 1984), with decreasing hydration
of the integument (Punzo and Jellies 1983; Hadley 1984), but - at the same saturation
deficit of the air - may be higher at high r.h. than at low r.h. (Humphreys 1975a;
Punzo and Jellies 1983).
Qualitative Aspects. The significant contribution of cuticular transpiration to total

water loss indicates that the spider cuticle (Dalingwater, this volume), as the arthropod
cuticle in general, is a more or less open system (Dalingwater, this volume). The out-
ward passage of water is not specifically restricted to certain areas, but is certainly
facilitated by countless fine openings, e.g. pore canals, pores of dermal glands or of
chemoreceptive organs. The fundamental role of the outermost cuticular layer, the
epicuticle, in restricting in tegumen tal water loss was recognized early in abrasion experi-
ments (Davies and Edney 1952; Nemenz 1955). Meanwhile, the spider cuticle and
epicuticle, respectively, have been subjected to detailed chemical analyses (Hadley
1978; Hadley et al. 1981), allowing the precise determination of qualitative and quan-
titative parameters of the cuticular waterproofing mechanism. The state-of-the-art in
this field has recently been reviewed by Hadley (1984). Here, only three basic aspects
shall be briefly outlined.
1. The resistance of spiders to desiccation (and thus, the striking interspecific vari·
ability in transpiration rates, see below) is largely determined by the amount and com-
position of cuticular lipids. The epicuticle seems to be crucial for the overall water-
proofmg properties of the cuticle (Edney 1977; Hadley 1978). The structure of the
46 R. PuIz
entire cuticle (e.g. heavy sclerotization) may additionally influence the resistance to
desiccation (Levi 1967b), however, it should be borne in mind that some Central Euro-
pean species carrying sclerotized scuta (e.g. Pholcomma gibbum, Ceratinella spp.) live in
humid or even moist microhabitats (Braun 1969). Striking differences in ultrastructure
and thickness suggest that the permeability of the cuticle varies between different parts
of the body (Edney 1977). It is interesting, therefore, that recent evidence indicates
direct transcuticular transfer of water from the female scorpion, Euscorpius [lavicaudis,
to the larvae on its back (Vannini et al. 1985).
2. It has been proposed that the spider cuticle, as that of many other arthropods,
undergoes an abrupt increase in permeability at certain 'transition temperatures'
(Davies and Edney 1952;Cloudsley-Thompson 1957;Stewart and Martin 1970; Humph-
reys 1975a). These may range from 18°-20°C (Eskafi 1974) to about 40°C (Davies
and Edney 1952) and are species-specific. The transition phenomenon has been a mat-
ter of controversy and has not yet been fully understood (Toolson 1978, 1980; Monteith
and Campbell 1980; Edney 1982; Hadley 1984; Wharton 1985). In the light of fluid
releases at high temperatures, abrupt increases in water loss, at least in live animals,
should be interpreted very cautiously. Moreover, such events might also be attributable
to increases in respiratory transpiration, provided the respiratory openings have not
been sealed (Yokota 1979).
3. Total water loss increases in dead spiders, even when the book-lung spiracles are
blocked (Davies and Edney 1952; Nemenz 1954). Active mechanisms, possibly located
in the epidermal cells, could therefore underly the control of cuticular water loss
(Edney 1977; Hadley 1984).
4.3.2 Respiratory Water Loss
Quantitative Aspects. Water loss from the respiratory organs (tracheae, book lungs;
Strazny and Perry, this volume) is low, provided the animals are resting in normal air
and at low temperatures. Under these conditions, respiratory water loss may contribute
only slightly more than 1% (Nemenz 1954) to 12% (Low 1983) to total water loss,
which would allow one to equate total transpiration with cuticular water loss (Hadley
et al. 1981).
The situation changes dramatically, however, when spiders are exposed to air con-
taining 6-10% carbon dioxide: the spiracles of the respiratory organs (which are normal-
ly almost completely closed) will then open and allow unhindered water loss from the
respiratory surfaces. Total water loss may increase by 43% (Davies and Edney 1952),
63- 86% (Cloudsley-Thompson 1957) or even 100% (Stewart and Martin 1970), de-
monstrating the efficiency of the spiracular closing mechanism in restricting respiratory
water loss (Levi 1967b). High water loss rates from the respiratory organs, in particular
from the book lungs, must be potentially expected: for instance, in a 10-g individual
of Eurype/ma ca/i/ornicum, an outer body surface of about 60 cm 2 compares with a
respiratory surface of about 40 cm 2 (Focke 1981). It is not surprising, therefore, that
Stewart and Martin (I970) measured such a high increase in total water loss when ex-
posing another theraphosid,DugesieZIa hentzi, to 10% carbon dioxide.
At constant temperatures, respiratory water loss might increase with activity (J.F . An-
derson and Prestwich 1985), but might decrease in dehydrated animals (Yokota 1979)
and possibly also in dry air (Humphreys 1975a; Edney 1977).
Qualitative Aspects. Respiratory organs and mechanisms of spiders are dealt with by
Strazny and Perry (this volume). We shall summarize here only those aspects pertinent
to respiratory water loss.
1. The respiratory system of spiders is both structurally and functionally adapted
to restrict water loss. Structural adaptations have been thoroughly discussed by Levi
(1967b). He argued that tracheae, when compared with book lungs, are physiologically
advantageous in providing for a more efficient gas exchange and in preventing water
loss (see Kestler 1985 for insects). In fact, highly branched tracheal systems extending
from the opisthosoma to the prosoma are found in many small spiders, potentially
endangered by heat and desiccation, but being largely resistant to these hazards (e.g.
Dictyna viridissima; Pulz, unpubl.). Physiological studies comparing book lungs and
tracheal systems with respect to respiratory efficiency and water loss are missing.
2. Functional adaptations closely resemble mechanisms known in insects (Edney
1977) and ticks (Kniille and Rudolph 1982) and are characterized as 'intermittent' or
'discontinuous' respiration. However, only few species have been studied so far, and
evidence is not conclusive.
In Araneus spp., the book-lung spiracles opened widely whenever the animals moved,
permitting burst releases of carbon dioxide (Robinson and Paim 1969) and possibly
also of water vapour. In Eurypelma californicum, on the other hand, the spiracles some-
times closed at the onset of strong activity, but opened fully in the recovery phase
thereafter (T. Ficke et al. 1986; Paul 1986; Paul et al., in press). Long-term recordings
of the transpiration of single book lungs in Phormictopus agree well with these observa-
tions: body movements were normally accompanied by an initial rapid drop and a
subsequent slow decrease of respiratory water loss (RWL). After activity, RWL re-at-
tained the former transpiration level only gradually (pulz and Ounas, in prep.). These
results do not support the hypothesis that book-lung ventilation (which is most intensive
during the recovery phase, Angersbach 1978; Paul et al., in press) causes significant
increases in RWL (Anderson and Prestwich 1985). Moreover, they indicate that in
moving theraphosids RWL can be restricted, most probably by a spiracular closing
mechanism. This is still a very tentative conclusion which awaits confirmation in other
species: according to observations in Araneus spp. (see above), spiracular control is not
likely to operate in all spiders during activity.
4.3.3 Liquid Water Loss
liqUid water loss in spiders occurs in the form of secretion (possible sources: poison
glands, rostral gland, gnathocoxal glands, mouth) and excretion (possible sources:
coxal glands, anus), as well as from the sexual opening during egg laying (Gertsch
1979; Humphreys 1983) (Fig. 13). Moreover, freshly moulted spiders may lose large
quantities of water due to the evaporation of moulting fluid (Hagstrum 1970b; Stewart
and Martin 1982). The relative and absolute importance of these water losses to the
water balance in spiders can only be roughly assessed.
Secretion. Water loss from the poison and rostral glands seem to be negligible, since
only small quantities of fluid are released. The function of the rostral gland is not clear
at present. It might be speculated that it adds secreta to cleaning and/or digestive fluids
(Sittertz-Bhatkar 1980).
48 R. Pulz
Water loss from the gnathocoxal (salivary) glands and from the mouth can be signif-
icant. At high temperatures (35 0 -40 °C), theraphosids evaporated 10-140!1g water/g/
min which was extruded from the gnathocoxal glands for evaporative cooling (see
above). At low T a these glands are activated during normal leg cleaning, which involved
similarly high rates of water loss. In both cases rates exceeded those of mere transpira-
tory water loss. Evaporative cooling and leg cleaning may involve elevated levels of
water loss for some minutes up to several hours (Pulz 1985b, 1986a).
Water loss from the mouth has not yet been quantified in spiders. However, it
is potentially high since spiders regularly extrude digestive secreta from the mouth
which mediate the extraintestinal digestion of prey (Collatz, this volume). Serious
problems may arise in those species which largely rely on water intake from the prey
(see above). The efficiency of prey water extraction can be significantly reduced: the
scorpion Paruroctonus mesaensis loses 0.37 ml of its own body water per milliliter of
prey water gained (Yokota 1984).
Excretion. Excretory water loss has also not yet been determined in spiders. After
feeding and drinking theraphosids were occaSionally seen with drops of a clear fluid
(volume: several microliters) hanging at the openings of one or more coxal glands, sug-
gesting both an osmoregulatory and an excretory function of these organs (Pulz,
unpubl. work). On the other hand, heat-stressed theraphosids were observed to extrude
coxal fluid which may spread on the joint membranes between coxae and sternum. This
could indicate a thermoregulatory, and again an excretory function of the coxal glands;
under heat load the rapid elimination of metabolic waste products may be crucial for
survival (Cloudsley-Thompson 1962). Fecal water loss generally occurs only on the
days following feeding. It seems to be negligible with respect to overall water balance,
since relatively more water is gained from the prey than lost with the feces (Nemenz
1954; Eskafi 1974).
Significant amounts of water may be lost with the moulting fluid: during ecdysis,
Dugesiella hentzi loses up to 50 times more water than non-moulting animals (Stewart
and Martin 1982). Osmotic problems potentially involved in such high increases in
water loss are minimized by excessive uptake of water prior to moulting (see below).
4.3.4 Patterns of Water Loss
Water loss in resting spiders exposed to low temperatures is continuous, a possible ex-
ception being respiratory water loss (see above). In moving spiders and at elevated tem-
perature, however, irregular patterns or periodic increases and decreases may appear.
Figure 14 illustrates the typical pattern of water loss of a theraphosid during locomotoric
activity. Subsequent short-term increases and decreases (peaks) occurred, clearly in-
dicating that the animal was continuously active. Even moving single appendages im-
mediately elicited transpiration peaks. Cheliceral spreading was accompanied by partic-
ularly large peaks (arrows in Fig. 14), suggesting bursts of water loss from the mouth.
These patterns of activity-related water loss are strikingly similar to patterns of car-
bon dioxide release both in active spiders (McQueen et al. 1979) and scorpions (Yokota
1979). They may be theoretically expected to result from discontinuous gas releases
from the book lungs (see above). However, measurements in animals with sealed book-
0.5
c LOCOMOT ION
'E To : 26°C
"-
en
E 0.3
1Il
1Il
~
...
<II
o
3 0 .1
o 10 20 30 40
Time (min)
Fig. 14. Discontinuous water loss in an individual of Phormictopus moving in a dry airstream (70 cm/
min). Arrows point to transpiration peaks elicited by spreading of the chelicerae. These large peaks
probably indicate the release of water vapour from the mouth
lung spiracles did not support this hypothesis (Pulz 1985a, 1986a, in press). Rather, it
appeared that transpiration peaks are mainly due to physical effects (turbulence) and
to water loss from other body openings (mouth, anus).
Water loss in active spiders is a dynamic process, leading to considerable increases in
rates relative to resting animals. Spiders agree well with insects in this respect (QUinlan
and Hadley 1982; Pulz 1986a). It follows that transpiration measurements in live
arthropods should always include observations on behavioural activity, and that the
latter should be stated together with the respective data on water loss.
Cyclic long-term increases and decreases in water loss (period length: several minutes)
occurred in resting theraphosids at high temperatures (35°- 37.5 °C), but also at normal
temperature during leg cleaning (Pulz 1985b, in press). These patterns apparently go
back to cyclic secretory activity of the gnathocoxal glands and reveal striking analogies
to the sweat glands of mammals (Whittow 1971; Gordon and Heath 1983).
4.4 Osmotic and Ionic Regulation
Osmotic and ionic regulation refers to various processes keeping fluctuations in body
water content and in the concentration and ionic composition of body fluids (see
Strazny and Perry, this volume) within tolerable limits. In spiders, research on these
problems is only in the first stage.
Osmoregulatory Mechanisms. Osmotic regulation has been demonstrated in connection

with spinning, moulting and desiccation.
The major ampullate duct, leading liquid silk from the major ampullate glands to
the spinneret serves also to extract water from the silk (Tillinghast et al. 1984). This
process is obviously accompanied by an exchange of sodium and potassium ions, the
drawn silk containing less Na+ and more K+ than the liquid silk within the gland. Water
extraction might be coupled to an active Na+/K+ pumping mechanism. It is most ef-
50 R.Puiz
ficient in minimizing water loss during web construction: an average web of Argiope
aurantia will be almost dry, containing no more than about 6% water.
In Dugesiella hentzi, water loss increased 12 days before ecdysis when the resorp-
tion of the old cuticle began, reached values 50 times higher than normal immediately
at the end of ecdysis and gradually declined to normal levels within 2 weeks (Stewart
and Martin 1982). However, excessive drinking and water storage began already 2 months
before ecdysis, raising the body water content from about 73% to 77% (Table 1) and
lowering the protein-nitrogen concentration of the blood from 13.2 to 9.4 mg/ml.
Despite large fluid losses during moulting, both values re-attained normal levels within
the first week after ecdysis, indicating that drinking is an effective means of 'preventive'
osmoregulation . Moreover, ingestion of water may also restore losses of body fluid
and volume, e.g. during fasting (Stewart and Martin 1970; J.F. Anderson 1974). It
remains to be studied how spiders then avoid fatal decreases in hemolymph osmolality
(J.F. Anderson 1974).
When desiccated , Argiope tri!asciata showed excellent osmoregulatory capabilities
relative to several other arthropods (Riddle 1985). Hemolymph osmolality did not
follow the predicted curve for absent osmoregulation, but rather remained fairly con-
stant over a wide range of body water contents (Fig. 15). Similar osmoregulatory
capabilities are found in desert arthropods , e.g . in the scorpion Parabuthus villosus
(Robertson et al. 1982).
Organs Involved in Osmoregulation. This is certainly the least understood and most
speculative aspect of spider osmotic and ionic regulation. Results of Veda (1974) point
to the rectal bladder (stercoral pocket) as a possible site of the active re-absorption of
water : osmotic pressure of the rectal fluid increased more than that of the hemolymph
when individuals of three species were starved and kept at low humidity. Moreover,
the Malpighian tubules, but also the coxal glands might play a role in osmoregula-
tion . These glands are generally considered to be excretory organs (Seitz, this volume).
mOsm
... ...
"' ......
"0 700
"0
.. . . . .
""..... . .
. - ..
E
~ SOD
.c
a.
---.-:.... .-. -_.. _-......
-..·:~~:·r~:·.: ... _-...::.-.
!.o.~ -:0-
~ 300
<5 Argiope trifasciata
E
!t! 100
1.S 2.S 3.S

Water content (mg H20/mgD.W,l
Fig. 15. Relation between body water content (mg H2 O/mg dry weight) and hemolymph osmolality
in Argiope trifasciata (redrawn from Riddle 1985). The solid line is a regression line fitted to the
data points. The dashed line predicts the osmotic response of standardized laboratory animals to
dehydration in the absence of osmoregulation. The more both lines diverge , the better the osmo-
regulatory abilities. Open symbols represent means of water content and hemolymph osmolality
for field collected (D) , standard laboratory (6 ) and rehydrated animals that had been desiccated be-
fore (0). Horizontal bars are 95 % confidence limits
However, this function should be re-investigated since large quantities of proteins oc-
curred in the coxal fluid of anaesthetized Phormictopus sp. (Pulz, unpubl. work). The
coxal glands of spiders might thus functionally resemble their homologoues in ticks
(Kaufman and Sauer 1982). 'Coxal glands' not related to the arachnid organs occur in
chilopods and are obviously osmoregulatory in function (Rosenberg and Seifert 1977).
5 Autecological Aspects
Parameters most frequently used to ecologically characterize spiders are temperature

and humidity preferenda (N ¢rgaard 1951; Engelhardt 1964; Hallander 1970); resistance
to heat (Holm 1950; Almquist 1970) and desiccation (Almquist 1971; Vollmer and
McMahon 1974), as well as rates of water loss (Nemenz 1954; Hadley et al. 1981;
Cloudsley-Thompson and Constantinou 1983; Baehr and Eisenbeis 1985). In general,
clear correlations are found between these parameters and the habitat characteristics
of a given species (Turnbull 1973). To allow quantitative interspecific comparisons,
however, the experimental techniques should be similar. Otherwise, confusion or even
contradictions may arise from the results.
Two examples illustrate this problem: (1) Herreid (1969) and Low (l983) both
measured transpiration rates in Eurype/ma californicum at 25 DC. Results, however,
differed by a factor of 10, possibly as a consequence of different experimental condi-
tions. (2) According to Vlijm and Kessler-Geschiere (l967), juveniles of Pardosa pul-
lata prefer somewhat drier habitats than the adults, whereas Hallander (1970) obtained
just the reverse result. Again, different methods may explain this puzzling contradic-
tion.
There are many more problems related to the measurement of autecological param-
eters which, however, cannot be considered here in detail (Engelhardt 1964; Krafft
1967; Turnbull 1973). I only refer to the fact that laboratory data - as a rule - are
obtained under conditions varying greatly from those in the field. For instance, spiders
may create favourable micro climatic conditions by burrowing (Engelhardt 1964;
Humphrey 1975a, 1978a) or just by web spinning (Toft 1980), rendering it difficult
to decide whether and to which extent autecological data measured in the laboratory
are transferable to field conditions.
5.1 Hygrothermal Preferenda and Resistance to Heat and Desiccation
Definitions and Background. 'Preferenda' are those temperature and humidity ranges
that are voluntarily chosen by spiders exposed to thermal or hygric gradients. Tem-
perature and humidity selection may depend on a variety of factors, the most important
being species-specific behavioural patterns (Engelhardt 1964; Krafft 1967), seasonal
variations and changes in the physiological state (N¢rgaard 1951; Lagerspetz and Jaynas
1959; Humphreys, this volume), as well as the availability of prey (Riechert 1985).
Heat resistance may be indicated by certain threshold temperatures where (1) escape
reactions cease due to a failure of leg movements (critical thermal maximum), (2) heat
stupor occurs (which is ultimately overcome in re-cooled animals), or where (3) heat
death occurs. Unfortunately, there is no consistent use of the latter terms in the litera-
52 R.Pulz
ture: some authors used heat stupor in the above sense (Schlott 1931; Buche 1966;
HAgvar 1973), others in the sense of heat death (Almquist 1970; Hallander 1970),
whereas another author used both terms synonymously (Nl/>rgaard 1951, 1956). The
problem is hard to solve, since the transition from stupor to death is not readily ob-
servable and is related to the time of heat exposure (Schlott 1931; Krakauer 1972).
The resistance to desiccation is indicated by the ability to resist water loss (as reflected
by the water loss rates) or to tolerate reductions in body water content (Edney 1977).
A critical state of the animal was inferred from the loss of righting response, curling
of legs under the body, along with responses to prodding and ultimate survival (critical
activity point; Vollmer and McMahon 1974).
Interpretation. At fIrst sight, no striking differences in resistance, but in preference

values exist in spiders (Tables 1 and 2). Resistance to heat is generally high, critical
temperatures being around or well above 40°C. Lethal water loss ranges roughly from
20-30% of body weight. Preferred temperatures, however, may range from values
around zero in winter-active spiders to more than 35 °c in surface-dwelling lycosids.
The following conclusions can be made:
1. Due to their siinilarity, resistance data per se are not very useful to ecologically
characterize a given species, i.e. to reveal ecophysiological adaptations. These will ap-
pear when resistances are related to other parameters: for instance, the rather uniform
limits of water loss contrast with strikingly variable rates of water loss, ranging from
less than 0.02% in theraphosids to 2.415% of body weight per hour in Argyroneta
aquatica (Stewart and Martin 1982; Cloudsley-Thompson and Constantinou 1983;
EskafI1974). Obviously, the ability to restrict water loss, and not to withstand reduc-
tions in water content, is a basic physiological adaptation to different habitats.
2. Lower thermal preferenda generally correlate with a lower heat resistance, how-
ever, winter-active spiders show that one is not allowed to generalize this conclusion
(Kirchner, this volume; Aitchison, this volume).
3. High humidity preferenda correlate with low temperature preferenda and high
rates of water loss, and vice versa (Engelhardt 1964; Hallander 1970). However, it is
doubtful whether hygric preferenda actually exist: the literature contains negative or
contradictory results (Palmgren 1939; Nl/>rgaard 1951; Nemenz 1954) as well as the
demonstration of clear-cut humidity preferences (Lagerspetz and Jaynas 1959; Engel-
hardt 1964; Hallander 1970). The sensory basis of hygroperception is unknown (Blu-
menthal 1935; Altner and Prillinger 1980), rendering the whole fIeld of hygric responses
in spiders open for future research.
4. The full signifIcance of ecophysiological data appears in comparative studies, in
particular, when species of the same genus or species living in closely adjacent micro-
habitats are compared. Examples have been demonstrated in numerous spiders, e.g. in
Amaurobius species (Fig. 132), Trochosa species (Engelhardt 1964) and Pardosa pul-
lata and Pirata piraticus (Nl/>rgaard 1951): ecophysiological data precisely reflect habitat
characteristics and contribute to a better understanding of the ecological distribution
of the respective species.
Table 2. Selected data on thermal preferenda and heat resistance of spiders. [See Holm (1950) and Almquist (1970) for more data on thermal paralysis] >-l
::r
~
Species Preferred Critical thermal Upper lethal Remarks References E!
temperature maximum (CTM) temperature .,::se.
(0C) Heat stupor (HS) (OC) 0.-
(0C) ~
~
Aphonopelma sp. B 27-31(a) CTM43 (b) (a) Burrow temperatures Seymour and Vinegar (1973)
...'"
:;0
(b) Not influenced by acclimation to '"~
10°C and 30 °C o·
Lycosa carolinensis 43-45 At constant temperature Moeur and Eriksen (1972) ~
>47 At fluctuating temperatures
Trochosa robusta 24.8 HS 40.4 Engelhardt (1964)
Trochosa spinipalpis 19.2 HS 39.3 Engelhardt (1964)
Arctosa alpigena 47.3 Holm (1950)
Pardosa spp. 43-47.5 Holm (1950)
(8 species)
Pardosa nigriceps 21-27 HS 39.7 Subad ult stages Almquist (1970)
Pardosa chelata 22.5-30 47.7 Hallander (1970)
Pira ta piraticus 18·-24(a) (a) Females without egg sac N</>rgaard (1951)
26-32(b) 35.3 (b) (b) Females with egg sac
Pardo sa pullata 20-27.5 45.1 Hallander (1970)
28-36 (a,b) 43.0(b) (a) Females without egg sac N.prgaard (1951)
(b) Females with egg sac
Pardosa sie"a 34.1 (a)/35.2 (b) (a) Females without egg sac
Pardosa ramulosa 26.8 (a)/29.3 (b) (b) Females with egg sac Sevacherian and Lowrie (1972)
Nephila clavipes CTM41.7 Krakauer (1972)
Cyrtophora citricola 25-29 HS 46 Blanke (1972)
Argiope trifasciata 48.8-58.6 2nd -10th Instars Tolbert (1979)
Agelena labyrinthica 22-25 HS 41-42(a) 45 (b) (a) Exposed;;' 1 h; (b) exposed 1 h Schlott (1931)
Agelena consociata 23 Krafft (1967)
Dictyna viridissima 48-48.5 Pulz (unpubl. work)
Theridion saxatile 47-49 N</>rgaard (1956)
Bolyphantes index 4.1 HS 39.6 Winter-active spider mgvar (1973)
Macrargus rufUS -1 to +3.5 HS 39.4-42.3 Winter-active spider Buche (1966) Ul
w
54 R. Pulz
5.2 Comparison of Developmental Stages
Juvenile stages may be expected to differ from adults in sensitivity to unfavourable

microclimatic conditions, mainly because of the smaller body size and the larger surface
to volume ratio. This assumption certainly holds for eggs and first instars within the
egg sac, being specially protected by certain behavioural patterns of the female (N</>rgaard
1956; Humphreys, this volume), and generally by the sheath of the egg sac. This serves
both to thermally insulate the brood (Hieber 1985) and to protect it from desiccation
(Schaefer 1976b; Ope1l1984; Baehr and Eisenbeis 1985).
In free-living juveniles, however, differences in sensitivity, but also in hygrothermal
preferenda cannot be predicted beforehand. Although the same tendency as in earlier
stages prevails, some interesting interspecific variations may also occur (Table 3). I was
particularly struck by the fact that juveniles of Pardo sa pu/iata (first instar after hatching
from the egg sac) lived more than 20 h in a desiccator (8-10% r.h.; 21°C), whereas
females died after 3-10 h (Pulz, unpubl. work). It is also interesting that juveniles of
some species select higher temperatures than adults (Table 3), which possibly favours
rapid development.
Juveniles may be exposed to more severe micro climatic conditions than those en-
countered by the adults. This is the case, e.g. in ballooning and vagrant spiderlings
Table 3. Comparison of subadult spiders with adult females with respect to hygrothermal preferenda
and resistance to heat and desiccation a
Species Temperature Humidity Resistance Resistance Reference

preference preference to heat to desicca-
tion
Trochosa spinipalpis + Engelhardt (1964)

Trochosa terricola Engelhardt (1964)
Trochosa ruricola Engelhardt (1964)
Trochosa robusta 0 Engelhardt (1964)
Pardo sa sierra Sevacherian and
Lowrie (1972)
Pardosa ramulosa Sevacherian and
Lowrie (1972)
Pardosa pullata +(a) +(a) +(b) (a) Hallander (1970)
(b) Pulz (unpubl.
work)
Pardosa chelata + Hallander (1970)
Clubiona similis Almquist (1970)
Clubiona trivialis Almquist (1970)
Loxosceles reclusa
at low r.h. Eskafi (1974)
at high r.h. 0 Eskafi (1974)
Theridion saxatile -1 Ncprgaard (1956)
Macrargus rufUS + + Buche (1966)
a The symbols refer to the situation in subadult spiders. In lycosids, only females without an egg
sac are considered for comparison: + higher; - lower; 0 no difference; 1 increases with age; 2 not
q uan tified.
landing and moving on hot surfaces, respectively. It could be particularly interesting to

determine how these juveniles differ from adults with respect to ecophysiological adapta-
tions, and whether the smaller stages must generally be expected to be the more vul-
nerable ones.
6 Conclusions
This chapter summarizes (1) temperature effects on basic physiological processes,

(2) mechanisms underlying the control of temperature responses, and (3) elements of
the water balance in spiders. The knowledge in these fields is far from complete, render-
ing future research highly desirable. Numerous attractive problems, like microhabitat
selection, responses to microclimatic factors in the field and mechanisms involved in
osmoregulation, await further investigations. A comparative approach, considering
interspecific, intraspecific and ontogenetic aspects, should prove particularly useful in
such studies. Results would then allow one to better characterize the ecophysiology of
spiders and would provide a detailed basis for comparisons between spiders and insects.
Like these, spiders show a striking ecological diversity, having successfully adapted to
the whole range of microclimatic conditions found in terrestrial habitats. Autecological
parameters, like temperature preference and resistance, often reflect these adaptations
and are briefly discussed in the last part of this paper.
Acknowledgements. My thanks are due to Dr. Rosemarie Schulz for valuable discussions and for
her assistance in typing the manuscript. Particular thanks are due to Rolf Laschefski-Sievers who
helped when my text computer refused to function. Part of my own work was supported by the
Deutsche Forschungsgemeinschaft (WU 63/7-1, 7-2).
IV Behavioural Temperature Regulation
WILLIAM F. HUMPHREYS!
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 56
2 Evidence of Temperature Regulation in Spiders .. . . . . . . . . . . . . . . . . . . . . . .. 57
2.1 Casual Field Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 57
2.2 Egg Sac Temperatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 57
2.3 Burrows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 58
2.4 Continuous Monitoring of Body Temperature .. . . . . . . . . . . . . . . . . . . . .. 59
3 Changes of Body Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 59
3.1 Daily Course of Body Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.2 Seasonal Changes in Thermoregulatory Activity . . . . . . . . . . . . . . . . . . . . .. 61
3.3 Change in Preferred Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4 Factors Affecting Body Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 62
4.1 Postural Adjustments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
4.2 Heat Shunting? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 63
4.3 Colour and Change of Colour. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 64
5 Ecological and Physiological Effects of Temperature Regulation. . . . . . . . . . . . . . .. 64
6 Prospect................................................ 65
1 Introduction
Most of the information on temperature regulation in spiders has been deduced from
their behaviour; this contrasts with other ectothermic animals, especially reptiles and
insects, on which many direct measurements of temperature have been made. This
chapter shows that spiders exhibit a broad spectrum of thermal behaviours and suggests
behaviours which might be expected with further study; less than 0.1% of spider species
have been examined in this context. The general background to thermoregulation has
been covered well in a number of recent reviews (Bligh et al. 1976; Gans and Pough
1982) as well as in works concerned specifically with invertebrates (Cloudsley-Thomp-
son 1970; Heinrich 1979,1981; Crawford 1981).
No indication of endothermy has been found in spiders, and is not to be expected
due to their low metabolic rate (J.F. Anderson 1970) and the lack of muscles capable
of high and sustained activity permitting endothermy, such as the flight muscles of
insects (Kammer 1981). In spiders we are limited to examining ectothermic temperature
regulation which is dependent on exogenous heat sources. Heat flux (conduction, con-
Department of Biogeography and Ecology, Western Australian Museum, Francis Street,

Perth, W.A. 6000, Australia
Behavioural Temperature Regulation 57
vection, radiation) can be influenced in various ways: behaviourally by basking in the

sun, moving to shade, tracking thermal gradients, postural change, thigmothermy,
avoiding boundary layers and stilting to aid convective cooling (A.J. Alexander and
Ewer 1958); physiologically by varying blood flow to radiating surfaces and by eva-
porative cooling; morphologically by colour, the nature of the integument, body size
and shape; shape constrains the extent of postural thermoregulation.
2 Evidence of Temperature Regulation in Spiders
2.1 Casual Field Observations
The first reference to behavioural thermoregulation in spiders seems to be the observa-

tions of Jean Henri Fabre in the late 19th century on a European wolf spider, Lycosa
narbonnensis (Fabre 1912). His observations are consistent with the behaviour of a
wide range of burrow-dwelling lycosids, behaviour which has been shown to have
thermoregulatory significance (Humphreys 1974, 1978a). Many species of spider sit in
the sun for part of the day, either on open ground (Edgar 1971b), at the burrow
entrance (Humphreys 1978a and unpublished), on ground level sheet webs (Riechert and
Tracy 1975) or on orb webs above the surface (Krakauer 1972; Robinson and Robinson
1974) and from these positions may shuttle between sun and shade (N</}rgaard 1951;
Riechert and Tracy 1975). Some web-building and vagrant spiders sun their egg sacs
(Eresidae: Millot and Bourgin 1942; Lycosidae: Vlijm et al. 1963; Vlijm and Kessler-
Geschiere 1967; Humphreys 1974, 1978a), and wolf spiders may change habitats as
adults to sun their egg sacs (Edgar 1971c). Wolf spiders frequently turn their egg sacs
while sunning them, especially when the eggs are young, possibly to prevent unilateral
overheating. .
2.2 Egg Sac Temperatures
Laboratory thermal gradient studies have shown that many spiders have distinct thermal
preferences (Lycosidae: W. Hackman 1957; Lane 1965; Almquist 1970; Hallander
1970; Agelenidae: Krafft 1967; Ctenizidae: Gray 1968; Pulz, this volume) and that
lycosids often select different temperatures according to whether or not they are carry-
ing egg sacs (Table 4)(N</}rgaard 1951; Lane 1965; Sevacherian and Lowrie 1972); this
has been confirmed by direct temperature recording from freely mobile lycosids and
their egg sacs in both the laboratory and the field (Humphreys 1974, 1978a). Compared
with non-breeding females, those carrying egg sacs may select higher (N</}rgaard 1951;
Sevacherian and Lowrie 1972) or lower temperatures (Humphreys 1978a) or show no
change in preferred temperature (N</}rgaard 1951; Vlijm et aI. 1963). With one excep-
tion (Pardosa sierra;Sevacherian and Lowrie 1972) the change in temperature preference
brought the egg sac temperature close to 30°C; it is possible that this temperature is
generally optimal for egg development in lycosids. In Trochosa the egg development
rate is related directly to temperature and ca. 30 ° C (range 29.5 °- 31.5 °C) is the maxi-
mum temperature tolerated by eggs, while adults survive to ca.40 °c (range 39°-42°C;
Engelhardt 1964). Similarly, the eggs of Theridion saxatile are more thermally sensitive
58 W.F. Humphreys
Table 4. Examples of temperatures at which spiders exhibit thermoregulatory responses
Family and species Temperature Method or behaviour Source

(OC)
LYCOSIDAE
Geolycosa godeffroyi
No egg sac 32-38 Continuous recording in Humphreys (1974,
With egg sac 28-32 laboratory and field 1978a)
Lycosa punctulata 29.3 (2.3) Continuous recording 6-day Humphreys (unpubl.)
24.1 (3.6) mean (range); next 6 days
Lycosa woodwardi 16 Lower and upper modes in Lane (1965)
29 thermal gradient
Pardosa ramulosa:
Species mean 26.3 Thermal gradient apparatus Sevacherian and
Female with egg sac 29.3 Lowrie (1972)
Female without egg sac 26.8
Male 25.0
Immature 24.1
Pardosa sierra:
Species mean 33.8 Thermal gradient apparatus Sevacherian and
Female with egg sac 35.2 Lowrie (1972)
Female without egg sac 34.1
Male 33.7
Immature 32.5
AGELENIDAE
Agelenopsis aperta 30.9 Moves from sun to shade; Riechert and Tracy
thermal balance model (1975)
ARANEIDAE
Cyrtophora citricola 28-31 Change in body posture Blanke (1972)
31-33
than the adult spiders (NI/>rgaard 1951). Exposure of egg sacs to the sun, however, can
serve to cool rather than heat the eggs; when the sun-exposed nest of T. saxatile heats
above 36°C the female moves the egg sac onto the web into full sunlight to prevent it
from overheating in the nest, returning the egg sac to the nest only when the nest has
cooled to 21 °c (NI/>rgaard 1956).
2.3 Burrows
Most work on the thermal relations of burrowing arachnids has been conducted in hot
climates (mostly deserts; Riechert and Tracy 1975; Crawford 1981), but more tem-
perate climates can produce rnicroenvironments that are thermally stressful to. spiders
(NI/>rgaard 1956; Humphreys 1978a). Solar heating of the ground surface during
the day and radiant cooling at night causes steep and temporally dynamic thermal
gradients in the upper layers of the soil, the temperature range declining with depth.
Spiders avoid thermal stress, both heat and cold, by withdrawing down the burrow
(Fig. 16; Seymour and Vinegar 1973) and the potential exists to track the most favour-
0000 1200 0000 1200 0000 1200

June 22 June 23 June 24
Fig. 16. Daily changes in the position of Geolycosa godeffroyi in its burrow in relationship to the
soil isotherms (solid curves). Arrows denote the direction of heat movement on either side of the
thermostatic (broken) lines. Screen temperatures : minimum -3.7 °e, maximum 15 .2 0c. (After
Humphreys 1978a)
able temperature as the isotherms move through the soil (Fig. 16; Humphreys 1978a),
as do some scorpions (Hadley 1970a). Burrows may be deeper in hotter locations (Gray
1968) or deepened during periods of extreme cold (Humphreys 1978a).
2.4 Continuous Monitoring of Body Temperature
Detailed records to temperature changes in spiders can be obtained from continuous

monitoring of temperature by means of implanted thermocouples or temperature
responsive radio transmitters (Humphreys 1974, 1978a). Plotting the temperature
of lycosids against air and substrate or 'model' spider temperatures typically gives
results like those shown in Fig. 17 (and Humphreys 1978a); the body temperature is
usually above the air temperature but, when the temperature is high, the spider tem-
perature is below the substrate temperature or the temperature of models of mass
similar to that of the spider.
3 Changes of Body Temperature
3.1 Daily Course of Body Temperature
Burrow-dwelling wolf spiders typically start basking as soon as the sun shines on the
burrow mouth and heat rapidly, the body temperature often overshooting (40°C,
Humphreys 1974) that which they maintain throughout the day (30°-35°C) when
conditions are favourable. If the weather is too hot, they change their orientation or
60 W.F. Humphreys
a 40
Fig. 17a,b. The prosomal tempera-
ture of Ly cosa tarentula, free-living
in their burrows plotted against
(a) air temperature and (b) the
~ temperature C C) of a 1-g clay
•
::I
OJ 30 'model' exposed to the sun 1 cm
Go
above the ground ; the spiders
E weighed between 0.8 and 1.6 g.
! Abbreviations : Tb temperature of
~20 the body, Ta air, Tm 'model'. The
o data were collected at Arnissa in
IZ!
northern Greece in 1978 using im-
planted thermocouples (Humph-
reys, unpubl. work)
15 20 25 30 35 40 45 50
A ir temperature
b TB : TM
40
: ..:.'...,i .:
.
~
!
! . ~i to it :
..
::I
..
~ 30 !~: I
Go ~
E
! ,
~20 J . '~
0
IZ! '."
20 25 30 35 40 45 50 55
Model temperature
seek shade , usually by withdrawing into the burrow (Humphreys 1978a, unpubl.). The
orb weaver Micrathena gracilis exhibits both short-term posturing and habitat-depen-
dent web orientation . In shaded areas the web is oriented to maximize , and in sunny
areas to minimize, the radiation incident on the body; the web orientation results in
significant elevation and depression respectively of body temperature in the middle of
the day (Biere and Uetz 1981). The inter-relationship between body temperature and
behaviour can be complex with some behaviours restricted to narrow temperature
ranges (Heath et al. 1971 ; Riechert and Tracy 1975 ; Casey 1981). In spiders these inter-
relationships have been studied seldom; but are nicely shown in Cyrtophora citrico/a
which exhibits courting behaviour only between ISO and 23 °C; there are two periods
of courtship during the day because the temperature is too hot in the middle of the
day and too cold at night to permit courtship (Blanke 1972). Although the duration
of copulation may be inversely related to temperature (Costa and Sotelo 1984), noth·
ing is known of the resulting reproductive success.
3.2 Seasonal Changes in Thermoregulatory Activity
Geolycosa godeffroyi avoided both lethally high summer and low winter temperatures
by withdrawing down its burrow. On favourable days throughout the year, however,
they basked in the sun at the mouth of the burrow. In winter they basked whenever
there was sufficient sunlight and this resulted in a single period of basking activity
around midday (Fig. 18). In summer the spiders basked in the morning and afternoon,
but withdrew down the burrow in the middle of the day to avoid excessive heat (Fig. 18),
as did Age/enopsis aperta on hot clear days (Riechert and Tracy 1975). The form and
magnitude of the temperature differentials shown in Fig. 18 are similar, both diurnally
and seasonally, to those from desert-inhabiting ectotherms (agamid lizards; Heatwole
1970).
3.3 Change in Preferred Temperature
Spiders may vary the temperature they select according to their breeding state (see
above), sex (Humphreys 1978a), stage of development (Sevacherian and Lowrie 1972)
and states of nutrition and hydration (Humphreys 1978a). In addition, there is evidence
suggestive of spiders having two temperature preferences under the same conditions;
bimodal frequency distributions may occur in a temperature gradient apparatus (Gray
1968) and body temperature may, repeatedly, change rapidly between two tempera-
tures (Humphreys 1978a). In addition, the two somata may be maintained at different
.-.-.,
Temperatures
12 SUMMER
.
Compared:
I
8 '~_o..:.::::
--_.------- _. ~""
4
0- _0".'1
,. . . . .
~ . . ( ..... . .
4''\
_/ '\ ""'' --''
.~..
"\
.' e--'" _ . Spider-ground
'-~---
Spider- air
._.-.-.-.-.".
; \
'. _.---.-.
......._1 '. _._--- \
, \'~ Spider- burrow
..... '
2 6 8 10 12 14 16 18 20 22 24
16 WINTER /\
___ -------------------'\ ,0///o~\\ ,,--..-._-- --....

.... --'..,'
... .' " #,1' ~ ..... .. Spider - ground
4
.~
' ._,_:\\0;'/... ;
,
./
a
l l ' 0-'-0,._ ._,_.--. Spid er - burrow
.. "... , . . ' ~ ,._-0---
'" --.....,._ .. Spider-air
o
• 6 8 10
Time
12
of
14
day
16 18 20 22 24
Fig. 18. Mean daily and seasonal changes in the temperature differences (Oe) between the prosoma
of Geo[ycosa godeffroyi and the air, ground surface and burrow temperatures. The data were col-
lected in the Australian Capital Territory. (After Humphreys 1978a)
62 W.F. Humphreys
temperatures and under different degrees of temperature control. Simultaneous record-

ings of the temperatures of the somata of live Lycosa tarentu/a with free access to a
radiant heat source show that the pro somal temperature is greater than that of the
opisthosoma [respectively x, (SD) N: 37.3 °c, (1.99), 101; 34.4 °c,(4.39),154] and
that the prosomal temperature is more carefully regulated (less variance) than that of
the opisthosoma (FSl.lS2 = 4.814 ,P = 0.029; Humphreys, unpubl.).
4 Factors Affecting Body Temperature
4.1 Postural Adjustments
Change in posture seems to be the main means used by web spiders to moderate their
body temperature. While a wide range of postural change is exhibited by different
species, depending mainly on their morphology, it serves to change the silhouette area
exposed to solar radiation and hence the equilibrium body temperature. Thin spiders
turn towards (Pointing 1965) or away from (Krakauer 1972; M.H. Robinson and B.C.
Robinson 1974, 1978), the sun and may track its apparent movement (Pointing 1965;
M.H. Robinson and B.C. Robinson 1974; Suter1981) irrespective of whether they are,
or not, on the sunlight side of the web (Fig. 19). Broad flat spiders, like Gasteracantha,
a S3
\
~
\
\
Fig. 19a-c. Thermoregulatory postures adopted by the

golden web spider, Nephila clavipes. The lines projecting
from three possible positions of the sun (S1 , S2 and S3)
indicate the corresponding orientations of the long axis of
the spider's body. Lateral views (a, c) and surface view (b)
of the web (W) showing the posture assumed in response to
a ventral (through the web) insolation; b lateral insolation ;
c dorsal insolation. (Redrawn from M.H. Robinson and
B.C. Robinson 1974)
expose their edge to the sun to achieve the same end (M.H. Robinson and B.C. Robinson
1978). Even very small spiders may reduce their equilibrium body temperature by
aligning their long axis with the sun's rays (by 0.5 °c in Frontinella communis weigh-
ing only 6 mg; Suter 1981).
Web orientation may depend on non-thermal constraints (Turnbull 1973) even for
spiders which show postural changes to sunlight (M.H. Robinson and B.C. Robinson
1974). The orientation of Argiope trifasciata webs, however, varies seasonally in rela-
tion to the midday angle of the sun and the radiant heat incident on the spider when
in its usual vertical resting position on the web. This results in thermal loadings which
are minimum in hot seasons because the spider is orientated with its long axis parallel
to the sun rays, and maximum in cool seasons as the venter is eXp6sed to the sun (the
venter is black which may aid heat absorption); the spiders postured only when the
web was not orientated appropriately (Tolbert 1979). Such 'appropriate' web orienta-
tion is seen also in Nephila clavipes during cold periods (Carrel 1978), although this
species postures to change its body temperature irrespective of web orientation (M.H.
Robinson and B.C. Robinson 1974, 1978) and it may exhibit evaporative cooling
(Krakauer 1972).
Wolf spiders change their orientation such that the maximum or minimum silhouet-
te area is exposed to incident solar radiation to moderate their body temperature; they
usually start basking with the opisthosoma exposed and later change to a prosoma ex-
posed posture, but may switch between the two postures at intervals (Humphreys
1978a and unpubl.). The effect of orientation on thermal balance in Lycosa tarentula
was generally similar to that found in Nephila clavipes (Carrel 1978); restrained live
spiders, rotated horizontally beneath a heat source, reached the highest prosomal
equilibrium temperatures when side-on to the heat source, had higher temperatures
when faced away from than when facing the heat source and higher temperatures were
reached when the heat source was directly overhead (Humphreys, unpubl.).
4.2 Heat Shunting?
Spiders may posture prosoma-to or prosoma-from the incident radiation to minimize

their heat loading; despite the equal silhouette areas of these postures are they the~mally
equivalent? Spiders pump blood from the opisthosomal heart, through the pedicel to
the front of the prosoma, from where it drains back through sinuses and the _pedicel
into the opisthosoma. The two somata are insulated from each other by their spatial
separation and probably by the position of the book lungs. This anatomy and the close
apposition of vessels in the pedicel offer the potential for counter-current heat ex-
change, as found in some insects (Heinrich 1976; Heinrich and Casey 1978). Depending
upon the posture, heat loading on the two somata will differ and haemolymph circula-
tion will probably provide the main means of heat transfer. The anatomy offers the
potential for drawing heat away from the hotter soma or pumping heat to the cooler
soma, hence the prosoma-to and prosoma-from the sun postures adopted by spiders
may not be equivalent despite their equal silhouette area.
Carrel (1978) thought that heat was dissipated by blood flow through the legs be-
cause dead spiders heated to higher temperatures than live ones. It is clear from his
data, however, that the disparity between dead and live spiders is greater at higher
64 W.F. Humphreys
temperatures; this warrents investigation as it suggests that blood flow is not only dis-
sipating heat, but that it dissipates relatively more heat at higher temperatures and is
thus in itself thermoregulatory (see also Pulz, this volume; Carrel, this volume).
Evidence suggestive of heat shunting comes from the rate of heat transfer between
the somata of live, but restrained, LycoSil tarentula when either the prosoma or opis-
thosoma was heated alone by a focussed microscope lamp, the other soma being
shielded by metal foil. Both somata heated at the same rate when exposed to the heat
source (prosoma 1.59 °cmin-I, opisthosoma 1.56 °cmin-I; Fs1 ,4 =0.049,P =0.836),
but the unheated soma heated at different rates (prosoma 1.15 °c min-I, opisthosoma
0.66 °c min-I; F s1 ,4 = 19.90,P= 0.011). Hence, heat is transferred forwards between
the somata more rapidly than it is transferred backwards (Humphreys, unpubl.); this is
in accord with anatomical considerations.
4.3 Colour and Change of Colour
Spiders may change colour between stadia, slowly within a stadium or suddenly (Holl,
this volume). The thermal significance of colour change is unknown, but it is suggestive
that Argiope argentata males loss their silvery-colourat the last moult, after which they
are no longer exposed to the sun (M.H. Robinson and B.C. Robinson 1978), while the
venter of Argiope trifasciata was orientated to the sun only after it changed from silver
to black at maturity (Tolbert 1979). Many spiders (especially araneids) have silvery or
light-coloured dorsal surfaces often in conjunction with black venters (M.H. Robinson
and B.C. Robinson 1978 (Fig. 73d,e). Even if the silhouette area is the same, heat may
be reflected by the light dorsum and absorbed by the dark venter and serve to decrease
and increase, respectively the heating rate and concomitantly the equilibrium tempera-
ture. Hence, the postures which prevent overheating (M.H. Robinson and B.C. Robinson
1974; Levi 1975b) or which increase radiant heating (M.H. Robinson and B.C. Robin-
son 1978) could be supplemented by cuticular properties which alter its reflectance of
radiant heat.
5 Ecological and Physiological Effects of Temperature Regulation
Geolycosa godeffroyi maintained its mean annual temperature 4.5 °c above the air
temperature by basking in the sun during the day and withdrawing into its burrow
during cold periods (Humphreys 1974, 1978a). The high mean temperature allowed
them to develop rapidly (Humphreys 1976), but this was associated with a respiratory
cost which was 152% of that which would have occurred in the absence of thermo-
regulatory behaviour (Humphreys 1977a, 1978b). Water loss was also influenced by
thermoregulatory behaviour and it may result, under natural conditions, in rates up
to 800% of those of non-thermoregulating spiders (Humphreys 1975a). The spiders
were also influenced physiologically by qualitative aspects of thermoregulation, name-
ly the resulting cycling of body temperature as opposed to temperature itself; changing
temperature caused an increase in the respiration rate (times 1.3 5) and the feeding rate
(Humphreys 1975b, 1977b).
Observations combined with sophisticated thermal balance models have shown that
desert-living funnel-web spiders, Age/enopsis aperta, selected web sites according to
their thermal characteristics and that these constrained the time available for handling
prey (Riechert and Tracy 1975). Web sites were actively selected in areas with the most
favourable thermal properties; those built in grassland depressions had favourable funnel
temperatures throughout the day, while those on the larva surface became too hot and
spiders occupied these sites mostly in spring. During the hot dry season the site selec-
tion was supplemented by the funnels being orientated such that they received the
least insolation. On hot days the spiders basked in the morning until they reached
about 31°C, after which they shuttled between sunlight and shaded areas to maintain
the body at temperatures suitable (21°-35°C) for prey capture and activities related
to it (searching for prey and laying silk on the sheet-web, sun basking and shuttling).
During the heat of the day, they left the funnel for the sheet-web only as long as
necessary to secure prey and repair the web; in very hot weather silk laying on the
sheet was confined to the evening. litter and habitat features providing shade (shrubs
and depressions) permitted the spiders more activity time on the web before they were
forced by overheating to withdraw into the funnel. The ability to capture prey at a
particular web site was constrained more by the thermal characteristics of the site than
by the abundance of prey at that site; different web sites encompassed an eight-fold
difference in the time thermally suitable for prey handling, but showed only two-fold
variation in prey abundance (Riechert and Tracy 1975).
6 Prospect
There is clearly ample scope for basic field observations and experiments to delineate
the range of families exhibiting behaviour consistent with thermoregulation. Further,
and more sophisticated, experimental studies are needed to determine the range and
complexity of behavioural and physiological temperature regulation in spiders. Examina-
tion of the ultimate effects on spiders of thermoregulation are compounded because
whole body attributes are the resultant of many rate processes which differ in their
thermal characteristics. We need to derme the relationships between temperature selec-
tion and those biological attributes (locomotion, physiology, growth, reproduction,
genetics and life history) more directly linked with ecology and ultimately with relative
fitness (Beittinger and Fitzpatrick 1979; Huey and Stevenson 1979).
Acknowledgements. Some of the work reported here was supported in part by a Royal Society
travel grant and the Australian Research Grants Scheme (No. 0181 15274 I).
V Behavioural and Physiological Adaptations to Cold
WALTER KIRCHNER 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 66
2 Ecological Aspects of Overwintering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 67
2.1 The Overwintering Habitats of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
2.2 Temperature Conditions in the Overwintering Habitats . . . . . . . . . . . . . . . . .. 67
3 The Cold Resistance of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 68
3.1 The Supercooling Point of Various Spider Species. . . . . . . . . . . . . . . . . . . .. 69
3.2 The Significance of Various Factors for the SCP . . . . . . . . . . . . . . . . . . . . .. 69
3.3 The Ability of Supercool and Winter Quarters . . . . . . . . . . . . . . . . . . . . . .. 74
4 Winter Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , 75
5 Aspects of Metabolism During Hibernation . . . . . . . . . . . . . . . . . . . . . . . . . . . , 76
6 Conclusions.............................................. 77
1 Introduction
Although spiders populate preferably the warmer climatic zones of the Earth, a number
of species have made their way into very cold habitats, such as the polar regions (Leech
1966; K. Thaler 1975) or upper regions of high mountains (Savory 1977; K. Thaler
1980). The air temperature in some of these habitats can fall as low as - 57°C (Leech
1966). The spiders of the lower reaches of the temperate climate zones do not have to
face such extremes; nonetheless winter air temperatures of -20°C and below can oc-
cur even there.
As is the case for other ectothermic animals, temperatures below 0 °c are problem-
atic for spiders for two reasons:
1. Sub-zero temperatures can cause the body fluids to freeze, entailing fatal damage to
the organism.
2. A reduction in body temperature necessarily slows the rate of metabolism, the
intensity of which is then insufficient to provide the energy needed for locomotion.
The resultant chill-coma prevents both feeding and the possibility of flight from
predators or other environmental dangers.
Lehrstuhl fUr Biologie (Angewandte Entomologie und Didaktik der Biologie), RWTH Aachen,
Sammelbau Biologiej42 C, Worringer Weg, D-5100 Aachen, FRG
Behavioural and Physiological Adaptations to Cold 67
These difficulties can be tackled by means of the following strategies:

1. Use of hibernacula reducing the effect of frost to an acceptable level.
2. Development of sufficient cold resistance.
3. Adaptation of the metabolism to winter conditions: either through a general reduc-
tion of body functions which, with the mobilization of reserves, enables the spider
to fast for some time, or by maintaining the metabolic rate at a level which allows con-
tinued movement and prey capture, even at temperatures several degrees below zero.
This chapter intends to describe the extent to which spiders have realized these
strategies.
2 Ecological Aspects of Overwintering

2.1 The Overwintering Habitats of Spiders
Our knowledge of the hibernacula of individual spider species is not complete, but they
are clearly very dependent on the respective biotope. More than 80% of spiders are to
be found in the winter in dense ground vegetation or in the ground itself (Schaefer
1977a). This is due to the fact that many spiders which live in the vegetation withdraw
to lower strata at the start of the cold season (Blake 1926). Nevertheless, a proportion
of spiders, 6.9% of the species in northern Germany (Schaefer 1977a), hibernate un-
protected in the vegetation.
Only part of the spider species spend the entire winter in the same place. Among
those that do are species which, like Araneus cornutus (Kirchner 1965), hibernate in
hollow plants, and some web builders which spend the winter in the herbaceous layer
(Duffey 1962). On the other hand, certain ground hibernators are more mobile (Aitchi-
son 1978); for instance, Pardosa /ugubris was observed, albeit under laboratory condi-
tions, to burrow deeper into the leaf-litter layer when the temperature was below 0 0 C
(Edgar and Loenen 1974).
Some spiders, particularly salticids, gnaphosids and clubionids, spend the winter in
a silken sac (von Kolosvary 1940; Jennings 1972; Duman 1979); others, .e.g. the Tro-
chosa species, line the hibernation chamber with silk (Engelhardt 1964). Spiders' eggs
are always protected by a cocoon during overwintering (cf. Schaefer 1976b,d); should
the young spiders hatch before winter, as is the case with Age/ena limbata (Kurihara
1979), they stay in the cocoon until the spring.
2.2 Temperature Conditions in the Overwintering Habitats
The spiders which are most exposed to the cold are those hibernating in the vegetation;
their body temperature can fall below -20 °C. Eggs are not adequately protected from
frost by the cocoon. Apart from affording protection against microbial attack, the co-
coon seems to be mainly a means of preventing desiccation and contact with ice
(Schaefer 1976b). Even winter quarters in hollow plant stems or under the bark offree-
standing trees are hardly warmer than the air, especially at night (Kirchner 1973;
Schaefer 1977a; Duman 1979). The bark of trees can,however, become relatively warm
through sun rays, which may explain the fact that 75% of hibernating philodromids
68 w. Kirchner
are to be found on the south side of the tree trunks (Duman 1979). This type of hiber-
nacula provides, first and foremost, protection against precipitation water (which is
liable to freeze) and predators.
Spiders which hibernate in the ground or in the vegetation near the ground are
usually not endangered by low temperatures when there is snow cover. A layer of snow
insulates well: the more air it contains, the better (new snow contains up to 97% air;
Niemann 1957). The temperature under a layer of snow more than 5 cm thick will only
fall slightly below zero even on cold days (Buche 1966). The insulating effect of snow
has been confirmed by a number of investigations (Holmquist 1931; MacPhee 1964;
Huhta 1965; Kirchner 1973). However, an older layer of snow becomes more and more
compact and thus increasingly conductive to cold (Niemann 1957); under extreme
climatic conditions, such as in central Alaska, temperatures of -11 ° to -18°C can oc-
cur even under the snow (Block 1979). Conditions can also become critical for ground
hibernators if there are strong frosts before the first snows or after the snow has melted
(Holmquist 1931; Huhta 1965; Block 1979). Finally, there are many regions where
despite very low winter temperatures, there is no permanent snow cover. Under these
circumstances the ground can freeze to a considerable depth. The ground litter contains
many insulating interstices, so that the variation in temperature in it is less pronounced
than that in the air above the ground (Edgar and Loenen 1974; Schaefer 1977a); non-
etheless the temperature in the upper soil strata can fall below - 5 ° C (Holmquist 1931)
or even to -15°C (Huhta 1965). In order to avoid the ground frost, spiders would have
to burrow deep into the earth; it seems doubtful that all species are capable of this
task. Coelotes terrestris, for instance, hibernates only 5 cm (TretzeI1961) and Trochosa
a mere 1-3 cm below the surface (Engelhardt 1964).
Stones on the ground provide only partial protection for spiders hibernating under
them. They are warmed relatively quickly by the sun, but on cold, sunless days when
there is no snow cover, the temperature on the underside can be lower than that at a
depth of 8 cm under the earth (Crawford and Riddle 1974). The microclimatic condi-
tions are better in a scree which has a thin covering of earth and vegetation: the tem-
perature there will not fall below -1°C (Kirchner and Kullmann 1975). Caves and
houses provide particularly good winter refuges as the temperature in them usually
stays well above zero. It can be seen from the above that with the exception of species
which hibernate in houses or caves, spiders of the temperate and cold climatic zones
must be able to withstand frost.
3 The Cold Resistance of Spiders
Investigations, particularly on insects (Danks 1978), have shown that an organism can
be damaged by cold even before the formation of ice crystals in the body fluids ("cold
injury"). More frequently, however, life-endangering damage occurs first when ice
formation sets in ("freezing injury").
Cold injury ist most likely to occur in spiders used to a warm climate: Kirchner
(1973) observed that some individuals of the social spider species Agelena consociata
from the tropics survived only a few hours at 0 DC, although the body fluids do not
freeze above -8°C. Almost all the other spiders which were investigated, mainly Middle
European species, survived to a temperature below 0 DC: that at which ice formation
in the body sets in. To date, no spider species has been found which can survive freezing
of the body, as has been observed for certain "freeZing-tolerant" insects (Miller 1969,
1978; Baust and Edwards 1979; Ring and Tesar 1980). Spiders are thus "freezing-
susceptible", sharing this feature with other arachnids, namely the mites (MacPhee
1961; SI/>mme and Conradi-Larsen 1977; SI/>mme 1978,1981; Block 1979; Schatz and
SI/>mme 1981) and the scorpions (Riddle and Pugach 1976), in fact, with most ofthe
insects (SI/>mme 1982).
Since freezing is fatal to spiders, the only course of defence open to them lies in
postponing the formation of ice for as long as possible. They achieve this by a varying
ability to supercool; i.e. the freezing of the body fluids occurs at a point below the
temperature to be exspected from the concentration of dissolved substances. In ex-
treme cases this "supercooling point" (SCP) may be as much as 20°C lower than the
normal freezing point. The SCP can be determined on an intact animal by means of a
thermoelement (for details of the method, see Kirchner and Kestler 1969). The SCP of
an intact animal is not identical with the freezing point of an isolated haemolymph
sample; the latter is much higher and lies significantly closer to the melting point of
the haemolymph (Husby and Zachariassen 1980). The difference between the freeZing
and melting points of the haemolymph is called "thermal hysteresis": the greater the
difference, the lower the SCP of the complete animal (Duman 1979). Since the deter-
mination of the SCP of terrestrial arthropods is so quick and simple, SCP has been
adopted by many authors (for summary see SI/>mme 1982) as a measure of cold re-
sistance. Little has been done towards establishing other methods, although the reliabil-
ity of the SCP as an indicator of cold resistance is subject to certain restrictions (see
below). Hence, the main theme of the following is the ability of spiders to supercool.
3.1 The Supercooling Point of Various Spider Species
To date, the winter SCPs of approximately 60 spider species from 14 families have been
determined (Table 5); they range from -4°C to -34°C. About one-fifth of the species
listed in Table 5 can be described as having a low cold resistance (SCP between - 4 ° C
and -8°C), two-fifths as having a medium resistance (SCP between _8° and -16°C)
and the remaining two-fifths as being highly resistant (SCP below -16°C). In contrast
to the mites, whose SCPs lie not infrequently in the range - 20° to - 30°C or even lower
(particularly in the Arctic, Antarctic and at high altitudes), the majority of spiders has
little resistance to cold. SCPs determined for scorpions lie mainly above -10°C (for
overview see SI/>mme 1982). As can be seen in Table 5, SCPs can differ widely within a
family, and indeed within a genus. Even within a species the variation can be large;
some individuals of Teutana castanea, for instance, freeze at -6°C, others at -15°C
(Fig. 20; Kirchner 1973). The spread of the SCP is greatest among species with a high
cold resistance.
3.2 The Significance of Various Factors for the SCP
Sex/Stage of Development. Sex appears to have no particular influence on the SCPo

The developmental stage, on the other hand, is clearly of importance: all spiders' eggs
70 W. Kirchner
Table S. Mean supercooling points of different species of spiders in winter
Species Overwintering Over- Decreased SCpb Reference

habitat wintering SCP during (OC)
stage a winter?
LINYPHIIDAE
Linyphia triangularis Leaf litter Egg -24.3 Schaefer (1976a)
Linyphia hortensis Leaf litter sub -16.4 Schaefer (1976a)
Linyphia peltata Upper vegetation sub -17.4 Schaefer (1976a)
Floronia bucculenta Upper vegetation Egg -30.5 Schaefer (1976a)
Bolyphantes index Snow crevices ad -15.3 Higvar (1973)
Bathyphantes approximatus Leaf litter ad -10.7 Schaefer (1976a)
Bathyphantes nigrinus Grass tussocks ad - 8.2 Schaefer (1976a)
Bathyphantes gracilis ? ad - 9.9 Schaefer (1976a)
Lepthyphantes cristatus Leaf litter ad -10.8 Schaefer (1976a)
Centromerus expertus Leaf litter ad - 7.1 Schaefer (1976a)
Centromerus silvaticus ? ad No - 5.2 Schaefer (1976a)
Centromerita bicolor ? ad No - 5.7 Schaefer (1976a)
Macrargus rufus Leaf litter ad - 7.0 Schaefer (1976a)
Microneta viaria Leaf litter ad -11.3 Schaefer (1976a)
Allomengea scopigera ? Egg -28.3 Schaefer (1976a)
Drapetisca sociJZlis Upper vegetation Egg -28.3 Schaefer (1976a)
Stemonyphantes lineatus ? ad Yes -15 Schaefer (1976a)
Erigonidium graminicola Hollow plant stems ad, sub -20.1 Schaefer (1976a)
Gongylidium rufipes Hollow plant stems sub -16.0 Schaefer (1976a)
Hypomma bituberculatum Hollow plant stems sub -11.0 Schaefer (1976a)
Oedothorax retusus ? juv -11.6 Schaefer (1976a)
THERIDIIDAE
Theridion pal/ens Upper vegetation sub -21.8 Schaefer (1976a)
Theridion impressum Upper vegetation juv -21 to Kirchner and
-24 Kullmann (1975)
Theridion sisyphium Upper vegetation juv -21 to Kirchner and
-24 Kullmann (1975)
Theridion denticulatum Inside houses juv -11.4 Kirchner (1973)
Theridion tepidariorum Inside houses ad, sub - 8.2 Kirchner (1973)
Teutana castanea Inside houses - 9.5 Kirchner (1973)
Teutana triJZnguiosa Inside houses ad -10.9 Kirchner (1973)
NESTICIDAE
Nesticus cellulanus Between rocks ad, sub No - 4.7 Kirchner and
Kullmann (1972)
ARANEIDAE
Araniella cucurbitina Upper vegetation juv,sub -20.2 Schaefer (1976a)
Araneus cornutus Hollow plant stems ad, sub Yes -22.8 Kirchner (1973)
Cyclosa conica Upper vegetation -21.4 Schaefer (1976a)
Singa nitidula Hollow plant stems sub -21.4 Kirchner (1973)
TETRAGNATHIDAE
Tetragnatha montana Upper vegetation sub Yes -18.4 Schaefer (1976a)
Meta segmentata Leaf litter Egg -24.0 Schaefer (1976a)
Meta menardi In caves ad - 4.0 Kirchner (1973)
Pachygnatha clercki Leaf litter ad No - 8.4 Schaefer (1976a)
Hollow plant stems ad - 5.8 Kirchner (1973)
Pachygnatha listeri Grass tusso cks ad -10.1 Kirchner (1973)
Table 5 (continued)
Species Overwintering Over- Decreased SCpb Reference

habitat wintering SCP during (OC)
stage a winter?
PHOLCIDAE
Pholcus phalangioides Inside houses ad No - 4.7 Schaefer (1976a)
AGELENIDAE
Cicurina cicurea Underneath rocks ad - 6.7 Kirchner (1973)
Coelotes terrestris Underneath rocks ad - 6.2 Kirchner (1973)
Coelotes spec. Underneath rocks sub No - 6.0 Kirchner (1973)
Histopona torpida Underneath rocks No - 6.5 Kirchner (1973)
SALTICIDAE
Euophrys frontalis Grass tussocks sub - 9.8 Schaefer (1976a)
Marpisl1ll muscosa Under bark ad -17.7 Schaefer (1976a)
LYCOSIDAE
Pardol1ll lugubris Underneath rocks sub No - 6.8 Kirchner (1973)
Pardol1ll pullata Grass tussocks juv, sub -13.1 Schaefer (1976a)
Arctol1ll perita In the soil ad No -11 Schaefer (1976a)
Pirata piraticus ? juv Yes -14.5 Schaefer (1976a)
PHILODROMIDAE
Philodromus spec. Upper vegetation sub -21.5 Kirchner (1973)
Philodromus spec. Under bark sub Yes -19.9 to Duman (1979)
-26.2
Philodromus aureolus Upper vegetation sub -24.7 Schaefer (1976a)
Thanatus strilltus ? juv Yes -24.6 Schaefer (1976a)
CLUBIONIDAE
Clubiona phragmitis Hollow plant stems ad Yes -20 Schaefer (1976a)
-16.1 Kirchner (1973)
Clubiona spec. Under bark sub -12.9 Kirchner (1973)
Clubiona spec. Under bark sub -15.9 Kirchner (1973)
Clubiona spec. Under bark sub -15.4 Duman (1979)
SEGESTRIIDAE
Segestrill senoculata Under bark ad -14.7 Schaefer (l976a)
ERESIDAE
Eresus niger In the soil ad -16.6 Kirchner (1973)
a ad = Adult spiders; sub = subadult spiders (big young specimens); juv = juvenile spiders (small
young specimens).
b SCP = supercooling point.
All spiders were studied in FRG with the following exceptions: Norway: Hftgvar (1973); USA:
Duman (1979).
investigated to date have proved to be highly resistant, a SCP below - 24°C is the rule,
whereas the values for juveniles and adults are rarely this low (Schaefer 1976a, 1977a).
The eggs of mites and of many insects also show a pronounced capacity for supercool-
ing (StPmme 1982); for instance, the eggs ofMitopus morio can withstand a temperature
of -20°C for at least 8 weeks (Tischler 1967). Within a species there mayor may not
72 W. Kirchner
n Fig. 20. Variability of supercooling point

Mela menard; Latr. in three species of spiders. The number of
individuals is given by the number of
squares (Kirchner 1973)
5
- 10 c
~n
o
E Pardasa sa/luaria L. Koch
'E
o
'0
~ 5
.0
E
::J
Z n
.., n
-5 -10 c
n Teu/ana caslanea Cl.
°0
~~~~~-_5~~~~~~~~~~~~---C
Supercooling point
be a significant difference between the cold resistance of adults and of juveniles: for
Theridion impressum and T. sisyphium the difference is marked (Kirchner and Kull-
mann 1975), but for other species the variation in the SCPs of the different develop-
mental stages is small (Schaefer (1976a).
As far as cold resistance is concerned, the egg stage would be optimal for survival in
frosty periods. Yet the proportion of spiders which hibernate in the egg stage is small;
Schaefer (1977a) estimates it at 7% of the North German species . Apossible explana-
tion would be that later developmental stages are better equipped than freshly hatched
spiders to deal with the circumstances (e.g. the food situation) which they face in the
spring.
Season. The question as to whether the SCP varies over the year has only been in-
vestigated with a relatively small number of spider species (Kirchner and Kestler 1969;
Kirchner and Kullmann 1972,1975; Kirchner 1973; Schaefer 1976a, 1977a; Duman
1979). At present, we can differentiate between two groups:
I. Species whose SCP remains more or less constant throughout the year; such species
have mainly little or medium resistance to cold (Table 5).
2. Species whose SCP becomes lower at the beginning or in the course of the cold
period. This can occur in two different ways (Salt 1961):
a) At the onset of winter one developmental stage may be followed by another with a
higher degree of cold resistance: "cold -hardiness" .
b) The influence of winter conditions brings about an increased ability to supercool
within the same developmental stage: "cold-hardening". This has been observed,
for instance, in subadult individuals of Thanatus striatus (Schaefer 1977a), in im-
mature Philodromus (Duman 1979) and in subadult and adult Araneus cornutus
(Kirchner and Kestler 1969). These species have a high resistance to cold. Although
it is too early for definitive conclusions, the indications are that a high resistance to
frost is developed only when needed.
Cryoprotectants. Particularly the phenomenon of cold-hardening suggests that the in-

creased ability to supercool is based on a qualitative and/or quantitative change in the
constitution of the haemolymph. That such changes take place can be shown by a
comparison of the melting point of the haemolymph in summer and winter animals:
in Araneus comutus the melting poin t of the haemolymph in summer is - 0.65 ± 0.07 ° C
(SCP of the intact animal: -8.3 ± 0.7 DC); in winter itis -1.76 ± 0.1 °c (SCP: -22.8 ±
1.9 0c) (Kirchner and Kestler 1969; for further examples, see Schaefer 1976a).
The fact that the melting point of spiders' haemolymph does not fall below about
- 2 OCto - 3 ° C in winter shows that their ability to increase their resistance to cold is
not merely due to an increase in the osmotic concentration of the haemolymph. The
secret lies rather in the accumulation of substances which in one way or another post-
pone the formation of ice. The question as to which substances may be responsible for
this in arthropods has been the subject of many investigations, particularly on insects
(cf. Salt 1961; Danks 1978; StPmme 1982), but also on mites (S</>mme 1965; S</>mme
and Conradi-Larsen 1977; Block et al. 1978; S</>mme 1979; Young and Block 1980;
Young 1980; Block 1980; S</>mme 1981; Schatz and S</>mme 1981 ; Block and S</>mme
1982). The following have been discussed as possible cryoprotective substances: gly-
cerol and other polyols, sugars, unsaturated fatty acids, inorganic phosphates, free
amino acids, proteins and glycoproteins. Of these substances glycerol has certainly re-
ceived the most notice. Whereas it has repeatedly been identified in mites (S</>mme
1982), Riddle and Pugach (1976) failed to find it in the haemolymph of the hibernating
scorpion, Paruroctonus aquilonalis. Few investigations have been conducted on spiders.
The young of Argiopeaurantia, which hibernate in cocoons, do not accumulate polyols
during that period (Riddle 1981). Glycerol was found in the immature Philodromus
spec. and Clubiona spec. during the winter (not in sU'llIller) together with a protein
which has the property of lowering the freezing point of the haemolymph in winter
(Duman 1979).
The araneid Araneus comutus accumulates considerable quantities of glycerol in
winter (2- 3% of the fresh weight), which is accompanied by a drop in the SCP of near-
ly 15°C. However, a direct connection between glycerol accumulation and increased
ability to supercool has not been established: if a hibernating spider is activated through
warmth, the accumulated glycerol is reduced more quickly than the resistance to cold
(Kirchner and Kestler 1969). Further investigations in this field are to be welcomed.
Gut Contents and Water Content. In contrast to a number of insect and mite species
(for overview, see S</>mme 1982), gut contents of spiders are not likely to influence the
74 W. Kirchner
SCP; for with spiders those particles that might act as nucleators in the formation of
ice are filtered out in the feeding process. As spiders do not appear to reduce their
water content in winter, this factor can be discounted as well.
Humidity Conditions in the Spider's Environment. Many spiders hibernate in damp

places (Schaefer 1977a), so that their exoskeleton can easily come into contact with
water (condensation, precipitation or melt water). If this water freezes, it may become
the starting point of an ice front which penetrates the animal's cuticula and initiates
the freezing of the body fluids ("inoculative freezing", cf. Salt 1963). It is not clear
whether this is the case with spiders.
Duration of the Cold Period. Salt (1950) is right in pointing out that the SCP is an un-
reliable measure of cold resistance. The SCP is defined as that temperature at which
the spontaneous formation of ice in an animal sets in when it is cooled at a controlled
rate, usually 1°C/min. The SCP which is determined in this way certainly does not
correspond to the temperature at which an animal would freeze in the field. There, the
formation of ice may set in at a temperature several degrees above the experimentally
determined SCP, if it is maintained for a long enough period. The length of time re-
quired at a particular temperature before ice begins to form in an animal varies from
individual to individual: with some it may take minutes or hours, with others days or
weeks. This applies also to spiders, as is shown in the case of A ran eus cornutus: although
this species has an average SCP of - 23 0 C, up to 30% of winter spiders do not survive a
2.5-day period at -18 °c (Kirchner and Kestler 1969). For further examples, see
Schaefer (1976a). In evaluating the SCP it should, therefore, always be taken into ac-
count that the temperature to be bome by the majority of individuals under natural
conditions over an extended period lies several degrees above the average SCPo
3.3 The Ability to Supercool and Winter Q~rters
The various types of hibernacula of spiders let in frost to different degrees; on the
other hand, the ability of spiders to resist cold varies from species to species (Table 5).
Is there a relationship between the degree of frost in a particular winter quarter and
the resistance to cold of the respective species? This question was the subject of in-
vestigations by Kirchner (1973) and Schaefer (1976a, 1977a).
Table 6 shows that spiders hibernating in places relatively protected from frost have
significantly less resistance to cold than those subjected to colder temperatures. How-
Table 6. Relationship between hibernacula and resistance to cold among spiders
Hibernacula Average supercooling point
Unprotected in vegetation Less than -10 °C, frequently around -20°C

In plant stems or under the bark of trees Less than -10 °C, sometimes less than -20°C
In or near the ground (low-lying vegetation, Normally between -5°C and -12°C, occasionally
litter layer, humus, under stones) lower
In houses or caves Between _4° and -12 °c
ever, even in temperate regions a hard winter can produce air and soil temperatures be-
low the average SCPs of various spider species. In such cases a certain number of spiders
will die from the cold, the more so as many individuals succumb to temperatures several
degrees above the SCP if the cold persists. The mortality rate due to frost can be de-
termined relatively easily in the laboratory (Engelhardt 1964; Kirchner 1965; Kirchner
and Kestler 1969; Almquist 1970; Schaefer 1976a). In the field, however, it is rarely
possible to establish more than the overall mortality rate. In woodland areas of Finland
it was observed that many spiders died when autumn frosts penetrated the ground: on
the other hand, if there was a permanent snow cover, the population remained constant
for a long period, or even throughout the winter (Huhta 1965). The mortality rate
among individuals of the winter-active species, Centromerus silvaticus, which have a
low resistance to cold (SCP :: -5.2 °C), is often high and can be correlated with the
hardness of the winter (Schaefer 1977b).
The question as to why some spider species with a low cold resistance nonetheless
survive very cold winters in sufficient numbers must remain a matter of conjecture at
present. Part of the answer is that every species produces some especially hardy in-
dividuals (Fig. 20). It may also be that some individuals find particularly well-sheltered
places to hibernate. Another possibility to be considered is a progressive lowering of
the SCP at times of protracted cold, as was established by Young and Block (1980) in
the mite, Alaskozetes antarcticus. However, as such acclimation is by no means the rule
among arthropods (Scf>mme 1979), further investigations on spiders would be desirable.
In addition, the effects of varying, unstable low temperatures on spiders should be ex-
plored, these being probably different from the effects of the constant sub-zero tem-
peratures normal in laboratories. Finally, some spiders have behavioural mechanisms
which cause their body temperature to remain significantly higher than that of their
surroundings (Humphreys, this volume).
4 Winter Activity
For their ability to move around, spiders, being ectothermic animals, are highly de-
pendent on the temperature of their surroundings. Increasing cold causes at some point
a chill-coma: for Trochosa this sets in as early as +2.5 °c (Engelhardt 1964), for many
other species between _4° and -8°C (Buche 1966; Aitchison 1978), for Bolyphantes
index (Fig. 94) not until -9.3 °c is reached (H~gvar 1973). The range -4°C to - 8°C,
in which the chill-coma sets in, corresponds well with that determined for mites from
the Tyrolean Alps (Schatz and Scf>mme 1981) and from the Antarctic (Block and
Scf>mme 1982).
Since many spiders only lose their ability to move at temperatures several degrees
below zero they can remain active at sub-zero temperatures. Thus, in Canada, among
other species, clubionids, linyphiids, lycosids and thomisids were caught in ground
traps between the snow cover and the ground surface (Aitchison 1978). In northern
Finland representatives of 102 spider species were observed running about on the snow,
though only on days when the temperature was not colder than -4°C (Huhta and
Viramo 1979). Reports on the winter activity of spiders have appeared in a large num-
ber of publications (von Kolosvary 1939; Tretzel 1954; Polenec 1962; Buche 1966;
76 w. Kirchner
H~gvar 1973; K. Thaler and Steiner 1975; Schaefer 1977a; Aitchison 1984a,b). Winter-
active spiders include not only vagabond species, but also web builders; Bolyphantes
index, for instance, builds its net in crevices on the surface of the snow cover (Fig. 94).
There may be several reasons for winter activity among spiders; of these the most im-
portant are probably the search for a micro climatically more suitable habitat, the search
for food and the search for a partner (Aitchison, this volume).
Of particular interest is the fact that some winter-active spiders reach maturity in
winter (Schaefer, this volume); in the area of Erlangen (southern Germany) such spe-
cies account for 2-3% of the total (Buche 1966). The advantage of winter maturity
probably lies in the reduced competition for food in the cold season. Winter maturity
is by no means coupled with a high resistance to cold (Buche 1966); as winter7maturing
spiders usually inhabit the litter layer, there is no need for this. On the other hand, the
thermopreferendum is strongly correlated to winter activity; among the winter-active
species investigated to date it normally lies between 0 0 C and +10 0 C, occasionally
slightly below 0 °c (Wolska 1957; Buche 1966;H~gvar 1973).
5 Aspects of Metabolism During Hibernation
During cold weather the majority of spiders is prevented from taking nourishment, the
exceptions being the winter-active species, about 9% of the total in northern Germany.
The resultant period of starvation can be overcome in two ways;
1. Through storage of reserves before the cold season. Thus, Collatz (1973) observed
in Tegenaria atrica a marked increase in lipids, proteins and carbohydrates during
the autumn. Fats appear to be the main source of energy in winter, as the lipid con-
tent falls rapidly towards the spring (Collatz 1973). Furthermore, the respiratory
quotient for hibernating Araneus cornutus lies between 0.6 and 0.7 (Kirchner 1965),
which is an indication of fat consumption.
2. Through reduction of the respiration beyond the point stipulated by the tempera-
ture; this was observed in Pisaura mirabilis (Dondale and Legendre 1971) and Ara-
neus cornutus (Kirchner 1973).
Many spiders enter a period of diapause in late autumn or early winter (Schaefer
1977a), during which the hunger tolerance is especially great (K. Miyashita 1969a).
The diapause is also important for another reason; it inhibits development and thus
ensures that certain developmental stages are not entered at unsuitable times of year.
Thus, in some species the young spiders are prevented by a winter diapause from hatch-
ing out of the eggs (Schaefer 1976b) and from leaving the protective cocoon (Kurihara
1979).
For the winter diapause to fulfill its function, it must begin at a suitable time. In
this respect the decrease in the length of the days appears to be of considerable im-
portance (Dondale and Legendre 1971; Schaefer 1976c; Kurihara 1979). The diapause
comes to an end already during the winter, sometimes as early as January; in spring
the spiders' activity is brought about mainly by rising temperatures. The situation is
different with a number of spider species which hibernate in the egg stage (e.g. Drape-
tisca socialis, Floronia bucculenta, Linyphia triangularis, Allomengea scopigera, Meta
segmentata); in this case the embryogenesis of eggs laid in the autumn is prevented by
higher temperatures: the egg diapause can only be interrupted by the effect of several
weeks oflower temperatures (Schaefer 1976b, 1977a, this volume).
6 Conclusions
In conclusions it should be pointed out that research on the ability of spiders to cope
with cold is still very much in its infancy. Results so far relate almost excluSively to
the overwintering habits of spider species from the temperate zones. The cold resistance
of tropical spiders and of spiders from polar regions, which sometimes face very low
temperatures even in summer, has not been as yet thoroughly investigated.
VI Respiratory System: Structure and Function
FRAUKE STRAZNy!,2 and STEVEN F, PERRY!
CONTENTS.
Comparative Anatomy, , , , , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 78
1.1 Historical Background. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 78
1.2 The Respiratory System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 79
2 Structure and Function of the Respiratory Apparatus. . . . . . . . . . . . . . . . . . . . .. 85
2.1 Structure of the Respiratory System. . . . . . . . . . . . . . . . . . . . . . . . . . . .. 85
2.2 Physiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
2.3 Functional Morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 91
3 Reconsideration of the Problem of the Dual Respiratory System . . . . . . . . . . . . . . . 93
1 Comparative Anatomy
1.1 Historical Background
A characteristic of spiders and other arachnids are the book lungs, which other arthro-
pods lack. In addition to (or in place of) book lungs, tube tracheae may be present.
The extent to which the latter are developed varies according to species (Table 7).
Since the book lungs of arachnids and the gills of the horseshoe crab, Limulus poly-
phemus (Merostomata), occupy homologous sites and are structurally superficially
similar, many attempts have been made to explain the possible transition from gills to
lungs. These "theories" fall into three groups:
1. The gill-lung theory, which maintains that book lungs are directly derived from gills
(Ray-Lankester 1905; McLeod 1884; Kingsley 1885; Purcell 1909).
2. The tracheal theory, in which book lungs derive from tracheae (Jaworowski 1894;
Demoll and Versluys 1922).
3. The theory that the respiratory apparatus of horseshoe crabs and arachnids developed
independently from an indifferent stage in a very primitive, common ancestor of
both groups (Heymons 1905; Reuter 1909; Dahl 1921; all cited in Versluys and
Demoll 1922).
1 Arbeitsgruppe Zoomorphologie, FB 7 Biologie der Universitat, Ammerliinder HeerstraBe 67 -99,

D-2900 Oldenburg, FRG
2 Present address: Universitat Osnabriick, Abt. Vechta, Fachbereich 13, DriverstraBe 22,
D-2848 Vechta, FRG
Respiratory System: Structure and Function 79
Table 7. Respiratory organs in the second and third opisthosomal segments of spiders. (After Kaest-
ner 1929, 1969)
Second opisthosomal segment Third opisthosomal segment Example
Book lungs Book lungs Mesotheiae, Orthognatha

Book lungs Short tube tracheae Filistata
Book lungs Long tube tracheae Dysdera
Book lungs Long entapophyseal tracheae Argyroneta
Book lungs Short tube tracheae and Araneidae, Lycosidae
short entapophyseal tracheae
Book lungs Short tube tracheae and Cryphoeca
long entapophyseal tracheae
Book lungs No respiratory organs Pholcidae
Sieve tracheae Long tube tracheae Caponiidae
Sieve tracheae No respiratory organs Symphytognathidae
Tube tracheae Long tube tracheae Telema
We consider the Merostomata to include the ancestors of the arachnids and thus
favor the gill-lung theory. As will become apparent in this chapter, book lungs resemble
inverted gills and their phylogenetic origin from the latter is probable, although the
exact mechanism of this transition is not known. We consider book lungs to be the
primary respiratory organ of arachnids and tube tracheae to be a secondary develop-
ment.
1.2 The Respiratory System
Many spiders possess two parallel sets of respiratory organs: book lungs and tracheae.
Since the functional morphology of these two systems is quite different, it is expedient
to consider first the structure of the primitive arachnid respiratory system and its rela-
tionship to the circulatory system and then to consider modifications of this primitive
pattern.
1.2.1 The Primitive Respiratory and Circuliltory System
The most primitive spiders, i.e. Mesothelae and Mygalomorphae, possess two pairs of
book lungs. These lie in the second and third opisthosomal segments, although in
adults these segments are recognizable only in the Mesothelae. The lungs lie ventro-
lateral and are externally characterized by the lightly curved and often hair-free chi-
tinous caps (Lungendeckel) which cover them. Each lung opens through its own slit-
like stigma ventrolaterally to the outside. Between the first stigmas lies the sex open-
ing. To the inside the stigma opens to a lung atrium from which the platelike lamellae
of the book lung extend cranially into the haemolymph sinus.
Haemolymph is drawn during cardiac systole between the book lung lamellae, where
it becomes oxygenated. It then flows through "pulmonary veins", which are actually
grooves in the digestive gland, to the pericardial sinus and enters the heart through the
80 F. Strazny and S.F. Perry
Fig. 21. Semischematic cross-section through the opis-

thosoma at the level of the book lungs. Arrows indicate
haemolymph flow . Abbreviations: D gut; H heart;
Ls lung sinus; Lv pulmonary vein; Md digestive gland ;
M longitudinal muscle; N opisthosomal nerves ; Pc peri-
cardial sinus. (From Foelix 1979)
first pair of ostia (Fig. 21). During a later systole the oxygenated haemolymph is forced
through arteries into the prosoma and the opisthosoma. From the prosoma, where it
loses its oxygen primarily to the nervous system and the leg muscles, it flows through
two lateral lacunae back through the petiole to the pulmonary sinus of the first book
lung pair.
The haemolymph from the opisthosoma probably becomes oxygenated by the second
book lung pair, then flows laterally over the respiratory lamellae and to the pericardial
sinus via the "pulmonary veins" . Since the pulmonary sinuses of both the anterior and
posterior book lungs are connected, a mixing of prosomal and opisthosomal haemo-
lymph is highly probable (Kaestner 1929).
Scorpions possess four pairs of book lungs and the distinctly metameric heart has
seven lung veins, which correspond to the segments of the preabdomen. The two most
cranial pairs and the most caudal "pulmonary veins" lack book lungs of their own, but
receive blood from the first and the fourth existing lung pairs, respectively . Kaestner
(1929) interpretes this anatomical configuration as evidence that lung pairs were origi-
nally present in all segments.
1.2.2 Variations on a Theme
Deviations from the primitive spider pattern affect mainly the second pair of book
lungs . These may be replaced by tube tracheae, which become so extensive in some
species that they extend through the petiole into the prosoma. These tube tracheae
are not necessarily homologous structures: some are derived from the second book
lung pair, while others are modified, hollow muscle insertion sites (entapophyses).
Modifications of the first lung pair are generally such that the lung atrium does not
give rise to respiratory lamellae, but rather to numerous, tubular tracheae (sieve tra-
cheae). Telema tene/la is the exception which proves the rule: its extensive tube tracheae
replace both the cranial and the caudal lung pairs (Table 7). In contrast to insects, in
which the tracheae enter cell groups directly, spiders always rely on haemolymph as an
oxygen-carrying medium. The tracheae do not make direct contact with cells, but rather
are always surrounded by a layer of haemolymph (Foelix 1979). In cases in which the
tracheal system is particularly well developed, the circulatory system is reduced ac-
cordingly. The heart is short and possesses fewer ostia and arteries than in species with
well-developed lungs (Crome 1952,1953).
1.2.3 Haemolymph
In spiders the haemolymph, as a gas transporter, is an integral part of the respiratory

system. But it also serves other life-sustaining functions, such as hydrostatic extension
of the extremities, transportation of metabolites and metabolic waste products and
haemostasis. In addition to various oxygen-carrying pigments (see Sect. 1.2.4), haemo-
lymph contains formed elements (haemocytes) and a circulating fluid, which is capable
of haemostasis.
The circulating fluid of invertebrate haemolymph has been classified into four
groups, according to its inorganic and organic constituents. Type I is dominated by
sodium and chloride as osmotic factors. In type II the concentration of circulating
sodium and chloride is lower in relation to potassium, calcium, magnesium and in-
organic phosphate. Type III is characterized by its low percent of chloride and sodi-
um, whereas amino acids and other organic components are present in higher con-
centrations. Finally, type IV is dominated by organic components, which determine
the osmotic and buffering behaviour of the haemolymph, whereas sodium and chloride
play only a subordinate role (Sutcliffe 1963; Cohen 1980; Punzo 1982).
Because of the strong dominance of sodium and chloride as osmotic factors, the
haemolymph of many terrestrial arthropods is classified as type I. It was assumed that
spiders as well would belong to this group; however, Punzo (1982) showed that in
lycosid spiders, sodium and chloride composed approximately 35% to 41% of the
haemolymph osmolyte and that potassium, calcium and magnesium as well as inorganic
phosphate were also present in low concentrations. Amino acids composed approximate-
ly 2.1%to 4%: free amino acid concentrations inhaemolymph, 66.1 to 202.3 mg/lOOmi.
The latter values, however, may vary considerably: Schartau and Leidescher (1983)
reported only 6 mg/l 00 ml in the North American theraphosid ,Eurypelma californicum.
The protein concentration may depend upon the nutritional, reproductive and moult-
ing stage of the animals and indicates that spider haemolymph may deviate strongly
from the typical type I pattern.
Of particular importance is the magnesium concentration. Schartau and Leidescher
(1983) measured 3.95 mmol/liter in Eurype/ma californicum. In Cupiennius salei
(Loewe and Linzen 1975) magnesium has been shown to affect the oxygen-binding
capacity of haemocyanin, although these results do not appear to hold for Eurype/ma
(B. Gross unpublished results cited in Schartau and Leidescher 1983). Similar magnesi-
um concentrations have been found in the haemolymph of the scorpion, Heterometrus
fulvipes (Padmanabhanaidu 1966) andHeterometrus swammerdami (padmanabhanaidu
1962): 4.9 mmol/liter and 5-25 mmol/liter, respectively.
82 F. Straznyand S.F. Perry
The free carbonate concentration is quite low: 15 mg/l00 ml and 13 mg/IOO rnl in
Eurypelma californicum and Dugesiella hentzi, respectively (Schartau and Leidescher
1983; Stewart and Martin 1970). The protein concentration in cell-free haemolymph,
5690 mg/IOO ml, consists of two protein fractions: haemocyanin (approx. 80%) and
another protein (20%) of unknown function (Marlel et al. 1976). Values for other
spiders (4700 mg/IOO ml for Cupiennius salei and 3300 mg/l00 ml for an araneid
spider) are lower. In comparison, the haemolymph of the scorpion Heterometrusful-
vipes contains 7350 mg/IOO ml protein (padmanabhanaidu 1966).
The total osmotic pressure of spider haemolymph ranges from 320 to 588 mOsmol/
liter (Cohen 1980; Pinkston and Frick 1973; Schartau and Leidescher 1983; Stewart
and Martin 1970; Sutcliffe 1963). This range does not differ conSiderably from that of
holometabolic insects (Sutcliffe 1963).
Of particular interest, with respect to osmotic balance, are two spiders which inhabit
the South African coast. Amaurobioides africanus inhabits higher shore regions "in the
zone extending from high water neaps to high water springs" (Moloney and Nicolson
1984): an area which is only infrequently flooded by salt water. The haemolymph of
this species has an osmolarity of 537 mOsmol/liter, which is in the upper range of other
terrestrial spiders. The second species, Desis formidabilis, inhabits "middle shores, be-
tween low water neaps and high water neaps" (Moloney and Nicolson 1984): a region
which is flooded with every change of tide. The osmolarity of Desis haemolymph is
930 mOsmol/liter and sodium and chloride make up 451 and 466 mmol/liter, respec-
tively. Thus, Desis can prey directly on marine crustaceans, the haemolymph of which
is practically isoosmotic with seawater (1.157 mOsmol/liter for Ligia oceanica; Gr.
Parry 1953). A maurobio ides , on the other hand, must dilute its crustacean meals with
freshwater or diluted seawater. The pH values of spider haemolymph vary according to
temperature and species between 7.25 and 7.9 (Angersbach 1978; Loewe and Linzen
1973; Schartau and Leidescher 1983; Stewart and Martin 1970). In vitro values tend
to be higher than in vivo measurements, possibly because CO 2 is lost to the air. Further-
more, the pH of centrifuged, cell-free haemolymph is higher than that of uncentri-
fuged haemolymph.
1.2.4 Haemocyanin
Oxygen-carrying or oxygen-storing pigments are present in invertebrates and in ver-

tebrates. These substances, which may occur intracellularly or extracellularly, fall into
the following groups: chlorocruorin, haemoerythrin, haemocyanin and haemoglobin.
For a comparison of the structure and function of these substances as well as their
distribution within the animal kingdom, the reader is referred to the recent symposi-
um "Respiratory Pigments in Animals" (Lamy et al. 1985a).
Haemocyanin is the oxygen-carrying pigment which is found dissolved in the haemo-
lymph of numerous arthropods (Crustacea, Arachnida, Xiphosura) and molluscs. It
gives the haemolymph its characteristic blue colour when oxygenated (Florey 1975).
In comparison to haemoglobin, the copper ofhaemocyanin is directly attached to the
protein and does not appear in a prosthetic group. One oxygen molecule is reversibly
bound on two copper atoms (prosser 1973) (Fig. 22).
d
Fig. 22a-d. The quaternary structur<l of chelicerate haemocyanin. a Model of the quaternary struc-
ture of Eurypelma ealifornicum (Theraphosidae) 37 S haemocyanin, composed of 24 subunits,
showing the topological distribution of the seven types of subunits. The subunits are present in the
proportions 4a, 2b, 2e, 4d, 4e, 4f, 4g. Light bars indicate the orientation of the dimers be and ff.
b The model standing upright on that side which is penetrated by the deep cleft (sideview image).
c, d Schematic representation of the Panulirus interruptus (Scorpiones) haemocyanin hexamer.
c Domaine 2, diamonds indicate copper ions. d The four helices surrounding the binuclear copper
side and providing the six histidine copper ligands. [a, b From Markl et al. (1981); c, d from Linztm
et al. (1985b). Science 225:522, copyright 1985 by the AAAS)
1.2.4.1 Comparison of Arthropod and Nonarthropod Haemocyanins. Just as haemo-

glob ins may differ within the same species, Significant variations in haemocyanin struc-
ture are also seen. Molluscan haemocyanin forms a cylindrical molecule of 10 to 20 sub-
units with a molecular mass of up to 9 million daltons. These subunits have eight re-
gions, each with two copper atoms and a mass of 50,000 daltons. The haemocyanin of
arthropods is smaller, from 500,000 to approximately 3.5 million daltons. The subunits
(each approximately 75,000 daltons) are composed of hexameres and oligohexameres
(Lin zen et al. 1985b) (Fig. 22). In recent years the structure of haemocyanin at four
different molecular levels of organization has been elucidated (Lamy et al. 1985b;
Linzen et al. 1985b). It now appears that in spite of great differences in structure, mol-
luscan and arthropod haemocyanins have a common molecular origin (Schneider et aI.,
cited in Linzen et al. 1985a).
1.2.4.2 Comparison of Chelicerate and Nonchelicerate Haemocyanins. Linzen et al.

(1985b) determined the amino acid sequences in seven haemocyanin subunits of two
crustacean and three chelicerate species. Of the amino acid sequences 17% are identical
among aU animals: 33% among chelicerates alone. If one considers the positions with
isofunctional residues, all investigated animals would have 29% congruity, and the
chelicerates alone, 46%. In particular, the regions with identical amino acid sequences
appear to lie in close proximity to one another, especially in the copper-binding region
(also oxygen-binding region). Crustacean haemocyanins demonstrate a loop with
21 amino acid residues which is completely lacking in chelicerate haemocyanins.
Similarly, at a different place the crustacean haemocyanin shows a loop with eight to
ten "extra" amino acid residues. Further differences are seen in the structure of the
active site and at the NH-terminal. Among chelicerates, the haemocyanin of limulus
has 48 chains and that of Eurype/ma, only 24 (Fig. 22). For Eurype/ma californicum
and Cupiennius salei the structure of the haemocyanin molecule and its physical and
chemical properties are known in detail (for references see Linzen et al. 1985b).
1.2.4.3 The Function of Spider Haemocyanin. The degree to which different spider
haemocyanins differ in function can be seen by comparison of the dissociation curves,
the oxygen-binding capacity of the blood and the effect of hydrogen ions (Bohr effect)
and magnesium ions on haemocyanin.
Angersbach (1975a) found that haemocyanin of Eurypelma helluo is deoxygenated
by only 10-12%. He calculated that only 1.2 m1 oxygen per milliliter haemolymph
would be released, if allhaemocyanin subunits react with oxygen in the same way. Near-
ly the same amount would be released from physically dissolved oxygen (approximately
0.8 ml oxygen per milliliter) in haemocyanin-free haemolymph. Thus, Angersbach
(1975a) postulated that haemocyanin in resting spiders may serve more as an oxygen
reservoir which releases oxygen to the tissue only when the Po 1 is very low.
In a later study, Angersbach (1978) found that in Eurypelma californicum, haemo-
cyanin in the lungs is only 52% saturated. The arteriovenous POl difference is 13.6 torr
such that the oxygen-binding capacity of the haemocyanin is only exploited to 47%.
In all, 9.3 ml O2 per milliliter haemolymph is released to the tissue, of which 95% is
transported by haemocyanin. Here, it is apparent that the haemocyanin-bound O2 is
more important than physically dissolved O2 , Haemocyanin may be important as an
oxygen reservoir when haemolymph flow is reduced (Angersbach 1978), which may
be the case when spiders are active and when hydrostatic pressure must be maintained
for repeated extension of the extremities.
Following activity, haemocyanin is completely oxygen saturated, the arteriovenous
pressure difference is 50.2 torr and the oxygen-carrying capacity of haemocyanin is
exploited to 82%. Thus, 15.6 ml oxygen per milliliter haemolymph are released to the
tissue, of which 89% is carried on the haemocyanin (Angersbach 1978).
During and following activity, the Bohr effect plays an important role. The ability
of haemocyanin to release oxygen in response to a lowering of the pH (Bohr effect)
differs from species to species and also according to the pH range in which this change
occurs. At high pH (pH8-8.5) the Bohr effect (.:llogPso/.:lpH) lies between -0.81 and
-1.2 in Eurypelrna helluo, Dugesiella californica and Cupiennius salei (Angersbach
1975b). In the pH range 6.9-8.4, however, the Bohr effect is reduced in Dugesiella
californica (Loewe 1972, cited in Angersbach 1975a) and below pH 6.4 it is reversed
in Cupiennius salei (Loewe and Linzen 1975). In this species the oxygen-binding af-
finity is reduced by magnesium ions.
The half-saturation in undiluted haemolymph of Cupiennius salei is reached at
26 torr (25 DC, pH 8.14; Angersbach 1975a), similar to the value for diluted (1 :20)
haemolymph in Eurypelrna californicurn: 27.8 torr (23 DC, pH 7.5; Angersbach 1978).
For Eurypelrna helluo, however, Angersbach (1975a) achieved half-saturation at 3 torr
when the pH was 8.14 (25 DC, undiluted haemolymph); whereas Loewe and Linzen
(1975) achieved half-saturation in Mg2+ free, diluted haemolymph in this species with
2 torr at pH 9, but required 17 torr at pH 6.4. The reason for these differences may lie
in the experimental technique.
1.2.4.4 Acid Base Balance. Although CO 2 gassing is a common laboratory method for
immobilizing spiders, little is known about CO 2 transport and bicarbonate buffering in
spider haemolymph. A single, excellent study (Loewe and Brauer de Eggert 1979)
showed that the Pco 2 directly affected the CO 2 concentration and the pH of Eurypel-
rna californicurn haemolymph. According to these authors, in Eurypelrna haemolymph
arterial CO 2 content and bicarbonate concentration are temperature independent,
arterial CO 2 partial pressure increases and arterial pH decreases with increasing tempe-
rature. The pH change is greater than would be predicted from the alphastat hypothesis
(Reeves 1972, cited in Loewe and Brauer de Eggert 1979), which states that the ratio
of [OIr] [H+] remains constant during temperature change. However, the pH change is
not as great as in an unbuffered system. The authors concluded that "the net charge of
blood proteins, mainly haemocyanin, is kept tolerably constant despite the fact that
.:l pHa/.:l t is low", but do not comment on the possible role of the haemocyanin mole-
cule as a buffer.
2 Structure and Function of the Respiratory Apparatus
When considering the dual respiratory system of spiders, one is continually confronted
with the questions why have so many spiders developed tracheae and, secondly, if
tracheae are more effective than the lung-circulatory system, then why have they not
replaced the latter system entirely. In the following sections first the structural dif-
ferences and then some aspects of respiratory physiology will be discussed. These two
aspects will then be integrated in Sect. 2.3.
2.1 Structure of the Respiratory System
2.1.1 Respiratory Organs
2.1.1.1 Book Lungs. As discussed above the lungs, which lie ventrolaterally in the
cranial opisthosoma, can be recognized from the outside by their shieldlike covers.
Fig. 23. Diagram of the left book lung viewed

caudally. White arrow indicates air movement;
black arrows, haemolymph flow. A air-filled
lamella. (From Foelix 1979)
Caudal to these covers lies the slitlike stigma, which can be opened by muscular action.
The stigma opens medially to the atrium, from the cranial wall of which slitlike per-
forations open into the book-lung lamellae. These form hollow, parallel plates which
extend into the haemolymph sinus (Figs. 23 and 24). All parts of the lung are covered
by a thin hypodermis, the nuclei of which may bulge into the haemolymph-filled inter-
lamellar spaces and form cellular bridges (Foelix 1979). The hypodermis secretes a
chitinous layer. This forms numerous, blunt-ended or sometimes branched spikes which
extend from the ventral surface of the pocketlike lamellae and prevent the collapsing
of these flexible structures under the hydrostatic pressure of the circulatory system.
In some cases these spikes may fuse with the dorsal wall of the respiratory lamellae,
giving the latter a grottolike appearance (Moore 1976). When spikes extend into the
atrium, they are usually branched (Kaestner 1929).
2.1.1.2 Tube Tracheae. Most spiders have one pair of book lungs in the second opis-
thosomal segment and, in addition, in the third opisthosomal segment a set of tube
tracheae replace the second book-lung pair (Table 7). In many species the tracheal
stigma lies just cranial to the spinneret, whereas in others it lies in the middle of the
opisthosoma. In some cases two separate tracheal stigmas are present.
The tracheal system varies immensely from species to species. If one may allow a
generalization, this stigma usually opens into a small vestibule from which two lateral
and two medial tubes exit. The lateral tubes are homologous to the second book-lung
pair, whereas the medial tubes are derived from cuticular muscle insertion sites (enta-
pophyses). The tracheae can be branched, but the degree of branching varies greatly
from species to species. Proximally, the tracheae are bundled, which results in a low
effective surface to volume ratio and enhances the oxygen transport function. Distally,
after separation from the bundles, individual tracheae may branch before reaching their
terminal gas-exchange sites (Kaestner 1929).
In many spiders, such as jumping spiders, and in Argyroneta aquatica, the tracheae
extend through the petiole into the prosoma, whereas in others (e.g. the common fun-
nel web spider, Tegenaria domestica) they remain confined to the opisthosoma. In
Argyroneta aquatica the tracheae begin with a circular cross-section lined with a chitin
intima in which numerous, fine spikes extend into the tracheal lumen. The spikes them-
selves possess chitin rings, such that the tracheae remain flexible, but cannot collapse.
Fig. 24a-d. The book lung of Tegenaria pieta. a Cross-section of a book lung; lateral body wall at
lower left; digestive gland at right (Semi-thin section; toluidine blue). Bar = 100 /.1m . b Scanning
electron micrograph. Bar = 10 /.1m. c Semi-thin section of the book lung, air lamellae are dark . Note
hypodermal bridge (B) across haemolymph spaces. Bar = 10 /.1m. d Electron micrograph of the lung
with spikes (S) and a hypodermal bridge (B). Bar = 1 /.1m
The tracheal capillaries are fine, unbranched structures, which possess a chitin-hypo-
dermal intima, but lack spikes. They are 5-7 11m in diameter (Crome 1952/1953).
2.1.1.3 Sieve Tracheae. A third, relatively uncommon respiratory organ are the sieve
tracheae. These may replace the first set of book lungs, and consist of a large number
of hollow, fingerlike projections from the atrial wall into the haemolymph sinus.
Morphologically, the sieve tracheae would represent an intermediate stage between tube
tracheae and book lungs (Fig. 25 and Table 7).
2.2 Physiology
2.2.1 Oxygen Consumption
Arachnids have, in general, a lower metabolic rate at rest than other poikilothermic
animals of the same size. Even during strenuous activity, spiders show only a limited
Fig. 25 . Diagram of sieve tracheae. Instead of lamellae,

tracheae extend from atrium into the haemolymph
sinus. (From Levi 1967b)
aerobic capacity, but their anaerobic metabolism is well developed (Prestwich 1983b).
The level of oxygen consumption depends on many factors and not just upon activity.
These factors include body mass, phase of development, sex of the specimen, moulting
stage , reproductive phase, ambient temperature , nutritional status, humidity, life style
(active or sit-and-wait predator) and acclimatization. The general, low metabolic rate
of spiders, itself, may be seen as an adaptation to a life style in which there may be
long waits between meals.
J.F. Anderson (1970) reports oxygen consumption rates ranging from 0.013 to
0.356 ml h- l g-l STPD (= standard temperature, pressure and dry) for spiders from
six different families: Ctenizidae, Theraphosidae , Lycosidae, Salticidae, Filistatidae,
Theridiidae). The jumping spider Phidippus, a diurnally active predator, shows the
highest metabolic rate, whereas sessile spiders, such as Filistata, show the lowest values.
For Tegenaria the oxygen consumption rate ranged from a low of 0.02 during rest
ml h- 1 g-l STPD to a high of 0.28 ml h- l g-l STPD during moulting (Strazny and
Perry 1984).
The intertidal spider, Desis marina , which lives in air spaces under kelp holdfasts,
may stay submerged up to 19 days and displays an oxygen consumption rate of only
half that of typical terrestrial sit-and-wait predators (McQueen and McLay 1983;
McQueen et al. 1983). This depressed aerobic activity appears to be an adaptation to
the limitations of the physical gill of the air bubble in which it lives to deliver oxygen.
The only spider which lives completely under water (Argyroneta aquatica) , also ap-
pears to show a depressed oxygen consumption (F. Braun 1931). Between 19° and
20°C it consumes 0 .01 ml O2 h- 1 g-l, which is approximately half the minimum
oxygen consumption rate recorded by Strazny and Perry (1984) for Tegenaria, one-
third the oxygen consumption rate of the wolf spider, Pardosa !ugubris (Edgar 1971 b)
and is similar to the low values measured by J .F. Anderson (1970) for theraphosid
spiders . This low oxygen consumption rate of the water spider becomes drastically
reduced at lower temperatures, allowing the animal to survive the winter under ice .
At 3°-5 °C Argyroneta aquatica consumes only 0.001 ml0 2 h- 1 g-l .
The effect of the nutritional state on the metabolic rate as indicated by CO 2 pro-
duction was studied for the wolf spider, Pardosa astrigera (Tanaka and Ito 1982). The
metabolic rate per milligram fresh body weight of well-fed adult females was approxi-
mately 80% that of males. At the beginning of fasting the metabolic rate increased,
but then fell in males to approximately half and in females to approximately three-
fourths the metabolic rate in the well-fed state. After 5 days' feeding both males and
females resumed the original metabolic rate. Similar results were obtained in earlier
studies for two other wolf spiders, Lycosa pseudoannulota and Lycosa T-insignita (Ito
1964; K. Miyashita 1969b; Nakamura, this volume).
Tanaka and Ito (1982) concluded that the apparent reduction in the metabolic rate
results from a change in the activity spectrum of the spiders. When active predators
hunger for long periods, they assume a sit-and-wait strategy, thus reducing the metabolic
rate. The data for the wolf spider, however, may not be applicable to some sessile
species, which, when hungering, tend to give up waiting and actively search prey "If an
adequate supply of prey was not forthcoming from the first site selected, the spider
abandoned that site and sought a new one, where another web was built" (Turnbull
1964). In addition, the change in CO 2 release may not accurately reflect a similar
change in oxygen consumption, since the respiratory quotient may not necessarily
remain constant, as assumed by Tanaka and Ito (1982).
The effect of changes in temperature on metabolic rate in the desert wolf spider,
Lycosa carolinensis, was studied by Moeur and Eriksen (1972). At high temperatures
(between 39° and 45°C) the metabolic rate of this inhabitant of the Arizona desert
increased very rapidly with increasing temperature (QlO between 2.58 and 4.21). Be-
low 39°C, however, the Ql0 was very low: between 1.0 and 1.54. The metabolic rate
in this species is relatively insensitive to temperature changes between 29° and 39°C.
Interestingly, the QlO of "June" spiders were all higher than those of "January" spiders
in the temperature range 39° to 45°C.
The foregoing results represent by no means the rule. Humphreys (1975b), who
studied temperature adaptation in another wolf spider that is capable of living at high
temperatures (Geolycosa godeffroyi), found that the Ql 0 decreased from approximately
2.3 to 1.6 when the animals were warmed from 10° to 40°C. Furthermore, he found
that the oxygen consumption rate of animals adapted to high temperature was lower
than that of specimens adapted to a lower temperature and momentarily raised to the
higher one. On the other hand, when animals adapted to a high temperature are cooled,
they show the same oxygen consumption rate as specimens adapted to the cooler tem-
perature. This metabolic overshoot during warming results in a mean respiration rate
under conditions in which the temperature cyclically varies from 10° to 30 ° C of 135%
of that shown by animals maintained constantly at the same mean temperature. Such
a response is not unique to Geolycosa, but is also shown by the wolf spider Lycosa
lenta and in particular by the jumping spider PhUlippus regius (J.F. Anderson 1970).
These data are not directly comparable with those of Moeur and Eriksen (1972) for
Lycosa carolinensis, since these spiders were acclimatized to the higher temperatures
during the day and cooled to 35 ° C at night. Not all spiders are capable of such adapta-
tion. J.F. Anderson (1970) demonstrated that it is primarily the more metabolically
active spiders that reduce their metabolic rate by up to one-third after 6 days at 30°C.
2.2_1.1 Oxygen Consumption During Rest, Activity and Recovery. J.F. Anderson
(1970) found that two species of wolf spiders showed rhythmic fluctuations in oxygen
consumption rate according to the time of day. Web spiders, on the other hand, did
not show this fluctuation. Strazny and Perry (I 984), however, reported a cyclic oxygen
consumption rate for the web spider Tegenaria (see also Cloudsley-Thompson, this
volume). McQueen et al. (1979) demonstrated diel patterns in oxygen consumption in
some, but not in all specimens ofthe wolf spider Geolycosadomi/ex.McQueen's (1980)
study centred upon the relationship between activity and oxygen consumption as
functions of temperature. He noticed that the respiratory rate during climbing in ver-
tical shafts of the cavities in which Geolycosa domifex lives is some ten times higher
than during rest, whereas activity above the ground and within the burrow increased
the oxygen consumption rate by only factors of 4.5 and 3.2, respectively. In addition,
the respiratory rate increases with increasing temperature until 18 °c,when it reaches
a plateau. The already high metabolic rate does not increase with temperature during
vertical work. McQueen (1980) further observed that small spiders undergo the greatest
increase in respiratory rate during activity at low temperatures.
Seymour and Vinegar (1973) found striking differences in the respiratory rate in
the theraphosid Aphonopelma sp.B. between activity and rest at 20°, 30° and 40 °c.
The respiratory rate during activity was usually approximately four times greater than
at rest. They found further that the spiders can only remain active for a short period
of time: the higher the temperature, the shorter the activity period (see also Pulz, this
volume).
2.2.1.2 Oxygen Consumption During Moulting. The funnel web spider Tegenaria
(Strazny and Perry 1984) consumed 3.5 times as much oxygen during a 4-h period in
which moulting occurred than during rest (0.278 ml h- 1 g-l and 0.79 ml h- 1 g-l,
respectively). This compares with an increase in oxygen consumption of only approxi-
mately 10% in the same species during recovery from that activity.
2.2.1.3 The Dual Respiratory System. When spiders with well-developed lungs and
modestly developed tube tracheae are forced to breathe through the tracheae alone,
they become phlegmatic and eventually die (Kaestner 1929). When the tracheal stigma
is blocked, however, they continue to live and (in Tegenaria) their oxygen consump-
tion rate is indistinguishable from that of intact animals (Strazny and Perry 1984).
Similarly, when the tracheal system is well developed and the lung stigma is closed
(Kaestner 1929), the spiders continue to live.
Davies and Edney (1952) found that the wolf spider Pardosa amentata consumes
more oxygen through the lungs than through the tracheae. Maya et al. (1981) showed
that a lycosid spider Hippasa pantherina of 128 mg takes up 18.0 ± 0.3 ml oxygen
during the day and 133 ± 0.3 ml oxygen during the night through the book lungs: the
tracheae provide 4.7 ± 0.2 and 2.8 ± 0.2 ml oxygen during the respective time periods.
The water spider, Argyroneta aquatica, can live several days under water with a totally
blocked respiratory system, presumably by skin breathing through the thin, opistho-
somal cuticle (F. Braun 1931).
2.2.1.4 Anaerobic Metabolism. Linzen and Gallowitz (1975) concluded on the basis
of enzyme activity that the muscles of Cupiennius salei are better suited for anaerobic
than for aerobic metabolism. In addition, they found only a small number of mito-
chondria in the muscle tissue, which implies a limited aerobic capacity.
In the theraphosid, Eurypelma califomicum, Angersbach (1978) found that the
oxygen partial pressure at the base of the legs (coxae) falls rapidly following prolonged
activity in spite of the relatively small number of mitochondria. He hypothesized that

the drop in Po, was due to stagnation of the haemolymph, which was not circulated
because of its hydrostatic function in the legs. At the same time, CO 2 enters the haemo-
lymph and reduces the pH. A second pH minimum then occurs, caused by the acid end
products of anaerobic metabolism. In the wolf spider Pardosa lugubris between 73%
and 81% (depending upon the sex) of the yearly caloric intake is respired aerobically
(Edgar 1971b), whereas in Aphonope/ma the major metabolic part appears to be an-
aerobic (Seymour and Vinegar 1973). The authors explained the anaerobic predomi-
nance during activity in Aphonopelma as a specialization for its more pronounced sit-
and-wait life style.
The by-products of D{- )-lactate dehydrogenase activity were the most common
intermediary metabolites in Eurype/ma californicum (prestwich 1983a). Lactate ac-
cumulated in concentrations up to 15 mmol g-l in the legs and in the prosoma, where-
as in the opisthosoma only 4 mmol g-l were found. During normal activity in spiders
it appears that those with the largest lung surface show the lowest lactate concentra-
tions. This would imply that the preferred metabolic mechanism is aerobic and that
the anaerobic metabolic mechanism is used only when the oxygen supply to the tissue
is inadequate. The lowest lactate concentrations were found in web-making spiders
which are normally not continually in motion.
Prestwich (1983b) found that during activity Filistata hibernalis, Lycosa lenta and
Phidippus audax depend to a large extend on anaerobic metabolism. Filistata derives
87% to 94% of its energy anaerobically, whereas Lycosa and Phidippus derive 65% and
55% of their energy from this pathway, respectively. During web construction,Neoscona
domiciliorum metabolizes primarily aerobically, but when it moves quickly, up to 40%
anaerobically (prestwich 1983b).
Unlike vertebrate muscles, spider muscles do not contain blood capillaries and the
mean diffusing pathway is at least twice as long as in vertebrate muscle (Angersbach
1975a). Thus, even if more mitochondria were present, it is doubtful in light of the
long diffusion distances, whether they could be adequately supplied with oxygen.
2.3 Functional Morphology
Using quantitative morphology, it is possible to measure the total respiratory surface

area, the thickness of the diffusion pathway between air and haemolymph and to de-
termine the morphological constitution of this diffusion pathway. If Krogh's diffusion
constant for oxygen in the various types of tissue is known, it is possible, using Fick's
first law, to calculate the conductance, or "diffusion capacity" of the lung for oxygen.
Secondly, if the oxygen consumption rate is known, the partial pressure difference of
oxygen across the lung can be calculated as the diffusion capacity divided by the oxy-
gen consumption rate. This factor, known as the driving pressure, provides valuable
information on the relative effectiveness of the respiratory organ: the greater the driv·
ing pressure, the more efficient the lung must be in order to meet respiratory demands.
Very large values for driving pressure indicate that the capacity of the lungs to deliver
oxygen is exceeded and that anaerobic metabolism or other sources of oxygen (e.g.
tube tracheae) must be of importance. Morphometric investigations, which allow a
quantitative estimate of lung function, have been carried out only by J.F. Anderson
(1970), J.F. Anderson and Prestwich (1980,1982) and Strazny and Perry (1984).
92 F. Straznyand S.F. Perry
2.3.1 Lung Surface Area
The lung surface area (SL) for specimens from four different families (Lycosidae, Salti-
cidae, Filistatidae, Theridiidae) increase with body weight (w) to the 0.842 power
(J.F. Anderson 1970). For specimens of the genus Tegenaria, the regression equation is
SL = 3.24 WO. 667 (SL in mm2 and W in mg; Strazny and Perry 1984). The difference
in exponents is probably due to the choice of specimens, but generally lies in the range
for mammalian lungs (Tenney and Remmers 1963). In general, the lung surface area
usually correlates with the metabolic rate, although Prestwich (1983b) noted that the
lung surface area is inversely proportional to the importance of anaerobic metabolism:
the more important anaerobic metabolism is, the smaller the lungs are. Thus, it may be
that lung surface area is an indicator only of the aerobic metabolic rate in spiders.
2.3.2 The Relationship Between Lung Surface Area and Lung Volume
The surface-to-volume ratio decreases with increasing body mass in Tegenaria (Strazny
and Perry 1984). Thus, for a 27-mg spider the surface-to-volume ratio is 245 mm 2
mm- 3 , but for a 342-mg spider, this parameter was only 66 mm2 mm- 3 . These values
are comparable with the same parameter in the sparrow lung or the rat lung: 387 mm 2
mm- 3 and 75 mm2 mm- 3 , respectively (Dubach 1981; Burri and Weibel 1971).
The lower surface-to-volume ratio in larger spiders can be explained by the fact that
the lung surface area increases with the square of body length, whereas lung volume in-
creases with the cube of body length.
2.3.3 Lung Volume
The total lung volume in Tegenaria is 4.3 to 7 ml g-1 . This value is not obtained along
a regression line extrapolated to a theoretical I-g spider, but rather calculated on a per
gram body mass basis for given animals. In contrast, Bridges et al. (1980) found a
tracheal volume in pupae of the moth Hyalophora which lies almost an order of mag-
nitude higher. The large volume of the tracheae is explained by their function in direct
oxygenation of the tissue, whereas in the spider this function is taken over by the
haemolymph. Although the surface-to-volume ratio of spider lungs decreases markedly
with an increase in body mass, the half-distance across the lamellar air spaces (maximum
diffusion distance from the center of a lamella) only increases by a factor of four
when the body mass has increased 900 times (J.F. Anderson and Prestwich 1980).
Thus, the diffusion conditions within the lung air space remain relatively favourable.
In addition, gas convection within the lung due to pulsing of haemolymph between the
lamellae is also possible (Herreid et al.1981). Spiders with two lungs tend to have larger
lamellae and intralamellar air spaces as do those with four lungs (J.F. Anderson and
Prestwich 1982).
2.3.4 Thickness of the Diffusion Ba"ier
Next to the total lung surface area, the harmonic mean thickness of the air-to-haemo-
lymph diffusion barrier is the most important limiting factor for gas exchange. This
diffusion barrier consists both of chitin and of hypodermis (Fig. 24), the substances
of which are very much different in their ability to allow diffusion of oxygen. Normally,
these two layers are of approximately the same thickness: hypodermis between 0.212
and 0.215 pm and chitin between 0.168 and 0.196 pm in Tegenaria. Strazny and Perry
(1984) found that this parameter did not vary with the size of the spiders, but in one
specimen the chitin layer was 42% and the hypodermal layer 177% thicker than in
other specimens. The authors attributed these changes to a presumed premoulting
stage. During moulting, the hypodermis undergoes cell growth and cell division, then
it secretes a new cuticle and absorbs the old one (Foelix 1979) (Fig. 33). Thus, as ob-
served by Strazny and Perry (1984), both of these layers must become thickened, and
the diffusing capacity of the lung must be reduced just at the time when the oxygen
consumption rate is at its highest.
2.3.5 Anatomical Diffusion Factor and Morphometrically Determined

Diffusion Capacity
The anatomical diffusion factor (ADF) is the ratio of lung surface area to the harmonic
mean thickness of the diffusion barrier (Perry 1978). This parameter is used as an
indicator of the anatomical suitability of a lung for gas exchange: a large surface area
or a small barrier thickness yields a large ADF. The values for spiders have been com-
pared with those for reptilian lungs (Perry 1983), since these are the only exothermic
vertebrates for which ADF has been calculated, and these values are lacking for other
invertebrate groups. Tegenaria appears to have approximately the same ADF per unit
body mass as does the monitor lizard Varanus exanthematicus (Strazny and Perry
1984). The morphometrically determined diffusing capacity is ADF multiplied by
Krogh's diffusion constant for oxygen in the tissue which the gas must cross. This
parameter is not identical with the physiologically determined diffusing capacity, since
it does not take into account diffusion within the air spaces, haemolymph or body tis-
sues. For reptiles and mammals the morphometrically determined diffusing capacity is
some lO to 20 times greater than the physiologically determined value. The comparison
of Tegenaria with reptiles shows a smaller morphometriC diffUSing capacity in Tegenaria
than in the monitor, since Krogh's diffusion constant for oxygen in chitin is much
lower than in vertebrate lung tissue.
The calculated oxygen driving pressure in Tegenaria during rest and activity lies be-
tween 1 and 3 torr, a value which in comparison with vertebrates indicates that the dif-
fusing capacity of the lung is not completely exploited. If one substitutes the values
for the moulting spider and for the increased oxygen consumption during this critical
phase, the oxygen-driving pressure is approximately twice as high as the highest values
reported for vertebrates (Strazny and Perry 1984). This implies that the lungs alone
may not be able to provide enough oxygen during moulting, when the tracheal system
could be a selective advantage.
3 Reconsideration of the Problem of the Dual Respiratory System
If we now reconsider the problem of the dual respiratory system in spiders, the fol-
lowing points can be cristallized. The structure of the book lungs is remarkably con-
94 F. Strazny and S.F. Perry: Respiratory System: Structure and Function
stant. In contrast, tracheae appear to be structurally very labile (Kaestner 1929). Since
the tracheal systems within different genera of a single family are often very different,
Kaestner (1929) assumed that the tracheae have originated several times. Thus, one
may hypothesize that under different conditions the tracheal system may have a quite
different significance. That is, for some spiders tracheae may provide a selective ad-
vantage over the book lung-haemolymph system which might be of no advantage or
even of disadvantage to a species with a slightly different way of life. Thus, basking
spiders tend to have better developed tracheae than those which live in a damp environ-
ment. Levi (1967b) pointed out the possible advantage of tracheae in reducing water
loss. Spiders with low body mass and high metabolic rate tend to have well-developed
tracheae (Levi 1967b). In jumping spiders and crab spiders the tracheae extend even
into the legs (Foelix 1979). According to J.F. Anderson (1970), the tracheal system is
more efficient in providing organs with oxygen than the book lung-haemolymph system.
Although the tracheae of very active insects are somewhat differently constructed than
those of arachnids, this shows just how efficient the tracheal system can be.
Since the presence of tracheae is not limited to either small or large spiders, the
water loss-metabolic rate hypothesis is not alone sufficient to explain the repeated oc-
currence of tracheae. The relatively large water spider Argyroneta aquatica, for examp-
le, has a well-developed tracheal system and lives in an extremely moist environment.
Here, the presence of tracheae may be more a matter of the center of buoyancy for an
animal living under water: the long tracheae extend into the prosoma (Levi 1967b).
When the tracheal system is well developed, the circulatory system tends to be re-
duced. Conversely, the functioning of the lungs is dependent upon a well-developed
circulatory system. In spiders the "arterial" and "venous" systems are not completely
separated (Willem 1917), so that a mixing of oxygenated and deoxygenated haemo-
lymph could negatively influence the efficiency of the respiratory system. In addition,
haemolymph also functions as a hydrostatic system. This means that just at the time
when the spider is most active, the blood stagnates in the legs. Gas diffusion through
air-filled tubes is several thousand times faster than in standing fluid (Weis-Fogh 1964).
Thus, the tracheal supply of the legs in jumping spiders and crab spiders can certainly
be seen as an adaptation for aerobic activity during the search for or the struggle with
prey.
According to Prestwich (1983b), Filistata derives its energy during activity mainly
for anaerobic metabolism, so that the function of the respiratory system in this species
is secondary. In wolf spiders and jumping spiders, on the other hand, aerobic metabolism
is important (Prestwich 1983b).
As pointed out by Strazny and Perry (1984), the possible insufficiency of the lungs
during moulting could also provide a selective advantage for tracheae formation. But
tracheae are also shed. The question still remains unanswered: if tracheae are so effective,
why have lungs been retained?
VII Heart Rate and Physiological Ecology
JAMES E. CARREL 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 95
2 Basic Cardiovascular Biology of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 96
2.1 Cardiovascular Anatomy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
2.2 Methods for Measuring Heart Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 96
2.3 Neural Control of the Spider Heart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
3 Body Size-Heart Rate Relationships in Spiders . . . . . . . . . . . . . . . . . . . . . . . . .. 99
3.1 Intraspecific Comparisons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 99
3.2 Interspecific Comparisons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.3 Relationship Between Heart Rate and Metabolic Rate . . . . . . . . . . . . . . . . . . . 101
4 Exercise and Heart Rate in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.1 Heart Rates After Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.2 Heart Rates During Exercise .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5 Temperature Effects on Heart Rate in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6 Effects of Arthropod Venoms on Heart Rate in Spiders . . . . . . . . . . . . . . . . . . . . . 106
6.1 Wasp Venoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.2 Milliped Venom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
7 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
1 Introduction
A spider's heart rate is a major indicator of its physiological state (Mikulska 1961 b).
Unfortunately, this concept is more an extrapolation from studies with a wide variety
of nonarachnid invertebrates and vertebrates than it is a conclusion drawn from studies
with spiders themselves. Forexample,J.C.Joneset al. (1971) and Ligon and Greenberg
(1971) summarized data showing that heart rate in a generalized invertebrate varies ac-
cording to the properties of the individual, its environment, and the experimental meth-
odology used to gather data. However, of the 100 or so publications cited by them,
some of which date back as far as 1837, none ostensibly deals with spiders.
The purpose of this chapter is to provide the first comprehensive review of cardiac
function in spiders as studied under different suites of environmental conditions and
then, as much as possible, to ascertain whether heart rate is a valid physiological in-
dicator in spiders.
1 Division of Biological Sciences, University of Missouri, 209 Tucker Hall, Columbia, MO 65211,
USA
96 J.E. Carrel
2 Basic Cardiovascular Biology of Spiders
2.1 Cardiovascular Anatomy
Spiders possess a partially closed circulatory system, consisting of a tubular heart and
arterial vessels, but no distinct capillaries or veins (Fig. 21). Extensive microscopic
examinations by Petrunkevitch (1922,1933) and R.S. Wilson (1967), which have been
summarized by Foelix (1982), showed that the heart is suspended dorsally in the
opisthosoma by many regularly arranged elastic ligaments attached to the exoskeleton.
When the circular cardiac muscles contract (systole), the heart tube becomes narrower,
forcing blood out through the main arteries, and the suspensory ligaments are stretch-
ed. When these muscles relax, the stretched ligaments shorten and passively expand the
lumen of the heart. During systole most blood in the heart is propelled forward in the
anterior aorta through the pedicel, but some blood is pumped laterally into opistho-
somal arteries and backward through the posterior aorta.
Circulation in the prosoma is complex (F oelix 1982). Blood from numerous ramifica-
tions of the anterior aorta bathes the central nervous system, skeletal muscles including
even those far out in the legs, and other prosomal tissues. Blood pools in ventral lacunae
and then seeps through the pedicel back into the abdomen where it mixes with opistho-
somal hemolymph. Partial recirculation is achieved in diastole when blood in the opis-
thosoma, partly reoxygenated by the book lungs and/or the tracheae, flows into the
lumen of the heart tube through several pairs of slit-shaped holes (ostia) located in its
walls.
2.2 Methods for Measuring Heart Rate
2.2.1 Optical Methods
Certainly the most simple and, therefore, probably the oldest method for studying
heart rate in spiders is to observe the beating organ through the body wall near the
front of the opisthosoma. Bristowe (1932) counted heart beats in Liphistius desultor
merely by watching it under ambient conditions. To enhance the visibility of spiders'
hearts, Mikulska (1961a,b) illuminated restrained individuals from below with a cool
beam of white light. R.S. Wilson (1967) stripped hairs off the opisthosomal cuticle,
painted it with immersion oil, and then illuminated the opisthosoma from the side to
see clearly the beating heart in Heteropoda venatoria. More recently, Carrel and Heath-
cote (1976) used a red helium-neon laser as a cool, but intense point source of light to
transilluminate the hearts of many unrestrained spiders (Fig. 26).
Continuous records of heart beats can be obtained using optocardiographic methods
that exploit the periodic fluctuations in cardiac opacity. Carrel and Heathcote built a
photocell detector system to monitor heart rates in transilluminated spiders. As shown
in Fig. 27, the photocell is usually placed directly above the spiders' opisthosoma. The
photocell detector can also be used to make measurements from a television monitor.
Hence, cardiac performance in a spider can be recorded on videotape and analyzed at a
later date. Similar systems have been independently developed for monitoring heart
rate in insects (Lhotsky et a1. 1975) and in gastropod molluscs (Feinstein et aI. 1976).
Heart Rate and Physiological Ecology 97
Fig. 26a,b. Visualization of the tubular heart (arrow) through the anterior surface of the opistho-
soma of Lycosa ceratio1l1 transilluminated by laser light. a Direct view; b Televised image after con-
trast enhancement. The pulsating heart dims during systole and brightens during diastole (Carrel
and Heathcote 1976)
98 J.E. Carrel
Video came r a
Photodetector
Stage ~ r--""\
micromete~
Video monitor
Chart recorder
Amplifier
Fig. 27. Cardiac monitoring system for spiders used by Carrel and Heathcote (1976)
2.2.2 Mechanical Methods
The beating heart causes rhythmic movements of a spider's opisthosoma. These move-
ments can be often observed and counted in unrestrained animals as Sherman and Pax
(1968) did with the wolf spider Geolycosa missouriensis. Alternatively, the rhythmic
excursions can be translated by a device and recorded , which usually necessitates re-
straining the spider. Willem and Bastert (1918) used a galvanometric device to record
opisthosomal pulsations in the cellar spider Pholcus . To record such pulsations in Hete-
ropoda venatoria, R.S. Wilson (1967) attached a force transducer to the opithosomal
cuticle above its heart. To record heart rate in Dugesiella hentzi, Stewart and Martin
(1974) attached a small, fluid-filled balloon to a pressure transducer and held it gently
against the dorsal anterior of the opisthosoma. They also cannulated spiders using
metal shafts of syringe needles connected to pressure transducers to make continuous
measurements of changes in blood pressure in the heart , the legs, the prosoma, and the
opisthosoma.
Cardiac pulsations also cause rhythmic movements of spider legs . These are most
noticeable when the leg of a resting spider is suspended in the air. Under these circum-
stances the tarsus or other leg segments will successively extend and flex in a small arc
(Plateau 1887; Gaubert 1892; Ellis 1943; Carrel, unpublished). The ventral, thin-walled
sacs at the femoropatellar and tibiometatarsal joints also balloon during the extensor
phase of leg movement (Ellis 1943 ; Carrel, unpublished). Studies with the cellar spider
Pholcus (Willem and Bastert 1918), the crab spider Misumena aleatoria (Ellis 1943),
and the wolf spider Lycosa ceratiola (Carrel, unpublished) confirm that rhythmic leg
extension is synchronous with systole and flexion is synchronous with diastole .
2.2.3 Electrochemical Methods
Beginning in the 1960's the widespread availability of sophisticated physiological equip-

ment made it easy to monitor the weak electrochemical fluctuations generated by
muscles in the beating spider heart. Sherman and Pax (1968) placed insect pin electrodes
mid-dorsally in the opisthosoma of the wolf spider Geolycosa missouriensis restrained
under a blanket of cheesecloth. They also made recordings of electrical activity in beat-
ing hearts totally isolated from the spider using extracellular platinum-iridium electrodes
or intracellular glass microelectrodes (Sherman and Pax 1968,1969).
Oxygen microe1ectrodes provide electrocardiograms when they are inserted in the
hearts of spiders. Angersbach (1978) discovered that each contraction of the heart pro-
duces a deflection of the electrode tip that is registered as a small blip in the otherwise
steady oxygen pressure Signal. He exploited this dual measurement system to study the
effect of exercise on oxygen transport and heart rate in EurypeJma californicum while
tethered to an orientometer.
2.2.4 Relative Merits of Various Methods
As noted by Carrel and Heathcote (1976) and Feinstein etal. (1976), the measurement
of heart rate in spiders and other invertebrates having hydrostatic skeletons is difficult.
Hydraulic pressure causes extensive bleeding from a wound created when a sensor is
implanted in a spider's body. Moreover, loss of blood or the presence of an. externally
attached sensor usually results in chronic elevation of a spider's resting heart rate,
probably because these treatments significantly disrupt feedback control of the cardiac
pacemaker system.
From a physiological perspective, therefore, noninvasive, optical methods seem to
be best for measuring heart rate in spiders and, for that matter, in other invertebrates
as well (Lhotsky et al. 1975; Feinstein et al. 1976). Optocardiographic methods meet
these physiological criteria, but they generally are limited to use with resting or with
restrained animals. Further miniaturization of optocardiographic assemblies that in-
corporate both a light source and a light sensor may solve most technological problems,
thereby permitting more widespread use of this method.
2.3 Neural Control of the Spider Heart
The heartbeat in spiders, unlike insects, is neurogenic in origin. From 50 to 100 neu-
rons located in an elongate ganglion on top of the heart, rather than specialized myo-
cardial cells, act as pacemakers. The structural and functional relationships between
the heart, the cardiac ganglion, and the central nervous system in spiders and scorpions
were recently reviewed by Sherman (1985). Readers are referred to his review for more
details.
3 Body Size-Heart Rate ReJationships in Spiders
3.1 Intraspecific Comparisons
Heart rates in spiders appear to decline as individuals grow and develop. For instance,
3-4-day-old spiderlings of the funnel-web spider Tegenaria agrestis have an average
100 I.E. Carrel
Fig. 28. Intraspecific variation in heart

rate (H) and metabolic rate (M) in
three spider species as a function of
body weight (W). The species identi-
~
ties and the exact equations for the
c:
regressio ns are:
E
-""
..... 100 - - - - - - - - - - -L 100 A: Atypusabboti, M = 82 W-O .0 6 ,
H =53 W- O. 13 ;
<II
<ll
.0
- - - - - - - - - - - - A
- - - - - - - -F
-
-=-<ll F: Filistata hibernalis, M =67 W- 0 . 06 ,
H = 57 W- 0 .27 ;
-""
~
<ll o L: Lycosa lenta , M = 145 W-o.os ,
H = 33 W- 0 . 14 .
-""
o
10 10 .0 (Data are taken from J.F. Anderson
....
<ll
""
<ll
and Prestwich 1982)
J: :::E
10 100 1000
Body weight ( mg)
heart rate of 106 beats/min at 20°C, which is nearly twice that of adult females (X =
59 beats/min) (Mikulska 1961b;Mikulska and Kokocinski 1965).
Intraspecific variation in spider heart rate is weakly related in a negative fashion to
body size. J.F . Anderson and Prestwich (1982), using the cool laser method of Carrel
and Heathcote (1976) , showed that this was true in three large American spiders, Atypus
abboti, Filistata hibernalis, and Lycosa lenta . Their data, summarized in Fig. 28, reveal
a small decline in heart rate on the order of 5 -1 0 beats/min in spiders differing in size
by one order of magnitude (100 versus 1000 mg body weight). However, they limited
their measurements to subadult and adult females weighing more than 20 mg. Examina-
tion of all life stages of these species, beginning with spiderlings that weigh about 1 mg,
would not likely alter greatly the parameters of their regression curves. But, on the
other hand, because of the logarithmic nature of the regression functions, one would
expect to find relatively high heart rates in spiderlings (50-60 beats/min or greater)
relative to 500-mg females (10-24 beats/min).
3.2 Interspecific Comparisons
Examination of many spider species from different families shows that resting heart
rate is predominantly a function of body size (Fig. 29). Carrel and Heathcote (1976)
reported a modest, but Significant , decline in heart rate with size in adult female spiders
representing 18 species ranging in weight from 300 mg to 15 g.
A negative relationship between body size and heart rate is also present in spiderlings.
Mikulska (1961 b) found that when she ranked 3- 4-day spiders belonging to five hetero-
generic species according to the size of their body silhouettes, the average resting heart
rate decreased as species size increased.
The heart rate-body weight relationships in adult and immature spider species for
all intents and purposes are the same (Fig. 29). Approximate weights of spiderings
were calculated for the largest species, the funnel-web spider Agelena IIlbyrinthica, and
the smallest species, Eugnatha striata, studied by Mikulska (1961 b) using her scale
10 00
,g 100
CII
Ql
E
....
o
10
1 L-____ ~ ______ ~ _______ L_ _ _ __ __ L_ _ _ _ _ _J __ _ _
0.1 10 100 1000 10000

Body weight ( mg )
Fig. 29. Interspecific variation in heart rate and metabolic rate in spiders and other poikilotherms.
Best fit regression equations for heart rate (H = beats/min) or for specific metabolic rate (M = III
0, /g/h) as a function of body weight (yY = mg) as published in seven studies are depicted. The de-
tails for each regression are;
1 Poikilotherm metabolism, M = 820 W- 0 .25 (Hemmingsen 1960).
2 Spider metabolism, M = 736 W- 0 . 29 (Greenstone and Bennett 1980).
3 Spider metabolism, M = 330 W- 0 .20 (J .F . Anderson and Prestwich 1982).
4 Spider metabolism, M = 947 W-O. 39 (LF. Anderson 1970).
5 Spider heart rate, H = 126 W- 0 .23 (Carrel and Heathcote 1976, all data).
6 Spider heart rate , H = 423 W- 0 .39 (Carrel and Heathcote 1976, data for primitive hunters and
weavers are omitted) .
7 Spiderling heart rate (a) at 27°C, H = 154 W- O.26 ; (b) at 20°C, H = 100 W- 0 .41 (Mikulska
1961b)
drawings and the equations of Greenstone et al. (1985) for converting spider body
dimensions to weight (1.3 and 0.3 mg, respectively). The resulting heart rate-body
weight values, in effect, define the limits of Mikulska's (1961b) data set for all five
species examined at two temperatures (20 0 and 27 a C). As indicated in Fig. 29, the
spiderling data straddle Carrel and Heathcote's (1976) regression curve for adult spiders.
Heart rate in spiders is also dependent upon foraging style (Table 8). Primitive
hunters and weavers, for example , on average weigh as much as small hunters, but the
former group has exceedingly low cardiac frequencies . Also, large hunting spiders have
heart rates that are approximately one-half those in large web weavers of comparable
size. Carrel and Heathcote (1976) suggested that low resting heart rates in both groups
reflect energy-conserving adaptations of spiders that invest little effort in prey capture
and, therefore, feed infrequently.
3.3 Relationship Between Heart Rate and Metabolic Rate
Heart rate in any animal is generally related to its metabolic rate . The logic underlying
this inference and examples supporting it have been summarized by S.F. Johnson and
Gessaman (1973). But, as pointed out by S.F. Johnson and Gessaman (1973) and Pros-
ser (l973), heart rate varies greatly with the activity level, stimulation, and internal
102 J.E. Carrel
Table 8. Heart rate in female spiders as a function of body weight a
Family Species Number of Heart weight Heart rate

body weights (mg) (beats/min)
tested
Theraphosidae Aphonopelma chalcodes 6 9,810 ± 530 13.0 ± 0.6

Dugesiella hentzi 5 12,910 ± 460 11.2±0.4
Pisauridae Dolomedes tenebrosus 3 796 ± 100 35.0 ± 4.6
Lycosidae Lycosa osceola 5 1,120 ± 81 20.6 ± 0.9
Lycosa ceratiola 5 422 ± 53 19.2 ± 2.5
Sparassidae Heteropoda venatoria 5 1,300 ± 109 36.0 ± 0.7
Agelenidae Agelenopsis pennsylvanica 5 479 ± 27 59.0 ± 4.5
Filistatidae Filistata hibernalis 8 389 ± 32 35.3 ± 4.2
Araneidae Argiope aurantia 5 427 ± 39 43.2 ± 6.4
Neoscona arabesca 5 301 ± 18 69.2 ± 3.3
Scytodidae Scytodes sp. 6 .80 ± 7 18.0 ± 2.0
Loxoscelidae Loxosceles reclusa 7 108 ± 7 12.4 ± 2.0
Thomisidae Misumenops asperatus 5 54 ± 13 53.6 ± 3.0
Misumenoides [ormosipes 5 98 ± 5 108.6 ± 7.1
Salticidae Metacybra undata 5 90 ± 15 106.2 ± 3.3
Phidippus audax 5 153 ± 5 92.4 ± 4.9
a Data (mean ± SEM) are from Carrel and Heathcote (1976 and unpublished results).
state of an animal. S.F. Johnson and Gessaman (1973) concluded that "heart rate can
be a fair to good index of the average hourly metabolism of some individuals at rest or
when exercising moderately."
Are heart rates and metabolic rates correlated in spiders? An unequivocable answer
"annot be given to this question because the necessary experiments have not been
done. A simultaneous record both of the heart beat and oxygen consumption in in-
dividual spiders must be obtained. Such experiments are underway (Carrel unpublished).
In the past decade several scientists have published evidence, all of which is indirect
in nature, either for or against the use of heart rate as a metabolic indicator in spiders.
The following paragraphs summarize their papers and other relevant material.
Carrel and Heathcote (1976) claimed that cardiac frequencies might be used to
measure standard rates of metabolism in spiders. They based their claim on the observa-
tion that the regression coefficient for the heart rate-body weight curve determined by
them is identical to the regression coefficient calculated using J.F. Anderson's (1970)
metabolism-size data (Fig. 29). Both studies used adult female spiders of many kinds
with the exception of primitive hunters and weavers. Carrel and Heathcote (1976), as
mentioned previously, suggested that spiders classified as primitive hunters and weavers
have unusually low resting heart rates for their size in order to conserve energy.
Regression analyses, the results of which are presented in Fig. 29, reveal that the
slope of the heart rate-body weight curve for all spiders studied by Carrel and Heath-
cote (1976) is similar to the slopes of two recently published metabolism curves for
spiders, including primitive hunters and weavers (Greenstone and Bennett 1980;
J.F. Anderson and Prestwich 1982). Moreover, all three spider curves have regression
coefficients comparable to Hemmingsen's (1960) metabolism curve for poikilotherms
(Fig. 29). It appears, therefore, that heart rate and metabolic rate in spiders are generally
correlated with each other.
Greenstone and Bennett (1980) disagreed with Carrel and Heathcote (1976) because
they found that respiration rates in primitive hunters and weavers are the same as those
in other small spiders. Greenstone and Bennett (1980) also reported that there are no
significant differences in the slopes of metabolism-body size regressions for nine families
of spiders. On the other hand, J.F. Anderson and Prestwich (1982) found that large,
web-weaving spiders have disproportionately higher rates of metabolism than other
large spiders. These authors tentatively concluded that large web weavers are adapted
to high energy availability and high energy expediture, which is in agreement with Carrel
and Heathcote (1976).
J.F. Anderson and Prestwich (1982) failed to find a simple linear or curvilinear pat-
tern between heart rate and oxygen consumption in three species of spiders. They ob-
tained their data by calculating heart rate and metabolic rate for idealized SOO-mg in-
dividuals of each species using the two sets of regression equations presented in Fig. 28.
Although these results certainly suggest that cardiac frequency is not related to inter-
specific variation in metabolic rates among spiders, no conclusions can be firmly drawn
from a study consisting of three data points. Many more species of spiders must be
studied and heart and respiration rates must be measured simultaneously in individual
spiders in order to properly evaluate this hypothesis.
A major and heretofore unrecognized problem in the field of spider energetics is the
disparity in the relationship between respiration rate and body size on an intraspecific
level. Some investigators, such as Ito (1964), Humphreys (197Sb), and McQueen et al.
(1979), obtained standard metabolic rate-body weight regressions for single species that
have slopes to equal to -0.18 or -0.19, values which are similar to interspecific metab-
olism and cardiac functions presented in Fig. 29. Other investigators, most notably
K. Miyashita (l969b) and J.F. Anderson and Prestwich (1982), obtained standard me-
tabolic rate-body weight regressions for each of nine species that all have virtually zero
slope (see Fig. 28 for several examples).
The implications of this dichotomy are tremendous. On the one hand, spider metab-
olism may scale directly in proportion with body size, whereas, on the other hand, it
may not. If the former idea is correct, than each interspecific metabolic rate-body
weight function presented in Fig. 29 in reality consists of a stair-step function generated
by a family of horizontal intraspecific curves whose y-intercepts decrease with spider
body weight. A consequence of this situation would be that the strength of any inter-
specific metabolism function, as measured by the correlation coefficient, would be
relatively weak, so the correlation between metabolic rate and heart rate in spiders
would almost certainly be even weaker.
In contrast, if metabolic rates in spiders actually do not scale with body weight
at the intraspecific level more or less as at the interspecific level, then the afore-
mention relationships are likely to be strong. More studies of spider metabolism need
to be made independently by various investigators to resolve this problem. If other,
well-studied invertebrate taxa are used as a bench mark (Prosser 1973), then one would
expect metabolism in most spider species not to scale directly with body size.
104 J .E. Carrel
4 Exercise and Heart Rate in Spiders
4.1 Heart Rates After Exercise
Spiders have much greater heart rates after strenuous exercise than before . Bristowe
(1932) noted this in Liphistius desultor, a tropical spider that apparently becomes ex-
hausted after a short dash out of its burrow. He counted the number of heart beats per
minute in one individual before and after it had run to exhaustion. His results , presented
in Fig. 30, show a rapid decline during the first 5-10 min of rest followed by a slow
decline lasting about 40 min until heart rate reached the resting (= pre-exercise) level.
The overall decline is from 120 to 27 beats/min. The post-exercise heart rate in Geoly-
cosa missouriensis, as determined by Sherman and Pax (1968), decays in a similar
fashion from a maximum of 176 to a resting value of 48 beats/min (Fig. 30). Compa-
rable results were obtained for three species of theraphosids (Dugesiella hentzi, Eury-
pe/ma californicum, and Brachypelma smithi) by Stewart and Martin (1974), Angers-
bach (1978), and J.F. Anderson and Prestwich (1985), respectively, under restrictive
laboratory conditions.
The decline in heart rate in spiders at rest after prolonged locomotion is not modeled
well by a single exponential decay function (Carrel, unpublished). The shape of the
curves in Fig. 30 suggest that a double exponential decay process may underlie post-
exercise cardiac recovery in spiders. The physiological basis of this apparently biphasic
recovery is not understood. The fast recovery phase , based on studies of other animals
(Prosser 1973), probably is neuroendocrine in origin. From the recent work of J.F. An-
derson and Prestwich (1985), one might logically infer that the slow component is as-
sociated with biochemical recovery from lactate build-up under conditions of anaerobic
metabolism caused by exhaustive exercise .
150
"2
~ 100
.,"'
<"a
o
.c
.,
00
o 0
~ •• • 0
....~
.... 50
00
.,os
:I:
Fig. 30. Decline of heart rate in spiders re-

covering from strenuous exercise. Data for
O L . . - - - ' - - - --'--_ _ .L......._-----' Liphistius desultor (e) and Geolycosa mis-
o 15 30 45 60 souriensis (0) are taken from Bristowe (1932)
T i me (min) and Sherman and Pax (1968), respectively
4.2 Heart Rates During Exercise
Monitoring the beating heart in active spiders has proved difficult. Most studies have
been performed on severely restrained animals. Under these conditions heart rate is
highly variable. In mild struggles it may increase greatly to post-exercise levels (Stewart
and Martin 1974). But in especially strong struggles heart rate may almost instantane-
ously decline below resting values and, in some instances, the heart may briefly stop
beating altogether (R.S. Wilson 1967; J.F. Anderson and Prestwich 1975). Presumably,
loss of cardiac function is the result of feedback on the heart from massive surges in
prosomal blood pressure, perhaps reaching 450 mmHg (= 60,000 N/m), that are gener-
ated when a spider attempts a quick escape (J.F. Anderson and Prestwich 1975).
Recently, J.F. Anderson and Prestwich (1985) demonstrated for the first time that
a spider's heart rate is elevated above resting levels during prolonged locomotion. Using
small pin electrodes, they recorded cardiac activity in Brachypelma smithi forced to
run for 10 min at different speeds on a treadmill. Under these conditions a spider's
heart beat rose in the first 2 min from 18 to 44 beats/min, then it increased slowly to a
sustained maximum of 58 beats/min.
One would expect heart rate to rise in any spider as it walks, runs, spins a web, or
engages in other activities. What remains to be investigated is the precise relationship
between heart rate and metabolism in unrestrained spiders at various stages of activity.
5 Temperature Effects on Heart Rate in Spiders
Heart rate in a resting spider increases as its body temperature increases. Heathcote
and Carrel (1979) demonstrated this in the wolf spider LyCOSll ceratiola, a species
endemic to southern Florida. They acclimated nine adult female spiders to 19 ° C for
several months in an environmental chamber. They subsequently changed the ambient
temperature every other day in a stepwise fashion by 3°-5°C, progressing from 19°C
up to 41 DC, then down to 11 DC, and back to 19°C. The spiders appeared to acclimate
within 1 day to each new constant temperature. Spider heart rates were not significantly
different when measured by laser transillumination (Carrel and Heathcote 1976) on
day 1 and day 2 after reaching a given temperature. Similar results have been reported
in metabolic studies of spiders and other poikilotherms (Moeur and Eriksen 1972).
Regression analysis of the heart rate-temperature data for L. ceratiola reveals a linear
pattern when an exponential transformation is used. The composite curve for all data,
as depicted in Fig. 31, has the formula
10gH=0.2597 +0.0371 T (r=0.98,P~0.001),
where H is the heart rate (beats/min) and T is the temperature (0C).

Heart rates in newly hatched spiders seem to respond to temperature more or less
as they do in adults. When tested at four temperature settings ranging from 20°C to
30.5 °C, Mikulska (1961 b) found a systematic increase in heart rate in Tegenaria agrestis
spiderlings from 106 to 272 beats/min. Regression analysis of her data reveals that the
exponential heart rate-temperature function is highly linear and that its slope is virtual-
ly identical to the regression curve for L. ceratiola:
log H = 1.2641 + 0.0380 T (r = 0.996,P< 0.001) .
106 I.E. Carrel
100
c:
.....E
!
'"
IV
.c
-'"
IV
.....
10
.....
'"
IV
::r
10 20 40
Te mperat ure (OC)
Fig. 31. Effect of temperature on resting heart rate in Lycosa ceratiola (N = 9). The mean, standard
error and range of observations at each temperature are given. Black: initial acclimation temperature;
white: ascending temperature program ; grey: descending temperature program. See tel(t for details
The relative effect of temperature on heart rate and metabolic rate in spiders can be
quantified by comparison of QIO values . The QIO is the factor by which a physiological
function is increased in response to a rise of 10 °c (Prosser 1973).
Comparisons of QIO values for heart rate and metabolic rate in spiders, as presented
in Table 9 , reveal three tentative patterns. First, in the 200 -30 °c temperature range
for which there is much information, QIO values for cardiac and metabolic rates in
spiders are similar, averaging 1.9 and 2.5, respectively. Second, the QIO values for heart
rates in spiders match those tabulated by Ligon and Greenberg (1971) for non arachnid
arthropods tested at 20 0 - 30 °C. Third, as pOinted out by Seymour and Vinegar (1973),
QIO values tend to increase at lower temperatures in spiders, such as arctic Lycosa sp.
and desert-dwelling Aphonope/ma sp, that are adapted to climates having large tempe-
rature fluctuations . Presumably, this is an energy conserving adaptation selected for in
animals dwelling in unstable environments. In contrast, QIO values are relatively con-
stant in spiders from thermally stable habitats , such as subtropical Lycosa ceratio/a
and tropical Lycosa sp . Physiological studies of more spiders from different habitats
across a wider range of temperatures are needed.
6 Effects of Arthropod Venoms on Heart Rate in Spiders
6.1 Wasp Venoms
Many stinging wasps employ paralytic venoms to immobilize spiders or insects so that
they may be used as food later on by developing wasp larvae (Coville, this volume).
The venom causes paralysis of the prey's locomotoric system that may be temporary
or prolonged without apparently damaging the circulatory system (Piek and Simon-
Table 9. QIO of heart rate and metabolic rate in spiders as a function of climate a .,"::I:...
....
Parameter Species Climate Temperature e C) Reference :;t!
~
0 10 20 30 40
.,"p
P.
."
Heart rate Tegenaria agrestis Temperate 2.4 2.4 Mikulska (1961h) ::r
~
Lycosa ceratiola Temperate 2.35 2.35 2.35 Heathcote and Carrel (1979) o·
0"
Ot>
Metabolic rate Lycosa sp. Arctic 4.6 4.2 2.7 1.3 Scholander et al. (1953) (S.
Geolycosa domifex Temperate 8.6 2.5 McQueen et al. (1979) e:.
Aphonopelma sp. Temperate 5.82 3.35 1.93 1.11 Seymour and Vinegar (1973) tr:I
Lycosa t-insignata Temperate 2.5 2.5 K. Miyashita (1969b)
8
0"
Ot>
Geolycosa godeffroyi Temperate 2.25 2.36 1.50 1.59 Humphreys (1975b) '<
Lycosa carolinensis Temperate 1.3 3.5 Moeur and Ericksen (1972)
Lycosa lenta Temperate 3.9 1.8 I.F. Anderson (1970)
Phidippus regius Temperate 2.4 2.7 J.F. Anderson (1970)
Filistata hibernalis Temperate 3.1 2.6 I.F. Anderson (1970)
Lycosa sp. Tropical 2.0 1.9 1.8 2.2 Scholander et al. (1953)
aWithin each parameter, species are ranked in order of decreasing latitude of locality.
o
-...l
-
108 J.E. Carrel
Thomas 1969). Unfortunately, until the mid-1970's all physiological studies of wasp
venom were performed with insect prey, particularly Galleria larvae.
Because spider heartbeats usually are neurogenic, whereas insect heartbeats are myo-
genic, Sherman (1978) questioned whether transmission across neuromuscular synapses
in the spider heart might be blocked by wasp venom in the same way as skeletal neuro-
muscular synapses. Upon investigation, he found that the heart and associated tissues
of the crab spider Misumena vatia were not affected by the venom of two sphecid mud-
daubers, Sceliphron cementarium (Fig. IlIa) and Chalybion cyaneum. Electrical activ-
ity of myocardial cells, which was monitored in situ by intracellular microelectrodes,
was identical in normal and paralyzed spiders. Furthermore, microscopic examination
failed to reveal any venom-induced alteration of the cardiac ganglion, the motor nerve
terminals, or the myocardial cells.
Two other lines of evidence confirm that the long-lasting paralysis induced by some
wasp venoms is largely restricted to the skeletal neuromuscular system in spiders. First,
heart rates in paralyzed crab spiders removed from nests of the mud-dauber S. cemen-
tarium are identical to those in unparalyzed spiders at rest (Carrel and Heathcote, un-
published results). Second, respiration rates in the wolf spider Geolycosa domifex often
remain at normal levels for 2 weeks after they are paralyzed by the pompilid wasp
Anop/ius relativus (McQueen 1979; McQueen et a1. 1979).
6.2 Milliped Venom
The viscous defensive secretion of the European pill milliped Glomeris marginata pro-
duces sedation in many spiders. Tests with wolf spiders, particularly Lycosa ceratiola,
demonstrated that sedation is attributed to two quinazolinone alkaloids, glomerin and
homoglomerin, present in sticky droplets oozed from glandular pores along the back
of a milliped (Carrel and Eisner 1984). Ingestion of a few droplets of secretion or the
equivalent amount of pure alkaloids is sufficient to sedate individual L. ceratiola for
several days; larger doses may prove fatal to spiders (Carrel and Eisner 1984).
The possibility that Glomeris alkaloids might generally depress neuromuscular
systems in the wolf spider Lycosa ceratiola was tested and rejected by Carrel and co-
workers. In one experiment (Carrel and Heathcote, unpublished results), heart rates
were monitored in wolf spiders before and after they were treated by injection with a
glomerin/homoglomerin solution sufficient to cause reversible sedation lasting 3 -4 days.
As a control, a second group of spiders was injected in a similar fashion with the saline
diluent alone in the fashion of Carrel and Eisner (1984).
Within 1 hall qUinazolinone-treated spiders, but none of the controls, were sedated.
Heart rates in sedated spiders were 95 beats/min, approximately four times the resting
rates they had before treatment (Fig. 32). Such high rates are characteristic of normal
spiders after strenuous exercise. As spiders recovered from the drug-induced lethargy
in the course of 4 days, their heart rates also returned to pretreatment levels. Heart
rates in the controls did not change Significantly during the experiment.
The cardiostimulatory effect of the milliped's defensive chemicals could possibly
arise from an effect either in the spider's central nervous system or peripherally at the
level of the myocardium. A second set of experiments was conducted to test for peri-
a .
1
I
~
100
E
.....
'"ra
jSOt tttt t t
,,
I
,
I
~~BO~--------~0------------~10~0~----------~2~01r0'-----------~3noon-
Time (hrl
b
100
.....E
'"ra
<II
!
.tl
<II 50
ra
....
~·B~0~------~0~-----------'lOO~----------~2noon-----------33000o
Time (hrl
Fig. 32a,b. Acceleratory effect of milliped defensive toxins on resting heart rate in Lycosa ceratiola.
a Spiders (N = 5) each injected at 0 h with 9 JJg synthetic glomerin/homoglomerin mixture; b con-
trols (N = 5) each injected at 0 h with carrier saline. Mean, standard error and range of measure-
ments are indicated (Carrel and Heathcote, unpublished results)
pheral inhibition of spider neuromuscular systems. Using the isolated spider leg prepara-
tion developed by Rathmayer (1965), Rathmayer, Carrel, and Eisner (in preparation)
failed to detect any effect of the two milliped substances on neuromuscular transmis-
sion or the excitability of leg muscle cells or axons of motor nerves.
110 J.E. Carrel: Heart Rate and Physiological Ecology
Thus, it seems likely that glomerin and homoglomerinproduce sedation and cardiac
acceleration in spiders by depressing the central nervous system. Sedatives of medicinal
value also act by depressing central nervous function in vertebrates, but a recent study
shows that arborine and methaqualone, two sedatives that are structurally related to the
Glomeris alkaloids, are ineffective in spiders (Carrel et al. 1985). The physiological basis
for observed taxonomic differences in drug susceptibility is not understood.
7 Concluding Remarks
Heart rate in spiders is a highly labile parameter, one subject to influences of mimy
external and internal factors. The pronounced lability of spider heart rates in a sense
is both good and bad. On the one hand, it means that heart rate should be a very sensi-
tive and broad physiological indicator in spiders; on the other hand, it suggests that
great care must be taken by experimenters to control for the effects of foreseen and
unforeseen variables in order to obtain meaningful results.
Although much more research is needed, several published studies indicate that
heart rate in spiders may be validly used as a physiological indicator. Most important-
ly, it is imperative that the relationship between heart rate and respiration rate be
documented simultaneously in spiders while they are resting or moving.
From an ecological perspective, the most intriguing questions left to be investigated
are those that deal with the effects of various substances on spider heart rates. What
are the influences of bulk nutrients, secondary natural products, and spider pheromones
on heart rates in spiders? Perhaps the spider heart can be used to detect the presence
of pheromones, which would be of great assistance in the chemical characterization of
the pheromonal substances. Furthennore, what is the interplay between exogenous
chemicals and endogenous neuroendocrine products in spiders? Chemical aspects of
the physiological ecology of spiders merit much more attention in the future.
Acknowledgments. I thank Mary Haskins, Jan Weaver, and Amy Slagle for reading drafts of the
manuscript.
VIII Moulting Hormones
JEAN-CLAUDE BONARIC 1
CONTENTS
Introd uction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
2 Existence of Moulting Hormones in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
3 Variations in the Levels of Endogenous Ecdysteroids . . . . . . . . . . . . . . . . . . . . . . 113
3.1 Intermoult Cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
3.2 Winter Diapause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
4 Effect of Exogenous Ecdysteroids on Development . . . . . . . . . . . . . . . . . . . . . . . 115
4.1 Ecdysteroids and Moulting Phenomena . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
4.2 Ecdysteroids and Diapause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5 Juvenile Hormones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
1 Introduction
The importance of ecdysteroids in the control of moulting phenomena is well known

in insects and crustaceans. Many publications show the wealth of our knowledge in
this field. However, research on the hormonal physiology of certain groups ofmandi-
bulate arthropods such as myriapods and most chelicerate arthropods began late and
only a few contributions have been published for these two groups. In 1968, Krishna-
kumaran and Schneiderman were pioneers in noting the action of exogenous ecdy-
steroids on the moulting process in the spiders Araneus comutus and Dugesiella hentzi.
But it took 10 years for steroid hormones to be detected and dosed for the first time
in an arachnid: the spider Pisaura miTabilis (Bonaric and De Reggi 1977). Since then,
the development of research has made it possible to extend our knowledge to the other
orders of arachnids, such as mites, harvestmen and, to a lesser degree, scorpions.
2 Existence of Moulting Hormones in Spiders
Using the technique recommended by De Reggi et al. (1975), Bonaric and De Reggi
(1977) estimated the level of endogenous ecdysteroids by radioimmunoassays (RIA)
carried out on total extracts of Pisaura miTabilis. From a quantitative point of view,
the levels of ecdysteroids noted in spiders vary as a function of the physiological state
of the subjects tested. The minimum values are close to 100 pmol/g (or SO p.g/g) ,
Laboratoire de Zoologie, Universite des Sciences et Techniques du Languedoc, Place Eugene

Bataillon, F-34060 Montpellier, France
112 J.-C. Bonaric
Table 10. Ecdysteroid levels in chelicerate arthropods a
Species Ecdysteroid Method References

levels
ARACHNIDA
Araneae
Pisauridae Pisaura mirabilis 750 ng/g RIA Bonaric and De Reggi
(1977)
Acari
Ixodidae Amblyomma hebraeum 236 ng/g RIA Delbecque et al. (1978a)
(14 ng/tick)
Argasidae Ornithodoros moubata (12 ng/tick) RIA Germond et al. (1982)
MEROSTOMATA
Xiphosura Limulus polyphemus 27 ng/g RIA J egla and Costlow
(1979a,b)
36 ng/g Bioassay Jegla (1974); Jeglaet al.
(1972)
PYCNOGONIDA Pycnogonum littorale 150 J.Lg/g RIA Behrens and Biickmann
(1983)
a The levels of ecdysteroids are the maximum values noted during development and expressed as
an ecdysterone equivalent.
whereas the maximum values reach 1000 to 1500 pmol/g (or 500 to 750 Ilg/g). The
combination of thin-layer chromatography (TLC) with RIA dosages carried out on
each fraction separated by chromatography, shows the existence of positive RIA
material in the zones corresponding to the a- and i3-ecdysones. i3-ecdysone or 20-0H-
ecdysone is the moulting hormone which is quantitatively predominant.
Moulting hormones have also been detected in other chelicerates (Table 10). Ecdy-
steroids are present in nymphs of the ticks Amblyomma hebraeum (Delbecque et al.
1978a) and Ornithodoros moubata (Germond et al. 1982) at levels comparable to
those found in Pisaura mirabilis. Romer and Gnatzy (1981) showed in vitro synthesis
of a- and i3-ecdysone by the oenocytes of the harvestman Opilio ravennae; these ecdy-
steroids are also present in the ovaries and tergites of the opisthosoma. Ecdysone
probably also occur in the hemolymph of scorpions (R. Stockmann 1985, pers. com-
mun.), but no precise values are known.
Jegla et al. (1972, Jegla (1974) and Winget and Herman (1976) have detected ecdy-
sones in Limulus polyphemus larvae by bioassay and confirmed it by RIA (Jegla and
Costlow 1979a,b; Jegla 1982). More recently, Behrens and Buckmann (1983) isolated
and identified two ecdysteroids from juveniles and adults of the pantopod Pycnogonum
littorale. These results filled a gap, as pycnogonids were the only class of chelicerates
within which moulting hormones had not yet been investigated. The ecdysteroid levels
found in pycnogonids (150 Ilg/g) are much higher than quoted in the other classes of
arthropods at present being tested [see Porcheron (1979) for insects and myriapods,
and Spindler et al. (1980) for crustaceans]. According to Behrens and Buckmann (1983)
such levels have only previously been found in adult female Locusta migratoria (Orthop-
tera) (Lagueux et al. 1977).
Moulting Hormones 113
3 Variations in the Levels of Endogenous Ecdysteroids
3.1 Intermoult Cycle
RIA analyses of extracts ofPisaura mirabilis , nymphal stage 8, show that ecd ysteroids are
constantly present during the intermoult cycle. The fluctuations in hormone levels are
shown in Fig. 33; they follow several phases. During exuviation El, the level of ecdy-
sones is relatively low (mean: 150 pmol/g); in this case, the analyses were made on
individuals sampled exactly at ecdysis. From exuviation to the fifth day of the inter-
moult period (Le. during intermoult stages A and B) the hormone level increases to
200 pmol/g. During intermoult stage C such high values are not observed. From the
17th to the 22th day (intermoult stage D) a main peak of 800 pmol/g appears. The
variance of the values between the 16th and the 22nd day of the intermoult period is
due to individual variability in the duration of the intermoult cycle, which produces a
shift of the main peak. The low values recorded at the beginning ofintermoult stage A
show that the level of ecdysone drops before ecdysis.
Variations in ecdysteroid levels have been correlated with the successive events of
the tegumentary cycle (Bonaric 1980, 1982). For Pisaura mirabilis a high hormonal
peak, consisting mainly of ecdysterone, is noted during the last third of each inter-
moult period. The pre-ecdysial synthesis which corresponds to the secretion of epi-
cuticle and exocuticle, occurs during intermoult stage D in a hormonal environment
rich in ecdysteroids (Fig. 33). The induction of apolysis, at the end of intermoult
stage C, could be linked to the increase in ecdysone levels.
Generally, pre-ecdysial cuticle synthesis and high hormonal levels are concomitant.
This fact has been demonstrated for Limulus polyphemus larvae (Jegla 1982) and tick
nymphs (Diehl et al. 1982; Germond et al. 1982) among chelicerates. These correla-
tions agree with data known for insects, crustaceans and myriapods (Willig and Keller
1973; Delbecque et al. 1978b; Joly et al. 1979; Sedlak and Gilbert 1979).
3.2 Winter Diapause
Using pooled spiders from the south of France, Bonaric (1980) monitored the fluctu-
ations in ecdysteroid levels during the winter diapause in Pisaura mirabilis. The persis-
tence of a low hormone level (300-400 pmoljg) from December to mid-February
characterises the diapause period (Fig. 34). This level is close to that which is usually
found during the intermoult stage C for active Pisaura nymphs (100-200 pmol/g,
Fig. 33). Towards the end of March, the hormone level suddenly increases, to attain
average values of 2000 to 3000 pmol/g. This increase may be caused by an increase in
temperature. The level of endogenous ecdysteroids then rapidly decreases to average
values of 700 pmol/g for sub adult males, and 1000 pmol/g for subadult females.
The existence of low endogenous ecdysteroid levels in diapausing nymphs of Pisaura
mirabilis agrees with result obtained during the diapause of butterflies, wasps and
crickets (Mauchamps et al. 1977; Claret et al. 1978; Ismail et al. 1979; C.N. Mc Daniel
1979). Wright (1969) working on Dermacentoralbipictus and Sannasiand Subramonian
(1972) on Rhipicephalus sanguineus, obtained a hormonal rupture in the larval diapause
114 J .-C. Bonaric
Intermoult ing
Ecdysteroids period
levels (days)
( pmol./g)
,, C(AI)/
.· ,
,
,.'t!
'
'
35
1500
,
~,
, ,
,, ,,
.,, 30
,, ''
..' ,1I "..", , , , . .f---- .
,.tr---- -----_ ..b,,'
..-
I /'
IT - - - - .....
,
~l
_
::
II
: :
I .
25
::I :
I :I T' ~(ECd.)tI
I .. I . :
! : : J.
20
1000 t~
: ______ ~ ........_______~r
11
I *, :
1 1
EX OGENOUS ECDYSTEROIDS 15
I
500
· .. 10
,
,
1
\J
'NOOG'NOUS ECOYOTEROIDS
r-t
5
111 1 r r•...J
}' -·J~""---' ... 1
1 _f I ... T
....,. "-'r" l--'r---t"'
T
1 1 -I
1·"t··l'--t·-·:
.............
l
I
I
OL-~~~~~~~~~~~~~-7,~~~~~~~~~~~~O
10 II 12 n ,. TI II 11 II " 20 IT :(d)
t· t
II
t
III IV
t V
':
Epi
TEGUMENTARY CYCLE
E1,
1
~.'. '
~. - , , , .
Ep
A_B_ _-----C----- - :
_0_
:
Ecdysis Apolysis Ecdysis
Fig. 33. Ecdysones and moult cycles of Pisaura mirabilis. Endogenous ecdysteroid levels (pmol/g
fresh weight, nymphal stage 8); effect of exogenous ecdysterone on the duration (d) of the inter·
moult period after injections (J - V) of 0.8 .ul of an alcoholic solution of ecdysterone (50 .ug/g)
(E) compared with a control (C) which received alcohol only; tegumentary cycle of nymphal stage 8
during the intermoult stagesA -D, tegumentary layers (Ep epidermis; En endocuticle;Ex exocuticle;
Epi epicuticle). (From Bonaric 1980)
Ecdysteroids
levels
( pmol'/g)
Jf. 3000
Temperature I,
(OCI
'I
' I
I I
,,
I I
,
\
I ,
I I
, I
20 , I ,
2000
,
\
I \
I
'\
I \ / .
15 I \ /
: y
I \ •
10
.-.-~.-'"·j \!} 9
f 1000
. /
" 7i 800
*d'
. _/\ I
• •......... / . I
V · "
600
I
*-- - - --- -*
200
o o
o F M A M
Fig. 34. Fluctuation of the levels of ecdysteroids during winter diapause in Pisaura mirabilis (broken
line). (From Bonaric 1980). Each sample point corresponds to the level of ecdysteroids in a batch
of 10 animals. The mean temperatures for periods of 10 days and the capture of spiders took place
in 1977-1978 (unbroken line)
of these mites atter tOPICal applications of ecdysteroids. These experiments allow the
prediction of a hormonal deficit in diapausing animals.
4 Effect of Exogenous Ecdysteroids on Development
4.1 Ecdysteroids and Moulting Phenomena
Krishnakumaran and Schneiderman (1968, 1970) first tested the effect of the moulting
hormones of insects (a- and (3~cdysones) as well as their synthetic analogues on cheli-
cerates. Injections ofthese hormones induced moulting in Araneus cornutus (Araneidae)
and Dugesiella hentzi (Theraphosidae), as well as in Limulus polyphemus. Pisaura mira-
bilis was used as a spider model and was studied more intensively. Knowledge of the
life cycle and the number and duration of the post-embryonic and intermoult stages
has made this very common species an ideal animal for physiological experiments. In
his thesis, Bonaric (1980) described experiments which show the effect of {3~cdysone
on moulting processes. The injection of an increasingly concentrated alcoholic solution
of ecdysterone made it possible to trace a dose~ffect curve (Fig. 35). The hormonal
effect (measured as the induction of moulting phenomena) is optimal at the relatively
strong dose of 50 Mg/g. Doses exceeding 100 Mg/g may prematurely trigger phenomena
116 J .-c. Bonaric
Ecdysterone (1'919)
•
..
Cl
C2
200 ......
100 .....
50
-
10
o 10 20 30 40 50 60 Post . p . (d)
Fig. 35. Effect of exogenous ecdysteroids on the duration of the moulting cycle in Pisaura mirabilis:
shortening of the postoperational period by injections of ecdysterone solution of increasing con-
centrations. (From Bonaric 1980). Post.p.(d) ,' Average duration expressed in days of the postopera-
tional period (i.e. the interval of time between the date of the hormone injection and that of ecdy-
sis induced by this injection). The horizontal lines represent the standard deviations ; Cl controls
without injection ; C2 controls which received an injection of ethyl alcohol (10%). Ecdysterone :
concentrations of alcoholic solutions of ecdysterone (/olg ecdysterone/g spider)
preparatory to moulting , but the exuviations are not normal and, in most cases, result
in the death of animals (Bonaric 1976, 1977a). The new cuticle is partly secreted, but
the rejection of the old cuticle is not ensured . Perturbations in the moulting cycle
resulting from a hormonal supply vary depending on the moment of injection (Fig. 33).
During intermoult stages A and B, any inj~ction of an ecdysterone solution (50 /1g/g
fresh weight) produces a lengthening of the intermoult period. The same dose of ecdy-
sterone injected during intermoult stage C, i.e. during the period of cuticle stabilisa-
tion, induces early moulting phenomena and ecdysis itself (Bonaric 1977b, 1978).
Early ecdysis caused by the hormonal treatment occurs in a synchronised manner with-
in the same experimental batch. Asignificant shortening of the post-operational period,
and consequently of the intermoult period , is observed. The same hormone supply in
premoult (intermoult stage D), when the endogenous hormone level is relatively high,
causes only a slight reduction in the intermoult period . Exuviation only seems able to
occur satisfactorily if the hormone level remains low. Exuviation is postponed or ab-
normal when the hormone level is maintained artificially high.
Among other chelicerates, mites and scorpions have also been subjected to hormone
treatment. According to R.S. McDaniel and Oliver (1978), topical applications of 1 /11
of .B-ecdysone solution on Dermacentor variabilis nymphs (Ixodidae) do not seem to
affect the duration of nymph development. However, these treatments cause perturb a-
tions at ecdysis, resulting in an increase in death rate (approximately 50% mortality

after a ~-ecdysone treatment for concentrations of 1 to 10 p.g/p.l). The same authors
did not discern any effect of ~-ecdysone on spermatogenesis and the testicle develop-
ment of mites. In adult female Ornithodoros moubata ticks, ~-ecdysone and a phyto-
ecdysone (ponasterone A) induce supernumerary moulting and the formation of per-
fectly viable "super-ticks" (Mango et al. 1976). According to Kumar (1980), the action
of a synthetic analogue of ecdysone on the moult of the scorpion Palamnaeus benga-
lensis is ineffective, no matter what hormone concentration is used. These early experi-
ments should be re-examined to confirm or refute these unusual and surprising results.
Following the initial research of Krishnakumaran and Schneiderman (1968), the
induction of moult by ecdysteroids has been clearly shown for the primitive chelicerate
Limulus polyphemus (Jegla and Costlow 1970; Herman 1972; Jegla et al. 1972). More
recently, a biochemical approach has been undertaken with the aim of specifying the
action of ~-ecdysone. Winget and Herman (1979) studied the effect of moult cycle
stage and ~-ecdysone on protein synthesis and found that ~-ecdysone injections pro-
duced an early stimulation of protein synthesis in the midgut gland followed within
48 h, by strong inhibition. The activation of moulting phenomena observed in cheli-
cerates is comparable to that noted in mandibulate arthropods and has been thorough-
ly studied (see, for example, Slama et al. 1974; Novak 1975;J.A. Hoffmann 1980).
4.2 Ecdysteroids and Diapause
According to Bonaric (1977c, 1980), applications of exogenous ecdysterone stop the

winter diapause and renew moulting phenomena in Pisaura mirabilis. In all cases, in-
jections of ecdysterone (doses of 50-100 p.g/g) induced an interruption of the diapause.
These experiments repeated at different periods of "diapause development" show a
seasonal variation in the sensitivity of spiders to the moulting hormone. Table 11, which
summarises results of these experiments, shows a significant reduction in the response
time of spider nymphs to the same hormone treatment between the beginning and the
end of diapause. These results are comparable to those that Bodnaryk (1977) obtained
on diapausing Mamestra conf;gurata butterfly pupae. The differences in sensitivity to
ecdysterone of diapausing animals reflect a gradation in the intensity of diapause during
this period and make it possible to derme several distinct states (prediapause-diapause-
postdiapause). The moulting hormones (a- and ~-ecdysone), by topical application or
Table 11. Variation in the response time of diapausingPisaura mirabilis nymphs (stage 7 and 8) to
an injection of ecdysterone (50 to 100 p.g/g) during winter diapause a
November December/January February
Duration of the 8.39 ± 1.01 (20) 7.68 ± 1.37 (17) 6.30 ± 2.63 (20)
postoperational
period (d)
X ± SO (total N)
a Significant differences exist between the duration of the postoperational period of the first and
third series (p < 0.05).
118 I.-C. Bonaric: Moulting Hormones
by immersion, also induce a rupture ofdiapause in larvae of the ticksDermacentoralbi-

pictus (Wright 1969) and Rhipicephalus sanguineus (San nasi and Subramonian 1972).
Although experimental conditions were different in most cases, an injection of ecdy-
sone also caused a rupture in diapause of insects (Chippendale 1977).
Generally, the results obtained on the hormone physiology of spiders or arachnids
are comparable with those for most arthropods. Ecdysones, originally regarded as
moulting hormones in insects alone, may now be considered moulting hormones for all
arthropods. According to Bonaric (1980), the ecdysial gland of spiders ("endocrine tis-
sue", Millot 1930a) could be one of the synthesis sites of ecdysteroids. In Pisaura mira-
bilis this prosomatic gland presents ultrastructural attributes of typical steroidogenous
tissues and a cyclic secreting activity which may be correlated with the hormonal peaks
recorded at different stages of intermoult. According to Romer and Gnatzy (1981),
other tissues, such as the oenocytes and ovaries, also playa role in the synthesis of
ecdysones in harvestmen and perhaps in arachnids in general.
5 J uveniIe Hormones
In insects, moulting hormones are often studied in association with juvenile hormones.
These two hormones are the main regulators of growth. Whereas research on the detec-
tion and determination of the role and metabolism of juvenile hormones is intensive
for insects (Pratt and Brooks 1981), it is much less so for other classes of arthropods.
Most are limited to studying the effect of exogenous juvenile hormone on growth and
reproduction. In che1icerates, the analogues of insect juvenile hormones are used in the
control of mites that are animal parasites or crop pest (Klein-Koch 1976; El Banhawy
1977; Gothe and Morawietz 1979). Treatments based onjuvenoids, carried out byim-
mersion or topical applications, cause perturbations in development, mainly due to
their antigonadotropic effect (Mansingh and Rawlins 1977; R.S. McDaniel and Oliver
1978; Pound and Oliver 1979). However, to our knowledge, juvenile hormones have
not been detected in spiders.
B Glands, Silk and Webs
I Glandular Aspects of Sexual Biology
ANDRE LOPEZ 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
2 Sperm Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2.1 Testes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
2.2 Vasa Deferentia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
3 Sperm Deposit Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.1 Sperm Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
3.2 Epigastric Glands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
4 Sperm Induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
4.1 Receptaculum Filling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
4.2 Palpal Chewing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
5 Courtship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.1 Silk Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2 Male Pheromones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.3 Female Pheromones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
6 Copulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.1 Precopulatory Operations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
6.2 Insertion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
6.3 Secondary Ejaculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
6.4 Post-Insertion Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
7 Post-Copulatory Glandular Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
1 Introduction
The sexual biology of spiders is dominated by the anatomical constraints on the male
to transfer its sperm, and by an obligatory courtship to bring about mutual arousal and
reduced aggressivity before mating can occur. There is total separation between the
male's genital tract, located in the opisthosoma, and the copulatory bulb carried by
the pedipalp, a prosomal appendage. As a result, the male must achieve a highly unusual
"sperm transmission" (Gerhardt and Kaestner 1937) that includes a full series of opera-
tions. The sperm cells produced in the testes and conditioned in the vasa deferentia
must firstly be discharged from the gonopore onto a sperm web (primary ejaculation)
and then sucked up (sperm induction) into the receptaculum seminis or spermophore.
1 Laboratoire Souterrain du C.N.R.S. Moulis, F-09200 Saint-Girons, France, and

Laboratoire de Pathologie Comparee, E.P.H.E., Universite du Languedoc, Place E. Bataillon,
F-34060 Montpellier, France
122 A. Lopez
This tubular organ, coiled inside the copulatory bulb, is formed by a cuticular wall and
an epithelium that appears as an integumentary invagination. The spermatozoa stored
in the receptaculum are finally injected (secondary ejaculation) into the female genitalia
during copulation.
The courtship that sperm induction precedes or accompanies, reduces predatory
tendencies and promotes the necessary mutual excitation, as pOinted out in M.H. Robin-
son's review (1982b). Schematically, it can be elicited by the direct contact of the male
with the female's body, on the occasion of a chance meeting, at a distance by volatile
pheromone or visual cues, and lastly, by a contact pheromone associated with the silk
(dragline, capture web, rest chamber), or simply with the substratum.
The spider's body incorporates an impressive quantity and diversity of gland cells,
located either in definite organs (intestine, tegument) or constituting a differentiated
secreting tissue as in the silk gland epithelium. It is not surprising that many of them,
nearly all epidermal in origin, are involved in "sperm transmission" and also regulate
the courtship process. Ignored in most recent books dealing with spider biology, they
confer on the spider's sexual life the partiCUlarly striking glandular aspects that anatom-
ical evidence supports and form a part of chemical ecology on the level of intraspecific
interactions.
2 Sperm Conditioning
Secretory processes can already be identified in the internal sexual organs of male spi-
ders.
2.1 Testes
The lumen of testes cysts encloses non-motile disc-shaped sperm cells characterized by
an unusual coiled axoneme with a 9 + 3 microtubules pattern. Moreover, they are
generally surrounded by individual proteinaceous capsules that they will retain unaltered
throughout the whole "sperm transmission". These capsules result from the condensa-
tion of a fine granular material which is produced by cyst wall supporting cells whose
cytoplasm includes a rough endoplasmic reticulum and Golgi apparatus in Telema tene!-
fa (Lopez, unpubl.), Oxyopes and probably in other spiders. The secretion elaborated
by the testes of various Mygalomorphae (De Carlo 1973), Segestria and Dysdera (Lopez
and Juberthie, unpubl.) is profuse, strongly reactive when using the periodic acid-Schiff
procedure, and encases several spermatozoa within a mucoproteinaceous common cap-
sule, clustering them into "spherules". They recall somewhat the spermatolophids of
Thysanura and the rounded sperm mass of Odonata (Insecta).
2.2 Vasa Deferentia
Glandular activity cues are seen again in the vasa deferentia. Those of Te!ema secrete a
genuine prismatic gutterlike spermatophore (Lopez 1977b) (Fig. 36a) in which the
above encapsulated spermatozoa are regularly piled. Epithelial cells are richly provided
Glandular Aspects of Sexual Biology 123
Fig. 36. a Vas deferens epithelium and spermatophore of Telema tenella: cross-section (x 5,500).
b Secretory epithelial cells of vas deferens in Telema (x 15,800). c Adenocyte of an epigastric gland
in Pholcus phalangioides (x 11,000). d Specialized epithelium of the receptaculum seminis in Hersilia
savignyi (x 8,800). For abbreviations see Fig. 37
124 A. Lopez
with dictyosomes and endoplasmic reticulum (Fig. 36b) elaborating a glycoprotein-

aceous material that integrates into the trilayered wall of the spermatophore (Juberthie
et al. 1981). It seems that testis secretion affords nutriments and true isolation to
spermatozoa, whereas the vas deferens is more likely to supply them with a wrapping
matrix or an elaborated spermatophore, thus forming a common sperm mass that can
glide and be handled better by the male.
3 Sperm Deposit Support
3.1 Sperm Web
The sperm mass is ejaculated onto the surface of the sperm web, a small silken struc-
ture, whose shape and thread number are as variable, according to' the species, as the
site (Lopez 1977a) that in Zoropsis may even include the female's body itself (Bristowe
1941). It is probably produced by the pyriform glands.
3.2 Epigastric Glands
Just after spinning and prior to ejaculation, the male completes its sperm web by the
deposit of an additional substance that is visible as a white sheet in Theraphosidae
(Melchers 1964). This deposit is considered, by extrapolation, as highly conceivable
in other spiders, either during sperm web construction (Blest and Pomeroy 1978), or
when abdomen and gonopore are rubbed against the web edge for primary ejaculation
(Lopez 1977a). The additional material issues from the ventral fusules of a median
gland cluster that is located in the anterior lip of the epigastric furrow. Qualified as
"epiandrous" (Marples 1967) or better "epigastric", these pregonoporal acinous glands
have been found in many spider families (Lopez 1977a). They are multicellular and ap-
pear closely related to silk glands in their histological peculiarities and rme structure
(Fig. 36c). In some families (Clubionidae, Segestriidae), they are replaced by or as-
sociated with unicellular glands, called "gonoporal", scattered or gathered in one pair
of clumps that open in the epigastric furrow via individual excretory dt:~ules (Lopez
1977a); their fine structure is still poorly investigated.
Undoubtedly, the sperm web represents an isolating support for the semen, but it
does not replace a spermatophore, for it is devoid of an organized wall. In contrast, the
secretion from the epigastric glands could be a shapeless equivalent of the spermato-
phore (Lopez 1977a), doubly so as these organs are missing in Telemidae. Its precise
role has not yet been elucidated. The similarity of the pregonoporal glands to the ventral
organs of Palpigradi is insufficient to allow us to establish a relationship since the bio-
logy of these Arachnida is practically unknown. The concept of nutriment supply to
already encysted sperm cells is no more convincing than pheromonal activity that labels
the substratum or attracts the female to the sperm deposit, as in the case of armex
lateral glands of male Pseudoscorpiones. A role affecting the male's organism itself by
reintegration at the time of sperm induction or later, on the female, is purely speculative
(Lopez 1977a).
4 Sperm Induction
4.1 Receptaculum Filling
This is the consequence of a secretion transfer in its wall. Histological and frne structure
studies (Lopez 1977a; Lopez and Juberthie-Jupeau 1985) indeed show here that the
epithelium is specialized and separated by an "outer palpal room" from the cuticular
wall. This wall is imperforate in Heptathela kimurai (Lopez, unpubl.) and Leptoneta,
and fenestrate in other spiders. Fenestrations either involve the whole depth (Telema)
or the endocuticle only (other spiders); in the latter case, the epicuticle shows tiny
apertures (NemesUz, Segestria, Hersilia). The epithelial cells, which also demonstrate an
intense secretory activity, are plentifully provided with apical microvilli and endo-
plasmic reticulum (Fig. 36d) between which endocytosis ensures a transit of fluid. They
can thus reabsorb secreta previously saturating the cuticle when imperforate (Hepta-
thela, Leptoneta) or flowing into all the spaces that underly and pierce it (other spiders).
So, the secreta bring about a depression in the lumen of the receptaculum in which the
seminal mass is drawn up, perhaps together with epigastric secretion when bulb filling
is performed through the sperm web silk (indirect induction). Such an active resorbtion
mechanism compromises the theory of "capillary forces".
4.2 Palpal Chewing
Various records on courtship preliminaries allude to palpal "chewing" (also called

"preening", "cleaning" or "nibbling") occaSionally prior to sperm web construction
itself (Blest and Pomeroy 1978), more often just before induction as in Stegodyphus,
or, preferably, after it (Blanke 1973b). This behaviour is found again during courtship
and copulation. The male spider applies a secretion from mouth parts on each paIp
drawn between the chelicerae. This secretion is undoubtedly provided by epidermal
cushions (Lopez et al. 1980) or by the acinous glands of the maxillae. Originally con-
sidered to be no more than salivary organs, these gnathocoxal or maxillary glands can
exhibit clear sexual dimorphism. In the males of certain Araneidae, of Linyphiidae
(Lopez 1977a) and Leptonetidae (Lopez and Emerit 1978), the majority of the glands
are tubuloacini called "sexual". They are distinguished from "classical" maxillary
glands, either by clear vesicles that fill a hyaloplasm showing free ribosomes in Lep-
ton eta (Lopez et aI. 1980), or by acidophilic secretion grains that, at least in Linyphi-
idae (Lopez and Juberthie-Jupeau, unpubl.) are strikingly complex (Fig. 37a) or whorl-
like, and elaborated by an extensive rough reticulum. Moreover, their large size makes
them protude beyond the gnathocoxae (Leptonetidae, Araneidae); sometimes, there is
even a surprising increase in the number of acini that spread far back into the prosoma,
and an elongation of their excretory ducts that are organized into bundles (Linyphi-
idae). Their excretory pores are differently located from "classical" gland pores in
Leptonetidae (Lopez and Emerit 1978). Such sexual dimorphism evokes to some ex-
tent that known for insect salivary glands in Panorpidae (Mecoptera) and in Stilbocoris
(Heteroptera, Lygaeidae). But it does not seem to correspond to the same behavioural
feature of a saliva ''wedding gift". The palpal chewing that is performed before or at
the time of sperm induction, could only moisten the bulb and allow its embolus to be
126 A. Lopez
Fig. 37. a Secretion grains in a maxillary gland cell of Lepthyphantes sonctivincentii (male) (x 8,800).
b Part of a pheromone gland cell in the cymbium of Nephila inaurata (male) (x 24,400). c Aleft
lateral view of the prosoma cephalic part in Argyrodes nephilae (male from La Guadeloupe Island)
(x 150). d Part of a clypeal adenocyte in Argyrodes argyrodes (x 18,000).
Abbreviations : Couter palpal room; D dictyosome or Golgi area ; EA end apparatus; EP epithelium;
G secretion grain ; J chelicerae; K knob; L lumen of the end apparatus; M mitochondria;N nucleus;
R endoplasmic reticulum;S sperm cell; T notch; V microvilli; W trilayered wall of the spermatophore;
Yeye
plunged more easily into the sperm mass just prior to suction. Pal pal chewing appears
as a rather common operation also performed by male spiders devoid of sexual maxil-
lary glands (Eresidae, Clubionidae, Lycosidae).
5 Courtship
Courtship involves tacto-chemical information that leads to a sexual encounter. Ac-

cording to various studies, such information consists of signals that are mediated by
silk and pheromone secretions, bound or separate. The partners perceive them by ol-
factory organs or by leg and palp chemoreceptors (see also Pollard et al., this volume).
5.1 Silk Utilization
The male often produces silky devices specially used to facilitate its approach to the
female and reduce her predatory tendencies, even when the species does not use silk in
predation. Mating or copulation threads (Araneidae) are secreted by ampullate glands
as are draglines. Partial wrapping of the female as though she were a prey (Meta seg-
mentata) and the wrapping of prey items offered to the female as a "wedding present"
(Pisaura mirabilis) are probably executed by the male with the silk of his aciniform
glands. This behaviour resembles the prenuptial gift custom of many empidids (Diptera).
In various Salticidae and mainly in Clubionidae, the male builds a chamber near the
female's nest and thus cohabits until mating. In Clubiona phragmitis (Wasowska 1969)
and Chiracanthium punctorium (Lopez, unpubl.), its silk could be supplied by special
glands ending in the fore spinnerets and distinct from pyriform glands.
5.2 Male Pheromones
There is rarely any evidence of sex pheromone production by male spiders. This phero-
mone has been said to be bound to the male's web threads in Latrodectus hesperus
(Ross and Smith 1979) or to the substratum that the male of Tegenaria walks on
(Leborgne et al. 1980). As we shall see further in females, the candidate secretory cells
could be located in the ducts of the silk apparatus and in the tegument, even of the
palpal cymbium (Fig. 37b).
The "cleaning" of the palps and forelegs in Araniella cucurbitina (Blanke 1973b)
seems again to denote a secretion from the gnathocoxal glands as early "long-distance
courtship". Similarly, a secretion of small sausagelike bodies appears from the male's
mouth of Teutana triangulosa (R. Braun 1956). These could carry a pheromone de-
posited on the mating thread or female's web, or, more simply, could denote the ex-
citation of the male triggered by the attractive pheromone of the female.
5.3 Female Pheromones
Adult females, virgins included, indeed elaborate attractive substances that elicit court-
ship behaviour in males. These may be volatile and aerially efficient at a distance as in
128 A. Lopez
the genera Cyrtophora (Blanke 1975a), Schizo co sa (Tietjen 1979) and perhaps Lepthy-
phantes (Kovoor et a1. 1981). Candidate glands for their secretion are located in the
legs, in the ventral tegument of the prosoma (chelicerae, labium, maxillae), of the
pedicle (Uroctea) and of the opisthosoma, namely on the epigynum, the lips of the
epigastric furrow and the spinnerets (Kovoor 1981 ;Kovoor et al. 1981; Lopez, unpubl.).
These glands are scattered between the epidermal cells or gathered into small clumps
readily showing a segmentary repartition in the prosoma of both sexes ("coxal glands"
sensu lato: Lopez 1983a). Each of them appears composed of an adenocyte that in-
vaginates into an extracellular cavity bordered with microvilli, by a sieve like walled
"end apparatus" (Fig. 37b), by an excretory ductule piercing the cuticle and by its
canal cell (Lopez and Juberthie-Jupeau, unpubl.). They are, therefore, related to class
3 glands as defined by Noirot and Quennedey (1974) in insects.
Generally, sex pheromones are of the surface or contact type, spread over the fe-
male's integument (Thomisidae, Clubionidae), deposited onto the substrate or associ-
ated with the silk of draglines (Lycosidae, Agelenidae), of chambers (Salticidae, Clubi-
onidae) and of the capture web itself (Krafft and Roland 1980; Leborgne 1981; M.H.
Robinson 1982b; Jackson, pers. commun 1985). In the first two cases, the abovemen-
tioned tegumental glands could be the origin of pheromones. On the other hand, silk-
labelling candidate glands are more likely to be located in the silk apparatus, a theory
substantiated by the presence of peculiar gland cells qualified as "lageniform" in the
epithelium of ampullate excretory ducts. These elongate bottle-shaped adenocytes
which interpierce the epithelial cells, have been observed in Cyrtophora and Argyrodes
(Lopez, unpub1.).
6 Copulation
6.1 Precopulatory Operations
Just before copUlation, males may bring secretions into play from silk glands or from
clypeal glands, apparently to restrain the female, already or not yet subdued by court-
ship.
6.1.1 Silk Glands
The male of various Xysticus species (Thomisidae) spins the "bridal veil": silk threads
that pyriform or aciniform glands provide, are drawn over the female's legs and pro-
soma, and fastened to the ground (Bristowe 1941). The male of Nephila maculata
(Araneidae) ties the leg bases together and coats the dorsal surface of its mate with an
intermittent thread deposition (M.H. Robinson and B.C. Robinson 1973a). The func-
tion of silk binding is not yet satisfactorily elucidated. It seems to provide an effective
hobbling that prevents the female from seizing the male, but mechanical constraint ap-
pears less plausible than a selective inhibitory effect of sensory hairs bent by the silk
threads.
6.1.2 Clypeal Glands
In various Argyrodes species (Theridiidae) and Erigoninae (Unyphiidae), the cephalic

part of the male's prosoma is often strikingly modified by deformations such as knobs,
notches and sulci (Fig. 37c) and were considered for a long time to be unprofitable
"nature fancies". In fact, as far as we know, the protuberances are gripped for copula-
tion by the chelicerae of the female. Thus immobilized, she seems to suck a trapped
fluid from the notches and sulci (Meijer 1976; Blest and Taylor 1977). Such behaviour
is somewhat reminiscent of Ischyropsalididae and Nemastomatidae (Opiliones) whose
females eat the secreta of the male's cheliceral gland. Similarly, females of various in-
sects, such as Oecanthus (Orthoptera) and Blatella (Blattodea), ingest the secreta of
the male's tegumentary glands. In male spiders, fluids are secreted by underlying "cly-
peal glands" described for the first time in Argyrodes (Lopez 1977a), either scattered
in the front epidermis and opening separately onto the surface, or gathered into bulky
organs provided with a system of confluent excretory ducts. As observed in Argyrodes
(Juberthie and Lopez 1980) and Erigoninae (Lopez, unpubl.), they present a common
basic fine structure, the glandular unit. Analogous to supposed pheromone glands and
again showing points of similarity with class 3 epidermal glands of insects (Noirot and
Quennedey 1974), each unit includes an adenocyte with an extensive granular reticulum
producing secretion, a sieved "end apparatus", an excretory ductule and canal cell(s)
(Fig. 37d). It probably also secretes an aphrodisiac that, once directly absorbed by the
female, could inhibit her aggressiveness. The secretion of a volatile pheromone still re-
mains possible when deformation is so eccentric that it makes fang grasp anatomically
unrealizable (Walckenaera). According to Blest and Pomeroy (1978), clypeal secretions
of Erigoninae could be derived from defensive substances produced initially by both
sexes, as in Mynoglenes.
6.2 Insertion
The male's palp could be directed to the female's copulatory duct by a pheromone
produced by spermathecal glands (Kovoor 1981) and detected by the chemoreceptors
of its cymbium. It is possible that consecutive bulb insertion would be facilitated by a
coating of its embolus with a "lubricant". This substance could be secreted either by
class 1 gland cells that border the copulatory ducts (Kovoor 1981), or by maxillary
glands during renewed "palpal chewing" that authors frequently report during copula-
tion in Araneidae, Unyphiidae, Lycosidae and Clubionidae. However, they point out
that the embolus is "chewed" after each palp first becomes loose from the female's
genitalia. One can thus infer that "lubrication" is only available in iterative copulations
(Linyphiidae, Araneidae) before the embolus is reinserted in the same female or others.
The secretion flow is sometimes so profuse (Lepthyphantes Zeprosus) that mouth clean-
ing with legs becomes necessary (van Helsdingen 1965). It does not seem to be eaten
by the female as is the saliva in Panorpidae (Mecoptera).
130 A. Lopez
6.3 Secondary Ejaculation
The sperm discharge and its forcing into the bursa copulatrix or copulatory ducts are
due to the secretory activity of the specialized bulb epithelium. Its cells, in all cases
devoid of individual excretory ductules, show, besides the extensive reticulum, a number
of conspicuous typical dictyosomes (Fig. 36d) more or less concentrated near the
nucleus. Round secretion grains with variable electron density, originate from these
organelles and are subsequently extruded into the outer palpal room. The fluid, which
they constitute, carries mucosubstances; it either saturates the receptaculum cuticle
that is swollen in proportion (L epton eta, Heptathela) or travels along the fenestrations.
In the latter case, it occasionaliy crosses the whole wall (Telema) and overflows into
the lumen as authors have generally reported. More often, it circulates in fenestrations
and presses against the epicuticle so that it more or less separates from the wall and
pushes forward (Lopez and Juberthie-Jupeau 1985). Whatever the case, hyperpressure
always occurs in the lumen, pushes on the sperm mass and forces it out of the recepta-
culum.
6.4 Post-Insertion Events
Several observers have emphasized the presence of an amorphous secretionlike material

blocking the epigynum of just-mated females, either as a film or as a variably coloured
and shaped plug. It has been observed in several families, such as Salticidae (Jackson
1980), Oxyopidae (Yoshikura 1982) and Theridiidae, for example Steatoda triangulosa
(R. Braun 1956) and the genus Argyrodes (Lopez, unpubl.). This "mating plug" or
"chastity belt" could prevent the female from further mating, in the same way as the
breaking of the embolus tip in her genitalia (Argiope, Nephila), or the sticking of the
cap that normally cloaks the virgin male's embolus tip in some Araneidae (Levi 1975a).
It can be considered as an adaptation related to "sperm competition" (Jackson 1980),
"sperm priority" and "paternity assurance" in spiders (Austad 1982), and is found
also in some insects (Diptera). One can assume that the mating plug is produced either
at the end of ejaculation by a part of the palp gland when this organ is bulky and com-
plex as in Argyrodes, or by maxillary epidermal cushions when these are extremely
well-developed as in Steatoda (Lopez, unpubl.). Maxillary acinous gland intervention
appears more unlikely as these organs are unmodified and even poorly developed in
such spiders.
Concerning maxillary glands, an important finding in Araneidae is the type of alterna-
tion taking place between them and the embolus cap (Lopez 1977a). "Sexual" glands
are present in the maxillae of the male when he lacks a cap and he can, therefore, re-
peatedly copulate with the same female or several partners (Araniella, Nuctenea, Acu-
lepeira, Kaira). Inversely, maxillary glands are banal in appearance when there is an
embolus cap and when the male can thus mate only once with each palp (Araneus,
Neoscona). The chances are poor that special maxillary secretion replaces the miSSing
cap. It looks as though this secretion is used either to coat the genital bulb that is main-
tained and lubricated for future insertions or, in Nuctenea, perhaps to attract the fe-
male which is said to solicit and approach the male (Levi 1975a).
7 Post-Copulatory Glandular Aspects
After single or reiterated copulation, all sexual activity disappears in the male spider;
the glands responsible for associated secretory processes degenerate or, at least, are
temporarily interrupted when the male lives for several years (Myga10morphae).
Glandular interferences still occur in the female's genitalia at the time of the precur-
sory steps of microscopic fecundation which has never really been observed up to the
present time. The spermathecal wall includes one to three types of secretory cells: uni-
cellular glands provided with individual ductules, epithelial elements belonging to class 1
epidermal glands (Coyle et al. 1983; Kovoor 1981; Lopez and Juberthie-Jupeau 1983).
They serve probably more than one function. Besides producing pheromone, they also
affect the stored sperm mass. They could promote a dissolution of the spermatophore
wall (Telemidae) and mainly, the lysis of the sperm cell capsules, making them motile.
A metabolic maintenance function, suggested by the presence of glycogen in secretions,
could also occur when the sperm must be kept viable in the spermatheca for several
months, as for instance in Antrodiaetidae (Coyle et al. 1983) and Mastophora (Lopez
et al. 1985); however, carbohydrates were more probably provided by the testis to each
sperm cell, prior to capsulation.
In fact, a mechanical role appears the most likely, being supported by fine structure
studies (Lopez and Juberthie-Jupeau 1983). In Telema, the epithelial gland cells of the
spermatheca show apical microlamellae and remarkably laciniated basal parts divided
by folds of plasmalemma into compartmen ts, including a great number of mitochondria.
Such fine structure implies a massive transport of water, the eruption of which in the
lumen could displace the sperm mass from the spermatheca to the uterus externus.
Here, the spermatozoa meet egg cells and mix with them more efficiently, prior to
fecundation and egg depOSition.
8 Conclusion
Glandular activities thus seem to ensure in spiders the normal course of sexual inter-
actions. These activities have been disclosed by behavioural observations and are sup-
ported by histological and ultrastructural studies. When the glands appear with the
penultimate moult and are the apanage of one sex only, they can be considered as se-
condary sexual characters (Lopez 1977a) to be quoted henceforth in zoology text-
books. This is the case, in the male, of palpal, "sexual" maxillary, clypeal, epigastric
and deferential glands and, exceptionally, specialized silk glands; and, in the female,
of epigynum and spermathecal glands. Their secretions, so far generally hypothetic,
play either a mechanical role as a supposed "lubrication", or a chemical role, and in
this case, they are relevant to pheromonal activity. Concerning the chemical role, class 3
glands found in the body and legs of both sexes could produce a pheromone qualified
as "symmetrical" (Krafft 1980). Moreover, it is incidentally worthy of note that class 3
glands abound in the forelegs of Mastophora (Lopez et al. 1985) (Figs. 63 and 86);
they probably elaborate the allomone mimicking sex attractant of female moths which
is proof of its possible origin in the spider's own sex pheromone. Kaira alba, another
Araneidae, also produces an attractant, but it seems this time to be supplied by giant
adenocytes ofthe aggregate silk glands (Lopez and Stowe 1985).
132 A. Lopez: Glandular Aspects of Sexual Biology
Very little information is currently available on the chemical nature of secretions.

It can only be said that the clypeal and maxillary glands, whose adenocytes are rich
in rough reticulum, seem to elaborate pure protein, whereas the palpal and epigastric
gland cells, abundantly provided with dictysomes and partly smooth reticulum, are
likely to secrete a glycoproteinaceous material. It appears that the stimulation of the
glands, mainly palpal and maxillary ,is brought about by an endocrine factor. The nature
of this mediator is not currently known; nor is it yet possible to ascribe its origin to
the testes, the neurosecretory complex or any other definite organ. The control of
sexual operations that activate glandular activities is essentially central, as demonstrated
for sperm web weaving, "palpal chewing" and sperm induction by experiments with
palpless Lycosidae and Linyphiidae males (Rovner 1966, 1967).
II Communication with Chemicals:
Pheromones and Spiders
SIMON D. POLLARD, AYNSLEY M. MACNAB, and ROBERT R. JACKSON!
CONTENTS
Introd uction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2 Salticidae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
3 The Phylogenetic Significance of Salticid Pheromones . . . . . . . . . . . . . . . . . . . . . 138
4 Dysderidae, Clubionidae and Araneidae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5 Sex Roles and Pheromones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
6 Pheromones and the Evolution of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
1 Introduction
Communication occurs when one individual, the sender, changes the behaviour of
another individual, the receiver, indirectly by providing it with a special stimulus, the
signal (see J.R. Krebs and Dawkins 1984; Jackson 1982a; W.J. Smith 1977). During
intraspecific interactions, spiders employ a wide variety of signals which may be ex-
changed over chemical, visual, vibratory, tactile or auditory sensory channels. Chemical
signals, or semiochemicals, appear to be the most primitive mode of communication in
the arachnids and have probably been retained in some form in all spider families. We
will not attempt to provide a comprehensive review of work which has been done on
spider semiochemicals (see Tietjen and Rovner 1982), but instead we will emphasize
more recent work on pheromones from our own laboratory. The term pheromone is
used to describe olfactory and contact chemicals that are used during intraspecific inter-
actions.
We will concentrate on two general questions.
1. How are pheromones used in the following groups of spiders: primitive and advanced,
visual and non-visual, web-building and hunting?
2. What role does silk, the "trademark" of the spider, have in chemical communica-
tion?
Spiders use silk in a great diversity of ways (see also Heimer and Nentwig, this vol-
ume). They may, for instance, use it in dispersal by ballooning, as a wrapping for their
eggs and their prey or for constructing a nest in which to take shelter; and, of course,
many species build webs, which are silken devices used in the detection and capture of
prey. By placing silk in its environment, the spider constructs for itself the medium
across which it will receive vibratory stimuli not only from potential prey, but also from
1 Zoology Department, University of Canterbury, Christchurch, New Zealand

l34 S.D. Pollard et aJ.
conspecific spiders who are potential mates or rivals. That is, spider webs and nests are
often arenas on which intraspecific interactions take place and over which the spiders
send their special vibratory signals (M.H. Robinson 1982b). The use of silk in intra-
specific communication parallels one of the important ways in which it is used in preda-
tion: it acts as an extension of the spider's sensory system (see Witt 1975).
Our attention will be given primarily to representatives of four families of spiders:
the Salticidae, the Dysderidae and the Clubionidae, most species of which are hunting
spiders, and the Araneidae, most species of which are web builders. The terms 'hunting'
and 'cursorial' are generally applied to spiders that capture their prey without using a
web, and this covers a wide range of predatory modes from actively stalking then leap-
ing on prey, as is typical of many salticids Gumping spiders), to waiting in ambush and
lunging on passers-by, as practised by many thomisids (crab spiders). Although most of
the salticids, dysderids and clubionids do not build webs, they all use silk in varied
fashions, including the construction of nests, and representatives of each of these families
are known to use silk for vibratory communication. The salticids are of particular
interest because they are the only spiders with highly developed acute vision (Blest
1985; Land 1985). The other hunting spiders are often referred to as the 'non-visual
hunters'. The Dysderidae belong to a relatively primitive group of spiders, the haplo-
gynes, whereas the Salticidae, Clubionidae and Araneidae are relatively advanced entele-
gyne spiders.
2 Salticidae
The jumping spiders (Salticidae) might be expected to use pheromones in only a minor
way, if at all, relying instead on vision for communication. Although this view used to
be prevalent, the Salticidae is now, ironically, the family of spiders from which we
know the most about pheromones.
The diverse and complex visual displays of salticids are well known (Crane 1949;
Jackson 1982a); but, as with the non-visual hunters, salticids often interact at their
nests. The salticid's nests, however, are often situated in places, such as under rocks
or between rolled up leaves, where visual displays are not feasible. The display reper-
toires of ten salticid species have now been described in detail (see Jackson 1986a),
and the males of each of these species practise courtship versatility. Males encounter-
ing females in the open, court using visual displays (Fig. 38), but if males encounter
females within nests, they perform vibratory displays (Fig. 39) on the nest before
mating. If the female in a nest is a subadult (one moult from maturity), he will co-
habit with her: the male waits in an adjoining chamber contiguous with the female's
nest; when the female moults, he mates with her, then leaves.
Males of all but 2 of 30 tested species discriminate between vacant nests of con-
specific and non-conspecific females, responding with vibratory courtship only on the
former. Males can also discriminate between nests of conspecific males and females,
but females do not respond in any distinctive way to nests of either males or females.
Certain species of wolf spiders (Lycosidae) are known to begin courtship or searching
behaviour upon contacting the draglines of conspecific females; some even following
draglines to locate the female (Tietjen and Rovner 1982). Although there is no evidence
Communication with Chemicals: Pheromones and Spiders 135
Fig. 38. Male Euryattus sp., a salticid, performing a visual courtship display to a conspecific female.
This display is 'primed' by olfactory pheromones produced by the female
Fig. 39. Male Cyllobe/lus ru[opictus, a salticid, performing vibratory courtship on the nest of a con-
specific female. This courtship behaviour is released by pheromones on the silk of the female's nest
136 S.D. Pollard et al.
that salticids follow draglines, the males of 4 of 24 tested species begin courtship on
contacting the draglines of conspecific females. In these instances, the displays that are
elicited are those normally used in visual courtship away from the nest. The draglines
and the majority of nests tested lost their effectiveness at eliciting courtship if left un-
occupied for several days. The nests of Phidippus johnsoni and Trite auricoma were ex-
ceptional, remaining effective for up to a month. Nests washed in polar solvents (iso-
propanol and especially water), but not a non-polar solvent (hexane), were ineffective
in releasing courtship.
Oden carried out some interesting studies on the use of pheromones by Phidippus
audax (Oden 1981). For example, he allowed female spiders to occupy one-half of a
partitioned cage and to lay down draglines. Next the female and the partition were
removed, and a male was allowed to move about freely in the cage. Oden recorded the
time the male spider spent on each side of the cage and found that more time was
spent on the 'female' side than on the control side of the cage. However, he did not
state whether the draglines of females of Phidippus audax elicited courtship from the
males.
Oden also tested males with draglines that had been immersed in solvents. Isopro-
panol and warm water, but not cold water or hexane, rendered draglines of Phidippus
audax ineffective at releasing male responses. Also, draglines of Phidippus audax when
kept isolated from the spider remained effective for up to 20 days. Oden suggested that
there is a mildly polar, high molecular weight pheromone, with a low vapour pressure,
on the silk of Phidippus audax; he speculates that, although unlikely to be a steroid,
the pheromone may be a protein.
As the nests of Phidippus johnsoni and Trite auricoma remained effective, when
isolated from the spider for up to 28 days, the pheromones on the silk of these spiders
may have a similar vapour pressure to those on the silk of Phidippus audax. Pheromones
on the silk of other salticid species, however, remain effective for a shorter period if
they are isolated from the spider and apparently have a higher vapour pressure than
the pheromones of Phidippus audax.
Cohabitation is an important mating strategy in the salticids (Jackson 1986a)
and apparently there are pheromones on the silk of adult and subadult females (Jackson,
in press). In the laboratory and in the field, cohabitation by male-female pairs of
Phidippus johnsoni usually occurs within 10 days of the female's final moult (Jackson
1978b), and other species seem to be similar in this respect (Jackson 1986a). In-
terestingly, in recent experiments on Phidippus johnsoni (Jackson 1986b), nests
of subadult females were shown to gradually increase in their effectiveness at releasing
male courtship in the period between 29 and 10 days before moulting. Nests of sub-
adults less than 10 days from moulting always released courtship.
Some species of salticids also use olfactory pheromones. Crane (1949), in some
simple experiments with the salticid Cory thalia xanthopa, showed that females probably
produce olfactory primer pheromones (primer pheromones lower the male's thresh-
old of response to visual releasers of courtship). Males of Cory thalia xanthopa readily
court recently decreased females, without contacting the cuticle, and then attempt to
mate with them. However, aged corpses failed to evoke any reaction from males.
Recently, 16 species of salticids were tested, for evidence of releaser and primer
olfactory pheromones (Jackson, in press). Special cages were used in which a con-
Table 12. Salticid species found to use olfactory pheromones
Species Cheliceral Subfamily Primer Releaser Mate location

dentition
Brettus cingulatus Plurident Spartaeinae Not tested Not tested Yes

Cory thalia canosa Unident Plexippinae Yes No No
Cyrba algerina Plurident Spartaeinae Yes Yes Yes
Euryattus sp. Fissident Cytaeinae Yes No Yes
Lyssomanes virldus Plurident Lyssomaninae Yes Yes Yes
Portia fimbriata Plurident Spartaeinae Yes No Yes
Portia labiata Plurident Spartaeinae Yes No Yes
Portia schultzi Plurident Spartaeinae Yes No Yes
Trite planiceps Fissident Cytaeinae Yes No No
specific female or some other (control) spider was confmed to an attached chamber,
from which the odour of the hidden spider could reach the male in the main part of
the cage. The odour of conspecific females released courtship from males of only 2 of
the 16 tested species; Cyrba a/gerina and Lyssomanes viridis (Table 12). These two
species performed courtship displays even when the conspecific female was removed
just before the test began, providing unequivocal evidence that the effective stimulus
was chemical. The displays released by the pheromone were those normally used by
the male in visual courtship.
The same type of test was carried out with the addition of a mirror at one end of
the cage, giving the test male the opportunity to display to his mirror image. Evidence
of primer pheromones was obtained for 8 of the 16 species tested (the male responded
sooner or for a longer period in the presence offemale odour). The displays given were
those that normally occur in male·male interactions, except that Lyssomanes viridis
males included courtship displays as well. Similar, but larger, cages were used to test
whether male salticids could locate females by means of olfactory pheromones (Jackson,
unpubl.). The hidden female was randomly assigned one end of the cage, and the male's
position in the main cage was recorded continuously for 30 min or at 5 min intervals
for 4 h. Although 41 species were tested, the males of only seven species were shown
to be able to locate females by means of olfactory pheromones (Table 12).
Although very little is known about the site of pheromone production in spiders,
some suggestions are available from an ingenious experiment of Oden's. Using the parti-
tioned cage method referred to earlier, he fastened a 'sled' to the underside of the ab-
domen of Phidippus audax females, which prevented them from depositing silk or
faeces in their cages. When the female was removed and the male tested, he still spent
more time on the side of the cage that the female had occupied, suggesting that the fe-
male produces pheromones from organs on the anterior half of her body, or possibly
the dorsal abdomen. Also, by 'milking' silk from females, placing it in the cage and test-
ing males, Oden found that silk is attractive to males as soon as it comes from the spin-
nerets. In another experiment, Oden covered the tarsi of female Phidippus audax then
repeated the tests as before; he found that males still spent more time on the 'female'
side of the cage. Unfortunately, Oden did not carry out tests using females with tarsi
and abdomens covered at the same time, so he was only able to conclude that the spin-
nerets are not the sole sites of pheromone production.
138 S.D. Pollard et at.
Certain salticid species have porelike structures on their dorsal abdomens: various
species of the Spartaeinae, including four of the species that have been shown to
use olfactory pheromones (Cyrba algerina and the three species of Portia listed in
Table 12) have mytilliform organs (Wanless 1984a);Holcolaethis and Sonoita, African
genera which have not yet been the subjects of behavioural studies, have pustiliform
organs (Wanless 1985). These organs require further investigation, but their structure
suggests that they might secrete pheromones (Wanless 1984b, 1985).
3 The Phylogenetic Significance of Salticid Pheromones
The spinning behaviour of most salticids is minimal, being restricted primarily to the
building of their nests and draglines, which are typically tubular in shape and not much
larger than the spider. This type of nest is built by all of the species listed in Table 12
except Cyrba algerina, Lyssomanes viridis, Euryattus sp., Portia fimbriata, Portia labiata
and Portia schultzi; and each of these species is aberrant in a number of other ways as
well.
Each species of Portia builds a large space web (ca. 200 mm across for adult females)
with a more or less funnel·like shape.Portia is also a genus of web-invading araneophagic
spiders that practise aggressive mimicry (Jackson and Hallas 1986a). Web building
and web invasion may be primitive (plesiomorphic) behaviours within the Salticidae
(Jackson and Blest 1982). Cyrba algerina belongs to the same primitive subfamily as
Portia (Spartaeinae), but does not build webs. However, Cyrba aIgerina moults and
oviposits on small silk sheets which might be described as rudimentary webs. It also
invades webs, is araneophagic, and uses aggressive mimicry (Jackson and Hallas 1986b).
Juveniles of Euryattus sp. indet. build space webs, but adult females build an unusual
type of nest (suspension nest), consisting of a curled·up dead leaf suspended by a few
threads from the vegetation or a rock ledge (Jackson 1985). Lyssomanes viridis be-
longs to another subfamily, Lyssomaninae, which has been regarded as primitive
(Wanless 1980); it builds a rudimentary sheet web (Hallas and Jackson, in press).
Is reliance on chemical communication primitive as suggested by Crane (1949)? On
the basis of cheliceral dentition, salticids can be placed into three groups (which do
not represent taxonomic units), plurident (primitive), fissident (transitional) and unident
(advanced), as has been discussed elsewhere (Wanless 1984a). Interestingly, the two
species for which olfactory releaser pheromones were demonstrated, Cyrba algerina
and Lyssomanes viridis, are plurident, belong to the subfamilies Spartaeinae and Lys-
somaninae, respectively, and have spinning and predatory behaviours that can be con-
sidered primitive. This suggests that the use of olfactory releaser pheromones is indeed
primitive in the Salticidae. However, it is unclear why the other tested Spartaeinae
(three species of Portia) did not provide evidence of olfactory releaser pheromones.
The phylogenetic significance of olfactory primer pheromones is also unclear. It is
interesting that this type of pheromone was demonstrated for each of the tested spar-
taeine species (Cyrba algerina and three species of Portia), Lyssomanes viridis, and two
fissident species (the behaviourally unusual Euryattus sp. indet. and Trite planiceps).
However, Cory thalia canosa which also uses primer olfactory pheromones is unident
and can be described, behaviourally, as 'ordinary cursorial salticids'. The species that
use olfactory pheromones in mate location are perhaps the most interesting: Lyssomanes
viridis, the only tested lyssomanine; Euryattus sp. indet, a web-building fissident; and
all of the tested Spartaeinae (Table 12). If olfactory mate location is especially well
developed in web-building spiders, such as araneids, and if salticids evolved from web-
building ancestors, then these species, which either build webs or at least rudimentary
webs and which have been considered to be primitive, are the very species in which ol-
factory mate location would be expected to occur.
4 Dysderidae, Clubionidae and Araneidae
Studies of the communicatory behaviour of the primitive haplogyne spiders, such as

the dysderids, are likely to be useful in understanding the evolution of spider behaviour.
Dysdera crocata is a distinctive red spider with large chelicerae. This cosmopolitan
spider is well-known not only for its formidable appearance, but also for its specialised
and unusual diet of woodlice (Bristowe 1941); but, until recently, little was known
about its mode of intraspecific communication. Although Cooke (1965) characterised
this species' courtship as "swift and unelaborate", in a more recent study (Jackson and
Pollard 1982) a complex repertoire of displays was described. Males court females both
at and away from nests, and their displays are released by contact with either the fe-
males' cuticle or the silk of her nest. Even a female who has been dead for up to
2 months will release courtship and copulatory attempts from males (Fig. 40). Necro-
philic males are thwarted by the tendency of their lifeless companions to tumble over
on their backs before the mating position can be assumed. However, males show no
interest in nests or dead females that have been washed in ether, evidently because this
solvent removes the courtship releasing pheromones from the female's silk and from
her cuticle. This primitive, non-visual hunting spider uses chemotactic stimuli to perceive
it's environment. When it contacts a woodlouse, it may feed, chemical stimuli from its
prey being partly responsible for the release of its predatory behaviour (Pollard 1986).
If its path intersects a conspecific or the nest of a conspecific, displays may be released.
One of the largest and most widely distributed families of non-visual hunting spiders
is the Clubionidae. Clubiona cambridgei, a New Zealand species, is the most thorough-
ly studied (Pollard 1984; Pollard and Jackson 1982, 1984). Matings seem to occur in
nests, and contact with the nests of adult and sub adult females, even in their absence,
releases courtship from the male. Nests that have been immersed in ether are ineffective
at releasing courtship. Apparently, Clubiona cambridgei, like Dysdera crocata, puts
pheromones on its silk. In contrast to females of Dysdera crocata, which are almost
always receptive to courting males, females of Clubiona cambridgei are evidently re-
ceptive only when they are virgins. Once they have mated, they treat males in a pre-
datory rather than sexual fashion. Consistent with virgin females being a valuable re-
source to them, a major precopulatory tactic of the males of Clubiona cambridgei is
to court and cohabit with sub adult females.
The webs of the females of several species of araneid are known to release male
courtship (Witt 1975). In this respect, the araneids resemble the hunting spiders we
just discussed. However, Blanke (1973a) has shown that females of Cyrtophora cica-
trosa, an araneid, also produce airborne or olfactory pheromones, by which conspecific
140 S.D. Pollard et ai.
Fig. 40. Male Dysdera crocata, a dysderid, performing vibratory courtship after contacting the cu-
ticle of a dead, conspecific female
males locate them before touching the web . As web-building spiders tend to be relative-
ly sedentary compared to hunters, perhaps olfactory mate location is especially im-
portant for this group of spiders.
5 Sex Roles and Pheromones
In some species, female spiders initiate intersexual interactions by producing phero-

mones that attract males into display arenas. Within these arenas male spiders usually
assume the more active (displaying) role (see M.H. Robinson 1982b). The female's
behaviour is analogous to that of an 'employer' advertising to attract an 'employee'.
The male responds to the advertisement, locates the female, and by courting 'applies
for the position' .
It is usually the female spider that produces both releaser and contact pheromones
that elicit courtship from conspecific males. However, the males of some species of
insects use pheromones to stimulate females during courtship (Thornhill and Alcock
1983). Comparable examples may occur in spiders, but more thorough investigation is
needed. For example, males of many species of Argyrodes (Theridiidae) have bizarre
knobs on their heads (Exline and Levi 1962), and at least some species are known to
have glands associated with these knobs (Lopez, this volume). Secretions from these
glands accumulate on the knobs of Argyrodes antipodiana males (Whitehouse, unpubl.
data); and females of this species spend considerable lengths of time 'mouthing' the
male's knobs during courtship and mating. The male may be producing pheromones,
but the influence of these pheromones on the female's behaviour is not clear. As primer
(rather than releaser) effects are the most likely, ascertaining the action of the pherom-
ones is likely to be difficult.
Even in the salticids, despite their acute vision, males may influence females with
pheromones. In certain species of the Spartaeinae males have distinctive organs on
their legs, the femoral organs (Wanless 1984a). It would be interesting to investigate
whether these organs produce pheromones that influence the female during courtship.
6 Pheromones and the Evolution of Spiders
Although the four families of spiders used as examples in this chapter have evolved
communication systems which emphasise different sensory channels, all use phero-
mones in intraspecific interactions. Pheromones are also known to be used by species
of several other families, including other web builders, such as amaurobiids, dictynids
and theridiids, and other hunters, such as lycosids and thomisids (Jackson 1982b;
Tietjen and Rovner 1982; Jackson, unpubl. data; Pollard, unpubl. data; Whitehouse un-
publ. data). There is anecdotal evidence that diplurid males locate females by phero-
mones (Hickman 1964), which is especially interesting because the diplurids belong to
the primitive suborder Mygalomorphae. That pheromones playa role in communication
is ascertained by experimentation, not simple observation, and the limited number of
examples of spiders known to use pheromones may be primarily a reflection of the
limited number of species that have been studied experimentally.
The path of spider evolution may have been complex and a large variety of species
and life-styles have arisen, but there is an easily recognised theme: all use silk. Spiders
often use silk to increase their perceptual range, and to amplify their effective size, but
silk can also be used to convey information. Pheromones bound to silk of webs, nests
or draglines become construction signals (see Jackson 1982a; W.J. Smith 1977), i.e.
chemical signposts left in the environment. The linking of pheromones with silk prob-
ably arose early in spider evolution and seems to be a important feature of this group
of animals.
Acknowledgements. We thank Curt Lively, Joe Waas and Mark Wells for their valuable comments
on this chapter.
III Spider Venoms and Their Effect
ZVONIMIR MARETIC 1
CONTENTS
The Venomousness of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
2 Biochemistry of Spider Venoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
2.1 Obtaining Spider Venoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
2.2 Biochemistry of Latrodectus Venom . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
3 Pharmacology of Latrodectus Venom . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
3.1 Neuromuscular Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
3.2 The Effect of Latrodectus Venom on Various Animals . . . . . . . . . . . . . . . . . . 147
4 The Venom of Loxosceles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
5 The Venom of Atrax robustus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6 The Venom of Phoneutria nigriventer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
7 Epidemiology of Spider Bites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
8 Clinical Problems of Some Forms of Araneism . . . . . . . . . . . . . . . . . . . . . . . . . . 151
8.1 Latrodectism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
8.2 Chiracanthism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
8.3 Lycosidism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
8.4 Loxoscelism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
8.5 Atraxism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
8.6 Phoneutrism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
1 The Venomousness of Spiders
In principle, with the exception of some groups of spiders which have no venom glands
at all (e.g. Uloboridae and Holilrchaea) all spiders with any kind of venom apparatus
must be considered as venomous, if not always for man, then perhaps for other animals,
or at least for insects, their usual prey. Vachon (1968) considered only about 100 spider
species as actually dangerous to man. According to Sutherland (1981), about 100 spe-
cies of Australian spiders may bite man, however, this does not, of course, mean that
all of them are dangerous. A spider is not necessarily venomous to man due to dif-
ferent reasons: (1) their venom does not contain toxic fractions which affect mankind;
(2) the quantity of the venom which the spider injects is insufficient; (3) their chelicerae
are not strong enough to penetrate the human skin; (4) they are too timid to come into
contact with man; (5) they have not had the opportunity of contact due to their life-
style, i.e. contact may have been accidental, therefore their venomosity has not been
recorded. Examples of the last consideration are the "new" venomous species which
1 Medicinski Centar, YU-52000 Pula, Yugoslavia

Spider Venoms and Their Effect 143
are still being "discovered", e.g. the theridiid Steatoda paykulliana (Maretic et al. 1964)
or the theraphosid Pterinochilus sp. (Maretic 1967). According to Russell and Gertsch
(1982) the number of such new venomous spiders in the United States has increased to
about 60 species.
The diagnosis of araneism is, in principle, difficult if the patient does not bring in
the preserved spider-culprit. One cannot depend on vague descriptions of the patient
or recognition of the spider from later. Besides, most physicians are unfamiliar with
spiders and spider bites. However, for a skilled physician the diagnosis is easy, e.g. dur-
ing an epidemic of latrodectism. According to Russell and Gertsch (1982), from
600 suspected spider bites with local signs and symptoms, recorded by them, 80% were
bites by other arthropods (ticks, bed bugs, fleas, etc.). Another significant percent of
these cases was due to other diseases with skin manifestations, thus only a minority
were real spider bites. In the diagnosis of spider bites, one should follow the same
principle as in all other diseases: the establishment of the etiological factors, i.e. the
cause, in this case, the spider-culprit.
The reactions of people with regards to the venomosity of spiders are often inade-
quate, since an exaggerated and baseless fear spiders often exists. An example is the
"tarantism" of former centuries, in which not only the uneducated masses, but also
the greatest intellectuals of these times believed. On the other hand, some authors of
later centuries, basing their opinions on incomplete observations, believed that spiders,
even those considered as most toxic, e.g. Latrodectus (Fig. 41a), were not venomous
and that only ignorance and superstition had created the myth of their toxicity. Similar
articles can sometimes be found in the daily press. Just as there are sensational reports,
which exaggerate the danger of spiders, sometimes one can also find reports minimaliz-
ing their venomosity. This may, in part, have its roots in inexact experiments of 19th
and early 20th century. In the United States in the early 20th century some profes-
sionals still believed that the black widow is harmless (Thorp and Woodson 1945).
McKeown (1936) says the same for Australia.
In conclusion, on the one hand, one should contest the unfounded fear and hatred
towards spiders which represent an important link in the ecologic chain, but on the
other hand, one should not forget that some species, though not very numerous, can
represent a real danger for the man.
2 Biochemistry of Spider Venoms
Spider venoms are compounds of low molecular weight, some of which are pharma-
cologically very active, and they are composed of proteinlike substances which are
mixtures of toxic, nonenzymatic proteins with various enzymes. The clinically most
important spiders, e.g. from the genera Latrodectus or Phoneutria (Fig. 41c), have pro-
teinaceous toxins like other venomous animals. The biochemical examinations of spider
venoms are relatively scanty in comparison to snake venoms. Snakes, in representation
of venomous animals, have attracted much greater attention; the quantity of their
venom is much larger, whereas the acquisition of spider venoms is much more difficult.
According to comparative studies, the quantity of dry venom injected by one bite of
Latrodectus amounts to 0.02-0.03 mg, while in a long nose viper, it has a weight of
144 z. Maretic
d Female Male
Fig. 4la-d. Poisonous spiders. a Black widow Latrodectus tredecimguttatus (Theridiidae), female;
b brown spider or violin spider Loxosceles /aeta (Loxoscelidae), female; c Phoneutria nigriventer
(Ctenidae), female, in a characteristic defensive-aggressive display; d Sydney funnel-web spiders
Atrax robustus (Dipluridae), female, male. (From Bettini 1978)
20 mg, whereas large, venomous snakes, such as Crotalus adamanteus, inject up to

437 mg (Kaiser and Michl 1958).
2.1 Obtaining Spider Venoms
The main methods for obtaining spider venoms are:

1. Collecting the pure venom directly from the chelicerae of living spiders.
2. Extraction from prepared venom glands.
3. Extraction from the whole cephalothorax or from its front part containing the
glands.
The pure venom, which is the most suitable for biochemical analysis, can be ob-
tained if one allows the spider to bite into a tampon of cotton (Lebez 1953) or into
paper used for electrophoresis (Muic et al. 1956). The venom can also be taken from
the irritated spider directly from the tips of the chelicerae with a micropipette. How-
ever, bearing in mind that the spider can voluntarily regulate the release of venom, it
is obvious that all these methods are imperfect. A total depletion of the venomous
glands can be achieved only by electrostimulation. With the tampon method one can
obtain, e.g. in Latrodectus, only small quantities of thefresh venom, i.e. approximately
0.5 mg. In contrast, Bucherl (1969) obtained 0.6-1.3 mg of fresh venom with electro-
stimulation. The dry venom amounted to 10- 20% of this quantity.
Phoneutria nigriventer, Atrax robustus or A. formidabilis can yield six to eight
times more venom than Latrodectus (Fig. 41). According to Bucherl (1956), from one
specimen of Phoneutria nigriventer one can obtain 1.8 mg dry venom in the winter,
and 2.5 mg in summer. Furthermore, due to the sexual dimorphism in some species,
differences in venomosity according to sex also exist. In Latrodectus only the female,
which is much larger than the male is venomous. In Loxosceles, both female and male
are equally toxic. Although the female of Atrax robustus is larger, the male is six times
more toxic.
It is clear that the extracts of the front parts of the cephalothorax with the venom
glands, or even of the whole cephalothorax contain many impurities, and that they are
unsuitable for biochemical examinations. Extracts from isolated venom glands were
used by several authors (Bettini and Toschi-Frontali 1960; Frontali and Grasso 1964;
McCrone and Hatala 1967, and others). Although this method is practical, a disad-
vantage is the contamination of the venom with numerous substances, foremost with
proteins from the gland wall. Impurities resulting from carelessness by the acquisition
of the venom (e.g. by digestive flUid) may also explain some of the proteolytic activities
of the venom, which have repeatedly been reported for several spider species (Collatz,
this volume).
2.2 Biochemistry of Latrodectus Venom
The venom of the most toxic European species, Latrodectus tredecimguttatus, is a

limpid, colourless solution in water which foams profusely if shaken. In summer, the
fresh venom is alkaline with a pH of 8.2 ± 0.3. At lower temperatures the pH becomes
146 Z. Maretic
Table 13. Analyses of Latrodectus venom
Result Species a Preparation Reference
Lipoids (lipoproteins) Pure venom Lebez (1954)

8 Protein fractions Pure venom Muic et al. (1956)
5 Protein fractions, Homogenates of venom Cantore and Bettini (1958)
hyaluronidase glands
5 Protein fractions, Extracts of venom glands Bettini and Toschi-Frontali
free amino acids (1960)
5 Protein fractions Homogenates of venom Grasso and Toschi-Frontali
glands (1962)
7 Protein fractions, m Homogenates of venom McCrone and Hatala (1967)
3 nonprotein frac- glands
tions
13 -1 7 Protein fractions m Pure venom Russell and Buess (1970)
5-Hydroxytryptamine Gland extract Pansa et al. (1972)
a t = Latrodectus tredecimguttatus; m = L. mactans.
more acid and the venom is less toxic (Vellard 1936). The fresh venom desiccates in
vacuum relatively quickly yielding flakes. Many authors have published results of elec-
trophoresis of the venom, but they are usually not comparable since neither the same
methods nor the same materials were used (Table 13). Muic et al. (1956) performed
electrophoresis of pure venom on filter paper; they obtained six protein fractions
which were then stained with bromphenol blue. With ninhydrin two further components
were detected. Cantore and Bettini (1958) and Bettini and Toschi-Frontali (1960) used
gland extracts, McCrone and Netzloff (1965), also using gland extracts, utilized disc
electrophoresis and proved that significan t differences exist between L. mactans, L. vari-
olus, L. geometricus and L. bishopi. Russell and Buess (1970) showed with results from
electrophoresis of pure venom that it can also be used for taxonomic purposes.
Frontali and Grasso (1964) obtained five peaks on Sephadex from which three were
toxic for the housefly and one for the guinea pig. Frontali and Grasso (1964) and
Vicari et al. (1965) isolated from venom glands of L. tredecimguttatus three toxic com-
ponents from which only one as toxic for vertebrates. McCrone and Hatala (1967)
isolated one toxic components from the gland extract of L. mactans. The principal dif-
ference in the results of both these groups was in the molecular weight, probably be-
cause they did not work with pure toxin, but with protein mixtures. Maretic (1955)
found in Latrodectus venom a "spreading factor" which Bettini and Cantore (1959)
identified as hyaluronidase. Besides proteins, the Latrodectus venom also contains free
amino acids (Bettini and Toschi-Frontali 1960). In fresh venom gland extracts Pansa
et al. (1972) found small quantities of 5-hydroxytryptamine. Lipid substances were also
identified (Lebez 1954).
In 1976, Grasso and Frontali et al., independent of other works, reported the electro-
phoretic separation of the pure Latrodectus toxin (Iatrotoxin). Chromatographic me-
thods were also used successfully and the results of both studies were partially identical
or supplemented each other. They also isolated a protein of 130,000 resp. 110,000 to
130,000 molecular weight from homogenates of the venom glands.
3 Pharmacology of Latrodectus Venom
3.1 Neuromuscular Effects
The Latrodectus venom acts upon neuromuscular transmissions which release the trans-
mitter acetylcholine from synaptic vesicles and extravesicular agglomerations, thus
enabling the transmission of the stimulus from the presynaptic to the postsynaptic
membrane, and from the presynaptic neurone to the receptor. It should be emphasized
that this toxin acts on the same site as the botulinus toxin, but only antagonistically
since botulinus toxin blocks the release of acetylcholine.
Longenecker et al. and Clark et al. (1970) showed the action of Latrodectus toxin
on a single neuromuscular junction of the Musculus sartorius of the frog. They moni-
tored the end-plate potentials (EPP) and the miniature end-plate potentials (MEPP).
Already after 5-10 min, the venom induced an increase of MEPP from 0.5 -1.5 s-1 to
300-1000 S-I. After reaching the peak the MEPP frequency falls to values less than
1 s-l. If the venom is applied further, the MEPP rate increases to 400,000 S-I. First,
the EPP amplitude increases several times and then falls to zero. The nerve-ending
spikes disappear at approximately the same time, probably due to depolarization of
the nerve ending. The authors attributed these findings to the reaction of the venom
with the nerve-ending membrane and the release of transmitters. The authors widened
the scope of this work with electron microscopic studies which showed a very "clean"
appearance of the neuromuscular endings which, in contrast to the controls, did not
show synaptic vesicles.
The mechanism of action upon the cerebral cortex is, in principle, the same as on
neuromuscular endings, which Frontali et al. (1972) proved by measuring the quantity
of acetylcholine in cortex slices which were treated with the venom and kept in a
phYSiological medium. Frontali (1972) extended these experiments on adrenergic nerve
endings using the rate iris as a model. With fluorescent and histochemical techniques
she monitored the depletion of catecholamines. The concept of a systemic derangement
of cholinergic and adrenergic nerve endings, with the release of transmitters, first in-
creased and then blocked, explains the mechanism of some signs and symptoms of
latrodectism. Fritz et al. (1980) confirmed that the pure latrotoxin also releases trans-
mitters in the neuromuscular transmission of mammals and provokes an eruption of
miniature potentials of motor end plates and the release of transmitters in the brain
cortex.
3.2 The Effect of Latrodectus Venom on Various Animals
The rat is moderately susceptible to this toxin. From 105 rats 75,2% died within 2 h
to 13 days (average 32 h) following the bite of one or two spiders. Two phases of
intoxication, Le. motoric restlessness and the paralytic stage, could be observed (Fig. 42).
Soon after envenomization, the animal rubs its snout, bends the neck, becomes shaken
by sudden jerks, throws itself from one side to the other, jumps and twitches in various,
bizzare postures. Abundant salivation occurs. After about 10 h the animal is quiet and
exhausted, with matted fur due to salivation and caudally due to spontaneous urinating
148 Z. Maretic
Fig. 42. Latrodectism in the rat: the stage

of motoric restlessness and the paralytic
stage
and defecation . Around the eyes, red rings of koproporphyrine from the Harderian
glands are seen. The whole body is trembling and the hind legs are in a paralytic posi-
tion. Within 72 h the animal may lose up to 32.7% body weight. The minimal intra-
peritoneal lethal dose for a rat of 50- 60 g is 0.032 mg (D'Amour et al. 1936).
The cat is very susceptible . Of 22 animals over 800 g, only two survived . The survival
time in others was 1-14 days (average 115 h) . Soon the cat becomes restless, meows,
crawls on the belly and shows severe paralytic symptoms (Fig. 43). Severe salivation
and lacrimation are present and finally, the animal quite exsiccated lies emaciated,
adynamic and flaccid (Maretic 1959).
From 26 guinea pigs of 250-720 g body weight, intoxicated in a similar way, all
died within 1- 56 h (average 6 hand 34 min) . There is a severe reaction of the respira-
tory system with dyspnea, irregular rhythm of respiration, polypnea, bradypnea, stridor
and finally, pulmonary edema and suffocation. Salivation and lacrimation are present
(Maretic 1959). The mouse dies from pulmonary edema after 20 min following the
bite of Latrodectus tredecimguttatus . The LDso for mice of 13-17 g body weight is
0.0135 mg of the dried gland injected intravenously (Bettini and Frontali 1960).
Fig. 43. A young cat 40 h following a Latrodectus bite in a state similar to coma. The animal is ex-
siccated with the fur glued by saliva and urine
Table 14. Toxicity of Latrodectus tredecimguttatus venom on

various animals. (From Bettini and Maroli 1978)
Animals LD,o values
Frog 2.18 ± 0.49 14.5 ± 32

Canary 0.085 ± 0.03 4.7 ± 1.7
Blackbird 0.42 ± 0.19 5.9 ± 2.7
Pigeon 0.11 ± 0.025 0.36 ± 0.07
Chick 0.19 ± 0.04 2.1 ± 0.57
Cockroach 0.0015 2.7
Housefly 0.000013 0.6
Guinea pig 0.028 0.075
Mouse 0.013 0.90
a mg of proteins (Lowry's method).
Rabbits, according to our own experience, are resistant to the venom. Also accord-
ing Finlayson (1936), the minimal intravenous lethal dose for a 1600 g rabbit is even
3.0 g fresh venom of L. indistinctus. The intoxicated dog suffers severe gastrointestinal
signs and symptoms with vomiting and diarrhea, but all authors agree that high doses
of the venom are necessary to kill it (Troise 1928; Sampayo 1942; Maretic 1959). Very
resistant are also sheep. In contrast, some large animals, such as horses or camels are
very susceptible to the venom. The extract of a single cephalothorax of a Latrodectus
caused the death of a horse (Stanic 1953), and the injection of a macerated 1.5 g spider
killed a large camel (Vellard 1936).
In insects the venom of Latrodectus first provokes paralysis, so that the spider be-
gins to suck their body juices while they are still alive. For LDso of various animals,
see Table 14.
4 The Venom of Loxosceles
According to Bucher! (1969), a Loxosceles rufipes yields 0.13-0.27 mg dry venom.

Fractionating by gel filtration, it was found that in the venom of L. reclusa two toxic
fractions exist: toxin 1 is responsible for the lesions produced in rabbits and is also
lethal for mice and rabbits; toxin 2 does not produce external lesions and is toxic only
for rabbits (Odell et al. 1975). A fraction obtained by gel filtration of L. laeta produced
qualitatively different lesions in rabbits (Suarez et al. 1971). Enzymatic activities were
not found in the venoms of these spiders. In the venom of L. reclusa hyaluronidase,
esterase, lipase and protease were found (Schenone and Suarez 1978).
The pathomechanism of lesions produced by Loxosceles venom, which is, in princi-
ple, proteolytic and hemolytic, is not well understood. Pizzi (1975) tried to explain it
as an anaphylactoid reaction. While intravenous application of the venom of L. reclusa
in dogs causes thrombocytopenia and hemolysis, in rabbits there was no deleterious ef-
fect. Even the enzyme composition of the venom cannot explain the pathomechanism
of lesions. The venom of L. reclusa produces a cytotoxic effect on cell cultures of
150 Z. Maretic
human epithelial cells (HeLa cells) and fibroblasts (Morgan and Felton 1965). Hemolytic
activity was registered in sheep erythrocytes incubated with venom of L. reclusa, but
only if serum was present (Kniker et al. 1969). However, in washed sheep or rabbit
erythrocytes, incubated either with the venom of L. reclusa or L. laeta (Geren et al.
1973; Biggemann 1971) there was no hemolytic effect. The rat is even resistant large
doses of Loxosceles venom. Neither an inactivator of the venom in its serum nor a
stimulator in the serum of susceptible animals were found (Biggemann 1971). There is
the possibility that the pathogenic effects of the venom are triggered by an initial dis-
turbance of the cell membrane function. Resistance and susceptibility seem to be re-
flections of genetically determined features of target cells rather than humorally medi-
ated processes (Schenone and Suarez 1978). The intradermal LDso of L. laeta venom
for rabbits is 1.5 glands (Schenone et al. 1975), which would correspond to about
0.2 mg dried venom. The intravenous LDso for dogs amounts to 2 glands which would
correspond to about 0.35 mg (Schenone and Suarez 1978).
5 The Venom of Atrax robustus
Besides large quantities of a spermine complex and four other fractions, which were
detected by using ultrafiltration membranes and which were of low molecular weight
and low toxicity, Sutherland (1973) found that an indole lactic acid-spermine complex,
rich in hyaluronidase and of a high molecular weight (15,000- 25 ,000), which formed
only 10% of the crude female venom, was the most toxic fraction. Its active component
was called atraxotoxin.
Sutherland (1972) established that the toxin of Atrax robustus acts upon neuro-
muscular endings, but differing from the toxin of Latrodectus, it stimulates the spon-
taneous action potentials acting directly on nervous fibers and so provoking an abundant
release of acetylcholine on motor end plates, thus releasing acetylcholine, adrenalin
and noradrenalin from the autonomic nervous system. There are also differences in the
toxicity and in the mode of action among the venoms of male and female spiders
(Morgan and Carroll 1977). Spence et al. (1977) registered on the isolated diaphragm
of mice spontaneous potentials of motor end plates as an effect of the venom. Similar
results were also obtained in isolated Nervus phrenicus. They thus concluded that the
venom of Atrax robustus changes the electrical field in nervous membranes.
The subcutaneous LDso for 200 g guinea pig from female crude venom is 2.4 mg
and from the male venom, 0.6 mg. The intravenous or subcutaneous male venom LDso
for Cynomolgus monkeys, which are particularly susceptible to Atrax venom, is about
0.2 mgkg- 1 (Wiener 1957).
6 The Venom of Phoneutria nigriventer
The venom contains histamine and serotonin, two polypeptides which contract the
guinea pig ileum, and other venom-active polypeptides, hyaluronidase and proteolytic
enzymes (Kaiser 1953; Diniz 1963). By electrophoresis only the flaccid, paralytic com-
ponent was separated (Pereira Lima and Schenberg 1964). The results of gel chromato-
graphy showed that the Phoneutria venom contains components of different molecular
weight, but which have identical pharmacological activities (pereira Lima and Schenberg
1964). They were shown to be responsible for erection, hyper salivation , toxicity (lethal-
ity), ileum contraction, distensive paralysis, flaccid paralysis, histamine, serotonin,
blood pressure rise and blood pressure fall. By ion exchange columns even 14 pharma-
cologically active components were found (Schenberg and Pereira Lima 1971). The
complexity of the venom is also evident from the fact that four different polypeptides
may induce the same effect. The subcutaneous Dosis certe letalis for the dog is about
0.0002 mg kg- 1 of dried Phoneutria nigriventer venom (Schenberg and Pereira Lima
1978). The intravenous LDso for a 20 g white mouse of dried Phoneutria venom is
0.00034 mg g-l (BiicherlI956).
7 Epidemiology of Spider Bites
The epidemiology of spider bites differs not only among various genera, but also with-
in the same genus, e.g. in Latrodectus. L. tredecimguttatus (Fig. 41a) is the most charac-
teristic representative of a "wild" spider, i.e. a nonurbanized species living only out-
doors, preferably in wheat fields, vineyards, under stones and bushes, forming some-
times whole colonies. The victims are most often farmers in summer during field work
and soldiers and other people dwelling outdoors. In contrast, L. mactans, also L. hes-
perus, is typically found in cellars, garages and corners of similar rooms, so that people
of different professions are bitten by accident. In Bogen's times (1926, 1932), when
the spider nested under the boards of outdoor privies penis bites were common. Transi-
tions in life-styles have occurred since then: L. hasselti and L. geometricus are now
found in various areas either indoors or outdoors.
Loxosceles bites occur most often (80%) indoors when victims are sleeping or dress-
ing after the spider has climbed into clothes or boots (Fig. 41 b). Sometimes it hides on
the wall behind pictures. It is most active in summer months (Gajardo-Tobar 1966).
The epidemiology of chiracanthism is similar. Most cases of Atrax robustus bites occur
in early summer, when the spider roams to houses and summer camps. Also, cases in
the bush have been recorded (Sutherland 1972).
The bites of Phoneutria nigriventer (Fig. 41c) have been reported mostly in winter
when the spider, due to cold weather, enters houses. It hides mostly in clothes, boots
or under bedcover. They are also called 'Banana spiders' since they hide often under
clusters of bananas. Accidents have also been recorded in Europe from people handling
ship cargos with bananas from Brazil (Schenberg and Pereira Lima 1978). Latrodectus
and Loxosceles are not aggressive at all, which is not quite true for Chiracanthium.
Phoneutria and especially Atrax are aggressive (Fig. 41c,d).
8 Clinical Problems of Some Forms of Araneism
8.1 Latrodectism
8.1.1 Manifestations
Latrodectism exhibits the features of the most important araneism, including not only
the severity of envenomization, but also sometimes the epidemic appearances, i.e., the
152 z. Maretic
frequency of cases, occurring over all continents and from antiquity. Thus, the most
comprehensive literature, dating from very ancient times, is available for this form of
araneism . Latrodectism is a syndrome, a clinical entity with complex manifestations.
Besides the very specific action on the nervous system, and through it on all other
organs, it is of significance also as a potent stressor which provokes a series of non-
specific responses.
The bite itself is slight, so that 58.2% of our 177 patients did not perceive it at all
(Maretic 1959). The latency is short, usually 10-60 min . Typical is the "early pain"
in regional lymph nodes which may become slightly swollen. The general condition
soon becomes poor . From the lymph nodes the pain extends to the waist and lumbar
region, the belly and thighs becoming insupportable . Cramps, convulsions and contrac-
tions of the musculature imbue the body, increase in frequency and the pains culminate
in paroxysms. The skin is hypersensitive and covered with abundant, cold and sticky
sweat. If the patients are able to stand, they are trembling, and due to pains and muscle
contractions they appear as if they are suffering from lumbago or intermittent claudi-
cation. Mostly they lie helplessly crying and lamenting. A significant loss of weight
might be observed. The body temperature is usually slightly elevated (Mare tic 1951,
1959). The "Facies latrodectismi" (or "latrodectismica") (Fig. 44), together with
motoric restlessness provide the patient with a peculiar aspect. Facies latrodectismi de-
notes a flushed face, covered with sweat and contorted in a painful grimace with swol-
len eyelids, a strong blepharoconjunctivitis, sometimes rhinitis, inflammation of lips
(cheilitis) and trismus of masseters, which one hardly observes in any other morbid
condition (Maretic 1955). Dilatation of pupils (mydriasis) and/or narrowing of pupils
(miosis) are frequently encountered.
Fig. 44. "Facies latrodectismi" (see text)

An accelerated heart rate (tachycardia) or a slowed heart rate (bradycardia) may

also be observed in patients (Bogen 1932; Gajardo-Tobar 1941). Tachycardia is usually
observed in the first hours of intoxication and later one finds bradycardia, in adult
males with 42 beats min- 1 (MaretiC and Lebez 1979,1985). In electrocardiograms of
the patients one can find a prolongation of the QT interval, high P 2 and P3 waves and
flattened T waves until negative. In guinea pigs, which use to have a heart rate of about
200 beats min- 1 , a sinus bradycardia of96 beats min- 1 could be observed. In some of
the animals Lucciani-Wenckebach's phenomenon was observed (Maretic 1963). A hyper-
tension of convergent type, even as high as 30/20 Pa (210/150 mmHg) is characteristic,
which is due to the increased secretion of catecholamines.
With respect to the respiratory tract, shallow and irregular breathing with an ac-
celerated rate is typical; sometimes dyspnea and stridor (Maretic 1959) or even suf-
focation (Finlayson 1936) occur. Difficulties in speech, hoarseness or even aphonia
have been recorded (Maretic and Lebez 1979, 1985). The venom acts also on the di-
gestive apparatus. A hyper- or hypofunction of the salivatory glands has been observed.
Some patients had only nausea, but nearly half of them vomited. Sometimes an en-
larged liver or jaundice (subicterus or icterus) can be found (Bogen 1926). Disgust of
food (anorexia) and obstipation are characteristic; sometimes an acute thirst is present
(Bogen 1932; Maretic 1959).
Patients bitten by a black widow pass little urine or none at all during the first 12 h
or more. The pathomechanism of this oliguria or anuria is complex. Frequently, there
is a retention of urine due to a disturbed innervation ofthe sphincter or rigidity of the
abdominal musculature. There may also be a significant loss of fluid by sweating,
vomiting and salivation. But also the toxin itself may damage the kidney and provoke
a toxic nephritis (Blair 1934; Beasley 1942). The urine is usually of high specific gravity
and frequently, albuminuria with leukocytes, erythrocytes and granular casts in sedi-
ment is found (Bogen 1926; Maretic 1951). A continuous erection (priapism) is men-
tioned by Bogen (1926), Gajardo-Tobar and Vildosola (1944), Bettini and Cantore
(1959) and Maretic (1959). Besides the action of the toxin on vegetative centers, the
increased blood viscosity may also playa role.
The venom of Latrodectus is a neurotoxin with a strong influence on the vegetative
and cerebrospinal nervous system. Consequently, a series of neuropsychic symptoms
have the tendency to dominate the clinical picture. The main symptom is an intense
pain which increases with the development of other symptoms. The pain is often pro-
found and muscular and does not irradiate towards individual nerves. According to
reports of patients, the agony of latrodectism is worse than any painful condition under
which they had suffered. Patients suffering from latrodectism in a milder form compared
it to cramps after physical exertion or with muscle aches felt after sitting in a draught.
The oppression, which is felt in the chest due to contraction of the thoracic muscu-
lature, perhaps also due to a decreased blood supply to the heart muscle together with
polypnea and dyspnea elicits in the victim a strong feeling of anxiety, even fear of
death (pavor mortis). It has been observed that otherwise healthy and strong people,
throw themselves on the bed or on the floor in a state of acute psychomotoric restless-
ness, moaning and performing various senseless movements. Sometimes they become
really psychotic and aggressive, so that without having an anamnesis of a spider bite,
some of them were admitted to the psychiatric department (Mare tic 1951). Barton
154 z. Maretic
(1938), still 2 months after the bite, suffered from amnesia and was unable to recognize
persons and objects. Insomnia is a further characteristic of latrodectism, not only in
connection with the severe pains, but also due to the influence of the toxin on the
hypothalamic centers (Bettini and Cantore 1959; Maretic 1959).
Usually after 20 h or more, the pains abate slightly; sometimes they are redu~t:d to
a series of muscle cramps. The pain descends into the lower limbs with a characteristic
burning of soles and then gradually disappears in the course of the next few days. The
pain is accompanied with tingling sensations (paresthesias). The tendon reflexes,
especially the patellar, were increased in most of our patients. The musculature ofthe
whole body is hypertonic. There is a rigidity of the musculature, especially of the ab-
dominal walls. Sometimes trismus and fasciculation of the musculature are present, as
well as clonic contractions. There may be a tremor, especially of the tongue and jaws,
so that chattering of teeth may be heard. All these symptoms are called the "myopathic
syndrome oflatrodectism" (Gaud and Delesalle 1949).
According to Sampayo (1942), the electroencephalograms of experimental animals
are similar to those in a state of excitement of the central nervous system, as in epileptic
convulsions and alkalosis. Maretic and Jela~ic (1953) proved that this toxin is one of
the rare agents which are able to break the blood-brain and the blood-cerebrospinal
fluid barrier. On the basis of headaches and changes which were found on the ocular
background, it was concluded that there must be an increase in the pressure of cerebro-
spinal fluid, which was proven also by lumbar puncture and manometry (Maretic and
Jela~ic 1953).
Though characteristic, the local signs of latrodectism are not very striking. A slight
redness (erythema) appears immediately following the bite. An erection of hairs (pilo-
erection) is also evident (Fig. 45). Sometimes red streaks are visible extending towards
the regional lymphatic nodes. In some patients a white papule like an urticaria was
present in the midst of erythema. Local sweating, prior to general sweating also some-
times appears. A few hours later at the site of the bite, one sees a pallid area of about
5 cm diameter sharply encircled by a reddish-blue ring. In the midst, one can now see
only one to two points, marks of the fangs. At this site changes in sensitivity develop:
i.e. a local painful sensation (hyperpathia or anaesthesia dolorosa) or a decreased sen-
sitivity, until total anesthesin (Maretic and Jela~ic 1953). In most of the severely ill
patients, who were not treated specifically, a rash appeared. It is polymorph, from
scarlatiniform to maculopapulous, or mince vesicles (Miliaria crystallin a), as a conse-
quence of profuse sweating, appear. The rash can be localized around the site of the
bite or generalized.
The toxin of Latrodectus is a strong stressor provoking in some instances a shock
(Blair 1934; Greer 1949; Finlayson 1956). However, more often the patients are in a
state of alarm reaction which would correspond to a countershock. One finds an in-
creased number of white blood cells (up to 26.2 X 109 ) with an increased number of
neutrophil white blood cells and a decreased number of eosinophil and lymphatic blood
cells, and slightly increased values of erythrocytes, hemoglobin and hematocrit. A slight
transient hyperglycemia, as well as increased values of blood urea were found. In elec-
trolytes, decreased values of sodium and chlorides were found, as well as increased
potassium.
Complications may appear in patients suffering previously from hypertenSion, renal
diseases, coronary disease or severe emphysema. Differentially diagnostically, the most
Fig. 45. Local latrodectism: piloerection with local sweating and gooseflesh (top). Erythema sur-
rounding the urticalike plaque which is elevated at the site of the bite itself (bottom)
important character is the rigidity of the abdominal walls, which can sometimes be
interpreted as an acute surgical condition, so that unnecessary laparatomies were al-
ready registered (Watson 1922; Bogen 1926; H. Wilson 1943). Renal failure , acute
psychoses, tetanus and meningitis must also be taken into consideration.
The duration of illness in untreated patients averages 1 week (Kobert 1901; Sampayo
1942; Maretic 1959). The convalescence is long, a few weeks or longer, and is charac-
terized by an asthenic and neurasthenic syndrome with loss of body weight, mental
changes and impotency. The lethality is now less than 1 %in the USA (Zahl 1971).
Among our 177 cases, there were no fatalities at all (Maretic and Lebez 1979). How-
ever, Bogen (1932) recorded a lethality of 6.5% and Taussig (1956) of 5% in the USA.
Earlier fatalities, obviously, were more frequent; modern statistics show only a low or
no fatality rate.
156 z. Maretic
8.1.2 Treatment
Since the oldest records of latrodectism date from 2.5 millennium, its treatment has
the longest tradition among all araneisms. A side from all the methods of folklore
medicine and the therapies of ancient authors, the most recent medical reports list
more than 100 various methods and drugs for treatment of latrodectism, which only
shows that most of them are useless.
Due to the very fast spreading of the venom from the site of the bite, there is no
first aid, and every local treatment is useless. Diaphoretics, hydrotherapy, diuretics and
similar remedies, as recommended by various authors, may have only a symptomatic
value. Morphine, only in heavy doses, can achieve modest results. Indeed, the first suc-
cessful treatment was the intravenous application of calcium salts introduced by E.W.
Gilbert and Stewart (1935). With the characteristic feeling of warmth the pains oflatro-
dectism also disappear. However, they usually return in 20 min or later, sometimes in
a milder form, however, the injection may be repeated. Corticosteroids have shown
some favourable effects (Maretic 1953).
Latrodectus antivenin is the treatment of choice in most cases. Our experience is
based not only on the specific anti- L. tredecimguttatus serum, which was first pre·
pared in Europe by Maretic (1951), but also on sera from the USA ("Lyovac"), from
the South African Institute for Medical Research and Australia (Wiener 1956), which
shows that the venoms of the whole genus Latrodectus should antigenically be about
the same. Following the intravenous application of a good antilatrodectic serum a
relief ensues already within 10-20 min, and in a few hours, complete recovery.
However, the best results were obtained by the combination of calcium and anti-
venin. If calcium is given intravenously first and then the antivenin, the signs and
symptoms disappear immediately due to the effect of calcium and do not return due
to the action of the antivenin. Thus, the patient, also in a very severe state, recovers
within the time necessary for manipulating the syringes. Russell et al. (1973), recom-
mended to treat all patients younger then 16 and older than 60 years, or those with
hypertension heart disease or severe intoxication, with antivenin, and all others with
the muscle relaxant methocarbamol.
The firm Sharp and Dohme (USA) began the production of antivenin in 1940, which
is today produced under the name Lyovac. Frozen extracts of macerated L. mactans
venom glands were used as antigens and horses as serum animals. Now, a concentrated
lyophilized powder is available, which is dissolved in 2.5 ml of physiological saline
solution. The content of one ampoule neutralizes the venom of 750 spiders. The anti-
latrodectic serum is available today also from Commonwealth Serum Laboratories,
Parkville, Victoria, Australia and from the Instituto Nacional de Microbiologia Dr. Carlos
G. Malbran, Buenos Aires, Argentina.
8.2 Chiracanthism
The clinical picture of bites of Chiracanthium punctorium, similar to the other generic
species, consists of marked local signs and symptoms (itching, erythema, edema, even-
tually slight necrosis, enlarged and painful regional lymph nodes) and moderate general
symptoms with chills, increased temperature, generalized aches and disturbed general
condition. In our own casuistry 1985 (unpublished data) there was one case of col-
lapsus, which was also registered by Habermehl (1974). Ori (1975) described nausea,
vomiting and even shock following the bite of Ch. japonicum. The treatment is symp-
tomatic (analgesic-steroid cremes, aspirin, eventually a relaxant).
8.3 Lycosidism
Apart from hysteric tarantism, the European Lycosa tarentula may occasionally provoke
local necrosis (Maretic 1959). Repeating the experiment, which had been performed
with venom of L. raptoria (Vellard 1936), the same authors proved in guinea pigs that,
similar to its South American relatives, the venom of L. tarentula may act cytotoxically
and provoke necrosis if applied intradermally (Fig. 46). In South America, e.g. in Sao
Paulo , where lycosids are frequent , similar necrotic lesions, due to intracutaneous bites,
are often seen, usually without general symptoms (Vellard 1936; Bi.icherl 1960). Treat-
ments is symptomatic (debridement, analgetics).
8.4 Loxoscelism
The experiments of C.W. Smith and Micks (1968) proved that Loxosceles ru/escens,
which occurs also in the Mediterranean area , has about the same venom apparatus,
quantity of venom and an effect like other "Violin Spiders", L. laeta and L. reclusa.
Gorham (1968) reported records of 126 cases in the USA and about 400 cases in South
America. Efrati (1969) described seven cases of loxoscelism in Israel, and Maretic and
Russell one probable case in Yugoslavia (1979). The toxin of Loxosceles can provoke
in man either the cutaneous or the viscerocutaneous form. The local skin necrosis,
which was called earlier "the gangrenous spot of Chile" (Macchiavello 1937) begins
with a violaceous plaque surrounded by edema and with blisters shanging to an eschar
which will slough, leaving a necrotic ulcer healing within 1- 3 weeks. In the viscero-
cutaneous form besides identical local manifestations, there develops within the first
24 h a hemolytic anaemia with jaundice, hematuria , hemoglobinuria, nephrosis, oli-
guria and even anuria (Fig. 47). This form occurs in about 13% of bites. While the
cutaneous form is not life threatening, the lethality of the viscerocutaneous form is
about 29.4% (Schenone and Suarez 1978).
The treatment must be prompt since when local or systemic lesions appear, the
therapy can only reduce their consequences. The treatment is divided into nonspecific
Fig. 46. Cutaneous necrosis,

probably into the inradermal
layers, following the bite of
Lycosa tarentula
158 Z. Maretic
Fig. 47. a Necrosis of the right thight fol-

lowing the bite of Loxosceles rufescens.
(Case of Maretic and Russell 1979). b Viscero·
cutaneous loxoscelism in 7-year-old girl, nec-
rotic eschar 21 days after the bite. (From
a Schenone and Suarez 1978)
and specific treatment. Nonspecific, corticosteroids must be given in large doses during
the first 2 -4 days, then reducing them later. The parenteral route is preferred due to
lesions of the digestive mucosa in systemic involvement (Schenone et al. 1959; Gajardo-
Tobar 1966). Depending upon the course of the disease, antibiotics, surgical treatment
of necrotic lesions, blood exchange transfusion and hemodialysis are taken into con-
sideration.
In Lima, Vellard (1954) produced the first Loxosceles (L. meta) antiserum on
donkeys. Furlanetto prepared an L. rufescens antivenin on horse in Instituto Butantan,
Sao Paulo, which is, at present, the only producer of Loxosceles antivenin. The opinions
on the success of Loxosceles serotherapy are very controversial. Furlanetto (1961) is
of the opinion that it is successful only if applied within 36 h following the bite.
8.s A traxism
In most cases the very toxic male spider injects very little venom. Thus, these patients
do not develop illness or a complete clinical picture is not present (Sutherland 1981).
Otherwise, the bite is very painful for hours, even days; but locally, only sweating,
erythema and piloerection, as in latrodectism, occur. Systemic symptoms may appear
within 10 min with nausea, vomiting, abdominal pain, diarrhea, profuse sweating, saliva-
tion, lacrimation, severe hypertension, dyspnea and sometimes pulmonary edema due
to acute intracranial hypertension, tachycardia, arrhythmia, hyperthermia, muscle
fasciculation and trismus of masseters. Investigations on monkeys disclosed excessive
catecholamine release and metabolic acidosis . Confusion, coma and an asphyxial death
can develop. In the second part of the syndrome the production of secretions subsides,
but the muscle activity continues. Until 1980 at least 14 fatalities due to A. robustus
male bites were recorded. The female does not cause serious illness in man. A. robustus
(Sydney funnel-web spider) is limited to an area of 160 km radius from the center of
Sydney. There are only two records of similar envenomization by A. formidabilis,
which is limited to some sparsely populated areas (Sutherland 1981).
First aid consists of applying a broad tourniquet and immobilization; atropine, fru-
semide and diazepam are given (Sutherland 1981). In 1980 Sutherland prepared a
purified male spider antivenin on rabbit. Using A-Sepharose, he obtained 1.1869 g im-
mune IgG from 55 ml serum. In vitro, 97 Ilg of this IgG was able to neutralize 1 Ilg of
the venom and protect experimental animals. In patients bitten by A. robustus and
also by A. formidabilis, and treated by the antivenin intravenously, a rapid improve-
ment was recorded. A pretreatment with antihistamines and hydrocortisone is recom-
mended (Sutherland 1981; Knight and Sutton 1982).
8.6 Phoneutrism
The pain due to the bite is so severe that it may cause shock if not combated immedi-
ately. It irradiates to the whole affected limb extending to the trunk. Locally, only a
punctiform mark or a mild erythema is seen General symptoms consist of sweating,
hypothermia, tachycardia and hypertension. In severe cases, erection was observed.
Since with the bite only small doses of this potent venom are injected, death rarely
occurs (Schenberg and Pereira Lima 1978).
Since severe local pain is the main symptom, one immediately applies morphine
intramuscularly, even intravenously, as well as local infiltration with xylocaine. How-
ever, contraindications for morphine are severe tachypnea and very young age, in such
cases morphine could affect even more the respiration. Important is peripheral vaso-
dilatation, which is attained with warm water, rubefacient ointments or even by intra-
venous doses of nicotamide. Since the venom contains histamine and histaminolibera-
tors, antihistamines must be given. In severe cases the antivenin must be given in large
doses, 1- 5 ampoules intramuscularly plus the same dose intravenously. The Phoneutria
antivenin was first prepared by Brazil and Vellard (1925), and now the Instituto Butan-
tan produces a polyvalent Phon eu tria-Lycosa-Loxosceles antivenin.
IV Comparative Structure and Histochemistry
of Silk-Producing Organs in Arachnids
JACQUELINE KOVOOR 1
CONTENTS
1 Introduction 160
2 Silk Production and Use in Arachnids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
2.1 Mites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
2.2 Pseudoscorpions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
2.3 S p i d e r s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
3 Silk Glands .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
3.1 Prosomai Glands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
3.2 Opisthosomai Organs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
1 Introduction
Among arthropods, silk is produced by insects, myriapods and arachnids, and extruded
as fine fibers before being processed into threads, ribbons, sheets, webs or cocoons
(Rudall and Kenchington 1971). The larvae of insects from many groups secrete a great
variety of silks, but only one kind is produced by a single individual. On the other hand,
the arachnid silks hitherto studied do not show as much structural and chemical diversity
as those of insects; although at least two different silks may be produced by a single
pseudoscorpion or mite, and no less than nine types of fibers are known to emerge
from the spinnerets of individual spiders of certain species.
Our knowledge of arachnid silks is still very scant and fragmentary. However, ex-
tensive and continuing studies on the silk-producing organs of arachnids over the last
10 years might renew the interest of specialists in the various fibrous proteins syn-
thetized and spun by these remarkable animals.
2 Silk Production and Use in Arachnids
Silk-producing arachnids belong to three orders: mites, pseudoscorpions and spiders.
1 Museum National d'Histoire Naturelle, Laboratoire de Zoologie-Arthropodes, 61 rue de Buffon,

F-75231 Paris, France
Comparative Structure and Histochemistry of Silk-Producing Organs in Arachnids 161
Mites from nine families of Pro stigmata (Acariformes) have been observed to spin silk,
but all do not use it for the same purposes. The well-known pest, Tetranychus (Tetra-
nychidae) for example, spins webs and feeds on cultivated plant leaves damaging them.
The nest is built throughout the life cycle and is used as a moulting chamber by mites
of both sexes (Hazan et al. 1974). Andre (1932) has observed Tetranychus making use
of draglines and their males spreading silk threads over the eggs laid by females; bal-
looning of Tetranychus in the wind has also been described (Fleschner et al. 1956).
Cunaxidae and Bdellidae moult in silken chambers, draw silk over their eggs and con-
struct special nets for capturing prey (Alberti 1973; Alberti and Ehrnsberger 1977)
(Fig. 49A). In some cases silk threads function as supports for spermatophores (in
Rhagidiidae; Ehrnsberger 1977, 1979) or as signposts to them (in Trombidiidae; Moss
1960). Spermatophore stalks, when present, might be made up of a fibrous silk-type
protein.
2.2 Pseudo scorpions
Silk produced by pseudo scorpions is used by the females to construct breeding cham-
bers within which they lay eggs. Juveniles hatch into protonymph in the maternal
chamber, then leave it and build their own silken chamber prior to each subsequent
moult. Adult males as well as females of Chthonius and Roncus spend the cold season
in hibernation chambers, which is not the case with Neobisium males (Wood and Gab-
butt 1979a,b).
In pseudoscorpions, sperm is transferred through the deposition of spermatophores
which are more or less complex, but always stalked, and in Chthonius ischnocheles
shown to be elaborated from a kind of silk (Hunt and Legg 1971; Legg 1973). Males
of the olpiid Serianus carolinensis deposit a spermatophore and then spin silk threads
from the top to bottom of the substrate in two rows, which indicate a path for leading
the female (Weygoldt 1966c).
2.3 Spiders
Silk production is of cardinal importance for spiders; it is a hallmark of its life, and of
all vital functions (Nentwig and Heimer, this volume; Decae, this volume). The physical
and chemical characteristics of silks produced diversify to correspond to these different
functions (Peters, this volume; Tillinghast and Townley, this volume); they may be
evident within individuals of a single species, but also noticeable among the sets of silks
produced by members of different systematic groups.
3 Silk Glands
Prosomal as well as opisthosomal silk-producing organs occur in the three orders of

arachnids. Prosomal glands of mites and pseudoscorpions fulfill several functions, while
162 J. Kovoor
Table 15. Glandular origin of silk in arachnids
Arachnid order Prosoma Opisthosoma
Mites Anterior pro somal glands in Accessory glands of the male genital
juveniles and adults of both tract for spermatophore production
sexes, in nine families of
Acariformes
Pseudo scorpions Prosomal glands in juveniles and Accessory glands of the male genital
adult females, degenerating tract for spermatophore production
in males of certain families Glandular part of the rectal pocket in
Serianus males
Spiders Prosomal glands in Scytodes of Epiandrous glands in males of most
both sexes, at all stages of families
development Vas deferens in males (spermatophore
of Telema)
Silk glands proper of the 4th and 5th
opisthosomal segments, in juveniles
and adults of both sexes
the silky product of certain opisthosomal glands related to the male genital tract is used
only for the elaboration of spermatophores. The most prominent and functionally sig-
nificant silk glands are found in the opisthosoma of spiders. The origins of the silk
fibers in the three orders are indicated in Table 15.
3.1 Prosomal Glands
3.1.1 Silk Glands of Mites
Prosomal silk-producing organs in acariform mites generally belong to the podocephalic

system, that is to the complex system of three or four pairs of glands opening into a
podocephalic canal. The latter is a common salivary duct which collects all the secretory
products and extrudes them into the cheliceral groove. Silk glands are thought to be
modified salivary glands.
In Bdellidae, they were described as acinous units comprising six to eight cells sur-
rounding a small lumen; the secretion of the infracapitular gland which empties through
its own duct, and that of the third podocephalic glands contribute to form the extruded
silk that is used to build moulting nests, egg cover and capture threads (Alberti 1973;
Alberti and Storch 1973) (Fig. 48A). In the latter glands, one of the six constituent
cells always contains very conspicuous secretory granules.
Silk glands of Rhagidiidae would be, according to Ehrnsberger (1979), the first and
the third podocephalic glands (Fig. 48B). It is unlikely that the infracapitular gland
takes part in silk secretion since the latter is quite unfunctional in larvae.
In Tetranychidae, silk glands consist of a single pair oflarge glandular cells extend-
ing laterally to the oesophagus and ending in an excretory duct that goes through a
hollowed hair at the tip of the pedipalp (Fig. 48C). In the case of these mites, none
Fig. 48A-C. The prosomal

gland system in mites (one side
only). A In Bdellidae (after
Alberti and Storch 1973,
modified). B In Rhagidia halo·
phi/a (Rhagidiidae). (After
Ehrnsberger 1979, modified).
C In Tetranychus urticae (after
Alberti and Storch 1974).
Abbreviations: cg coxal gland;
ig infracapitular gland; pg 1,
2, 3 podocephaJic glands;
sg silk gland; tg tracheal gland .
Silk glands are finely dotted
of the podocephalic glands, reduced to three pairs, can be considered as a silk gland
(Alberti and Storch 1974 ; Mothes and Seitz 1981).
3.1.2 Histochemical and Chemical Data on Mite Silk
No obvious difference appears in scanning electron microscopy between the secretory

product used for the nest and those for the capture threads of Bdellidae. Freshly de-
posited threads adhere to the ground or to the cuticle of prey and lose their stickiness
after some time. The spinning ability of Cunaxidae has been demonstrated in two spe-
cies and a careful observation of the capture nest built by Cunaxa revealed two kinds
of threads : thick, smooth threads form the scaffolding of the nest, while thin threads
bearing equally-spaced nodules cross the other threads in an irregular meshwork (Alberti
and Ehrnsberger 1977) (Fig. 49). Whether these thin threads are sticky has not yet
been established. Bdellid infracapitular and third po do cephalic gland secretions are
proteinaceous: neither acidic mucosubstances nor lipids have been detected histo-
chemically; both proteins are rich in tyrosine (Alberti 1973). Through the transmis-
sion electron microscope, the secretory granules of the infracapitular glands appear at
first as tubulofibrils in the dilated portions of the endoplasmic reticulum, but on ma-
turing, look compact and no longer exhibit this structure (Alberti and Storch 1973).
164 J. Kovoor
Fig. 49. A Capture nest of Cunaxa mites; B detail of the silk threads. (After Alberti and Ehrnsberger
1977, redrawn)
According to Andre (1932), nests of Tetranychus are never adhesive. In Tetranychus,

silk gland cells are filled with electron-lucent granules which originate from the endo-
plasmic reticulum, the most prominent structure of these cells. It can therefore be as-
sumed that the secretory product is proteinaceous. The amino acid composition of the
carmine spider-mite (Tetranychus cinnabariumj silk is known from the study of Hazan
et al. (1975); it contains a higher percentage of aspartic and glutamic acids than other
arthropod silks. This characteristic together with the amount of non-polar amino acids,
small side-chain residues, and tyrosine content recalls the prekeratin of vertebrate epi-
dermal tissues. Spider-mite silk resembles pseudoscorpion silk, but differs in its higher
proline content.
3.1.3 Pseudoscorpion Silk Glands
Pseudoscorpions are provided with two to about twenty prosomal silk glands (With
1906; Vachon 1966). Their excretory ducts open either in a sclerotized area, on the
tip of the movable finger of the chelicerae, the spinning tubercle of many Neobisiidae
and all Chthoniidae (except one species), or in the galea, a ramified extension of the
cheliceral finger, each ramification containing one excretory duct. The galea exhibits
sexual dimorphism: in adult neobisiid and garypid males, it is atrophied along with the
atrophy of the male silk glands (Bertkau 1887; Kaestner 1927). In Chthoniidae, tubular
silk glands are located only in the prosoma, while in Neobisiidae, they can be very large,
extending far into the opisthosoma (Vachon 1949).
3.1.4 Chemical Data on Pseudoscorpion Silk
Apart from a few general histological characteristics, nothing is actually known of the
histochemical reactivity of these glands. An amino acid analysis of the silk from Neo-
bisium maritimum made by Hunt (1970) shows it to be made up of a protein recogniz-

able as a silk from its relatively high glycine and serine contents. The presence of re-
ducing groups in a moderately high proportion distinguishes it from lepidopteran and
spider silks which contain only small amounts of cysteine. On the other hand, pseudo-
scorpion silk, with increased contents of basic and acidic amino acids, resembles the
epidermal prekeratin of mammals.
3.1.5 Cheliceral Glands of Scytodes Spiders
The only prosomal glands which are known to produce silk in spiders have been found
in the genus Scytodes. The peculiar behaviour of these spiders when catching prey
amazed observers such as Monterosso (1928), Bristowe (1931b) and Dabelow (1958),
who described it in detail. A gluey silk is projected at a distance of 1.5 - 2 cm from the
chelicerae onto the victim, which thus stuck to the spot. This silk originates from the
venom glands (Millot 1930b) which are especially wide in Scytodes and made up of
several lobes and a large pouch where the silk is produced and stored (Fig. 50A).
A histochemical and ultrastructural study of the glands (Kovoor and Zylberberg
1972) showed the gluey silk to be a fibrous glycoprotein. Each elementary fiber consists
of an axial tubule (proteinaceous) on which thin, ondulated glycoproteinaceous fibrils
are radially attached. There are two kinds of fibers differing by the length of the radial
fibrils (Fig. 50). Before extrusion, the fibers are packed in the apical part of the glan-
dular cells in a paracristalline form.
3.2 Opisthosomal Organs
Although no opisthosomal silk gland has yet been described in adult females of mites
or pseudoscorpions, the most elaborate spinning apparatus lies in the opisthosoma of
female spiders.
3.2.1 Accessory Glands of the Male Genital Tract
In many edaphic mites and all pseudoscorpions, glandular parts of the male genital
tract produce fibrous proteins. A similar process has been recently discovered in the
male of telemid spiders (Lopez 1977a; Legendre and Lopez 1978; Juberthie et al.
1981; Lopez, this volume), so that at least in spiders silk glands seem to be derived
from epidermal invaginations.
3.2.1.1 Mites. The genital tract of mites and pseudoscorpions is a complicated assem-
blage comprising the testes, the vas deferens and various accessory glands as diverticula
of the genital atrium (Henking 1882; Michael 1896; Thor 1904). Both direct and in-
direct sperm transfer occur in mites. The secretion of a fibrous material is associated
with the production of differently shaped spermatophores. In Rhagidiidae, the sper-
matophore is only an envelope for the sperm packed deposited on a silk thread (Ehrns-
berger 1977). Allothrombium, thrombidiid mites, deposit their spermatophores, simple,
long-necked bottles, on the ground and enlace them with a few silk threads (Moss 1960).
166 1. Kovoor
The structure of erythraeid and oribatid spermatophores is more complicated (Wood-

ring 1970; Witte 1975; Waitzbauer 1983); silk is absent as supporting or fastening
threads. The stalk of the spermatophore itself consists of a fibrous substance drawn
out of the genital aperture as a translucent fluid which quickly hardens in air. In most
cases, the stalk and the crown are supposed to originate from the glandular part of the
testes (Witte 1975) and be secreted by the vas deferens in Bdellidae according to Alberti
(1974).
Histochemical studies by Witte and Storch (1974) on erythraeid genital organs have
shown the glandular part of the testes secreting three substances, one of which is a
protein rich in tyrosine, amino-terminal and reducing groups; the same histochemical
tests give very similar results on the stalk of the spermatophore. The crown contains
an acidic mucosubstance, indolic and sulfhydryl groups which are also found in another
secretory product of the testes.
32.1.2 Pseudoscorpions. Pseudo scorpions do not actually mate. Males transfer the
sperm indirectly by means of spermatophores deposited on the ground in the presence
or absence of females. Spermatophores are produced by accessory glands of the male
genital tract and shaped by the genital atrium (Weygoldt 1966a,b, 1969a). The struc-
ture of the spermatophores ranges from a simple, thin stalk bearing the sperm packet
as a small ball at the tip in the Neobisiidae, Garypidae, Olpiidae and Cheiridiidae
(Chamberlin 1931 ;Kew 1930; Weygoldt 1969a) to a complex shape involving the deve-
lopment of a bipartite collar (Chthoniidae) or spinelike structures (Tridenchthoniidae).
The spermatophores of the Cheliferidae exhibit the most complicated structure (Wey-
goldt 1969b); in the Chernetidae a drop of liquid is added to the stalk under the sperm
packet; it is absorbed by the wall of the latter which swells and causes the release of
the sperm. A liquid droplet is also present on the stalk of the spermatophores of chthoni-
ids, but, in these cases, it is supposed to contain a pheromone attractive to females.
The stalk of the spermatophores has a somewhat fibrous appearance. Each part of a
spermatophore is secreted by one kind of accessory gland (Vachon 1938; Legg 1971).
Its shape is moulded in the elaborate diverticula of the genital atrium through which
the secretory products have to pass. The stalk, collar and support of the sperm packet
originate from the secretion of the posterior dorsal glands. These glands are the most
prominent of all accessory glands. They consist of a certain number of club-shaped
lobules with a thick secretory epithelium extruding the product in a central lumen;
each lobule is connected to an excretory duct emptying in an invagination of the genital
atrium. The secretory product looks heterogeneous in Chthonius ischnocheles (Legg
1973) and homogeneous in Cheridium museorum (Legg 1974a) and Hysterochelifer
meridianus (Bois sin et al. 1970). Histochemical tests show the product to be mostly a
protein containing sulfhydryl groups, tyrosine, and arginine; besides the protein, a
lipidic substance is found, especially in Chthonius, that is responsible for the hetero-
geneous aspect of the secretion. The same tests applied to the spermatophores give
Fig.SOA-E. Silk secretion in the cheliceral gland of Scytodes. A Sagittal section of the gland show-
ing the silk secreting pouch (s) and the anterior venom lobes (v). Bar = 100 /.1m. B Portion of two
neighbouring cells elaborating different fibers. Bar = 0.5 /.1m. D, E Details of the two kinds of fibers
in transverse section. Bar = 0.2 /.1m (Kovoor and Zylberberg 1972)
Fig. 50
168 J. Kovoor
similar results on the stalk. Mucosubstances and polysaccharides are absent in both the
spermatophore stalk and the secretion of the posterior dorsal glands.
Chemical characterization of the protein of the spermatophore indicates that the
amount of small side-chain amino acids (glycine, alanine and serine) nears 50% and that
acidic amino acids exceed basic ones. These two characteristics are typical of silk fibro-
ins (Hunt and Legg 1971). The structural protein of the spermatophore differs from
that of the nest whose silk is richer in proline and basic amino acids.
3.2.2 EpUlndrous Glands of Male Spiders
Epiandrous glands were discovered in males of mygalomorphs as numerous small acini-

form glands, each opening through a simple spool on the anterior lip of the epigastric
furrow. Their secretory product seems to be added to the special net built by males for
depositing the semen, which then will fill their palpal bulbs (Me1chers 1964; Legendre
1972). Similar glands have also been found in male spiders of other suborders, liphisti-
ids as well as araneomorphs (Marples 1967; Lopez 1974; Lopez, this volume). They are
lacking in some families (Lopez 1977a). Only two of these glands are found in the
Pholcidae, Scytodidae and Theridiidae. The shape and histochemical characteristics of
the epiandrous glands are about the same in all spiders studied. The silken nature of
the acidophilic secretion is inferred only from the somewhat fibrous interior of the
glands and the positive histochemical tests for proteins in general, tyrosine and reactive
amino, reducing, and carboxyl groups (Kovoor, unpubl.).
3.2.3 The SerUlnus Case Among Pseudoscorpions
In the olpiid SerUlnus carolinensis, adult males secrete silk from the glandular portion
of their rectal pocket (Weygoldt 1966c, 1969a). It is the only known case of a silk
gland derived from an endodermal region of the digestive tract. Histochemical data on
this gland and the fibers it produces are presently lacking.
3.2.4 Silk Glands Proper of Spiders
Silk glands of spiders are probably derived from epidermal invaginations of the fourth
and fifth segments of the opisthosoma which bear the spinnerets. According to Bristowe
(1932), these glands could have possibly developed from segmentary coxal glands
primitively associated to appendages which have become adapted into spinnerets after
losing their ambulatory function. Ancient spiders most probably had four pairs of
similarly functioning spinnerets bearing spinning tubules, the spools, which were simple,
modified hairs. Evolution has led to the great diversity in the spinning apparatus of
present-day spiders seen from structural, functional, physical and chemical aspects of
the silk secreted and spun by members of the various systematic groups (e.g. Fig. 71).
Anatomical and histological studies of silk glands should have priority before histo-
chemistry, followed by the chemistry and physico-chemistry of silks. The first chemical
approaches to spider silks did not take into account the anatomical data provided by
arachnologists at the end of the 19th and the beginning of the 20th centuries. Chemical
analyses were, therefore, limited to a few species, mostly common araneid spiders of
the genera Araneus or Nephilo. and to the entire web or threads pulled from the spin-
nerets. In 1970, Andersen performed valuable analyses on the secretory products from
each kind of silk gland isolated from Araneus diadematus and demonstrated the actual
variety of silks produced by a single species (Table 18). During the last 15 years, histo-
chemical methods have been applied to the study of the silk glands of a large number
of spiders representing at present 20 families from about 90 that are included in the
araneomorph suborder. In the case of liphistiomorphs and mygalomorphs, precise
histochemical reports on their silk glands began appearing only a few years ago (Palmer
et al. 1982; Palmer 1985; Haupt and Kovoor, in prep.).
The silk glands of a spider are basically a sac, the open end of which is connected to
an excretory duct running into a spool, or spigot, on a spinneret. The sac comprises a
secretory epithelium lying on a thin basal lamina and a surrounding lumen were the
secretion extruded from the cells is stored. Only the excretory duct is lined internally
by a cuticular intima produced by its cells and fusing with the cuticle of the spool near
its opening at the tip. The types of spinnerets in the three major groups of spiders are
indicated in Table 16.
The structural variation of the spinnerets and their spools is an external sign of the
even more extensive variation of the silk glands (Fig. 51). Spiders can be classified as
generalists or specialists according to the structure and composition of their spinning
apparatus, and correlatively to the use made of the silk fibers produced. The generalist
spiders produce one to three kinds of fibers which are indifferently used for the construc-
tion of their retreats or burrows as well as egg cases or sperm webs. Silk glands of the
specialists are more diversified: four to nine kinds of fibers are produced and each type
of construction has its characteristic fiber composition. Furthermore, any construction
Table 16. Variations of the spinnerets in the m~or groups of spiders
Spinnerets Suborder
Liphistiomorphae Mygalomorphae Araneomorphae
Anterior lateral 2, large, pluri- 2, small, uni- or bi- 2, biarticulated, functional

articulated, func- articulated, func-
tional tional or non-
functional or
absent
Anterior median 2, small, uni- absent 2 or 1 (cribellum) func-
articulated, non- tional, or colulus non-
functional or functional, or absent
absent
Posterior median 2 or 1, very small, 2, uniarticulated, 2, medium-sized or small,
uniarticulated, functional or non- uniarticulated, func-
nonfunctional or functional or absent tional or nonfunctional
absent or absent
Posterior lateral 2, large, pluri- 2, large, pluri- 2, biarticulated, functional
articulated, func- articulated, func- or nonfunctional or
tional tional absent
170 J. Kovoor
of these spiders comprises several types of silk fibers. The only exception lies in the
draglines of araneomorphs which seem to be composed of fibers emerging only from
the spigots of ampullate glands.
3.2.4.1 Silk Glands of Liphistiomorphs and MygaIomorphs. The simplest silk gland
system may be expected among the spiders exhibiting the largest number of primitive
(plesiomorphic) anatomical characteristics. These spiders are presently the liphistio-
morphs or Mesothelae, a name which indicates the location of the spinnerets on the
median part of the opisthosoma (Fig. 51A). In fact, Liphistius are provided with three
kinds, and Heptathela and Ryuthela with four kinds of silk glands (Haupt and Kovoor
in press).
The simplest situation has been described by Palmer et al. (1982) for the genus
Antrodiaetus which spend their life in burrows and possess only one kind of very
nwnerous, small-sized silk glands of the acinous type. These glands feed only two pairs
of spinnerets (posterior) and secrete two different proteins in their distal and proximal
halves, respectively. The distal cells produce a basic protein rich in sulfuydryl groups,
and the proximal ones an acidic protein with numerous C-terminal groups. The silk
fiber has, therefore, a double composition: a core and an outer coating, both devoid of
a polysaccharide component detectable histochemically.
The silk glands of Euagrus, diplurid spiders with two pairs of posterior spinnerets
which build extended sheet webs connected to a tubular retreat below stones or logs,
are also of a single kind (palmer 1985), although variable in size and shape. Each gland
comprises three secretory regions, distal, middle and proximal, which elaborate a hydro-
phobic, an amphophilic and a hydrophilic protein, respectively. In the last mentioned,
free aldehydes and a neutral polysaccharide are detected together with sulfuydryl and
disulfide groups.
Other diplurid spiders (e.g. Hexathele hochstetteri) with six functional spinnerets
were studied by Glatz (1972), who found three types of silk glands. The anterior spin-
nerets bear the spools of bipartite acinous glands which secrete a basophilic product in
their proximal cells, while the posterior median and lateral spinnerets are fed with large
and small tubular-acinous glands. An acidophilic substance is secreted by the distal
cells, and a basophilic one in the proximal third of these glands. No histochemical data
are available for this species.
Atypus, atypid mygalomorphs living in a tube of thick silk woven pardy under-
ground, and whose anterior spinnerets are not functional in adults, show two kinds of
silk glands corresponding to the large and small tubular ones of Hexathela (Glatz 1972,
1973). But the large silk glands of Atypus are made up of a single type of secretory
cells, the cyanophilic product of which reacts rather weakly to the general tests for
proteins and for reducing groups and tyrosine. On the contrary, the small glands which
secrete two proteins give moderate positive reactions in the distal cells, while the prox-
~
Fig.S1A-F. Spinnerets of spiders. A Liphistius malayanus (Liphistiidae) (Photo J. Haupt). Bar =
0.5 mm. B Linyphia triangularis (Linyphiidae). Bar = 100 101m. C Filistata insidiatrix (Filistatidae).
Bar = 200 101m. D Pholcus phalangioides (Pholcidae). Bar =50 101m. E Segestria florentina (Segestri-
idae). Bar =0.4 mm. FScytodes thoracica (Scytodidae). Bar = 100 101m.
Abbreviations: AS anterior spinneret; CO colulus; CR cribellum;MS median spinneret;PS posterior
spinneret; A T anal t~bercle
Fig.51
172 J. Kovoor
imal ones are very poor in tyrosine and reducing groups though appearing very rich in
carboxyl groups (Kovoor, unpubl.).
Silk glands of the theraphosid spiders have been studied only in one oriental species,
Melopoeus albostriatus (Kovoor, unpubl.) which dig burrows about 30 cm deep. Their
two pairs of posterior spinnerets are fed with two and three kinds of glands, respective-
ly the large tubular glands widened in the proximal region, the small tubular ones, and
the fewer acinous ones. In all glands, two distinct proteins are produced in the distal
and proximal parts, the expansion of which differs according to the kind of glands.
The secretory products of the acinous glands are obviously different in their histo-
chemistry from that of the other two kinds. The basophilic protein of their prominent
prOximal partis rich in tyrosine, but does not contain reducing groups; a polysaccharide
component seems to be added to the protein. Histochemical tests give closely similar
results in large and small tubular glands. The two proteins differ in their reducing groups
and tyrosine contents, much higher in the proximal than in the distal parts of the glands
which contain an acidic mucosubstance . The fibers resulting from the binding of the
two proteins forming their core and outer coating, respectively, differ both in their
diameters and the relative proportions of their two components (Fig. 53).
This mode of fiber differentiation which does not involve a chemical modification
of the proteins produced also appears in the trapdoor spiders Cyrtocarenum and Ne-
mesia. The anatomy of the spinning apparatus in the two species of these genera was
studied by Glatz (1972, 1973); histochemical data on the same species were added
recently (Kovoor, unpubl.). Three kinds of silk glands are distinguished at first as large-,
medium- and small-sized and open in spools of three different shapes (Fig. 52). In
Cyrtocarenum, three different proteins are elaborated in the three regions of all glands.
Histochemical differences from one type of gland to another are not at all obvious.
The protein of the distal region is rich in reducing groups which are lacking in the other
Fig_ 52A,B. Spools and spigots on a posterior spinneret of (A) a mygalomorph, Cyrtocarenum sp.,
and (B) an araneomorph, Tetragnatha extensa. Bars: A = 100 Slm; B = 20 Slm
Fig. 53. Longitudinal sections of mygalo-

morph silk glands and the fibers originat-
ing from their secretory products.
(Explanation in text)
n IEII
Cyrtocarenum
Melopoeus
g
t
two regions where carboxyl groups are especially abundant. Polysaccharides, other than
glycogen, have not been detected in any part of the glands. In both cases, the extruded
fibers differ in the proportions of the three proteins from which they are composed
(Fig. 53).
The small group of liphistiomorph spiders also dig burrows operculated by a trap-
door; the whole construction is lined with silk and Liphistius, furthermore, spin fishing
lines in front of the burrow (Bristowe 1932). These spiders offer other aspects of a
limited diversification of the silk produced. In Liphistius malayanus, the anterior lateral
spinnerets are supplied with two kinds of glands and the posterior ones with only one
kind. All these glands, derived from the acinous type, comprise three different regions
(Fig. 54A,D). The secretory products of the distal and middle regions appear in the cells
as granules composed of a central part and a cap which are histochemically distinct.
The distinctive features of the glands, apart from their location in the opisthosoma, are
based on the more or less extended collar; moreover, the histochemical characteristics
of the composite granules differ from one region of the glands to another, and within
the same region from one kind of gland to another. In the collar and the middle part
of the anterior glands, the proteins are associated with an acid polysaccharide compo-
nent; in the posterior glands, the same region secretes a glycoprotein. The polysac-
charide component is always detected in the outer cap of the composite granules; reduc-
174 J. Kovoor
ing groups decrease from the distal part to the collar (Fig. 54D). In Heptathela kimurai
and Ryuthela nishihirai, each pair of spinnerets is supplied by two kinds of glands
opening through long simple spools. The anterior glands comprise a major part of cells
full of composite granules and a more or less extended collar secreting histochemically
different smaller granules. The posterior glands, on the other hand, consist of a single
category of secretory cells (Fig. 54B).
All the secretory products are proteinaceous and have a double composition that is
obvious for the intracellular granules made up of two unequal distinct portions. The
granules decrease in size from the first (anterior) to the fourth (posterior) kinds of
glands. In the anterior glands of Heptathela kimurai an acidic mucosubstance is as-
sociated with the protein and forms the outer layer of the composite granules (Fig. 54C),
while the center is a more or less reducing protein. Mucin is the main component of
the collar secretion. Among the posterior glands, one set secretes an acidic glycoprotein
and the other set, a weakly acidic one. Silk glands of Ryuthela nishihirai do not appear
to produce any acidic mucosubstance; one kind of gland only secretes a reducing glyco-
protein containing terminal-amino groups (Fig. 54B).
The anatomical differences between the glands are, in these cases, less important
than the variations of the histochemical characteristics of their secretory products
from which fibers mostly differentiate. Various combinations of anatomical variations
of the glands and chemical variations of the secretory products are the origin of dif-
ferentiation in the spinning apparatus of araneomorph spiders.
3.2.4.2 Silk Glands of Araneomorphs. The suborder araneomorphs includes the great
majority of the spiders, about 35,000 species in 90 families distributed all over the
world and demonstrating their wide possibilities of adapting their structure and be-
haviour to various ecological conditions. The spinning apparatus often undergoes adap-
tative variations, but some basic familial characteristics usually remain unchanged. The
major trend of the evolution of the spinning apparatus in araneomorphs seems to be a
functional specialization of each pair of spinnerets and the afferent glands, involving
precise modifications of the shape of the spinnerets, the number and morphology of
spools or spigots and the anatomical and histochemical characteristics of the correspond-
ing glands.
In araneomorph spiders the spinnerets, as in mygalomorphs, are located near the
end of the opisthosoma, the last segments of which are confluent in an apparently un-
divided and shortened region. As indicated in Table 16, one to four pairs of spinnerets
are present in araneomorphs. The lateral anterior pair is the only one to be functional
in all species. The median anterior spinnerets shown by spiders of certain lineages have
been named cribellum by Blackwall (1841) in view of their flattened shape and ap-
parently porous surface (Fig. 51). The only spiders possessing eight functional spin-
nerets are, indeed, the cribellate araneomorphs. The colulus is a nonfunctional rem-
nant of the median anterior spinnerets (Fig. SIB).
The araneomorph spinning apparatus is a combination of large and small glands,
the former being generally fewer. Silk glands of araneomorphs are better characterized
than those of the mygalomorphs or liphistiomorphs. Apstein (1889) studied the
anatomy of the spinning apparatus in a number of families of araneomorphs and clas-
sified the silk glands according to their shape and which spinnerets they supply, into
Fig. S4A-D. Silk glands in liphistiomorphs. A, D Liphistius maJayanus cameroni; A glands of the
posterior spinnerets; general reaction for proteins of decreasing intensity from the distal to the
proximal regions of glands; D glands of two distinct kinds opening on the anterior spinnerets; more
or less positive reaction for reducing groups (black or grey). Both bars = 100 ",m. B Ryuthela nishi-
hirai; portion of two kinds of glands feeding the posterior spinnerets, positive reaction for polysac-
charides, black. Bar = 25 ",m; C Heptathela kimurai; portions of two kinds (a"ow and arrowheads)
of glands opening on the anterior spinnerets, all secretory granules have a double structure, black
positive reaction for acidic mucosubstances. Bar =50 ",m (Haupt and Kovoor , in prep.)
176 J. Kovoor
categories that are still valid with slight improvements or modifications with respect to
recent studies. For more comprehensive coverage of earlier literature on araneomorph
silk glands, see Kovoor (1977a).
Variations in the Composition of the Spinning Apparatus. In some families, the dif-
ferentiation of silk glands is limited to the three categories that are found in almost all
araneomorphs: ampullate, aciniform and pyriform glands. In the very simple case of
Scytodes and Loxosceles the only pair of large-sized and tubular glands corresponds to
the ampullate type of Apstein (1889); two other types of glands are found which con-
serve the primitive acinous morphology and small size: the so-called pyriform and the
aciniform glands. Dysdera, Diguetia (Lopez 1983b) and Telema (Kovoor and Lopez
1983a) all probably related spiders, also show a single pair of ampullate glands, as well
as pyriform and aciniform glands, the latter being diversified into two kinds. Salticid
spiders of both sexes are provided with the same three kinds of glands, but the ampul-
late ones are in four pairs, a couple of them opening on each of the an terior lateral and
posterior median spinnerets (Millot 1935).
Another type, namely the tubuliform glands which provide silk for egg cases, ap-
pears in female Agelenidae and Lycosi"dae, whose spinning apparatus otherwise shows
the same composition as that of Salticidae (Kovoor 1976). Further differentiation oc-
curring in other families, such as Segestriidae, Clubionidae, Gnaphosidae or Thomisidae
concerns the ampullate glands whose characteristics differ according to the spinnerets
to which they are connected. Hersiliidae show, in addition, two kinds of aciniform
glands opening on the posterior lateral spinnerets (Kovoor 1984). Furthermore, Arane-
idae and related families, which comprise the most specialized spiders, have developed
two other types of silk glands, the aggregate and flagelliform glands corresponding to a
functional unit on the posterior spinnerets used to build capture threads (Sekiguchi
1952; Peters 1955; Kovoor 1972) (Fig. 52B). In cribellate spiders, the cribellum and
the median and posterior spinnerets are involved in elaborating special capture threads
through the functioning of the cribellar, paracribellar and/or pseudoflagelliform glands
(Peters, this volume).
Ampullate Glands. The term ampulla refers to the shape of the proximal part (sac) of
the ampullate glands in many araneomorphs, the distal part being a long, convoluted
tube (tail). But Telemidae and Oecobiidae, for example, have globular ampullate glands;
in Hypochilidae and Dysderidae, these glands are simply tubular. A crescent-shaped
ampulla is a peculiarity oftheridiid spiders (Kovoor 1977b; Kovoor and Lopez 1983b)
and an S-shaped one, that of Hersiliidae and Nephilinae (Kovoor 1984, 1986). The
Fig. 55A-D. Ampullate glands of araneomorphs. A Tegenaria sp. (Agelenidae), diverticula ted distal
part, two secretory products, in the proximal half of the ampulla presence of a polysaccharide com-
ponent (black); B Cyclosa conica (Araneidae), Azan staining; Al major ampullate gland, secretion
of three proteins; A2 minor ampullate gland; Ac aciniform glands; Ag aggregate glands; F flagelli-
form gland; P pyriform glands (Kovoor and Lopez 1980). C Tubular ampullate gland of Amaurobius
sp. (Amaurobiidae), abundant reducing groups in the major part of the gland (black), but not in
the short proximal funnel. D Ampullate glands of Uroctea durandi (Urocteidae), moderately react-
ing reducing groups in the whole glands except in the proximal collar where they are lacking. Bars
= 100 !lm (Kovoor 1979)
Comparative Structwe and Histochemistry of Silk-Producing Organs in Arachnids 177
Fig. 55
178 1. Kovoor
distal part of the ampullate glands of Tegenaria (Agelenidae) is ramified (Fig. 55A).
This peculiarity perhaps indicates the original separation of the ramifications as isolated
glands, later forming a single, large gland.
Most ampullate glands comprise two categories of secretory cells, each occupying
the distal or the proximal part of the glands. Usually, the proximal part is less extended
than the distal one; sometimes it is reduced to a collar as in certain cribellate spiders,
such as Hypochilus Filistata, Dictyna or Amaurobius (Fig. 55C) and the ecribellate
spiders, e.g. Telema, Pholcus, Uroctea (Fig. 55D) and Linyphia. On the other hand, the
proximal category of secretory cells may occupy nearly the whole ampulla in certain
Araneidae (Cyrtophora, Argiope) and Uloboridae (Polenecia, Miagrammopes). Numer-
ous ampullate glands of different sizes, but secreting the same product, open on the
anterior spinnerets of Uroctea (Fig. 55D) and on the anterior lateral and median poste-
rior spinnerets of Eresidae. The large glands feeding the anterior spinnerets (major
ampullate glands) often differ morphologically and histochemically from those open-
ing on the median spinnerets (minor ampullate glands). None are connected to the
posterior spinnerets.
In certain Araneidae (Cyrtophora, Cyclosa, Gasteracantha), major ampullate glands
exhibit three different secretory regions (Kovoor and Lopez 1980, 1982) (Fig. 55B).
It is also the case of the only pair of ampullate glands in Dysdera and Diguetia (Lopez
1983b). One to three proteins are thus secreted by the ampullate glands. Their distal
part generally secretes a substance more acidophilic than that of the proximal one,
which may be only weakly acidophilic, amphophilic or basophilic. Histochemical tests
for tyrosine (Fig. 56C) and reducing or sulfhydryl groups are very strongly positive in
the protein of the distal part, and negative or weakly positive in that of the proximal
region (Fig. 55C). The protein of the latter is often accompanied by an acidic polysac-
charide (Fig. 55A) which might form a protective or water-regulating outer layer for
the silk fiber. In some families (Filistatidae, Urocteidae), both regions of the ampullate
glands elaborate a protein and an acidic mucosubstance the acidity of which is weaker
in the proximal than in the distal region.
Histochemical tests for proteins, their reducing groups and tyrosine give the most
strongly positive results in the distal part of the minor ampullate glands of Araneidae,
related families and Uloboridae (Fig. 56B). Amino-terminal groups are generally not
detected in the proteins of either part of the glands; a polysaccharide component as-
sociated with the product of the proximal region may be present in a few cases. A long,
coiled excretory duct is a characteristic of the ampullate glands which produce long
and strong elastic fibers.
Fig_ 56. A Intense reaction of the two proteins in a convoluted ampu1late gland of a male Hersilia
savignyi. Bar = 200 /.1m (Kovoor 1984). B General reaction for proteins in different glands of Zosis
geniculatus (Uloboridae), intense reaction in the minor ampullate (A2) tubuliform (T) and proximal
half of the pyriform glands (J», moderate in the pseudoflagelliform (PS), rather weak in aciniform
and cribellar glands (AC, CR) (Kovoor 1978). Bar = 100 /.1m. C Reaction for tyrosine, intense in a
small ampullate gland and moderate in the aciniform glands (AC) only in the distal part of Filistata
insidiatrix (Filistatidae). Bar = 100 /.1m. D General view of the posterior region of the opisthosoma
in Nephilengys malabarensis (Araneidae), reaction of carboltyl groups in proteins, intense in the
proximal region of pyriform glands (arrowheads), slight in aggregate (AG) and negative in acini-
form glands (arrows). Bar = 0.5 mm (Kovoor, in press). E Pyriform glands of Cyrtophora citricola
(Araneidae) with short or large proximal part strongly positive to PAS reaction for polysaccharides.
Bar = 100 /.1m (Kovoor and Lopez 1982)
Fig. S6
180 J. Kovoor
Pyriform Glands. Besides the ampullate glands, pyriform glands open on the anterior
lateral spinnerets of all spiders. The pear-shape is not general; in many families (Eresidae,
Urocteidae, Gnaphosidae, Araneidae etc.) these glands are elongated tubes. Millot
(1949) qualified them as "mi-parties" referring to their distal and proximal halves, e.g.
in Salticidae or Segestriidae, secreting distinct proteins. But in Dysdera and Gnaphosi-
dae, the distal region is very small; in certain Araneidae (Cyrtophora, Nephilaj, on the
contrary, the same region may extend to three quarters of the pyriform glands (Fig. 56E).
Oecobius and Leucauge show tripartite glands. Especially involved in the capture of
prey, pyriform glands of Gnaphosidae are less numerous, but much larger than usual.
Those of Uroctea durandi show a single category of secretory cells elaborating two
substances, a protein and an acidic glycoprotein relatively increasing from the distal to
the proximal parts of the glands (Kovoor 1979). In Pholcidae, usual pyriform glands
are absent; the posterior lateral spinnerets are not functional (Fig. 51 D). It seems,
therefore, that the anterior lateral ones playa major part in the capture of prey. Two
large globular glands with a bent neck show the histochemical characteristics of pyri-
form glands and are thus supposed to represent them on the anterior spinnerets (Kovoor,
in press).
The opposite staining affinities of the proteins secreted by the distal and proximal
cells, respectively are striking. Variations observed in spiders belonging to different
families concern the anatomy of the glands rather than the histochemistry of the pro-
teins. The general function of the pyriform gluey silk, in different thread connections
and the fixation process, results in a relative chemical uniformity. In the distal region
of the glands, reducing groups and tyrosine are detected, reactions beingin the moderate
range. The protein secreted by the proximal cells, on the other hand, is very rich in
amino-terminal and carboxyl groups (Fig. 56D) and is often associated with a polysac-
charide (Fig. 56E). An ultrastructural study on pyriform glands of Araneus diadematus
showed this combination to result in complex-structured granules within the cells
(Kovoor and Zylberberg 1980).
Aciniform Glands. All the numerous, small-sized glands opening on both the posterior
median and lateral spinnerets are classified as aciniform glands. Their shape and struc-
tural and histochemical characteristics vary extensively from one systematic group to
another. They were not detected in Zorospidae and Eresidae, both cribellate families;
in the latter, ampullate glands are numerous and of different sizes, and may supplant
the aciniform ones (Kovoor, unpubl.).
In most families, cribellate or ecribellate, these glands are composed of two cate-
gories of tall secretory cells, each occupying the distal or proximal glandular region
..
Fig. S7A-F. Aciniform glands. A Large and small aciniform glands in Uroctea durandi; strong reac-
tion of reducing groups (black) in the distal, main part of the glands, no reaction in the proximal
part. Bar = 100.um (Kovoor 1979). B In Filistata insidiatrix, the distal part of the aciniform glands
does not contain strong acidic groups that are present in the proximal part (black). Bar =50 .urn.
C One kind of aciniform glands in Telema tenella secrete two proteins, in the proximal part, the
secretory granules are composite. Bar = 20 .urn (Kovoor and Lopez 1983a). D Aciniform glands A
and B in a male of Cyrtophora citricola; very different reactivity of the proteins. Bar = 25 .urn
(Kovoor and Lopez 1982). E Large aciniform glands in Hersilia savignyi secrete two proteins and
an intermediary mucin. Bar = 25 .urn (Kovoor 1984). F Aciniform glands A and B of a female Argiope
bruennichi; reducing groups are moderate in glands A (grey) and absent in glands B (arrowheads).
Bar = 100.um (Kovoor 1972)
Fig. 57
182 J. Kovoor
which may be more or less extended (Fig. 57A,C). But Hypochilus and Dysdera show
quadripartite, and Hersilia, tripartite aciniform glands (Fig. 57E). On the other hand,
Gnaphosidae, which use silk from their anterior spinnerets to enlace prey, possess few
of these glands and those too of a single category of secretory cells. This is also the case
in spiders of the large group of araneid related families among which the aciniform
glands of certain Araneidae are diversified into two kinds (A and B), one with a thin
secretory epithelium, the other, much less numerous, with taller cells (Fig. 57D,F).
The latter are the only ones in Nephilinae, Metidae, theridiid and linyphiid spiders.
Globular aciniform glands of fllistatids secrete a strongly acidic mucoprotein in
their proximal third and another one, weakly acidic, in their distal region (Fig. 57B).
A mucosubstance is also secreted in two of the four very distinct parts of dysderid acini-
form glands, while in Hersilia, a thin layer of an acidic mucin ensures the linkage be-
tween the two different proteins produced by the glands, after their extrusion into the
lumen (Fig. 57E) (Kovoor 1984).
Histochemical characteristics of aciniform gland secretion may be related to those
of ampullate glands as in Urocteidae (Fig. 57 A), Filistatidae (Figs. 56C and 57B), Sal-
ticidae or Clubionidae. Spiders swathing their prey and also constructing elaborate egg
cases (Hersiliidae, Uloboridae, many Araneidae) generally possess a large number of
aciniform glands whose secretory products share some histochemical characteristics
with those of tubuliform glands (Fig. 57D,F) (see next section); their acidophilic pro-
teins contain only a few tyrosine, while carboxyl, amino-terminal and reducing groups
may often be absent. They are also usually devoid of a polysaccharide component.
Amino-terminal groups are always detected in the aciniform glands of theridiids and
linyphiids, corresponding to the aciniform B glands of the Araneidae (Fig. 57D,F),
Hersiliidae andPolenecia (Uloboridae) (Kovoor 1972, 1977a, 1984 and unpubl.).
Tubuli/orm Glands. Females of most araneomorphs have developed the ability of build-
ing special silken cases where eggs are laid and spiderlings spend, after hatching, one or
two stages of their life cycle. Egg cases are made up of several kinds of silk fibers, but
the main ones and also the thickest originate from tubuliform glands which start func-
tioning only at the same time as vitellogenesis begins in the ovary.
The number of these glands, which open on the posterior median and lateral spin-
nerets through generally large spigots, varies from three pairs in Dictynidae, Thomisidae,
Gnaphosidae, Araneidae and related families, four pairs in Agelenidae and Amaurobi-
idae to more than twelve pairs in Lycosidae, Heteropodidae, Ctenidae, Hersiliidae,
Urocteidae, Zoropsidae, Eresidae and Uloboridae.
The secretory cells, of a single category, cuboidal at the beginning of the function-
ing, become very flat when the tubes are full of secretion, i.e. when the female is about
to lay eggs. One or two proteins are elaborated in the same cells. In Uroctea durandi,
for example, whose egg cases are white, tubuliform glands secrete a single protein re-
acting rather moderately to the histochemical tests used for such substances: it is poor
in reducing (Fig. 58B) and carboxyl groups and does not contain amino-terminal groups;
tyrosine, proline and basic amino acids are detected (Fig. 58A). The histochemical
characteristics of this protein are very similar in tubuliform glands of spiders belonging
to different families. A coloured accessory product, often added to the main secretion,
is generally a reducing protein very rich in tyrosine. Anatasiu-Dumitresco (1937, 1942)
Fig. 58A,B. Transverse sections of tubuliform glands of Uroctea durandi. A The single product re-
acts moderately to a general test for proteins. BThis substance does not contain sulfhydryl groups,
unlike the aciniform gland (AC) products. Bar =100 }lm '(Kovoor 1979)
described the successive genesis of both products in the same cells of tubuliform glands
of a theridiid and an araneid spider. Coloured fibers of the egg cases originate from this
double secretion.
Filistatidae , Scytodidae, Pholcidae , Dysderidae, Segestriidae, Salticidae and Club-
ionidae lack these glands; their eggs are either simply held together by a few threads
(Scytodes and Pholcus) or laid within a retreat and covered with layers of silk threads
similar to those used for the retreat in other families.
Flagelliform and Aggregate Glands. Araneid spiders and related families form their cap-
ture threads or the sticky spiral of the orb webs from the silk of one pair of flagelliform
(or coronatae) and two pairs of aggregate glands whose functional unit was first de-
scribed by Sekiguchi (1952), then by Peters (1955) in Araneidae (Fig. 61). This unit is
lacking in Cyrtophora and residual in Mecynogea whose webs do not harbour sticky
spirals (Kovoor and Lopez 1982; Lopez and Stowe 1985). Adult males are, in all cases,
devoid of these glands (Sekiguchi 1955a,b). The three spigots close to each other are
located on each of the posterior lateral spinnerets (Fig. 52B).
The flagelliform glands of Araneidae resemble ampullate glands for which they were
long mistaken. They comprise a long, convoluted tube which is ramified in Nephila
and Nephilengys and a proximal cylindrical part with a thick epithelium (Fig. 55B). In
linyphiids and theridiids, flagelliform glands are much smaller than in Araneidae and
their two parts are not so distinct; the distal one , particularly short in Latrodectus and
Steatoda (Theridiidae), secretes a protein rich in amino-terminal groups, while in Ara-
neidae, tyrosine, reducing and carboxyl groups are abundant in the protein of the distal
tube . In all cases, the proximal part produces an acidic glycoprotein. The secretory
products has a typical elastic aspect within the lumen of the glands. The axial fibers of
the sticky spiral originate from the flagelliform glands , they are covered with the glue
of the aggregate glands (Fig. 60) .
184 J. Kovoor
The large, multilobed aggregate glands might have been formed by the coalescence
of a number of small glands. In Araneidae, Tetragnathidae and Linyphiidae, both pairs
of these glands are similar in their anatomy and histochemistry. Their single product is
a weakly acidic glycoprotein which will be extruded as a gluey, finely fibrillar liquid
(Fig. 56D). The multilayered, nodulated excretory duct modifies the final composition
and hydration ofthe glue through its activity (Kovoor and Zylberberg 1979).
Theridiid spiders possess two different pairs of aggregate glands (Fig. 59G), one of
which resembles that of Araneidae, but secretes a more acidic glue (Kovoor 1977b;
Kovoor and Lopez 1983b). Glands of the other pair, qualified as an atypical aggregate,
are connected to a wide and very short excretory duct; they produce a mixture of a
glycoproteinaceous substance and proteinaceous grains, as do the cheliceral silk glands
of Scytodes which are also involved in the immobilization of prey.
Pseudoflagelliform, Paracribellar and Cribellar Glands. Cribellate spiders belong to dif-

ferent, unrelated families; their capture threads differ in structure and composition,
and, in any case, are more elaborated than those of the ecribellate spiders (Peters, this
volume; Figs_ 65-67)_ The posterior median and lateral spinnerets, together with the
cribellum, are usually involved in the construction of these threads which are considered
to originate from three kinds of glands (peters and Kovoor 1980).
A pair of pseudoflagelliform glands has been described earlier in uloborids (Kovoor
1977c), and Peters (1984) demonstrated that the fibers which emerge from their two
spigots on the posterior lateral spinnerets are the axial support of the cribellar wool.
They are long and sinuous in Zosis and Uloborus, shorter in Hyptiotes and Po len ecia,
and consist of a single category of secretory cells which elaborate a protein devoid of
tyrosine and reducing groups, but in which carboxyl groups are histochemically de-
tectable. Similar, shorter glands have been identified as pseudoflagelliform in amaurobi-
id, eresid, psechrid and zoropsid spiders; in these case, the protein contains both car-
boxyl and amino-terminal groups (Fig. 59F).
This kind of gland has not been found in Dictynidae and Filistatidae; ampullate
glands are more likely to provide the axial fibers of their capture threads. Moreover,
Filistata insidiatrix shows three pairs of acinous glands, one opening on the posterior
median and two on the posterior lateral spinnerets through special long spools. These
glands produce a protein resembling that of the flagelliform glands of Araneidae; rich
in tyrosine and reducing groups (Fig. 59B,C), this elastic silk may appear as curly fibers
within the calamistrated ribbons. Capture threads of thaidid, dictynid and uloborid
~
Fig. 59. A Posterior part of the opisthosoma of the cribellate spider Zosis geniculatus; cribellar
glands secrete an acidic mucosubstance (black). Bar = 250 ~m (Kovoor 1978). B, C Special glands
of Filistata insidiatrix female supposed to provide the curly fibers of the capture threads. B The
protein is rich in reducing groups. C Strong general reaction of the protein. Bar =25 (B) and 100 ~m
(C). D Cribellar glands (CR) of Filistata insidiatrix female secrete a strongly acidic mucosubstance
(grey); the proximal half of pyriform glands (P) is stained by hematoxylin (black). Bar = 0.5 mm.
E Cribellar (CR) and paracribellar (PCR) glands are distinguishable from the carboxyl group reac-
tion of their secretory products, stronger in PCR than in CR (Zosis geniculatus, Peters and Kovoor
1980). Bar = 25 ~m. F Pseudofalgelliform gland in A maurobiusferox female; the protein is rich in
terminal amino groups. Bar =25 ~m. G Typical and atypical aggregate glands in the theridiid Latro-
dectus mactans female; PAS-positive reaction in the atypical gland, the typical gland secretes a more
acidic mucosubstance. Bar =100 ~m (Kovoor 1977b)
Comparative Struct.lre and Histochemistry of Silk-Producing Organs in Arachnids 185
Fig. 59
186 J. Kovoor: Comparative Structures and Histochemistry of Silk-Producing Organs
spiders comprise curly fibers as well, originating in the paracribellar glands which open
through long articulated spools on the posterior median spinnerets (Peters and Kovoor
1980). The number of these small acinous glands, located in the vicinity of cribellar
ones and the median posterior spinnerets, varies from about 40 in Thaida to 80 in
Hyptiotes and Mallos (Peters 1983). Their secretory product reveals histochemical
similarities with both pseudoflagelliform and cribellar gland secretions; tests for pro-
teins are weakly positive; the anionic groups detected are less abundant than in the
pseudoflagelliform gland product (Fig. 59E).
The cribellum offemales functions during the whole life cycle; it becomes nonfunc-
tional in mature males. The corresponding glands, the pseudoflagelliform and para-
cribellar glands, and their spools and spigots disappear after the last moult. This involu-
tion may be sometimes incomplete. Cribellar glands, of which there are thousands in
certain species, are the smallest of all described silk glands. Typically acinous, they
secrete an acidic mucin; both sulfate and carboxyl groups are detected in the cribellar
secretion of Filistata insidiatrix (Fig. 59D), while carboxyl groups predominate in that
of other cribellate spiders studied (Fig. 59E). The proteinaceous part of the secretory
product does not contain tyrosine and reducing groups.
The histophysiological approach to the study of arachnid silk glands leads to an overall
view of silks and their chemistry centred at their very sites of elaboration. It gives a real,
though approximative idea of the structural diversity and chemical singularities expected
in the silk fibers themselves. Thus, the presence of a glucidic component associated
with the protein in the silk glands of a number of species is noteworthy in this respect.
The case of spiders is striking in their adaptation to a wide range of environmental con-
ditions based on a multiple choice combining the different types of fibers which they
have at their disposal for the constructions so vital to them.
Acknowledgements. The author is indebted to Mrs. D. Guillaumin and J. Andre for their technical
assistance in scanning electron microscopy, and to J. Rebiere who prepared the drawings.
V Fine Structure and Function of Capture Threads
HANS M. PETERS 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
2 Gluey Capture Threads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
2.1 Typical Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188
2.2 Atypical Capture Threads in Araneids . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
3 Cribellar Capture Threads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
3.1 Autonomous Capture Threads in Uloborids . . . . . . . . . . . . . . . . . . . . . . . . . 194
3.2 Heteronomous Capture Threads in Uloborids . . . . . . . . . . . . . . . . . . . . . . . . 194
3.3 Other Cribellate Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
3.4 Comparative Remarks on Cribellar Silk . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
4 Comparative Functional Aspects of Capture Threads . . . . . . . . . . . . . . . . . . . . . . 201
4.1 Persistence of Stickiness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
4.2 Influence of Rain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
4.3 Physical Basis of the Stickiness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
1 Introduction
Capture threads may be defined as threads whose peculiar properties cause prey to ad-
here to them. With respect to these properties two types of capture threads may be
distinguished:
1. Capture threads whose suitability for holding prey is based upon a covering that
consists of a viscous glue. These threads are sometimes called sticky or adhesive
threads. However, these characteristics can also be attributed to the other type of cap-
ture threads. Therefore, capture threads belonging to this first type will be called
gluey capture threads in this review.
2. Capture threads whose suitability for holding prey is based upon a covering that
consists of innumerable, extremely fine fibrils. They are often called hackled bands
or hackled silk because of their characteristic appearance. Since the fibrils originate
from the cribellum, these capture threads are also spoken of as cribellate threads or
cribellar silk. However, these terms are ambiguous: cribellate spiders produce a variety
of threads, which are not at all capture threads, and cribellar silk denominates only
Institut fijr Biologie III der Universitiit, Abt. Verhaltensphysiologie, Beim Kupferhammer 8,
D-7400 Tiibingen, FRG
188 H.M. Peters
a component of these capture threads. Therefore, the threads in question will be

called cribellar capture threads in this review.
2 Gluey Capture Threads
Gluey capture threads have been observed in the webs of representatives of many
taxonomic groups most of which may be classified with the vast family Araneidae and
their relatives, namely in: anapids, araneids, linyphiids, metids, nesticids, symphyto-
gnathids, tetragnathids, theridiids and theridiosomatids (see Coddington 1986). In
these cases, the glue is typically arranged in the form of droplets along an axis. This
type of capture thread has been studied in more detail almost exclusively in araneids
(Tillinghast and Townley, this volume).
2.1 Typical Structure
Figure 60 shows some examples. In orb weavers the capture threads are suspended be-
tween the radii of the web. The droplets are more or less regularly arranged (Fig. 60a, b).
The axis consists of two axial fibers that lie next to each other and that are together sur-
rounded by a thin covering that seems to be a continuation of the glue that forms the
droplets (Fig. 60b). Figure 60c shows a capture thread· interwoven in the sheet of a
web of a linyphiid spider. Figure 60d represents a capture thread of a theridiid spider.
Due to the applied method, the two axial fibers can clearly be seen in this case. Further-
more, a highly liquid outer layer and an optically denser core can be distinguished in
the droplets. The droplets of araneids have a corresponding internal structure. The ad-
hesivity of the droplets is probably due to the central core (Richter 1956, on Araneus).
The capture threads of araneids originate from six glands (three on each side ofthe
spider), namely two flagelliform glands and four aggregate glands (Fig. 61). The ex-
cretory ducts of these glands end on the posterior spinnerets. Their spigots are arranged
in a very characteristic form called 'triad' by Kovoor and Lopez (1982) (Fig. 61). This
arrangement is closely related to the function of the triads. While the secretion of the
flagelliform glands is extruded, forming the axial fibers, these fibers become simulta-
neously covered with the glue produced by the aggregate glands (Fig. 61). Soon after
extrusion, this covering splits up into droplets, the details of this process are little
known.
Sekiguchi (1952) was the first to describe this arrangement and its function for
Araneus and Nephila; he observed it as well in Argiope, Cyclosa, Leucauge and Tetra-
gnatha. Independently from Sekiguchi, I also described it for Nephila (1955). In these
spiders the appearance of the flagelliform glands is somewhat different from that, e.g.
in Araneus, and they were named glandulae coronatae. Both names have been adopted
in the literature, but since these glands are essentially the same in all cases, they should
uniformly be called flagelliform glands for conformity.
Triads occur also in linyphiid spiders (Hopfmann 1935; Kovoor and Peters, unpubl.).
Capture threads with droplets have often been mentioned from representatives of this
large family. Their relative number and the size of the droplets vary very much with
the genera and species; in some cases they seem to be lacking (for details and references,
Fine Structure and Function of Capture Threads 189
Fig. 60a-d. Gluey capture threads of various araneoid spiders. a Nephila madagascariensis, adult 9,
dark field . b Araneus diadematus, adult 9, SEM; the size of the droplets and the distances between
them can also be much more regular. c Linyphia triangularis, adult 9, SEM; in this species the drop-
lets are very small and scattered, a grain of pollen is seen adhering to this thread . Bars = 50 /lm.
d Arachaearanea tepidariorum, phase contrast ; note the optically denser core of the droplets and
the two axial fibers. (d Courtesy of E. Kullmann)
see Wiehle 1956; Kullmann 1971). Where they are present, it can be concluded from
the anatomical data that they are Similarly produced as in araneids. However, it seems
likely that, in contrast to araneids, the axis of some of the gluey threads consists of
one fiber only (thus originating from only one triad), whereas others are double fibered
or run together with ordinary threads of the web (peters, in prep.).
As to theridiid spiders, gluey capture threads occur in the webs of many, but not
all, of them. They are differently arranged in the different web types (e .g. Wiehle 1937).
It is doubtful whether the composition of these threads always corresponds to the
example shown in Fig. 60d.1t is not yet clear how the capture threads of theridiids are
produced , although some information about their spinning apparatus is available
190 H.M. Peters
Fig. 61. The production of a capture

thread in araneid spiders. (Example:
Nephi/a, Peters 1955)
posterior
spinneret
gl andula
flageillformis
(Kovoor 1977b; Kovoor and Lopez 1983b). This apparatus is more or less comparable
to that of araneids.
Beyond the limits of araneid spiders or their nearest relatives, gluey capture threads
seem to be very rare. Such threads have recently been observed in the webs of the
pholcid spider Modisimus sp. (Briceno 1985; Eberhard and Briceno 1985). The sticky
balls of Modisimus turned out to be water soluble. It is interesting that Kirchner
(1986) did not find gluey capture threads in the webs of Pholcus pha/angioides, in con-
trast to Hopfmann (1935), who reported on such threads with reference to Pholcus
opilionides. In this case the glue is not divided into droplets.
A brief note from Schlott (1931), referring to Age/ena /abyrinthica, is interesting
in this context. Schlott reported that at least some areas of the sheet of the Age/ena
web may be sticky, and suggested that there are threads in these parts of the web which
are "totally covered by a sticky varnish". The stickiness disappears quickly when the
web is exposed to dry, warm air. The source of that "varnish" could be a mucous se-
cretion of the aciniform glands (see Kovoor 1976). It is obvious that if the glue is not
divided into droplets, gluey capture threads are more likely to be overlooked as capture
threads .
2.2 Atypical Capture Threads in Araneids
2.2.1 Poecilopachys australasia and Pasilobus sp.
The webs of these spiders appear as highly simplified orb webs (see Clyne 1973 for
Poeci/opachys, M.H. Robinson and B.C. Robinson (1975) for Pasilobus). The web of
Pasilobus consists of two sectors only, with a few obliquely suspended capture threads
that hang well below the plane of the web (Fig. 62). The structure of these capture
threads differs from that of other araneids insofar as the glue is placed upon founda·
tion lines. While Poecilopachys and Pasilobus are producing such lines, they simulta-
neously coat much of the length of these lines with the glue that forms the droplets.
The uncoated parts of them appear as sections of dry silk at the ends of the capture
threads (Fig. 62). Unfortunately, the fine structure of these threads is not known.
However, since Poecilopachys and Pasilobus are araneids, one may expect that their
spinning apparatus functions similarly as in other araneids. Therefore, I suggest that
the glue is extruded together with two axial fibers, while it is placed upon the founda·
tion line. The ampullate glands are good candidates for the production of these lines.
The combination of the components (two axial fibers and glue), constituting freely
suspended capture threads in other araneids with fibers from the ampullate glands,
must lead to very strong capture threads. This interpretation accords well with that
known of the function of this peculiar type of capture threads. When prey is caught,
these threads break at one end. It was observed in Pasilobus that this regularly occurred
at the outer end of these threads where there is a "low-shear joint". Even a heavily
fluttering large moth is unable to free itself (Fig. 62).
2.2.2 Bolas Spiders (Mastophora and Related Genera)
These spiders have attained the penultimate stage of web reduction before spinning
devices for catching prey has completely disappeared, as in Celaenia and related genera.
Bolas spiders, while hanging to a horizontal line , the trapeze, hold the end of a single
capture thread in the claws of one foreleg (Fig. 63a). This capture thread consists of a
line of dry silk, the pendulum line or bolas line, and a large ball of glue which is fixed
at the end of this line. Occasionally, there may be several smaller balls above the large
Fig. 62. The web of Pasilobus sp. with

a moth that has struck a capture thread .
(Courtesy of M.H. Robinson , modified
from M.H. Robinson and B.C. Robinson
1975)
192 H.M. Peters
a b
Fig. 63. a Mastophora hanging to the trapeze line (1) with its viscid ball at the pendulum line (2)
(Gertsch 1947). b Ball of Mastophora dizzydeani lowered onto a glass slide; layers of material of
different viscosities appear near the edges; the regularity of folding of folded fiber is overemphasized.
(Courtesy ofW.G. Eberhard, from Eberhard 1980b)
drop (Longman in McKeown 1936 : 102, on Dicrostichus). When prey passes, Masto-
phora swings the ball at it with high speed and with a high degree of accuracy, the ball
fastening the insect very tightly (Fig. 86c).
When the spider produces such a capture thread it first fixes the beginning of the
pendulum line at the trapeze and draws it out to its full length. At the end of this line,
the gluey mass is combed out by alternating pulls of the hindlegs until a large globule
is formed (for more details, see Gertsch 1947; Eberhard t 1980b). In the center of these
balls is a mass offolded fiber (Fig. 63b) (Eberhard 1980b). Work (in Eberhard 1980b)
described these fibers as "noodle like ribbons which are nearly rectangular in cross sec-
tion with only slightly rounded corners and slight bulges on their flat side" . Their
dimensions in two samples were 10.7 X 3.4 f.1m (Mastophora dizzydeani) and 11.4 X
4.3 f.1m (a related species).
Lopez and Stowe (1985) and Lopez et al. (1985) recently studied the spinning ap-
paratus of Mastophora . This apparatus corresponds to the typical araneid pattern. This
means that there are two flagelliform glands and four aggregate glands ending with two
triads. According to Eberhard's results, the authors concluded from these anatomical
data that Mastaphara extrudes the glue together with two axial fibers, and correspond-
ingly interpreted the fiber mass inside the balls as representing such fibers . This means
that these fibers, structurally transformed as they are in Mastophora , instead of sup-
porting the glue along their axis as is usually the case with araneids, these fibers are
curled within one large drop . This peculiarity can be related to the fact that in Masto-
phara the axial fibers are attached to a supporting thread only at one end instead of
being stretched between two such fixing points. As to the glandular origin of the
pendulum lines, there is little doubt that they originate from the ampullate glands
(probably the major ampullate glands, see Work in Eberhard 1980b).
As to the function of the fiber mass inside the balls, Eberhard writes that it permits
"quick, reversible elongations which extend the spider's striking range and perhaps also
serves to hold prey once it is hit."
3 Cribellar Capture Threads
Cribellar capture threads are produced in a totally different manner than gluey capture
threads. Typically, the cribellum is an obliquely directed plate in front of the anterior
spinnerets (Fig. 64a). In many cases this plate is subdivided into a right and a left field.
Rarely are these halves again subdivided into two parts, in the longitudinal direction
of the spider's body (Dresserus) or transversally (Zoropsis, Bertkau 1882). The cribel-
lum is covered by numerous, very fme spigots, each connected with a small gland of its
own . The fibrils are brushed out from the cribellum by means of the calamistra, one
on the metatarsus of each fourth leg (Fig. 64b). The structure of the calamistra varies
very much with the families. Cribellar capture threads occur in spider families of
the large heterogeneous assembly of the so-called Cribellates. In most cases by far
the fine structure of these threads is not or only superficially known. Thus, we must
confine ourselves to some better known examples.
Fig. 64a-c. Hyptiotes paradoxus, adult 9, SEM. a Cribellum viewed from the front side. b Calamis-
trum; bar = 100 J.Lm. c Cribellum spigots under high magnification; bar = 2.5 J.Lm
194 H.M. Peters
3.1 Autonomous Capture Threads in U1oborids
In the typical uloborid orb web the capture threads are attached between the radii,
just as in the orb webs of araneids. Each such section contains two axial fibers (Fig. 65a).
Peters (1984) has shown that these fibers originate from two glands, named pseudo-
flagelliform glands by Kovoor (1978). The spigots of these glands lie on the posterior
spinnerets. The axial fibers are surrounded by the cribellar silk. In the dark field this
silk appears as a periodically distended cloudy mass. The distensions, called puffs,
result from the activity of the calamistra. It has often been suggested, but not yet proven,
that one puff corresponds to one backward swing of a calamistrum.
There is a third component within the uloborid capture thread, the paracribellar
fibrils. They are named after the paracribellum (Peters and Kovoor 1980) from which
they originate. The paracribellum lies behind the cribellum on the median spinnerets.
The spigots are similar to the cribellum spigots, but much longer and much less numer-
ous (roughly one-hundredth of the number of the cribellum spigots). In the example
shown in Fig. 65a, the paracribellar fibrils are arranged in two strands along the axial
fibers and are connected by bridges. The possible function of this substructure is dis-
cussed in Peters (1984).
The nature of the cloudy mass on the capture threads of cribellate spiders remained
for a long time a mystery. Most arachnologists considered it as a viscous glue, although
Bertkau, as early as 1882, had clearly concluded, mainly from the anatomy of the
cribellum and its glands, that those masses "must" consist "of closely packed fine threads
corresponding in number to the number of the cribellar glands". Then, in 1956(a,b)
Lehmensick and Kullmann began studying the mysterious clouds with the electron
microscope and so confirmed that they are, in fact, composed of innumerable, fine
fibrils.
The arrangement of the cribellar fibrils can best be seen in capture threads of very
young uloborids (Fig. 65b) because they are much less numerous than in capture threads
of adults. These fibrils appear highly curved and wound, but apparently keep a distance
from another. The paracribellar substructure is poorly developed in such young stages.
It is suggested that the few fibrils running straight on through the thread belong to it.
3.2 Heteronomous Capture Threads in U1oborids
The structure of the capture threads described above may be regarded as typical for
most uloborids. These capture threads were named autonomous capture threads (Peters
1983). However, the capture threads of some uloborids are more complex. In these
cases the material constituting freely suspended capture threads in the former cases is
placed upon and so combined with fibers of the web's scaffolding. In this way another
type of capture threads, with different functional properties, is brought forth. This
type was named heteronomous, in contrast to the autonomous capture threads. For
the compound that in heteronomous capture threads is placed upon a foundation line,
the name calamistrated strand was proposed (Peters 1983).
. I .
mp . .
f 100 J-Im
a a ~
Fig.6Sa-c. Structure of autonomous capture threads in uloborids. a Uloborus walcke1ll1erius,

adult '( ; section of a capture thread in the horizontal plane viewed in the dark field. pu puffs, af one
of the two axial fibers; Is lateral strand of the paracribellar substructure, on the other side largely
covered by the axial fiber; mp one of the median patches (bridges) of the paracribellar substructure.
b Uloborus walcke1ll1erius, section of a capture thread of a very young specimen; left side inter-
mediate zone between puffs; right side beginning of a puff, TEM, native preparation. cZosis plumi·
pes, adult '( , cribellum fibrils under high magnification, TEM, native preparation
196 H.M. Peters
3.2.1 Polenecia and Miagrammopes
Wiehle (I 931) was the first to observe that in Polenecia (previously Sybota) producta
the cribellar material is connected with the radii of the web and parts of its framework.
Fig. 66a-d. Structure of heteronomous capture threads in uloborids. aPolenecia producta, adult 9,
section of a capture thread consisting of a radius of the web (arrow) and a calamistrated strand
with (brightly shining) two axial fibers and the paracribellar substructure, dark field. b Polenecia
producta, adult 9, section of a web with a captured Drosophilil, the arrows indicate some of the
points where the calamistrated strands are fixed at the radius. cMiagrammopes sp., adult 9, section
of a capture thread viewed in the dark field, in the middle the foundation line, accompanied by the
two axial fibers of the calamistrated strand, the brightly shining patches between the puffs probably
belong to the paracribellar substructure. dMiagrammopes, same species, adult 9, section of a foun-
dation line not covered by a calamistrated strand, SEM
Peters (1983) described the fine structure of these capture threads. Figure 66a shows
that the composition of the calamistrated strands accords with the composition of
autonomous capture threads in uloborids. The webs of Miagrammopes consist of one
single long thread or of some few threads only (e.g. Lubin et al. 1978). In these webs,
too, the calamistrated strands are placed upon foundation lines. One may say that
these traps are almost entirely built up of capture threads. The fine structure of these
threads has recently been studied by Peters (in prep.) (Fig. 66c). Apart from the cribel-
lar fibrils, there are two axial fibers in the calamistrated strands. The paracribellar sub-
system is difficult to identify in this case. However, since these spiders dispose of well-
developed paracribella, there is no doubt that such subsystems exist (Fig. 66c).
Both Polenecia and Miagrammopes first draw the foundation lines of their construc-
tions and then, after this work has been accomplished, cover them with calamistrated
strands. The glandular origin of the ground lines of the Polenecia web is not fully clear.
May be they are a common product of the major and minor ampullate glands. In Mia-
grammopes they are produced only by the major ampullate glands (Peters in prep.). In
this case, there are always two fibers only and, as an oddity, these fibers are twisted
around each other (Fig. 66d). As to the attachment of the calamistrated strands to the
ground lines: in Polenecia these strands are periodically fixed by means of an adhesive
(Fig. 66b). In Miagrammopes they simply become periodically pushed down upon those
lines. It is obvious that the ground lines take off much of the load which in autonomous
capture threads is assigned to the axial fibers alone. Thus, it is not surprising that the
pseudoflagelliform glands of Polenecia and Miagrammopes appear very small and the
axial fibers relatively thin in comparison with their homologues in other uloborids
(Kovoor and Peters, in prep.).
3.3 Other Cribellate Spiders
Apart from uloborids, autonomous capture threads seem to occur very rarely in the
webs of cribellate spiders. Capture threads that obviously belong to this type were ob-
served in the webs of Titanoeca albomaculata (Szlep 1966). Certainly, the great major-
ity of Cribellates produce heteronomous or semi-autonomous capture threads (for the
latter type, see Fig. 67). In most cribellate spiders the cribellum is bipartite. In these
cases the calamistrated strands consist of two symmetrical half-strands lying close to
another or keeping some distance from another (Fig. 67). Apart from axial fibers and
cribellar fibrils, components of a third category that vary conSiderably with the families
seem always to be present in these strands. However, in constrast to the uloborids, the
glandular origin, neither of the axial fibers nor of the fibers of the third category, is
known. Thus, it is not possible to order these components according to the principles
of homology.
Highly curled, undulated fibers have often been observed in capture threads of
cribellate spiders (Fig. 67). These fibers are much thicker than the axial fibers and their
number varies. There is, for instance, one pair of them in the capture threads of Amau-
robius (Comstock 1948), Filistata insidiatrix (Lehrnensick and Kullmann 1956a) and
Oecobius annulipes (Zimmermann 1975). Three pairs are present in Stegodyphus sara-
sinorum and four in Stegodyphus lineatus (Kullmann 1975). Whereas in uloborids the
198 H.M. Peters
Fig. 67a-c. Stegodyphus Iineatus, adult 9. a Section of the web demonstrating semi·autonomous
capture threads; parts of these capture threads are freely suspended between threads of the web's
scaffolding, others are attached along these threads over shorter (thin bars) or longer (thick bar)
sections, near 1:1. b Capture threads splitting off into their two half-strands. c Fine structure of a
half-strand viewed in the dark field;A axial fiber; C mass of cribelium fibrils on undulated fibers
cribellar silk is supported by the axial fibers, in the present cases, this silk seems to be
supported by the undulated fibers (Fig. 67c).
It seems likely that the undulated fibers can act as tension springs, tending to con-
tract the strands if there are no counter forces that stretch them. This may help to ex-
plain the peculiar shape of the capture threads found, for instance in Filistata (Fig. 68).
Fig. 68. Filistata insidiatrix, adult 9. Section of a

capture thread; arrows indicate two points where
the caiamistrated strand (consisting of two half-
strands) is fixed at the foundation line
In these cases the calamistrated strands (consisting of two half-strands) surround the
foundation lines in loose loops. In Filistata the foundation lines radiate from the spi-
der's retreat. When such a line becomes covered with a calamistrated strand, the spider
- slowly moving along the line - fixes the strand from time to time at this line. Since
the distances between the fixing points are much shorter than the corresponding sec-
tions of the strands, these strands are laid down in loops.
Paracribellumlike structures have also been found in the spinning apparatus of some
cribellate spiders outside the Uloboridae (Ixeuticus, Mal/os, Thaida, Peters 1983; Jack-
200 H.M. Peters
son 1982c). It is possible that such spiders produce fibrils that are comparable to the
paracribellar fibrils found in uloborids.
3.4 Comparative Remarks on CribeUar Silk
With higher magnification it becomes clear that the cribellar fibrils of uloborids are
provided with small nodes along their axis (Fig. 65c) (Hyptiotes, Miagrammopes, Po-
lenecia, Uloborus, Zosis, see Peters 1984 for references;Miagrammopes, Peters, in prep.).
Similar structures have been mentioned with reference to Dictyna (Lehmensick and
Kullmann 1956b) and Oecobius (Zimmermann 1975). In Filistata arizonica (Friedrich
and Langer 1969) and in Filistata insidiatrix (Lehmensick and Kullmann 1956a) nodes
are lacking. In these cases the cribellar fibrils have a ribbonlike shape; they were about
15 nm thick and about 100 nm wide (Filistata insidiatrix, Lehmensick and Kullmann
1956a).
As to the diameters of the cribellar fibrils reported in the literature, the respective
data must be regarded with great caution. No one knows what happens to these ex-
tremely fme fibrils while they are prepared and studied with the electron microscope.
Disregarding possible species-specific differences, one may roughly estimate that the
diameters often lie in the order of a magnitude of 20-30 nm in adult or near-adult
cribellate spiders; the nodes are always thicker.It is possible that the diameters increase
with the spiders' age. However, no data are available. Most authors did not even inform
the reader about the developmental stage with which they dealt. There is little doubt
that the effectivity of cribellar silk is related to the number of fibrils fOrming one sec-
tion of a cribellar capture thread. However, it is unknown how many fibrils are simulta-
neously secreted. One may expect that these numbers are related to the number of
spigots. These spigots are related to the size of the cribella and so, at least partly, to
the size of the spider (Table 17).
The size of the cribella and the number of spigots, and apparently also the relative
number of fibrils, increase very much during the spider's life span. For instance, when
an Oecobius annulipes begins web building, it disposes of about 70 cribellum spigots;
Table 17. Number of cribellum spigots of different species of cribellate spiders (adults) in relation
to the size of the spiders and the size of their cribella a
Species Body length Transversal Number of Reference

of spiders diameter of spigots
(mm) cribella (mm)
Oecobius annulipes 3.4 0.120 550 Zimmermann (1975)

Zosis piumipes 4.5 0.460 5,400 Peters (1984)
Stegodyphus 10 0.980 16,000 Kullmann (1968b)
sarasinorum
St~odyphuspacificus 20 1.600 40,000 Kullmann (1968b)
a The cribellum of Zosis (Uloboridae) is undivided; the cribella of the other species are bipartite.
In the latter cases the transversal diameter of the cribella denotes the sum of the transversal dia-
meters of the two parts covered with spigots; the number of spigots denotes the total number (all
data are approximative).
when it is an adult, it has about 550 (Zimmermann 1975). The corresponding numbers
for Hyptiotes cavatus are about 1100 and 4500 (OpellI982). One may expect that the
distances between the cribellar fibrils, to some extent, depend on the density of the
cribellum spigots. It is interesting in this respect that Opell (1982) found in Hyptiotes
cavatus that this density remains fairly constant throughout the lifetime. It is also
noteworthy that the density of the cribellum spigots in Stegodyphus paci/icus differs
only very slightly from that in Stegodyphus sarasinorum, although the differences be-
tween their numbers is very great (Table 17) (Kullmann 1968b).
4 Comparative Functional Aspects of Capture Threads
4.1 Persistence of Stickiness
It has often been emphasized that cribellar capture threads retain their adhesive pro-
perties for longer periods than gluey capture threads. The latter are subject to drying
up and so are likely to lose their stickiness soon. However, there are great species-
specific differences in this respect (see also Tillinghast and Townsly, this volume).
Eberhard (1980b) referred to some araneids whose capture threads remain adhesive
for a long time. The same author drew attention to the sticky balls of the bolas spiders
(Fig. 63) as an extreme on the opposite side. These drops, due to their high water con-
tent, lose their stickiness very quickly by evaporation. This explains why they are
periodically ingested and become replaced, if the spider did not catch prey. In Masto-
phora dizzydeani these ingestions took place on an average of about 0.5 h. As to the
persistence of stickiness of cribellar silk, judging from occasional, personal observa-
tions, I expect considerable species-specific differences as well. Eberhard (1980a) re-
ported an example of long persistence. This author left a web of a mature U/oborus
diversus for 3 months in low humidity. At the end of this period he tested the sticki-
ness of the capture threads and found "no obvious difference compared to new silk".
4.2 Influence of Rain
Since the viscous coat of gluey capture threads is water soluble, one may expect that
these threads are damaged by rain. However, it seems that, in general, rain destroys
the entire webs rather than damaging only the capture threads (e.g. M.H. Robinson
and B.C. Robinson 1972). Cribellar capture threads, too, are endangered by rain. Re-
ferring to Dictyna sp., Wiehle (1949) claimed that the cribellar mass is dissolved in
water, but provided no evidence. I doubt that Wiehle is right. Cribellar fibrils, at least
of U/oborus walckenaerius (adult females), are not soluble in water (tap water, 20 DC,
24 h, unpubl. observations). When these fibrils come in contact with water, they do not
lose their stickiness, because they become dissolved, but because the puffs immediately
collapse. This collapse is irreversible.
4.3 Physical Basis of the Stickiness
If one refrains from the details, the stickiness of gluey capture threads does not offer
great problems. On the other hand, the physical basis of the stickiness of cribellar silk
202 H.M. Peters: Fine Structure and Function of Capture Threads
is unknown. While locked up in the glands, the cribellar secretion is mucouslike (e.g.
Kovoor 1978 for uloborids). But what happens to this mucous while it is being trans-
formed into fibrils? Kullmann (e.g. 1975) described these fibrils as "dry", Eberhard
(1980a) as "probably dry". These words do not tell us much. Cribellar silk (at least
that of uloborids) adheres firmly even to the smooth surface of glass. Static electric
charges of the fibrils have been taken into consideration (peters 1984). This idea was
recently discussed in more detail with reference to the sheet web of the diplurid spider
Euagrus by Palmer (1985). However, this is a field wide open for further research.
Considering the function of a capture thread, one has to take into account that such a
thread rarely acts on its own, independent of its equals. Generally, capture threads are
more or less closely associated with another and so act in cooperation. The capture
threads of Po len ecia and even more so the long threads of Miagrammopes, for instance,
represent exceptions to the general rule. It is certainly not by chance that these capture
threads are heteronomous, i.e. strengthened by foundation lines. Capture threads, such
as those of Poeci/opachys and Pasi/obus, must also be classified with the heteronomous
type, although the mode of their production differs from that found in Polenecia and
Miagrammopes. The unique capture device of the bolas spiders can be regarded as a
further development of the Pasi/obus type (e.g. M.H. Robinson 1983). One may expect
that further study will reveal many more heteronomous capture threads, for instance
within the theridiid family with its great variety of web constructions.
It is an open question as to how far the stickiness of individual capture threads is
influenced by the mode of the association of these threads with their equals. For in-
stance, the web of the cribellate spider Dresserus sp. is an extremely dense sheet. One
can easily observe that an insect coming in contact with such a web becomes imme-
diately entangled within the maze of fibers, and one has the impression that it is more
this entanglement which holds the prey than the stickiness of the weave.
Finally, one has to consider that the structure of the threads used by spiders for
prey capture is always closely related to the spider's specific capturing behaviour. Pos-
Sibly, even ordinary silk can be used for catching prey. An example of this has been
nicely described by Gertsch (1979) for Ariadna. These spiders live in tubes from the
mouth of which a series of dry lines radiate outward. "The spider sits just within the
tube, its six legs directed forward, in position to leap. The touching of one of the trap
lines brings it out with surprising swiftness, like a j ack-in-the·b ox , to the spot where
the unlucky insect has tripped. It seizes the victim, then carrying it instantly back into
the tube."
VI Chemistry, Physical Properties,
and Synthesis of Araneidae Orb Webs
EDWARD K. TILLINGHAST and MARK TOWNLEY!
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
2 Water Soluble Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
3 Water Insoluble Components (Fibroins, Elastomers, Adhesives, and Junctional Cements) . 204
3.1 Radial Fibers and Hub Spiral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
3.2 Adhesive Spiral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
3.3 Junctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
4 Synthesis of Orb Web Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
1 Introduction
It was the distinguished organic chemist, Emil Fischer (1907), who clearly demonstrated
the protein nature of spider silk. Only much later (F .G. Fischer and Brander 1960)
was it realized that the orb web contains low molecular weight, water soluble com-
pounds as well.
2 Water Soluble Components
Nearly half the orb web of an araneid spider is water soluble. The water soluble sub-
stances include potassium dihydrogen phosphate, potassium nitrate, 'Y-amino butyr-
amide, taurine-related compounds and often other amines as well, depending on the
species (F.G. Fischer and Brander 1960; Schildknecht et al. 1972; C.M. Anderson and
Tillinghast 1980). Most of the water soluble compounds are confined to the adhesive
(sticky or viscid) spiral, as has been made evident by feeding isotopes (Kavanagh and
Tillinghast 1979; Anderson and Tillinghast 1980; Tillinghast et al. 1984), and these are
likely confined to the viscid droplets. The bulk of the water soluble compounds are
probably the products of the aggregate glands and their ducts (Kovoor, this volume).
The exact role of the water soluble compounds has never been established unequivo-
cally, but as 'Y-amino butyramide is hygroscopic and potassium dihydrogen phosphate
is deliquescent, they may function, in part, to absorb and retain water, without which
Department of Zoology, College of Liberal Arts, University of New Hampshire, Spaulding Life
Science Building, Durham, NH 03824, USA
204 E.K. Tillinghast and M. Townley
the sticky spiral would no longer be adhesive (Schildknecht et a1. 1972). In addition,
potassium dihydrogen phosphate may prevent bacterial degradation (Schildknecht et al.
1972).
3 Water Insoluble Components (Fibroins, Elastomers, Adhesives, and

Junctional Cements)
3.1 Radial Fibers and Hub Spiral
Major ampullate fibers, which predominate in both radial fibers and the hub (non-
sticky) spiral, have been the most extensively studied spider silks. Glycine and alanine
account for about 55-69% of their amino acid composition, depending on the species
(Zemlin 1968; S.O. Andersen 1970; Tillinghast 1984; Tillinghast and Christenson 1984;
Table 18). The small side chains of these amino acids allow for the tight packing pre-
sent in the antiparallel-chain pleated sheets which form, as first proposed by Pauling
and Corey (1953), the crystalline domains of silk fibroin . The existence of axial mole-
cular orlen tation in the polypeptides results in major ampullate fibers being birefringent
(Zemlin 1968; Work 1977b); that is, visible when placed between crossed polaroids
(Fig. 69). It should be noted that minor ampullate and fine fibers are also highly bire-
fringent (Work 1977b, 1981a).
Measurements made of some of the physical properties of major ampullate silk have
helped to reveal why these fibers are well suited to functioning as structural supports
in the orb web and elsewhere. Of major importance are the high tensile strengths of
these fibers. Values obtained by several workers range from about 0.6-2 GPa (DeWilde
1943; Lucas 1964; Zemlin 1968; Denny 1976; Work 1976, 1977b), and it has been
shown (von Weimarn 1932, in DeWilde 1943; Denny 1976) that the exact tensile
strength value is dependent on the rate at which the fiber is extended, as would be
expected of organic fibers. Thus, the faster the fiber is extended, the greater its tensile
strength will be, and the greater the force it will withstand before breaking (von Weimarn
Fig.69. An adhesive spiral-radial junction under normal light (left) and with crossed polaroids
(right), demonstrating the birefringent character of radial (R), and thus, major ampulIate, fibers.
SS = adhesive spiral. Bar = 100 J.Lm
("l
Table 18. The amino acid composition (mol/100 mol) of the contents of various Araneus diadematus spinning glands, as well as of the silk gland contents of ::r
(1)
the silkworm Bombyx mori. [Data from S.O. Andersen (1970) for Araneus and Lucas et al. (1958) for Bombyx; cf. also Fig. 71]
~.
......
Amino acid Araneus diadematus Bombyx mori :::
."
::r
Glandular origin '<
[
Major Minor Flagelliform Aciniform B Aciniform A Pyriform Aggregate :4'
0
ampullate ampullate (median spinneret) (posterior spinneret) 'c:l
(1)
...::l".
(1)
Asp 1.04 1.91 2.68 8.04 8.61 10.52 9.18 1.3 !f'
Thr 0.91 1.35 2.48 8.66 9.04 4.44 7.58 0.9 I»
::sc:>-
Ser 7.41 5.08 3.08 15.03 16.45 14.84 6.80 12.1
CIl
Glu 11.49 1.59 2.89 7.22 7.80 10.38 9.75 1.0 '<
::s
Pro 15.77 Trace 20.54 2.99 3.58 7.78 10.81 0.3 E}
Gly 37.24 42.77 44.16 13.93 13.20 7.76 14.48 44.5 ~.
Ala 17.60 36.75 8.29 11.30 10.74 9.87 6.19 29.3 0
..,
Val 1.15 1.73 6.68 7.37 6.86 5.35 5.75 2.2 ~
I»
Cyt 0.2 ::s
Met 0.1 Sl.
c:>-
I»
He 0.63 0.67 1.01 4.27 4.24 3.71 4.71 0.7 (1)
Leu 1.27 0.96 1.40 10.10 8.21 5.39 5.50 0.5 0

...c:r
Tyr 3.92 4.71 2.56 1.99 1.56 2.20 2.22 5.2 ::;:::
(1)
Phe 0.45 0.41 1.08 2.79 2.56 2.29 3.83 0.6
g'
Lys 0.54 0.39 1.35 1.90 2.57 9.01 7.43 0.3
His Trace Trace 0.68 0.31 0.59 2.83 2.38 0.2
Arg 0.57 1.69 1.13 4.09 4.01 3.63 3.40 0.5
tv
o
u.
1932, in De Wilde 1943; Denny 1976). In addition, major ampullate silk has a breaking
extension ratio (BER = length of fiber when maximally extended/original length) of
about 1.2-1.4 (DeWilde 1943; Lucas 1964; Zemlin 1968; Denny 1976; Work 1976,
1977b). For a fiber with a tensile strength of about 1 GPa, this range is close to the
optimal BER, in Denny's (1976) model, for supporting the maximum static force pos-
sible.
The resilience (energy recovered upon relaxation following extension/energy ex-
pended for extension) of major ampullate silk and its implication, as measured and dis-
cussed by Denny (1976), should also be mentioned. He found the initial resilience to
be low (0.35), and as with other organic fibers, some of the energy loss occurs as a result
of a component within the fiber which recovers only slowly following extension. That
is, within the time allowed for one extension-relaxation cycle, complete recovery of
such fibers does not occur and, thus, an energy loss results. The significance of this
slowly recovering component to spider silk is that much of the kinetic energy of an
impacting insect is slowly dissipated by the web rather than being transferred back to
the insect, potentially catapulting it back out of the web (Denny 1976).
An interesting property of axially unrestrained major ampullate fibers, discovered
by Work (1977b), is their supercontraction, or axial retraction, in water. As Work
(1977b, 1981b) pointed out, while supercontraction in some natural or man-made
fibers can be produced by harsh chemical means and/or high temperatures, major am-
pullate silk is unusual in terms of the comparatively mild conditions (i.e., water at
room temperature) which will induce it to supercontract. One possible function of
water-induced axial retraction which has been proposed (Work 1977b, 1981 b) is to
support the added weight of water on a web as a result of rain or dew. The same re-
tractive forces which produce super contraction in unrestrained wetted major ampullate
fibers would also oppose the extensional forces imposed by the water droplets' weight
on the structure of the in tact web.
From X-ray diffraction studies, Work and Morosoff (1982) found that the individual
crystalline domains of supercontracted major ampul1ate silk, other than being rotated
as a whole, are not affected by supercontraction. Thus, it would appear that alterations
within the noncrystalline, amorphous domains are responsible for the macroscopically
observed supercontraction. Whatever alterations may occur, Work (in Work and Moro-
soff 1982, 1985)has hypothesized that supercontraction does not produce an order-to-
disorder transformation in the amorphous domains. A mathematical treatment of super-
contraction has been published (Fornes et al. 1983). Gosline et al. (1984) have sug-
gested that about 85% of the axial retractive force of wetted major ampullate silk is a
result of conformational entropy changes in the protein chains composing the amor-
phous regions. As this is a characteristic of rubbers, they propose that supercontracted
silk is a protein rubber in which crystalline inclusions are embedded.
Of interest is the observation that minor ampullate fibers, produced by the glands
which are the most morphologically similar to the major ampullate glands, do not super-
contract in water (Work 1977b). Neither do the fine fibers observed by Work (1981a),
which is somewhat perplexing considering that these fibers may be products of the
major ampullate glands (Work and Peters, in Work 1981a).
Chemistry, Physical Properties, and Synthesis of Araneidae Orb Webs 207
3.2 Adhesive Spiral
The adhesive spiral is composed of paired, core fiberlike elastomers upon which ad-
hesive droplets are deposited. The core fibers of the adhesive spiral are not birefringent,
or are only weakly so (Denny 1976, Fig. 69), and have a much higher BER than major
ampullate fibers (DeWilde 1943;Denny 1976). These two observations tend to support
the view that crystalline components, to the extent seen in major amp~llate silk, do
not occur in the adhesive spiral's core elastomers (Denny 1976). Speculatively, this
could be due to the distribution of bulky side-chain amino acids throughout the protein,
which would not favor pleated sheet and, thus, crystallite formation (Denny 1976).
The functional significance of such a high BER is that the adhesive spiral is well
suited to decelerating impacting prey over a relatively large distance, thus decreasing
the force transmitted to the other web elements at impact (Denny 1976). The ability of
the web to stop impacting objects is further enhanced by the conversion of some ad-
hesive spiral-radial junctions (SS-R, using the nomenclature of Jackson 1971) to a
"pulley" form as a result of stress on adhesive spiral segments (Eberhard 1976b). That
is, before an adhesive spiral segment reaches maximal extension, the SS-R junctional
cement may be detached from the surface of the adhesive spiral fibers. The cement,
however, remains attached to the radius and, apparently, continues to surround the
adhesive spiral. Thus, the length of the adhesive spiral segment, as delimited by the two
transecting radii, is increased as adjacent portions of the adhesive spiral are pulled
through the junction. The increased length provides an increased decelerating capacity
(Eberhard 1976b). Denny (1976) has also reported that the initial resilience of the ad-
hesive spiral is about the same as for major ampullate silk. Its importance to prey cap-
ture is the same as discussed for major ampullate fibers.
The adhesive surrounding the core elastomers is composed of protein as well as the
water soluble compounds discussed earlier. The amino acid composition of the contents
of the aggregate glands is comparatively low in small, side-chain amino acids such as
glycine and alanine, but high in polar amino acids such as glutamic acid and, thus, their
amino acid composition is more nearly that of a glue than a silk, according to Andersen
(1970; Table 18). Kovoor (1972) has observed that the contents of the aggregate gland
are rich in carbohydrate, which may correspond to a glycoprotein shown to exist in
the viscid cover of the adhesive spiral (Tillinghast et al. 1981). This glycoprotein's
principal carbohydrate moiety is N-acetyl galactosamine (Dreesbach et al. 1983) or
galactosamine (Sinohara and Tillinghast 1984).
3.3 Junctions
The hub spiral-radial (NS-R) junctional cement is susceptible to proteases (Tillinghast

et al. 1981), and is fibrous and birefringent (Work 1981a). The proposal that NS-R
junctions and attachment disks have a common glandular origin is supported by the
observation that attachment disks are also birefringent (Work 1977a). In contrast, the
SS-R junctional adhesive is neither susceptible to most proteases employed to date
(Kavanagh and Tillinghast 1979), nor is it birefringent (Work 1981a).
4 Synthesis of Orb Web Components
Spider silk synthesis has been most studied in the major ampullate glands. These glands
are composed of a tail , an ampulla, and a duct (Figs. 70 and 71). The long coiled tail
is the site of 90% of the polypeptide synthesis and is composed of a single layer of
columnar epithelial cells (Bell and Peakall 1969). Unlike the cells composing the silk
glands of the silkworm Bombyx mori (Akai 1971), these cells apparently lack Golgi
apparatus (Bell and Peakall 1969), suggesting fundamental differences in the processing
of silk fibroin following its synthesis. Although the ampulla is the storage site of pro-
genitive silk, it also participates in protein synthesis (PeakaIl1968). The protein secreted
into the lumen, however, is not the same in all parts of the ampulla. The distal third or
fourth of the ampulla is composed of a different cell type from the proximal two-thirds
or three-fourths (Kovoor 1972). The ampulla's distal cell type is the same as that com-
posing the tail. The two cell types synthesize distinct proteins, resulting in the composite
nature of the fibers produced (Kovoor and Zylberberg 1980; WorkI984). (For details,
see Kovoor, this volume).
Two means by which protein synthesis has been induced in major ampullate silk
glands are (1) exposure of the glands to cholinergic agents and (2) removal of silk from
the silk glands by pulling from the spinnerets. Following Witt's (1963) observation that
Fig. 70. An excised major ampullate gland from Araneus cavaticus, composed of a tail, ampulla
(AMP), and duct. The tail, compactly folded within the abdomen, has been unraveled in this photo-
graph. Bar = 1 mm. (From Tillinghast and Townley 1986)
Chemistry, Physical Properties, and Synthesis of Araneidae Orb Webs 209
A 8
c D
Fig. 7IA-D. Location of silk glands in the abdomen of Araneus sericatus (female). A Major ampul-
late (left), flagelliform (middle), and minor ampullate (right) glands. B Aggregate glands. C Cy-
lindrical glands. D Aciniform glands opening on median and posterior spinnerets and the more
ventrally located pyriform glands opening on anterior spinnerets. (From Mullen 1969)
physostigmine fed to spiders increases web weight, Peakall (1965) demonstrated that
this anticholinesterase agent stimulates the rate of incorporation of amino acids into
glandular protein in vivo and in vitro, and that this stimulation could be blocked by
atropine. Peakall (1964, 1965) also observed that the mechanical pulling of silk like-
wise stimulates protein synthesis . This form of stimulation, however, could not be
blocked by atropine indicating that two distinct mechanisms are involved in regulating
protein synthesis. Both means of stimulation have been confirmed by Candelas and
Cintron (1981).
In addition, Peakall (1966) observed that cholinergic stimulation using acetylcholine,
as well as stimulation by mechanical pulling (PeakaIl1969), results in two basic events.
First, preformed protein droplets (fibroin) are secreted by the epithelial cells of the
major ampullate gland into the lumen. Then new protein is rapidly synthesized. The
protein synthesis inhibitors, actinomycin D (at transcription) and puromycin (at trans-
lation), block the synthesis of new protein, but do not prevent the secretion of pre-
formed fibroin into the lumen. Candelas and Lopez (1983) have reported that a wave
of RNA (mainly mRNA) synthesis precedes the wave of new protein synthesis follow-
ing stimulation.
The precise mechanism by which the silk gland senses a silk deficit is as yet un-
known. Peakall (1969) has presented evidence for a sensory structure at the junction
of the ampulla and duct of the major ampullate gland. Kovoor (1972) and Foelix (1971)
have confirmed the presence of the structure, but find no evidence for a sensory func-
tion. Even so, the existence of a sensory device remains an attractive idea, for when
210 Tillinghast/Townley: Chemistry, Physical Properties, and Synthesis of Orb Webs
silk is pulled more from one gland than from its pair, protein synthesis is more rapid
in the gland with the greater deficit (Tillinghast and Townley 1986).
It has been noted by Bell and Peakall (1969) that the major ampullate duct is far
too long to serve merely as an inert conduit for silk transport. At least two related pro-
cesses apparently occur in the duct: molecular alignment to form pleated sheets and
desiccation of the pro genitive silk.
Work (1977a) has observed that whereas the major ampullate fibers are birefringent,
the stored aqueous solution of progenitive polypeptide in the glands is not. Thus, its
conversion to birefringent, water insoluble silk fibers occurs in the duct, but not neces-
sarily in the same location within the duct under all conditions (R.S. Wilson 1962b;
Work 1977a). It may occur anywhere from the extreme end of the spigot to the cuti-
cular pits region (Work 1977a). [See R.S. Wilson (1962a) for a description and the
location of the cuticular pits region.]
The majority of the information on the nature of the stored polypeptide solution
and its conversion to water insoluble fibers has thus far been obtained using B. mori.
Presumably, the situation in the major ampullate glands of spiders is at least similar.
Kobayashi et al. (1980) have reported that the conformation of silk fibroin in aqueous
solutions of less than 5% is a random coil. As the concentration is increased above 5%,
there is an increasing amount of conversion to a loose, fourfold helical conformation.
This helix was first proposed by Konishi and Kurokawa (1968) to describe the poly-
peptide conformation present in the silk I form of B. mori fibroin. (The silk I form is
obtained by removing the progenitive silk from the silk glands and allowing it to dry
undisturbed. Silk II is the pleated sheet form produced when silk is pulled from the
spinnerets.) This loose helix has a residue translation along the molecular axis of
0.227 nm. The ex-helix, by comparison, has a residue translation of 0.15 nm. Since the
silk fibroin concentration is about 30% in the middle portion of the silk glands of
B. mori, a substantial quantity of the stored silk is presumably in the loose helical con-
formation. Fibroin concentrations in major ampullate glands are similarly high; for ex-
ample, about 41 % in Argiope aurantia (Tillinghast et al. 1984).
The change from loose helix to pleated sheet may be accomplished by a process like
that proposed by Iizuka (1983) as a result of work with B. mori. Briefly, the proposal
is that the shearing force which results from pulling silk extends molecular folds by
rupturing the hydrogen bonds which maintain the folds via water molecules. Once ex-
tended, the molecules can readily form pleated sheets and, thus, crystalline domains.
Water is removed from the progenitive silk during its passage through the duct
(peakall 1968) and this is accompanied by an exchange of K+ for Na+ (Tillinghast et al.
1984). Whether the ion exchange reflects the mechanism of desiccation and/or is
essential to the orientation of polypeptide chains to pleated sheets is unknown.
Acknowledgments. We owe special thanks to Dr. Robert Work for his exhaustive critique of our
manuscript and yet acknowledge that what errors may remain are our own. Thanks also to Dr.
Charles Walker, Nancy Cherim, and Margaret Tillinghast for technical assistance. This chapter is
dedicated to the memory of Theresa Townley and Sharon Beth Tillinghast.
VII Ecological Aspects of Spider Webs
WOLFGANG NENTWIG 1 and STEFAN HEIMER 2
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
2 The Protective Function of Silken Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
2.1 Tubes and Retreats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
2.2 Egg Sacs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
3 Silk as a Prey-Catching Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
3.1 The Evolution of Webs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
3.2 Wrapping Behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
4 Stabilimenta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
4.1 Occurrence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
4.2 Functional Explanations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
4.3 Mechanical Function as a Moulting Platform . . . . . . . . . . . . . . . . . . . . . . . . 220
4.4 Antipredator Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
4.5 Stabilimenta as Relicts? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
5 Silk Use and Communication Between Conspecifics . . . . . . . . . . . . . . . . . . . . . . . 221
6 The Use of Silk and Webs by Heterospecifics . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
6.1 Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
6.2 Other Animals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
1 Introduction
When silk leaves a spider's silk gland it enters a new sphere, the spider's environment.
Here, silk is used to form various structures such as retreats, snares, cocoons, draglines,
etc. Since nearly all interactions between spider and environment are silk-mediated,
this substance deserves treatment from the demecological or synecological point of
view.
2 The Protective Function of Silken Structures
2.1 Tubes and Retreats
Originally, spiders probably used silk to cover the insides of the burrows which they
dug into the ground. Eggs were deposited in the burrows and covered with silk. All the
1 Institut flir Zoologie der Universitat, Universitatsstra£e 31, D-8400 Regensburg, FRG
2 Museum flir Tierkunde, Augustusstra£e 2, DDR-8010 Dresden, GDR
212 W. Nentwig and S. Heimer
various silken structures which spiders use as living quarters are tubes or can be derived
from tubes. This structural similarity can be explained by the way the spider spins. The
observations presented here use Pharmictapus (Theraphosidae) as a model, but they
can readily be made on other theraphosids or other families as well. Phormictapus has
four spinnerets which produce very thin and dry threads simultaneously. They are
neither fused together (e.g. by glue) nor distorted (e .g. by movemen ts of the spinnerets).
Several hundred threads together form ribbons which are attached to the uneven surface
of the substrate (Fig. 72a). Contact is probably made by adhesion alone; no special at-
tachment points could be detected. Perpendicular movements of the opisthosoma
yield a silken carpet of many ribbon layers. Slow forward movement enlarges this
carpet to a sheet which, inside a narrow burrow, acquires a curved shape. As the spider
progresses, it tends to maintain contact with its silk (positive tigmotaxis), with the
result that the V-shaped sheet finally closes to a circular structure which forms a tube.
The radius of this tube approximates to that of the radius of the moving spinnerets.
A silken tube has special physical properties. It is waterproof (but not airtight); in
its tube a spider can survive flooding after heavy rainfall or in semiaquatic environ-
ments (Schaefer 1976a). Some spiders live permanently in aquatic habitats within their
silken tubes, using the air inside as a physiological gill; for example, Argyroneta aquatica
in freshwater (F. Braun 1931), several desid species and other seashore spiders (Bristowe
1923; Lamorall968; McQueen and McLay 1983) and the linyphiid, Bathyphantes tar-
rentum, in mountain creeks (Heimer unpubl.). Temperature and humidity inside a
silken tube are relatively stable compared with the outside conditions (Humphreys,
this volume). These features combined with the general antibiotic properties of spider
silk [the reasons for which are not understood, but may be due to the fact that the
macromolecular structure of silk (Tillinghast and Townsley, this volume) is so inert
that most enzymes cannot decompose it, which makes fungal or bacterial attacks im-
possible] make such tubes an ideal protection. Many spiders use silken tubes through-
out their lives (e.g. Liphistiidae, Atypidae, Migidae, Segestriidae) or as a retreat in
periods of inactivity (e.g. Salticidae, Clubionidae, Gnaphosidae). In many cases resting
spiders or those which moult inside their tubes close them from the inside for optimal
protection (e .g. Salticidae, Clubionidae). The retreats of aerial web builders are perhaps
not homologous to the plesiomorph tubes, but they have the same properties and func-
tions.
2.2 Egg Sacs
After oviposition the eggs are held together by a liquid which slowly evaporates. Prob-
ably all spiders, therefore, use silk to protect their eggs against physical and biotic
threats. Spiders never lay eggs singly [the single egg in the vulva of Oanaps and Har-
pactea reported by Brignoli (1974) is a misinterpretation of a complex gland (Heimer
1986b, exceptions, however, are Telema and a barychelid which consistently deposit
only one egg per cocoon (Roth and Roth 1986)], so silk has an additional function as
a container to keep the eggs together. The silken egg case can be used to attach the co-
coon to particular places, as a camouflage device or for transport. Some spiders fix
their egg sacs to their spinnerets (e.g. Lycosidae, Nesticidae), or carry them in their
Ecological Aspects of Spider Webs 213
Fig. 72. a Phormictopus (Theraphosidae) spinning a "carpet". (Photo: Heimer) ; bArgiope argentata
(Araneidae) wrapping a prey item. (From Foelix 1979)
214 w. Nentwig and S. Heimer
chelicerae (e.g. Pisauridae). In some cases the silken layer around the eggs consists only
of a few threads (e.g. Pholcidae, Scytodidae). Spiders which actively transport their
cocoons may do so in order to find optimal thermal conditions for their eggs (e.g.
Lycosidae; Humphreys, this volume). Also, the early contact between mother and off-
spring may favour the development of social structures (Buskirk 1981).
The construction of an egg case starts with a basal plate on which the eggs are laid.
Afterwards the spider covers the egg mass with a network of silken ribbons which are
attached to the basal plate. Many spiders, which subsequently care for their offspring
or which live in habitats with stable environmental conditions (e.g. caves), do not add
further layers. Most spiders, however, cover the complete egg case with one or more
additional layers of cottony silk. Quite often the outer layer is of solid silk which pro-
vides mechanical protection (e.g. Lycosidae, Gnaphosidae). This layer may be covered
with dirt particles, exuviae or prey remains (e.g. Cyc/osa, Tegenaria, Agroeca, Hyptiotes).
It may also be made of coloured silk which camouflages the egg case perfectly against
a background of the same colour (e.g. Tetragnatha). Argiope builds yellow, green or
whitish cocoons which are suspended in grassy vegetation, whereas the cocoons of
Nephila clavipes are yellow to brown and are found on bark.
Some cocoons are covered with loops of fairly strong silk (e.g. Episinus, Ero). While
some threads are used for the suspension of the cocoon, the complete loop layer may
be a special protection against egg parasites. Nevertheless, small hymenopterans and
dipterans exist which parasitize even these cocoons (Valerio 1984) and probably no
type of cocoon is completely free of parasites.
3 Silk as a Prey-Catching Device
3.1 The Evolution of Webs
The retreat of a spider is most effective as a protective structure when it is closed on

all sides. This, however, limits the possibility of perceiving and catching prey. There-
fore, spiders have evolved two different strategies: either they leave their silken retreat
to catch prey (e.g. Clubionidae) or they have modified the tube to be both retreat and
snare at the same time. In the latter case parts of the opening of the retreat are enlarged
to form a signalling device. Potential prey items which touch or run onto the structure,
produce vibrations which inform the spider of the presence, kind, size and speed of the
victim.
The plesiomorph situation is certainly represented by single threads which radiate
from the opening of the spiders' tube. Sometimes, these threads are elevated onto small
silken socles (e.g. Liphistius, Segestria, Uroctea; Murphy and Platnick 1981). Some-
times the retreat is closed by a lid which is only opened when the spider is lying in wait
for passing prey (e.g. Ctenizidae, Liphistiidae; Haupt 1979).
Connecting threads between signal threads prevent passing arthropods from escaping
quickly and represent a simple kind of snare. Many cribellate and ecribellate spiders
produce such irregular webs (e.g. some Dipluridae, Amaurobiidae). Further extensions
may form a funnel web (where the complete snare is a modified opening of the tube)
or a sheet web (where only the upper or lower half of the tube is enlarged into a snare).
The effectiveness of snares is increased remarkably by the addition of threads to form

a three-dimensional web. These webs are more successful in coping with the vegetation
stratum which contains a wide range of potential prey, particularly flying insects. From
the functional point of view the various types of three-dimensional webs operate more
or less in the same way. A general distinction can be made, however, between spiders
which run over the sheet (e.g. some Agelenidae) and those which hang below it (e.g.
some Linyphiidae), but there is strong evidence to show that these web types have
envolved several times independently.
Three-dimensional space webs are effective in catching prey, but they require much
material and time for their construction. Therefore, several reductive trends with re-
markable parallels can be observed in many spider groups. A general trend leads from
three-dimensional space webs to two-dimensional webs. The regular arrangement of
threads in a web also facilitates the location of entangled prey items. In many spider
groups such regular thread arrangements tend towards radial symmetry with the spi-
der in the center of the web. The best-known examples of radial symmetry are the orb
webs and similar structures produced by many families (Araneidae, Tetragnathidae,
Metidae, Sym phytognathidae, Anapidae, Theridiosomatidae, Uloboridae). Use of sticky
substances (e.g. in Araneidae) or cribellate silk (e.g. in Uloboridae) prevents entangled
insects from getting free and allows the spider to reduce further the total number of
threads per web.
Radial thread arrangement can also be found, however, in other web types (e.g.
Dictynidae, Amaurobiidae, Filistatidae) (Fig. 67a). Finally, it should be mentioned
that nearly all families which have evolved webs of radial symmetry also include spe-
cies with a more reduced web and completely (secondary) webless species. The araneid
Pasilobus (Fig. 62) and the uloborid Hyptiotes build sector webs; one-thread webs are
built by the uloborid Miagrammopes; adult Pachygnatha are secondary webless as are
some araneid spiders which use pheromone traps to attract moths (Figs. 63 and 86;
Lopez, this volume). More information concerning the evolution of spider webs is given
in Kaston (1964), Heimer and Nentwig (1982), Nentwig and Heimer (1983) and Levi
and Coddington (1983).
3.2 Wrapping Behaviour
Most spiders use silk to wrap prey items when subduing them (attack wrapping) or
after capture (postimmobilization wrapping). Wrapping is important for web spiders
which have to overcome insects as soon as possible after the prey has entered the web.
Spiders wrap with alternating movements of their fourth pair oflegs by throwing large
silk ribbons onto their prey. After this first immobilization the item is quite often held
with the third pair of legs and is rapidly rotated in front of the spinnerets. The spin-
nerets and the fourth pair of legs then work together to wrap the prey completely
(Fig. 72b). This behaviour is highly developed in pholcids, araneids and uloborids
(Eberhard 1967; M.H. Robinson and Olazarri 1971; M.H. Robinson 1975). After wrap-
ping some items do not need further treatment by a venomous bite; indeed, uloborids
and some other spiders have reduced their poison glands completely (Opell1979).
Most theridiids throw a silk mass and glue droplets onto their prey items. Drassodes
and Uroctea run around their victims and fix them to the ground with silk ribbons.
Hersilia and some diplurids display similar movements, but they use their unusually
long spinnerets to throw a large silken "carpet" onto the prey (Bristowe 1958; Stern
and Kullmann 1975).
Wrapping among free-hunting spiders has not been adequately investigated, but
observations have been made which confirm a wrapping behaviour or similar , behaviour-
al patterns for most families of non web-building spiders (Melchers 1963; Rovner and
Knost 1974; Heimer, unpubl.; Decae, pers. commun.). Wrapping behaviour follows no
family-specific pattern. Many mimetids (e.g. Australomimetus, Arcys, Celanor, Mime-
tus) wrap their prey regularly (Cutler 1972; Heimer 1984, 1986a), whereas Ero species
never wrap (Heimer 1986a). Cupiennius and Dolomedes do not wrap harmless insects,
but only dangerous prey items or insects with long legs or wings (e.g. grasshoppers).
Phormictopus intensifies its wrapping movements after the prey mass has been thor-
oughly chewed and is disintegrating (Heimer, unpubl.). Lycosids, which live in higher
vegetation strata wrap their prey, whereas closely related ground-dwelling species do
not. All spiders exercise a size-specific discrimination: small items are not wrapped,
but the larger a victim is, the more likely it is to be wrapped (M.H. Robinson and
Olazarri 1971; Nentwig 1985b). Only a few families appear never to wrap prey items,
among them Salticidae, Thomisidae, Philodromidae and Heteropodidae.
These examples show the multifarious functions of prey wrapping in spiders. There
are spider-specific, prey-specific and habitat-specific patterns. Web-building spiders
wrap fairly often, free-hunting species seldom. Wrapping in both groups yields the ad-
ditional advantage that an enwrapped item cannot escape or be lost if the spider has
to move (e.g. to catch a second item or to escape an enemy). Therefore, prey wrapping
in spiders should be considered as a multifunctional and complex behavioural sequence,
but not as a taxonomic pattern.
4 S tabilimen ta
4.1 Occurrence
After having built a snare, some orb weavers add more silken structures to the web, so-
called stabilimenta (Fig. 73). These structures are not necessary for the trapping func-
tion of the web and they can be found only in the araneid and uloborid orb web (Hing-
ston 1927). The cruciform or shield-shaped stabilimenta of Argiope species are well
known (Fig. 73b-e) (see below), but other spiders build such structures too, most of
them, however, are less conspicuous. Cyclosa constructs regular, species-specific silk
structures which form dots, lines or circles (Bristowe 1941; Rovner 1976) and include
parts of exuviae, prey remains, faeces or dirt particles. The spider sits motionless in the
middle of the stabilimentum. Its body and egg sacs have the same colour and the same
surface texture so that they are hardly detectable and represent a good example of
mimesis. The regular arrangements of silken dots in the orb web of Salassina crassispina
may have a similar mimetic function or it may deflect a predator's eye from the spider
which usually sits on the hub (Fig. 73). However, the silk is conspicuously white and
the spider is completely black. Casteracantha and Micrathena species add small silk
flocks to their web or close a few web meshes with a thin silk layer (Fig. 73f- h). The
Fig. 73a-j. Stabilimenta in orb webs. a Nephi/a c/avipes; b, cArgiope savignyi; d, e Argiope argen·
tata; f, g Gasteracantha cancriformis; h Micrathena sexspinosa; i Salassina crassispina; j unknown
uloborid
form of these silk structures is highly variable. Uloborids (Philoponella, Uloborus, Zosisj
also build very variable stabilimenta. The most common type seems to be one (or two)
linear silken bands which extend through the web's hub, usually with a central gap
about as long as the spider's body. Some species build a circle or spiral of dense silk
close to the hub of their webs or cover the hub with a fine sheet (Fig. 73j) (WieWe
1927; Eberhard 1973; OpellI979). The spider normally sits in the central gap in a rest-
ing position with extended legs and appears to be "fused" to the stabilimentum so
that stabilimenta, among uloborids, may generally be considered as a protective device
against visually hunting predators (Eberhard 1973). Quite often this device camouflages
the spider very well, but the silk itself is so conspicuous that some doubts remain.
4.2 Functional Explanations
The functional explanations for the cross and shield structures of Argiope species have
been summarized and discussed in M.H. Robinson and B.C. Robinson (1970b, 1973b;
B.C. Robinson and M.H. Robinson 1974, 1978). They cite the following possible func-
tions of stabilimenta in Argiope:
1. Antipredator function: a) Direct concealment of spider
b) Disguise of spider
c) Deflection from the spider
d) Advertisement of the web
e) Shielding against penetration
2. Mechanical functions: a) Adjustment of web tensions
b) Moulting platform
3. Other functions: a) Visual attraction of insects
b) Protection of the web against too large insects by mak-
ing it visible
c) Sunshield
The uncertainty with respect to stabilimenta can also be seen in the number of
names they have been given: stabilimentum, device, decoration, adornment, structure,
etc. The main problem in deviSing functional explanations is the extreme variability of
stabilimenta. M.H. Robinson and B.C. Robinson (1970b) found 65% of the webs in a
population of Argiope argentata to be without stabilimenta, only 3% had complete
stabilimenta. Similar figures were recorded in comparable investigations (Marson 1947;
Marples 1969; B.C. Robinson and M.H. Robinson 1974; Lubin 1975; Nentwig 1986b).
There is no obvious correlation between stabilimentum form and other factors such as
season, habitat and height of the web above ground. However, older spiders build fewer
stabilimenta than younger ones (Ewers 1972; Nentwig 1986b).
An experiment in which artificial stabilimenta were added to sticky traps to test
their use as a visual attraction for insects proved negative (Nentwig, unpubl.). A thermo-
regulative effect of the stabilimentum as a sunshleld may be excluded as well. Measure-
ments in the laboratory gave an order of magnitude of 0.05°-0.2 °c of shade effect
(Nentwig unpubl.). This is certainly negligible, especially since Argiope species have a
wide range of possibilities for behavioural thermoregulation (B.C. Robinson and M.H.
Robinson 1978). In contrast to many other authors, we think that cross or shield
stabilimenta are very conspicuous structures, so that a function, such as for conceal-
ment, disguise or camouflage against predators is unlikely. The colouration of Argiope
females (white , yellow and black) seems to be advertising (aposematic); in its natural
environment, however, we think that the colouration is more likely to be mimetic. The
contradiction presented by an advertising stabilimentum and a concealed spider is dif-
ficult to explain and may indicate that the stabilimenta of Argiope have different func-
tions.
3 4ro
:0 0:0[0]0:0.0:0.0:0:0:0:0.0.0:0:0:0
ro
o@]o :o 0:0 °
530 40 6 50
7
::
60
. . . XXIXIX : :
80 100
: : // / kC: ::::
11 0 120
130 140 150
160 170
. ,
1&0 190 200
: :+1 Argiope argenlale
4
0:0:0.O@]O:O ,0: 0:0: 0:0: 0: 0:0:0:0 ,0.0.0,0: 0,0:0.0:
10 20
5 6
:0:0:0@l0:0:0:0:0:0:0:0:0:0 0 o :o :o :o@]o ° :0:
30 40 50
7
:0:0:0:0:0~0:0 0:0 :0:0:0 0 00 : :0°0:0:00°
60 70
8 9 -a dul t
: °'§;:'§\~ 0 : 0: 0:: 0:: 0XZXZSSSJSIL 7S7X
80 90 100
7. 140
0: 0: : : : : : 0:: : : :.£.10:
160
Argiope s8vignyi
Fig. 74. Changes in stabilimentum form during ontogenesis of Argiope argentata and A. savignyi in
the laboratory. Each box represents 1 day, large numbers denote successive moults, small numbers
denote the age of the spider in days. The stabilimentum form is given as a symbol: circle = shield
stabilimentum; X = cross stabilimentum; bars = parts of a cross stabilimentum
4.3 Mechanical Function as a Moulting Platfonn
Whereas most orb weavers moult in their retreat or on a few nonsticky threads, Argiope
moults in the orb web. During and after moulting, the spider is defenceless for nearly
half a day and very susceptible to all kinds of disturbance. Therefore , many spiders do
not rebuild their webs for several days before moulting or reduce their webs to small
and stable platforms. Under laboratory conditions it was found that Argiope species
(anasuja, argentata, bruennichi, savignyi, trifasciata) of all ages were able to build sta-
bilimenta, but juveniles build them more frequently. A few days before moulting,
stabilimenta are built more frequently and/or more perfectly, so that spiders often
moult on these stabilimentum-stabilized webs (see also B.C. Robinson and M.H. Robin-
son 1978) (Fig. 74). Individuals change the stabilimentum form (cross, shield, parts
and combinations of both) during ontog(~nesis under constant conditions without
detectable reason; A . savignyi builds more shield stabilimenta, whereas A. argentata
builds more cross stabilimenta (Fig. 75). In general, Argiope builds in the laboratory
slightly less stabilimenta than in the field.
4.4 Antipredator Functions
This expression summarizes several phenomena which may vary according to spider
species, predators, habitats, etc. (1) Web shaking : Argiope spiders can set their webs in
vibration by rapid extension and retraction of the legs. Spider and web thus swing back
and forth, "disappear" in a shade of white silk and are difficult to locate by predators
Ishield I I cross I
% webs with stabiliment
100 savignyi anasuja
t-t-t- r-
60
20
100 r r
argentata argentata
60
20
[
-8 -6 -4 -2 m +2 +4 +6 +8 -4 -2 m +2 +4 time/d
Fig. 75. Changes in stabilimentum building frequency in Argiope species which moult (m) on a
shield or cross stabilimentum in the laboratory. A few days before moulting stabilimenta are built
more frequently, after the moult spiders construct them less frequently
(Tolbert 1975). (2) Especially younger instars with relatively large shield stabilimenta
may gain mechanical protection against predators by darting rapidly through the free
zone of the web with surprising speed on the far side of the shield. (3) The conspicuous
stabilimentum may advertise the position of a web to predators (e.g. birds) which can
then avoid the web. M.H. Robinson and B.C. Robinson (1970b) and Horton (1980)
discuss orb webs as a noxious stimulus to birds.
There is, however, no convincing experimental proof of a selective avoidance of orb
webs with stabilimenta. M.H. Robinson and B.C. Robinson (1970b) performed experi-
ments with model stabilimenta which indicated that birds could use stabilimenta as an
aid to prey location. This would, of course, be disadvantageous for the web resident.
Horton (1980) showed that both the web and the stabilimentum inhibited predation
by blue jays. Eisner and Nowicki (1983) gave evidence that webs without stabilimenta
are more quickly damaged than webs with stabilimenta and discussed flying predators
as possible causes. In this respect stabilimenta could be advantageous for the web re-
sident.
4.5 Stabilimenta as Relicts?
The high degree of variability of stabilimenta in the field was not taken into considera-
tion in the above-mentioned bird experiments. The fact that spiders in the laboratory
build preferably a stabilimentum before moulting, that these structures are highly
variable and that spiders can also moult without them, indicates a high flexibility. It
may be that araneid spiders had to develop a stabilimentum during the evolution of
orb webs because these first, day-active orb weavers moulted on their orb webs. Only
stabilimenta provided a safe moulting platform away from sticky threads and struggling
prey items. Changing moulting circumstances (e.g. moulting in the retreat) and/or
ameliorations of the static conditions of the orb web have had the result that the
moulting platform is no longer necessary and could, therefore, be reduced. Thus, we
can regard it as a relict which occasionally appears during moulting even in spiders,
which otherwise never build stabilimenta (Nephill1, M.H. Robinson and B.C. Robinson
1973b) (Fig. 73a) or under circumstances which are probably nonfunctional (Gastera-
cantha, Micrathena, Fig. 73f,g). In some cases the stabilimentum has changed its func-
tion during evolution and is used today in a protective context. This corresponds to
the conditions found among Cyclosa and Sall1ssina, perhaps also to some extent in
Argiope species (Fig. 73b,i). Generally, we interpret the high degree of variability in
form and frequency of stabilimenta (Fig. 74) as the result of low evolutionary pressure
and functionality.
5 Silk Use and Communication Between Conspecifics
Among most spiders, silk is used as a means of communication between sexual partners.
Quite often spiders contact through silk -mediated pheromones (recent reviews in Tietj en
and Rovner 1982; Pollard et aI., this volume). As an alternative (or additional) mecha-
nism, vibration signals may be emitted. Spiders have evolved a wide spectrum of me-
thods to produce vibrational stimuli: Many araneid males pull at peripheral threads of
the webs of conspecific females, Agelena and Amaurobius drum with palps or abdomen
on the sheet of the web; many linyphiids and theridiids possess special stridulation
organs as vibration sources (for a recent review, see Krafft 1982).
Before courtship, male spiders build a sperm web. This is a rather small, often tri-
angular web onto which the sperm is deposited from the genital porus. Then, the co-
pulatory organs, the palps, absorb sperm and store it (Lopez, this volume; Kovoor, this
volume).
Silk may be involved in the courtship behaviour of many spiders in different ways.
Some males wrap a prey item and offer it to the conspecific female which feeds on the
"nuptial gift" (e.g. Pisaura mirabilis, Meta; Blanke 1974; Nitzsche 1981). Thomisids
fetter a female with a few threads before copulation (Bristowe 1941); some araneids
build a "nuptial bridge" (Foelix 1979).
The substrate on which spiders copulate are usually sheets. Platnick (1971) and von
Helverson (1976) gave evidence that the type and size of the web may have had some
influence on the evolution of the position of spiders during copulation. Primarily
hunting spiders copulate on the substrate, on a simple sheet spun over the substrate
(e.g. Theraphosidae), or in front of the opening of the female's tube (e.g. Heptathela,
Haupt 1979). Spiders which hang in their web copulate in the same position (von
Helverson 1976; Heimer and Nentwig 1982), though. the silken structure may be reduced
to one "nuptial thread". Copulation inside silken tubes has led to copulation positions
which are determined by the lack of space. Sometimes these positions have been main-
tained even after the spiders changed their place of copulation or reduced their web
completely during further evolution (e.g. some lycosids; Platnick 1971; von Helverson
1976).
Most spiders which take care of their offspring are web-building spiders and the
most highly evolved social spiders are exclusively web builders (Buskirk 1981; Nentwig
1985c). Silk-mediated communication is common between mother and offspring.
Guarding females are able to guide juveniles to freshly killed prey through vibration or
to warn them of danger (e.g. Coelotes, Theridion;N(jJrgaard 1956; Tretzel 1961; Kull-
mann 1974). Spiders which live in communities of 1,000 to 10,000 individuals (e.g.
Agelena consociata or Anelosimus eximius; review in Buskirk 1981) certainly need a
wide spectrum of communication skills (Krafft 1982).
6 The Use of Silk and Webs by Heterospecifics
6.1 Spiders
Many spiders use the webs of heterospecific spiders to live in and to catch prey (com-
mensalism) or to steal prey (kleptoparasitism) or to prey on the resident spider (preda-
tion) (Vollrath, this volume). Some spiders regularly enter the webs ofheterospecifics
and prey on the resident spider. The pholcidHolocnemus is known to enter a web (e.g.
of Cyrtophora) where it locates the resident spider and kills it (Blanke 1972). Most
mimetids are specialized predators of web spiders. Gelanor, Mimetus andAustralomi-
metus (Heimer 1986a) enter the webs of araneids, theridiids and other web builders
and kill the resident spider. Ero stays at the periphery of the host's web and pulls at
the threads in a manner similar to a courting male (Czaika 1963) or an entangled in-
sect. The resident spider approaches the source of vibration and is immediately bitten
and killed by Ero. A few observations have been made in which even salticids or tho-
misids, otherwise free-hunting spiders, are able to move into araneid webs where they
kill the host (M.H. Robinson and Valerio 1977; Nentwig, unpublished).
6.2 Other Animals
Many insects live more or less constantly in or near webs though in many cases the
advantage of this behaviour is not completely clear. Especially webs which are con-
structed to last for weeks or months attract many other arthropods as exemplified by
the following case study on Anelosimus eximius.
The large webs of A. eximius are maintained for many years. The webs not only
catch prey, but collect vegetation debris as well. This forms a decomposing layer on
the web floor, where algae, mosses and lichens grow. This world within a world is in-
habited by many arthropods. Detritophagous species (e.g. Psocoptera, Collembola)
feed on prey remains and the debris layer; carnivorous species (e.g. ants, Staphyli-
nidae) may prey on them. Some phytophagous insects (e.g. beetles or leaf-cutting ants)
probably use the web only to reach their next food plant (Table 19). Most of these
species are certainly not specialized on A. eximius webs, they live on bark or on the
ground and occasionally extend their range to the spider webs. Some species, however,
are more closely associated with the webs of A. eximius.
The mirid bug Ranzovius moerens was found exclusively in the webs of A. eximius
and of the closely related A. jucundus (Wheeler and McCaffrey 1984; Nentwig and
Christenson 1986) where it feeds on prey remains. The bugs move very fast on the web
Table 19. Arthropods (except spiders) "associated" with the webs ofAnelosimus eximius in Panama
(Nentwig, unpubl.)
Abundance a Life style b
Formicoidea +++ p/c

Lepidoptera ++ d
Coleoptera: Staphylinidae ++ c/d
Curculionidae +++ p
Chrysomelidae + p
Others +++ d/p
Diptera: Cecidomyiidae ++ r
Larvae + d
Heteroptera: Miridae +++ c/d
Psocoptera +++ d
Collemboia +++ d
Acari +++ d
Opiliones ++ c/d
lsopoda + d
Hirudinea + c
a + Occasionally; ++ more often; +++ frequently.

b c =Carnivorous; d =detritophagous; p =phytophagous; r =resting.
threads, never get entangled and were never found in the spider's diet. More than half
of all A. eximius webs regularly contained the bugs. The behaviour and life cycle of
the noctuid moth Neopaitis madata is closely linked to seasonal changes in the webs of
A. eximius. The larvae feed on remnants of the spider's prey. Pupae can frequently be
found in the debris layer of the webs and the adult moths could be observed flying
around the webs (M.H. Robinson 1978).
Cecidomyiid and other midges are common around webs of many spider species
and occur sometimes in great abundance. The cecidomyiids use exposed threads as
resting places, where they hang with one or two front legs and stay there for hours
(Fig. 76). When disturbed, the midges flyaway, but never get entangled in the web.
Lahmann and Zuniga (1981) studied the resting behaviour in Costa Rica and found
the midges to be equally abundant on different threads of a web and on webs of dif-
ferent species and that they are equally prepared to use cotton threads; This suggests
that protection from predators and the possibility of rapid departure from an exposed
thread may be the main reasons for this strange resting behaviour. It could also be
observed in other long-legged Nematocera (C.P. Alexander 1936; Scott 1958; Sivinski
Fig. 76. Insects resting on spider threads. Female gall midge (Cecidomyiidae) in Middle Europe.
(Photo: Heimer)
, em
,.......---,
Fig. 77. Flight of one Microphor anomalus female (broken line) near the orb web of Zygiel/a x-no-
tata. The signal thread (S) leads from the hub to the retreat where the spider sits. Open circles re-
present prey items (mainly aphids) , flight duration 30 s. (From Nentwig 1985e)
and Stowe 1980) and other insects, such as emisinid bugs or moths (M.H. Robinson
and B.C. Robinson 1976;Fig. 76).
Numerous other Diptera species (mainly flies) are frequently found feeding on prey
items entangled in webs. Female Microphor flies are obligate kleptoparasites in orb
webs (Nentwig 1985e) (Fig_ 77), other species probably steal prey from a web only
occasionally. For a recent review on Diptera, associated with spider webs, see Nentwig
(1985e).
A wide spectrum of other animals uses spider webs more or less regularly. Panorpa
(Mecoptera : Panorpidae) lands in the webs of orb weavers and steals prey (Thornhill
1975). Tropical giant damselflies (Odonata: Pseudostigmatidae) are specialized preda-
tors on small web-building spiders and their prey (0. Fincke 1984). The polistine wasp
Mischocyttarus can walk on web threads , including sticky threads, owing to its special
tarsal lobes (Jeanne 1972). The wasp steals prey and preys on the resident spider. Ver-
tebrates users of spider webs include neotropical hummingbirds which pick insects and
spiders from the web, but also steal silk material (Fig. 97), small European songbirds
and Australian birds (A.M. Young 1971 ; Nicolai 1973; E. Thaler 1976).
C Feeding Ecology
I Structure and Function of the Digestive Tract
KLAUS-GUNTER COLLATZ 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
2 Two Methods of Prey Handling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
3 Structure of the Alimentary Tract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
3.1 The Mouth Cavity and the Gnathocoxae . . . . . . . . . . . . . . . . . . . . . . . . . . 230
3.2 Adaptations of the Foregut . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
3.3 The Mesenteron: Adsorption and Storage Function . . . . . . . . . . . . . . . . . . . . 232
4 Composition and Function of the Digestive Fluid . . . . . . . . . . . . . . . . . . . . . . . . 234
4.1 Enzymes in the Digestive Fluid of Tegenaria . . . . . . . . . . . . . . . . . . . . . . . . 234
4.2 Venom and Digestion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
4.3 Regurgitation Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
4.4 Excretion and Water Transport from the Alimentary Tract . . . . . . . . . . . . . . . . 238
1 Introduction
Spiders are extraordinary in their mode of digestion which is known as extra-oral.

Digestive fluid is poured over and into the prey and the dissolved material is swallowed
into the intestine where it is subjected to further and final digestion. Together with this
peculiar method of food consumption, we find a number of anatomical adaptations
and physiological adjustments, the functional relationship of which has been studied
since the early works of Bertkau (1885).
2 Two Methods of Prey Handling
In general, arachnids completely lack biting mouthparts. The function of grasping and
the eventual rupturing of the prey is attributed to the structure of the chelicerae. In
Araneidae the chelicerae as well as the gnathocoxae are serrated, whereas in Theridi-
idae or Thomisidae they are not. Therefore, spiders like Tegenaria, Araneus or A vicu-
laria are able to knead their prey during the sucking process until nothing, but an amor-
phous mass remains. The remnants of a thomisid's meal, in contrast, appear as well-
preserved carcasses in which small holes have been pierced by the chelicera to inject
the digestive fluid (Gerhardt and Kaestner 1937).
1 Institut fUr Biologie I (Zoologie) der UniversWit Freiburg, Albertstrafl.e 21a, D-7800 Freiburg i.Br.,
FRG
230 K.-G. Collatz
3 Structure of the Alimentary Tract
Up to the first molt the spider lings are incapable of feeding themselves and during the
early nymphal stage, the digestive tract remains only incompletely developed. The
resorption of yolk from the digestive tract, together with the formation of the mid-
intestine, establish the ability of external food digestion (Legendre 1978).
The alimentary tract of the adults does not differ principally in its gross morphology
from that of other arthropods (Fig. 78). It is composed of an ectodermal stomodeum
(pharynx and esophagus), an entodermal mesodeum (midintestine) with numerous
diverticula forming a "liver" or hepatopancreas and an ectodermal proctodeum which
is enlarged to form a cloaca. Both stomodeum and proctodeum are usually chitin-con-
taining structures. The epithelium of the midintestine, however, is not covered with a
peritrophic membrane. The absence of this protective device, so characteristic for in-
sects, crustaceans and other arthropods, is obviously a result of their feeding habits.
3.1 The Mouth Cavity and the Gnathocoxae
The mouth cavity lies behind the chelicerae and is equipped with bristles and setae of
different origin, which together form an effective filter system. Attempts have been
made to determine the maximum size of particles which can pass through the filter ap-
paratus (Bartels 1930). It could be shown that the maximum diameter for an unrestrain-
ed passage was about 1 tim. This, however, does not rule out the intake of larger particles
into the mouth cavity. In orb weaving spiderlings the uptake of particles as pollen,
fungus spores or other microorganisms seems to play an important role in their energy
budget (R.B. Smith and Mommsen 1984).
A closer look at the structures which form the mouth cavity show a ventral labium
(derived from a sternite) and a dorsal rostrum (or epistome) bordered to the left and
right by the gnathocoxae (specially formed coxae of the pedipalps). Within the different
orders of Arachnida the gnathocoxae are subjected to manifold variations leading to
very differently shaped mouth cavities, especially pronounced, for instance, in Scor-
piones, Pseudoscorpiones, and Solifugae and fused to form a single concave plate in
Uropygi, Schizomida, and Ricinulei (Savory 1977). The functional significance of these
morphological variations, however, is mostly unknown and their value in taxonomy is
better appreciated. This was extensively reviewed by Kaestner (1952,1953). All mouth-
parts contain glands whose functions are not entirely known. But Sittertz-Bhatkar
(1980) attributed the special role of oral cleansing to the gnathocoxal glands. A histo-
chemical examination revealed the glandular secretions to be mucopolysaccharides.
They bind the unfiltered particles into a bolus which may he discharged after a meal.
3.2 Adaptations of the Foregut
After passing through the sieve of the rostrum, the predigested liquid food reaches the
distensible pharynx, flows through the very narrow esophagus and enters the sucking
stomach. The pharynx and sucking stomach part of the foregut, but not the esophagus,
are well supplied with muscles allowing an extensive dilatation especially of the normal-
Structure and Function of the Digestive Tract 231
Fig. 78. The different parts of a spider's

digestive system in the foregut region:
CH chelicera; R rostrum; GN gnathocoxa;
L labium; E esophagus; EVesophageal
valve; DL dorsal dilator ; RI ring muscle;
LO lateral dilator; MI mid·intestine;
SS sucking stomach
-·--- R
LO___ •
R______ - - ~~~'E';j
__• ______________ _ __ HI
ly inflated and double V-shaped sucking stomach. Since the esophagus passes through
the prominent nerve mass of the epi- and hypoesophageal ganglia, the muscles attach-
ing at the foregut can be divided into two portions: those in front and those behind
the nerve mass. All these dilator and constrictor muscles, together with a number of
other muscles, which are not directly inserted on the foregut, but operate indirectly to
open the pharynx and probably effect rostral movement, amount to more than 20 bund-
les.
Three groups of striated muscles insert at the four chitinous plates of the sucking
stomach. These are dorsal dilators which are mediodorsally attached to the prosomal,
lateral dilatators, attached to the endosternite and ring muscles. A more detailed de-
scription of the foregut muscles was recently given by Marples (1983). He studied
muscles of the foregut of 43 species of spiders belonging to 29 families, but found only
minor differences in the pre cerebral part and no variations in the region of the sucking
stomach.
Two rostral chitinous esophageal valves situated at the border between esophagous
and sucking stomach, together with the action of the ring muscles, prevent the reflux
232 K.-G. Collatz
of the engorged, predigested fluid (Legendre 1961a, 1978). Contraction of the ring
muscles forces the valves to close and the swallowed liquid food moves in a one-way
fashion to the midintestine. Relaxation of the ring muscles and contraction of dilator
muscles lead to an opening of the esophageal valves and powerful suction. The co-
ordinated muscle movement results in a peristaltic wave and considerably facilitates
food ingestion.
Millot (1936) observed differently situated cell groups in the pharynx region of
Zoropsis, Atypus, Pisaura, and several other spiders. According to their innervation
and structure as chemoreceptors, he supposed them to the gustatory cells. However, as
Foelix (1979) noted, this has yet to be confirmed and the possible existence of tarsal
chemoreceptors cannot be ruled out.
3.3 The Mesenteron: Adsorption and Storage Function
The following mesenteron is the most striking organ system in the spider. It sends
numerous diverticula not only into the opisthosoma, but also through the petiole into
the prosoma passing to the coxae of the legs. The opisthosomal part of the diverticula
are intimately connected with interstitial tissue and together form both the digestive
and storage organ. Authors of past and present have engaged in the study of the struc-
ture and function of this organ system and have given it so widely different names such
as "liver", "fat body", "pancreas", "chylusstomach", "chylenteron" or - as already
noted - "mesenteron" (Nawabi 1974). Each term only partially describes the func-
tion of this central organ of the spiders.
The complex interactions of the cell types of the mesenteron also include the func-
tioning of Malpighian tubules which are embedded together with hemolymph vessels
in the interstitial tissue. To gain insight into the function of this organ system, mainly
histological and histochemical investigations were done. Only a few studies deal with
quantitative biochemical aspects of the digestion.
The cells of the arachnid mesenteron mainly appear as a dimorphous epithelium
consisting of absorptive cells and secretory cells, the former being more numerous than
the latter. Berkau (1885) was the first to describe this anatomic feature and since then
a number of examinations of arachnids as well as mites, pantopods, and Xiphosura have
confirmed this structure (summarized in Nawabi 1974; Alberti and Storch 1983).
A third type of mesenteron cells, referred to as basal cells (Millot 1926) seems to be
the nondifferentiated embryonal cell reservoir for both absorptive and secretory cells,
but has not been mentioned in all of the species descriptions.
It should be noted, however, that these cells types are not only restricted to Cheli-
cerata, but also occur in coelenterates, plathelmints, and molluscs and in combination
with other intestinal cell types also in crustaceans. A detailed description of the cyto-
physiology of digestive and storage cells in the "liver" of a scorpion was recently given
by Goyffon and Martoya (1983).
The quantitative prevailing absorptive cells appear as typical transporting epithelial
cells with a prominent basal labyrinth and an excessive brush border. The mitochondria
are regularly ordered. The secretory cells are characterized by their well-developed,
rough endoplasmic reticulum and numerous vacuoles. Ii. basal tubulus system is lack-
ing. The different epithelial cells are connected by septate desmosomes. In most, but
not all, of the investigated species the interstitial tissue represents a storage organ with
numerous fatty and glycogenic inclusions. The intimate coupling of the epithelial cells
with fingerlike extrusions into the storage cells facilitates the substance transport be-
tween them (Nawabi 1974).
3.3.1 The Cell Cycle of Mesenteron Cells and Food Adsorption Time
Both the interstitial cells and epithelial cells undergo a characteristic cell cycle prior to,
during, and after food consumption (Fig. 79). A short time after prey capture the ab-
sorptive cells are rich in food vacuoles and partly dislodge the lumina of the midgut
diverticula. The secretory cells are emptied at the end of food ingestion. The resynthesis
of digestive enzymes takes about 24 h and is linked to an enlargement of the endo-
plasmic reticulum .
A remarkable similarity in the cytologically observed kinetics of food ingestion was
revealed by plotting the amount of reducing substances as reaction products of the most
prominen t enzymes (amylase, chitinase, N-acetylglucosaminidase) against the time after
beginning of feeding (Mommsen 1977). As short as 1 min after prey capture, the amount
of glucose equivalents rises, reaches a maximum between 10 to 60 min and decreases
thereafter within 250 min. After 24 h only traces of reducing substances are detectable
indicating that all predigested food was absorbed .
:g '"c: !a",
:g
.,
Q)",
~ .~ ~~
.~'" >~
0 1 -5~1 1 .9
a.>uQ.
~~g~'Ca J" ~~ ~
~oa~-U) =Bo
U) .....
~~~£ ~ 0. C. 0
- .... ro:l ~o
~Q)C:o-C:('Q
~ 'g, B . ~ B~E~ BE~5~~ ~~",t'l0~

u 'S"O
a>o~!.hgoQ.)
('):JQ)"OW°.c
eE
>Q)-~="8 >~;:(1)gg
>E~-o=
8_~
~-o.c > ... E Q) . ~ . - ~~ .~ ~! ~~~!.U~:
00)
_ > ';0
0. _ o ":.= '8 08 E'g 9~Eo.;~
(,)
9 :> o..B.E
Q) > .... 11) - -
g~ E~ ~ ,Q '"
~:: ~ -= .~a. ~ ~~.E e~ ~~.:; ~ ~\...~OO'C
~!ij~u~
~ 0 ~-& J, ~a:1_
Q) co ""0 --
150 en a. co ~ n:I ~~~:5 ~ '§~~ '§~ ~ -E ~-t3 .~ 8 (/)<1)_(0"'0_ (j)E~~:O ~~~~8~~~
0 0 0 0 0 om
o
.
5 10
,
15 20
Fig. 79. Time course (h) of food ingestion and adsorption in arachnids : I variation of reducing sub-
stances in the digestive fluid of Tegenaria (values in /Jg X 10-1 glucose equivalents per microliter
digestive fluid) ; II oxygen consumption of Paruroctonus (values in microliter O 2 per gram wet
weight and hour); III cycle of secretory and adsorptive cells in the mesenteron of Stegodyphus
(I adapted after Mommsen 1977; II Bradley 1982; III Nawabi 1974)
234 K.-G. Collatz
The increase of the basal metabolic rate of mammals after a meal has been long ob-
served. Several authors reported the same effect in insects and spiders (summarized in
Bradley 1982). The amount of oxygen consumption during the first hour after feeding
is remarkably similar in both spider and scorpion and reaches a threefold increase. The
length of time between food ingestion and return of oxygen consumption to basal
metabolic rates can be taken as a criterion for determining the assimilation time. In
the scorpion Paruroctonus the uptake of predigested food into digestive cells is com-
pleted after about 6 h, which approximately agress with gut passage times of other
arthropods (see Bradley 1982 for further details).
3.3.2 Minor Influence of Starvation on Mesenteron Cells
Starvation does not alter the ultrastructure of the epithelium of the midintestine to a
great extent. Mitochondria as well as the brush border and the basal membrane remain
unaffected. This is in contrast to other arthropods. Even after a starvation time of more
than 1 year, the interstitial storage tissue of the scorpion Buthus still contains glycogen
and lipids (Alberti and Storch 1983). The adaptation to long starvation periods is clearly
demonstrated by both the integrity of tissue structures and the remarkable ability to
economize the consumption of body reserves via a drastic reduction of the metabolic
rate (Collatz and Mommsen 1975; J.F. Anderson 1970).
4 Composition and Function of the Digestive Fluid
Despite apparent opportunities to study the occurrence and action of digestive enzymes
in the cell-free, regurgitated digestive fluid, namely, a defined ionic composition and
pH value, and the lack of special cofactors, only a few studies exist which reveal more
than the qualitative description of the presence of such digestive enzymes. The most
complete information at hand is the careful and thorough survey from Mommsen on
the hydrolytic properties of the digestive juice of the house spider Tegenaria africa
(Mommsen 197 8a- c) and its comparison to that of the hunting spider Cupiennius salei
(Mommsen 1978c, 1980). We, therefore, shall refer to these results in more detail.
Earlier investigations on the action of digestive enzymes cover the lytic activity of
digestive juice on protein, starch, or fat and are mainly of historical interest (Bertkau
1885; Schlottke 1936). The work of Pickford (1942) offers slightly more information
on the conditions for the activity of proteolytic enzymes in both tissue extracts and
gut fluids of spiders. The special role of alkaline protease in web digestion will be men-
tioned later.
4.1 Enzymes in the Digestive Fluid of Tegenaria
The digestive fluid of Tegenaria is relatively rich in protein (7% wIv) and nearly all of
this protein, separated into about 20 distinct bands after electrophoresis, exhibits
enzymatic activity (Table 20). Calculating the total body protein of Tegenaria (Collatz
and Mommsen 1974b), together with the amount of regurgitated digestive fluid after
Table 20. Specific activities, pH optima and activities at the pH of digestive fluid (7.45) of Tegena·
ria atrica digestive enzymes. Only enzymes with a specific activity of more than 1% of that of the
amylase are listed (13 other enzymes with lesser specific activities are not considered). (After
Mommsen 1977)
Enzymes Specific activity Activity at pH of digestive pH Optimum

(% amylase activity) fluid (7.45) (% pH opti-
mum activity)
O!-Amylase 100 100 7.5

Carboxypeptidase A 43.4 79 8.4
Esterase 25.1 88 8.3
Tributyrinase 18.4 70 8.0
f3-N -acetyl-glucosaminidase 15.9 13 5.5
Esterase 12.5 100 8.0
Lipase 12.4 70 8.5
Chitinase 1.3 83 6.8
exhaustive electrical stimulation, it becomes evident, that up to 3.5% of the spider

protein accounts for digestive enzyme protein.
The protein content of Cupiennius salei digestive juice reaches even higher values
of 10% to 12% (Neumann 1977). This protein content is almost ten times higher than
that of vertebrate duodenal or pancreatic juice. It can be seen that the high enzyme
concentration of the digestive juice is effectively adaptive to the mode of feeding since
it will be diluted after entering the prey.
The enzymes in the regurgitated fluid appear to be fully active since no evidence for
the existence of zymogens has been found. Surprisingly, the most active enzyme of all
was an ex-amylase which exists in two or three isozymes. This was, as Mommsen (1978d)
pointed out, in contrast to the prevailing hypothesis that the capability of polysac-
charide digestion is only weakly developed in carnivorous invertebrates (Barnard 1973).
4.1.1 Amylase of Tegenaria is "Cold Adapted"
A comparative investigation of the Cupiennius and Tegenaria amylases indicated that

the enzymes of both species are nearly identical in their kinetic parameters, the tem-
perature dependence of activity, however, is remarkably different. Using an Arrhenius
plot analysis, Mommsen (1978d) was able to show that below the transition point of
30°C, which is common to both enzymes, the Tegenaria enzyme exhibits lower energies
and enthalpies of activation. Together with the observation that Tegenaria even feed at
temperatures of about 4 °c (Collatzand Mommsen 1974a), the advantage of this "cold
adapted" enzyme becomes clearly evident. The Central American spider Cupiennius
lives in higher temperature ranges and does not survive 4°C for longer than about 24 h.
236 K.-G. Collatz
4.1.2 The Chitinolytic System
Besides the amylase activity, the chitinolytic system is remarkable due to its high spe-
cific activity and other kinetic parameters which allow an effective digestion of chitin
(Mommsen 1980). Two enzymes, a single chitinase and two isoenzymes of (J-N-acethyl-
glucosaminidase (chitobiase), are responsible for the chitinolytic action. The chitinase
shows a broad pH-optimum and a very low apparent Michaelis constant in comparison
to chitinases from insects or fungi. The pH-optimum of the chitobiase differs from that
of the chitinase and is sharper. Thus, under optimal conditions for chitinase, the chito-
biase exhibits only 13% ofits maximal activity. Nevertheless, even the reduced activity
of the latter enzyme is about twice as high as that of the former under these conditions.
The rate-limiting step in chitin degradation seems to be the chitinase (Mommsen 1977).
According to the observed activities, one can show that the chitinase-chitobiase con-
tent of Tegenaria digestive juice will easily account for the digestion of prey even under
in vivo conditions. Other carbohydrases exhibit only minor activities.
4.1.3 Carboxylic Esterases and the Role of Emulsifiers
The search for carboxylic esterases in Tegenaria reveals six isozymes with nearly iden-
tical molecular weights. They are not very different from lipid digesting enzymes in
other invertebrates and hydrolyze numerous different ester-substrates. As in the case
of chitinase, the activity of lipid -digesting enzymes cover a broad pH-spectrum and are,
therefore, well adapted to catalyze the digestion of substrates of different origin.
The digestion of the fat requires the participation of emulsifying substances. In
vertebrates these are the well-known esterified bile acids (glycocholate, taurocholate).
Emulsifiers of the steroid type apparently do not occur in invertebrates. Nevertheless,
the low surface tension of invertebrate digestive juice indicates the existence of emulsi-
fying substances and this was long known even for the digestive juice of spiders as the
study ofBertkau (1884) shows. He observed that the extra-oral juice of different spiders
was able to emulsify olive oil immediately and permanently, but in the absence of bile
acids. Since then, the structures of the corresponding substances are known to be mix-
tures of fatty acyl-amino acids (Collatz and Mommsen 1974b) with high proportions
of glutamic acid and fatty acids of medium (C S -C 1S ) chain length. The spectrum of
both amino and fatty acids in the emulsifiers is very similar for Tegenaria and Cupien-
nius, but it differs from the corresponding amino and fatty acid composition in total
body homogenates and from similar emulsifiers from other arthropods.
4.1.4 Proteases and Web Digestion
Since orb-weaving spiders are capable of digesting their own web, the properties of
proteases are of special interest. Tegenaria digestive juice contains a single protease
exhibiting a broad pH-optimum range. Unlike many other invertebrate proteases, it
is acid-stabile and attacks native protein at considerably high rates.
Several other peptidases and amidases with "chymotrypsine" and "trypsine"-like
activities are also present, in which a carboxypeptidase A was identified as one of the
most active hydrolytic enzymes (Mommsen 1977). The observation that Araneus di-
adematus regularly consumes its own web and then rebuilds a new one (Breed et al.
1964) led to the study of the fate of the ingested protein and the corresponding
enzymes. According to Peakall (1971), the digested amino acids of proteins are re-
incorporated into the new web. In Argiope two alkaline protease fractions were identi-
fied as responsible for the digestion of silk fibroin (Tillinghast and Kavanagh 1977)
and were separated from two other major proteases without activity against the fibers
of the large ampullae. It is possible that one fraction of the fibroin-digesting proteases
may contain more than one protease, whereas the second fraction consists of only a
single protease (Kavanagh and Tillinghast 1983; Tillinghast and Townly, this volume).
Spiders are able to incorporate a food mass which exceeds their own body weight.
The 200-mg Tegenaria africa, for instance, can ingest the predigested body mass of six
flies (40 mg each) within 3 h! Nevertheless, the in vitro determined activities and con-
centrations of its digestive enzymes show that it is possible to treat such enormous
amounts of food.
4.2 Venom and Digestion
The question has been often discussed whether the venom assists in digestion. Mter
the first demonstration of caseinolytic activity in two South American spider venoms
(Kaiser 1953), a number of authors reported proteinolytic, lipolytic, or general, non-
specific hydrolase activity in the venom of different spider species.
Despite the high 25% protein content of Tegenaria venom, Mommsen (1977) was
unable to detect any Significant enzymatic activity in the venom except for traces of
esterase and hyaluronidase activities. However, he did show that there is an extremely
high content of potassium (360 mVal/l) which exceeds that of the hemolymph by
50-fold. A careful reexamination of the venoms of three theraphosid species revealed
the earlier assumed proteolytic enzymes to be impurities resulting from less care in the
dissection of the venom gland or by obtaining the venom by electrostimulation which
easily results in a concomitant regurgitation of digestive fluid (perret 1977). This was
confirmed by Rekow et a1. (1983). However, an appreciable sphingomyelinase activity
was found in extracts of the venom apparatus of the brown recluse spider Loxosceles
by these authors (see also Maretic, this volume).
4.3 Regurgitation Feeding
The special instance should be mentioned in which the digestive fluid does not serve
the digestion of prey. Members of the cribellate (Eresidae) as well as those of ecribel-
late (Theridiidae) spiders feed their offspring by regurgitation of intestinal juice (sum-
marized in Kullmann 1968a; Nawabi 1974). The preparation and composition of this
nutritious juice differs somewhat from the pure digestive fluid as shown by Nawabi
(1974) (Fig. 79). The predigested food, which is taken up by the females after cocoon
building, does not enter the adsorptive cells. Moreover, the absorptive cells are sub-
jected, together with the secretive and interstitial cells, to partial lysis. The dissolved
products become liberated into the lumen of the intestine, mixed with the predigested
238 K.-G. Collatz: Structure and Function of the Digestive Tract
food, and thus, the offspring's nutrition is fonned. As a consequence of this autolytic
process, the females of the corresponding species die shortly after successful reproduc-
tion. It would be of great interest to evaluate the physiological and the possible hor-
monal events leading to this "programmed" death in greater detail.
4.4 Excretion and Water Transport from the Alimentary Tract
The structure and function of the alimentary system and its adjacent diverticula can-
not be fully understood without a short look at its role in excretion. As mentioned
above the Malpighian tubules of Arachnida are embedded in the interstitial tissue of
the opisthosoma (Fig. 82). The partly perforated basal membrane of the tubule cells is
tightly attached to the interstitial cells pennitting an intensive substance flow via endo-
cytosis into the tubules. In contrast to myriapods and insects, these diverticula of the
digestive tract are of entodermal origin and, therefore, insert at the end of the midin-
testine (Seitz, this volume).
Despite this different ontogenetic origin, the structure of arachnid Malpighian tubules
is remarkably uniform and resembles generally that of the little differentiated types of
apterygote insects with three segments (Seitz 1975, this volume). A striking feature,
however, is the development of a cellular plug which is able to block the openings or
terminal ends of the tubules (Fig. 83). This leads to a blockade of excrete especially of
guanine crystals. The characteristic storage of guanine in different parts of the spider's
body obviously depends on the stopping of excretory flow and it results in the white
color at the place of deposition which serve secondarily as species-specific signals (Holl,
this volume, Fig. 8).
In addition to the extensive storage excretion, the arrangemen t of Malpighian tubules
and midintestine allows an osmotically driven, backward transport of water from the
alimentary tract to the intestinal tissue and hemolymph. Up to now only one study,
however, has dealt with water reabsorption in arachnids. Using a perfused scorpion
ileum, Ahearn and Hadley (1976, 1977) were able to show that the desert scorpion
Hadrurus exhibits two, parallel-operating, water-transfer pathways, including an inter-
cellular and a transmural transport. These transport mechanisms are similar to those
known from insects. They require a paracellular standing osmotic gradient and an active
sodium transport across the apical brush border of tubulus cells which can be inhibited
by potassium. Starvation causes a sharp rise in luminal sodium which favors the uptake
of water. The same effect could be shown during dehydration.
Acknowledgments. The critical reading of the English manuscript by Dipl.-Biol. John Plant is grate-
fully acknowledged.
II Excretory Organs
KARL-AUGUST SEITZ 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
2 The Coxal Glands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
3 Nephrocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
4 The Interstitial Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
5 Guanocytes of the Intestinal Diverticula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
6 The Malpighian Tubules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
6.1 The Initial Section (Inferior Tubule) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
6.2 The Main Section (Superior Tubule) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244
6.3 The Terminal Section (Anterior Tubule) . . . . . . . . . . . . . . . . . . . . . . . . . . 244
7 The Stercoral Pocket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
1 Introduction
Arachnida have developed various systems to remove excreta. The following have been
identified and have been investigated to a greater or lesser extent: in the prosoma, the
coxal glands and the nephrocytes; in the opisthosoma the interstitial celis, the guano-
cytes of the intestinal diverticula (Seitz 1972), the Malpighian tubules and the epithelia
of the stercoral pocket. The nephrocytes, guanocytes, interstitial cells and, indeed, in
some cases also the cells of the hypodermis all collect excreta, but are not capable of
removing them from the body. This is the function of the Malpighian tubules (Seitz
1975), the coxal glands and the stercoral pocket. The following excreta have been
identified: guanine, uric acid, adenine and hypoxanthine (see also Holl, this volume).
2 The Coxal Glands
The coxal glands are inconspicuous organs, which are usually situated segmentally on
either side of the endosternites in the haemolymph current and which have external
openings at the first and third coxa of the walking legs. They are well developed only
in primitive spiders; in labidognath species they are either rudimentary or missing al-
together: therefore, they playa relatively minor role in a general treatment of excre-
tion among spiders. There is a lack of microscopic research data on the coxal glands
1 Fachbereich Biologie/Zoologie, Universitiit Lahnberge, D-3550 Marburg, FRG

240 K.-A. Seitz
and that which is known shows many structural variations; however, in general, a com-
mon feature is their subdivision into four sections:
1. The excretory saccule (sacculus): the sacculus lies opposite the endosternite of the
coxa of the first leg and has an excretory function.
2. The collecting duct: this structure connects the sacculus with the labyrinth.
3. The labyrinth: the labyrinth is a tightly coiled tube normally extending over four
coxae. Its epithelium contains membranous folds which suggest resorptive activities.
4. The exit tube: this channel sometimes begins with a bladder like expansion, con-
tinuing as an initially straight tube; its opening lies on the first or third coxa.
3 Nephrocytes
The nephrocytes are the largest cells in the spider's body, with a diameter ranging from
30 to 80 11m. In general, they are to be found as individual cells or groups in certain
parts of the prosoma. Such parts include the region of the chelicerae, the underside of
the suboesophageal ganglion and areas of the petiolus. These cells are probably special-
ised cells of the intersti tial cells which have become independent and capable of absorb-
ing the products of metabolic processes or injected particles from the haemolymph.
They can then restructure the substances within their cytoplasm and finally store them.
The cells usually have a close packed periphery. The nucleus lies in the light-coloured
centre of the cell surrounded by a large number of normally, yellowish-brown crystals.
As in the case of the guanocytes, the crystals are formed with the participation of the
smooth endoplasmic reticulum and mitochondria. It is not clear how the nephrocytes
could pass off their excreta to the coxal glands or the Malpighian tubules. However,
the guanocytes of older animals do not contain more excreta than those of younger
ones.
4 The Interstitial Cells
Deposits of guanine and uric acid can be identified in the opisthosoma by means of
the Gomori silver technique (Pearse 1968). High concentrations of excreta can be
found in the Malpighian tubules, parts of the intestinal diverticula, interstitial cells
and in the stercoral pocket (Fig. 80a- d). Here again the interstitial cells absorb the
products of metabolic processes from the haemolymph or neighbouring tissue, trans-
forming them with the help of the smooth endoplasmic reticulum and mitochondria,
and store the crystalline end products. At present, it has not been possible to establish
whether these crystals are then passed on to excretory organs.
5 Guanocytes of the Intestinal Diverticula
Particularly in the region of the peripheral intestinal diverticula, resorption cells suc-
cessively change into guanocytes. In this process the cells initially lose their stored sub-
stances; this is accompanied by the lysation of some parts of the organelles. During this
Excretory Organs 241
MDD
Fig. BOa-d. Guanine deposits in various regions of the opisthosoma of Cupiennius salei: a peripheral
opisthosoma; b central opisthosoma; c hindgut region; d stercoral pocket.
Abbreviations: Stercoral pocket (KT), Malpighian tubules (MG), intestinal diverticula (MDD), inter-
stitial cells (ZG)
242 K.-A. Seitz
breaking down phase, the first excretion crystals appear, although they share no similar-
ity with the guanine crystals which come later. The latter appear only after the cells
have become interlaced with an endoplasmic reticulum. The tubules of the reticulum
expand into cisternae which are connected to the numerous endocytotic vesicles. The
combination of endocytotic vesicles and cisternae gives rise to crystalline saccules.
Crystallisation starts in the walls of these saccules with the participation of mitochon-
dria: at first, fibrous structures appear which project in coils into the lumen of the sac-
cule. Crystalline platelets form parallel to the fibres.
As the crystallisation process progresses, an electron dense layer forms under the
membrane of the saccule which appears to differ in form from that of the crystalline
centre. In the distal areas guanocytes and resorption cells form a combined body,
whereas the proximal guanocytes extend deep into the lumen of the diverticula. These
cells have the highest activities in the storage of guanine (Fig. 81). The guanocytes lose
part of their load of excreta when the apical cell periphery becomes detached. How-
ever, most of these excreta are stored in the distal cell regions and serve, for example,
colouring (Holl, this volume). The storage capacity of the guanocytes and the inter-
stitial cells appears, in analogy to the storage capacity of the fat body of insects, to be
the most important limiting factor determining the longevity of spiders. During the
reproductive phase in female spiders and insects, the interstitial cells or the fat body
can accumulate guanine. If the storage capacity is reached, we cannot observe any
further activities in nutrition and production of eggs. Thus, the life phase of senescence
starts (Seitz unpublished).
6 The Malpighian Tubules
The Malpighian tubules represent the most important excretory system of spiders.
They consist of two, occasionally four, entodermal tubes which begin at the sides of
the stercoral pocket and, branching frequently, extend throughout the opisthosoma
embedded between the interstitial cells. Their diameters vary, with a tendency to be-
come less with increasing distance from the origin. Three different sections can be
identified in the Malpighian tubules, not only externally, but also on histological and
microscopic examination. These sections each have specific functions. In analogy to
the corresponding morphological sections in the Malpighian tubules of insects (although
these are ectodermal), they are called the initial (inferior), main (superior) and terminal
(anterior) sections.
6.1 The Initial Section (Inferior Tubule)
The lumen is surrounded by up to three cells which form a tube of 15 to 25 Ilm dia-
meter. These sections are found mainly in peripheral body regions. The tube is usually
~
Fig. 81. Guanocytes from beneath the hypodermis near the peripheral midgut diverticula: Guano-
cytes (Ge), hypodermis (HY), endoplasmic reticulum (ER), nucleus (K), vesicle formation (VA).
Groups of distal guanocytes in direct contact with another, proximal guanocytes forming spurs,
surrounded by resorption cells (RZ), glycogen (GL). Guanine crystals in the gut lumen (MD), ex-
cretion crystals (GU), microvilli (MV), rest of cuticle (KU)
Excretory Organs 243
Fig. 81
244 K.-A. Seitz
free of formed excreta. The lumen or apical border of the cells comprises a series of
microvilli. At the base of these microvilli vigorous exocytotic actions start. Next to the
nuclei in the cytoplasm of the wall cells, broad cisternae of the smooth endoplasmic
reticulum are found which contain finely granulated material. The cisternae are sur-
rounded by normal, large mitochondria and also others which show all stages of de-
generation. The boundary to the interstitial cells is formed by the basal membrane,
which contains fine pores. Beneath these, vesicles form and migrate into the cytoplasm.
In the cytoplasm they are found mainly near the smooth endoplasmic reticulum. The
basal membrane has attached fine muscle cells, which are presumably responsible for
the indirect transport of excreta. The wall cells are connected by tight junctions
(Fig. 82a). Certain initial sections contain a pluglike structure which is formed by one
of the three wall cells expanding into the lumen of the Malpighian tubule. The lumen
is greatly constricted and the microvilli at this point are displaced. The cytoplasm of
these special cells contains fewer organelles than other wall cells of the initial section;
otherwise their composition is the same (Fig. 82b).
The pluglike structures exist also in other chelicerates. The physiological significance
is still unknown. However, we can postulate that the plugs are the morphological con-
dition for excreta storage or they guarantee the selective resorption in this part of the
Malpighian tubules (Seitz 1972; Mothes and Seitz 1980).
6.2 The Main Section (Superior Tubule)
Here, a cross-section contains twice as many cells as in the initial section and reaches
30 11m in diameter. The wall cells have basal infoldings which are typical for cells with
a high capacity of resorption. The border of microvilli is sparser than that of the initial
section, although individual microvilli may be branched. Here, too, exocytotic processes
take place at the base of the microvilli. The cytoplasm of the wall cells contains reserve
substances and particles of excreta, but few organelles. Again, the smooth endoplasmic
reticulum is well developed; its cisternae contain excreta. These depots of excreta are
always surrounded by mitochondria which, however, usually show signs of degenera-
tion. The great majority of the normal mitochondria lie in the basal infoldings, usually
surrounded by particles of glycogen. The wall cells are connected by tight junctions.
Various types of excreta can be found in the wall cells: (1) mainly oval bodies in the
cisternae of the smooth endoplasmic reticulum; (2) free crystals in the cytoplasm and
in the lumen of this part of the Malpighian tubules; (3) irregular concretions. The
boundary to the interstitial cells is formed by the basal membrane, to which again fine
muscle cells are attached. Beneath the pores of the basal membrane, a variety of dif-
ferent endocytotic vesicles can be found (Fig. 82c).
6.3 The Terminal Section (Anterior Tubule)
The closer the Malpighian tubules are to the region of the hindgut or stercoral pocket,
the flatter the wall cells are and the greater the capacity and content of the lumen. The
flattish wall cells have only a few microvilli which, however, are very long. The lumen
contains crystals, concretions and vesicles filled with granules and cell debris. The sur-
rounding tissue differs from that of the other sections in that there are fewer interstitial
Excretory Or$ans 245
•
••
... • •
••
d
Fig. 82a-d. Cross section of various regions of the Malpighian tubules. a Initial section; b initial sec-
tion with cytoplasmic plug; c main section; d terminal section.
Abbreviations: Basal membrane (BM), amorphous excretion crystals (E), excretion crystals (EK),
endocytotic vesicles (EV), conglomeration of excreta (EKO), vesicle with fine granular content
(FV), dictyosome (GF), smooth endoplasmic reticulum (GER), nucleus (K), lumina (L), mito-
chondria (M), infoldings of the membrane (MF), muscle cell (MU), intestinal diverticula (MOD),
wall of the Malpighian tubules (MW), microvilli (MV), oval enclosures (DE), reserve substances (RS),
irregular condensations (UV), interstitial cells (ZG), cistern of the smooth endoplasmic reticulum
(ZGER)
246 K.-A. Seitz
Fig_ 83. Longitudinal section of the superior tubule with cellular plug (ZVP). In the lumen of the
tubule (L): excretion crystals (EK), cell debris (ZR) and free cells (FZ). Microvilli bent in direction
of the excretion streaming (arrows) . Cells of the wall flattened, separated from the interstitial tissue
by the basal membrane (BM). Towards the intestinal diverticula (MOD) another basal membrane
(BM); nucleus (K)
cells and more muscle cells. The space between the midgut diverticula and the wall cells of
the Malpighian tubules is much smaller (Fig. 82d). Irregularly placed along the terminal
section are valve mechanisms. Thus, the interstitial cells protrude into the lumen of
the Malpighian tubules, while still covered by a thin layer of wall cells. In this way a
cellular plug is formed. Opposite this is a hollow in the wall, the edge of which pro-
trudes as a ridge into the lumen as well. The interaction of this ridge and the plug
forms an effective two-way valve (Fig. 83). It is difficult at present to correlate the
ultrastructure of the cells with their function . But the apical microvilli border, and the
other organelles indicate secretory activities in the inferior tubule. In general, the cells
of the tubules absorb waste matter from the haemolymph, transform it in their cyto-
plasm and excrete it in the form of guanine On the basis of the similarity in the ultra-
structure of Malpighian tubules in spiders and insects, the physiology of excreta forma-
tion will occur in the same manner as mentioned by Rockstein (1974) .
7 The Stercoral Pocket
The structure of the wall of the stercoral pocket is similar to that of the main section
of the Malpighian tubules. The cells have deep basal infoldings and their lateral mem-
branes are extremely indented. Toward the lumen two cell types can be identified: the
excretory cell I and II. The type I cell contains densely packed excretion crystals, con-
Excretory Organs
Fig. 84. The wall of the stercoral pocket. Lumen

(LU) with agglomerations of the faeces (KH) and
excretion crystals (G). Wall epithelium with excre-
tory cells of type I and type II (EX I and EX II).
Basal infoldings (ZM). The boundary is formed by
the basal membrane (BM), to which a thick muscle
layer is attached (M). Next to this is situated a syn-
cytial organised interstitial tissue (SY) . Nucleus (K),
vesicle (V), excretion crystals (U)
248 K.-A. Seitz: Excretory Organs
cretions and vesicles. This type is very similar to the guanocytes. The type II cell has
very few organelles and no deposits of excreta. The lumen of the stercoral pocket con-
tains large agglomerations of faeces which consist of a fibrous shell with amorphous
matter within, in which cell debris such as degenerated nuclei can be identified. The
agglomerations are closely surrounded by excretion crystals (Fig. 84). The boundary
to the interstitial tissue is formed by the basal membrane, to which a thick layer of
muscle tissue is attached. This layer is embedded in syncytial organised interstitial tis-
sue. The stercoral pocket has all the morphological characteristics of an actively func-
tioning excretory organ. At the same time, the spider uses it as a storage depot for large
quantities of excreta and faeces, which can then be expelled suddenly in stress situa-
tions.
III The Prey of Spiders
WOLFGANG NENTWIG 1
CONTENTS
1 The Potential Prey of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
2 The Diet of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
2.1 Web-Building Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250
2.2 Nonweb-Building Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
2.3 Prey Specialists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254
2.4 Scavenging Spiders? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
3 The Size of the Prey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
4 The Defensive Measures of Potential Prey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
4.1 Behavioural Avoidance After Visual Detection of a Web . . . . . . . . . . . . . . . . . 259
4.2 Morphological Characteristics Aiding Prey Escape from a Web . . . . . . . . . . . . . . 259
4.3 Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
4.4 Aggressivity and Chitinisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
4.5 Chemical Defence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
1 The Potential Prey of Spiders
The range of prey theoretically available to a given species of spider is called its poten-
tial prey. It is usually determined by the method used to trap or catch all arthropods
in the habitat in which the spider species lives. For methods, see Southwood (1978).
The potential prey of epigeous spiders, for example, consists mainly of Collembola,
Diptera (Brachycera), Coleoptera, Hymenoptera (Formicoidea), Araneae and, to a les-
ser extent, Hemiptera. Spiders hunting on vegetation have a potential prey spectrum
of Hemiptera (Aphidina, Cicadina), Diptera, Coleoptera, Araneae and Collembola, but
less Hymenoptera (mainly parasitic wasps). Diptera (Nematocera), Hemiptera (Aphidina,
Cicadina), Hymenoptera (parasitic wasps) and Coleoptera, but less Collembola or Ara-
neae (order according to decreasing abundance) are available to web-building spiders
which filter their prey from the air (Nentwig 1981).
1 Institut flir Zoologie der Universitat, Universitatsstral1e 31, D-8400 Regensburg, FRG
250 W. Nentwig
2 The Diet of Spiders
2.1 Web-Building Spiders
The actual prey of web-building spiders is relatively easy to investigate by collecting

the prey items or their remains directly from the webs, from the retreat of the spider
(some Theridiidae), from a subsheet (some Linyphiidae) or from the ground (some
Araneidae). A major problem is the collecting selectivity of many researchers, who
tend to collect mainly large prey items and overlook the smaller ones. Nentwig (1985b)
discussed this point and showed that small midges, thysanopterans or aphids were as
important in the diet of big spiders as heavy grasshoppers or beetles. The small insects
which are caught in a web are often too weak to free themselves. They die undetected
and remain attached to the web (Nentwig 1982a). Those web-building spiders (e.g.
Araneidae) , which rebuild their web daily and eat the old silk in order to recycle the
web proteins, also eat these attached small insects (Peakall197l). Such behaviour pro-
vides a most economic way to obtain small insects and other microscopic particles of
nutritional value. In a study of Argiope argentata, small insects comprised up to 23%
of the total prey dry weight (Nentwig 1985b) and represented a part of their diet which
cannot be neglected (Fig. 85).
By the orb-weaving spiders, sticky webs are very suitable for filtering "aerial insect
plankton", such as Nematocera, Aphidina or Thysanoptera. Studies on various araneid
and tetragnathid species (Kajak 1965 ; Nyffeler 1982; Nentwig 1983) in the temperate
zone yielded similar results as to the composition of the diet of tropical orb weavers,
such as species of A /paida, Argiope, Erio phora or N ephi/a (Shelly 1983; Nentwig 1985b).
Nematocera and Homoptera (Aphidina, Cicadina) constituted the main prey groups
(60-90%), but Hymenoptera, Brachycera and Coleoptera were also caught. Due to
the different composition of the potential prey , tropical spiders tend to catch more
Hymenoptera, Coleoptera and Psocoptera, whereas spiders of the temperate zone catch
more Homoptera and Diptera. Differences below order level, however, may be more
pronounced. In the tropics, the Hymenoptera caught consisted mainly of winged ants,
whereas in the temperate zone they were mostly parasitic wasps; Hemiptera in the
tropics are mainly Cicadina and Psyllina, in the temperate zone, Aphidina.
%
Fig. 85. Cumulative presentation of
90
/' the prey size of 11 species or families
of web-building spiders (curve A,
70
A mean ± SD of species data, number
of prey ca. 15 ,000 (Araneidae, Tetra-
50 gnathidae, Linyphiidae, Theridiidae,
Agelenidae, Pholcidae, Amaurobi-
30 idae, Dictynidae, Filistatidae), com-
pared with prey size of the social
10 theridiid Anelosimus eximius (cur-
ve B, number of prey =573). (From
3 5 7 9 11 13 mm Nentwig 1985c)
prey length
The Prey of Spiders 251
Large insects, such as butterflies, bees, grasshoppers or dragonflies, may contribute

a major part to the total prey biomass of an orb weaver (M.H. Robinson and B.C. Ro-
binson 1970a, 1973a; Vetz et al. 1978; Murakami 1983; Howell and Ellender 1984).
Grasshoppers contributed 55% to the biomass of individually wrapped prey items of
Argiope argentata (Robinson and Robinson 1970a). Honey bees constituted 75% of
the prey biomass of Araneus quadratus (Nyffeler and Benz 1978) and the investigation
of Howell and Ellender (1984) yielded similar data for Odonata in the diet of Argiope
aurantia. Such large or "difficult" prey items (e.g. insects with strong chitinisation,
chemical protection or aggressive behaviour, see below) are overwhelmed by araneids
(and many other web-building spiders) using a highly evolved attack-wrap behaviour
which enables them to immobilise prey without direct contact (M.H. Robinson et al.
1969; M.H. Robinson and Olazarri 1971) (Fig. 72b). The prey spectrum of araneids
and other orb weavers, therefore, is extremely wide (Table 21).
Linyphiids build their space webs mainly without sticky threads (Peters, this volume)
and seem to compensate for the overall lack of sticky threads by a high thread density
in their webs. They yield, therefore, a prey spectrum rather similar to araneids (Tab-
le 21). This agrees with the results of Wise and Barata (1983), who found a very similar
prey composition in the sticky orb web of Metepeira and the non sticky space web of
Mecynogea (both araneids). In the prey of linyphiids, however, small items (Nemato-
cera, Homoptera, Thysanoptera, Collembola) predominate and more agile insects, such
as small hymenopterans or bugs, play only a minor role. Due to the lack of a wrapping
behaviour, linyphiids cannot cope with large or mechanically protected insects, such as
Lepidoptera or Coleoptera.
Theridiidae may be considered as a group which combines the web-building and be-
havioural properties of araneids and linyphiids. A typical theridiid web is a space web
with sticky threads, which are, however, locally separated from the web center. Ad-
ditionally, theridiids exhibit an attack behaviour in which they throw sticky silk masses
upon their victims. This enables the spider to overwhelm many prey types possessing
specific defences (ants, bees, beetles) so that these types may constitute a comparatively
high percentage of a theridiid diet (Table 21). This silk-throwing behaviour probably
represents a preadaptation for specialised preying on difficult items (see below).
Agelenids build extensive funnel webs which they use merely as a running platform.
Prey is seized with the spider's chelicerae, but is not wrapped with silk or held with the
legs. The prey of agelenids consists of a wide range of taxa, but very small or strongly
chitinised prey is usually absent. Some specialised agelenids, however, bite into the
intersegmental membranes of beetles or millipeds and kill them in large numbers
(Tretzel 1961; Riechert and Tracy 1975; Riechert and Cady 1983). The lack of sticky
threads requires a fast response of the spider to insects hitting an agelenid web. Butter-
flies, bees, wasps and flies may provoke an attack which quite often will be successful.
These groups consequently comprise a high percentage of the agelenid diet (Table 21).
Pholcidae (Pholcus) and Scytodidae (Scytodes) are two unrelated families with webs
consisting of only a few Signal threads. Both are long-legged, but handle their prey in
different ways. Scytodes, the spitting spider, subdues prey by spitting a mucilaginous
substance from its specialised poison glands (Millot 1930b; Dabelow 1958; C. Gilbert
and Rayor 1985; Kovoor, this volume). This probably limits the prey spectrum under
natural conditions primarily to soft-chitinised Nematocera and Araneae. Pholcus ex-
N
Table 21. Composition (%) of the main prey groups of spiders (+ indicates 0.1-0.4%) Ul
N
Spider species Prey groupa Reference
Nem. Brach. Hym. Col. Hom. Het. Thys. ColI. Others
Ecribellate web-building spiders

Araneidae
Araneus diadematus 59 --- 16 20 1 + 3 Nyffeler (1982)
Araneus quadratus 75 --- 9 1 13 + 1 2 Nyffeler (1982)
Argiope argentata 50 4 25 7 6 1 1 + 6 Nentwig (1985b)
Argiope bruennichi 50 ---- 14 1 30 + 2 4 Nyffeler (1982)
Nephila clavipes 56 6 23 5 4 2 1 + 4 Nentwig (1985b)
Zygiella x-notata 29 1 2 + 65 + 2 1 Nentwig (1983)
Tetragnathidae, Metidae
Meta menaridi 7 --- I 29 63 (a) Piitzsch (1966)
Meta segmentata 43 --- 5 48 + 4 Nyffeler (1982)
Tetragnatha extensa 81 --- I 1 8 + 9 Nyffeler (1982)
Linyphiidae
Various Erigoninae 7 --- 4 28 1 3 46 12 (b) Nyffeler (1982)
Linyphia triangularis 13 3 + 3 80 + + + 1 Nentwig (1983)
Various Linyphiinae 38 2 2 2 44 2 2 8 Nentwig (1983)
Theridiidae
Achaearanea riparia 6 --- 36 21 14 1 23 (b) Nyffeler (1982)
Anelosimus eximius + 1 18 25 10 14 + + 31 (c) Nentwig (1985c)
Theridion impressum 28 --- + 43 9 20 Nyffeler (1982)
Theridion pictum 19 33 6 34 3 1 5 Luczak and Dabrowska-Prot (1969)
Agelenidae
Agelena lab yrinthica 16 --- 29 8 7 2 38 (d) Nyffeler (1982)
Tegenaria ferruginea 4 22 25 26 4 + 4 15 (e) Nentwig (1983)
Pholcidae
Pholcus phalangioides 5 4 66 8 5 + 1 + 11 (f) Nentwig (1983) ~
Scytodidae Z
CD
Scytodes longipes 2 1 1 2 22(f) ::s
68 4 Nentwig (1985a)
?
dO·
Oecobiidae o-l
Uroetea durandi 8 12 59 9 + 9 + 3 Nentwig (1981) it
Cribellate web-building spiders :4'
'<
Filistatidae
'"0
....,
Filistata nana 31 5 16 30 3 6 8 3 Nentwig (1982b) til
"0
Eresidae IS:
Stegodyphus manicatus 5 7 11 11 22 15 22 7 Nentwig (1982b) ~
Dietyna arundinaeea 23 28 13 4 21 Heidger and Nentwig (1985) '"
9 3
Dietyna montana 90 4 2 1 + + 2 + Nentwig (1982b)
Amaurobiidae
Amaurobius ferox 1 74 1 21 3 Nentwig (1981)
Nonweb-building spiders
Lycosidae
Pardosa lugubris 13 19 4 20 8 2 34 (0 Edgar (1969)
Pardosa purbeekensis 20 --- 3 2 39 46 (0 Schaefer (1974)
Pardosa ramulosa 22 --- 7 24 11 2 34 (0 Yeargan (1975)
Pirata piraticus 12 --- 7 4 43 34 (0 Schaefer (1974)
Thomisidae
Misumena vatia 32 51 17 (e) Morse (1981)
Xysticus (on foliage) 65 --- 17 3 1 1 4 9 Nyffeler (1982)
Salticidae
Phidippus johnsoni 12 18 9 3 12 46(0 Jackson (1977)
Several species 7 9 18 4 18 2 42 (g) Nentwig (1986a)
Oxyopidae
Peueetia viridans 15 41 4 9 34 (e, 0 Turner (1979)
a The abbreviations of the prey groups refer to the following taxa: Nematocera, Brachycera, Hymenoptera, Coleoptera, Homoptera, Heteroptera, Thysanop-
tera, Collembola, others.
(a) 27% Isopoda, 16% Araneae, 11 % Myriapoda;
(b) many unidentified items;
(c) 12% Blattodea, 8% Orthoptera, 7% Lepidoptera,
(d) 15% Lepidoptera, 10% Orthoptera, 6% Trichoptera;
(e) many Lepidoptera;
(0 many Araneae; N
U1
(g) 20% Araneae, 16% Orthoptera. Vl
254 w. Nentwig
hibits a wrapping behaviour comparable to araneids, which enables the spider to subdue
a wide spectrum of "difficult" prey, such as ants, beetles or millipeds (Table 21).
The special adhesive properties of cribellate silk (Peters this volume) has always
provoked speculation about the trapping effectivity of cribellate webs (Eberhard 1980a,
Nentwig 1982b, Heidger and Nentwig 1985, Strohmenger and Nentwig 1986). Most
cribellate webs consist only of a few threads (e.g. Filistata, Dictyna, Amaurobius), but
because of their adhesive properties, these webs are able to catch relatively large insects
(flies, beetles). Such insects may become trapped in these snares even after being only
partially entangled, for instance, with one leg only. Under natural conditions, however,
the diet of cribellates (Table 21) consists mainly of small prey items. The prey oflarge
cribellate spiders also includes many difficult insects, such as beetles, parasitic wasps
and isopods (Nif>rgaard 1941, 1951).
2.2 Nonweb-Building Spiders
These spiders usually are capable of handling only one prey item at a time. The hunting
success is relatively low, while the ingestion time is short, which results in a low per-
centage of prey-{!arrying spiders in a given population. Estimations range from 0.4%
(Jackson 1977) to 6-8% (Edgar 1970; Schaefer 1974; Nyffeler and Benz 1981). There-
fore, an analysis of the diet of nonweb-building spiders always deals with low prey
numbers. Much of the information available is on lycosid spiders, whose main prey
groups are Collembola, Diptera, Cicadina, Aphidina and Araneae. Prey records for some
salticids and pisaurids (D.S. Williams 1979; Nitzsche 1981) are similar and confirm that
most of these spiders prey on small and soft-chitinised insects that do not possess spe-
cial defensive mechanisms (Table 21). Ground-hunting saltidids, thomisids and oxyopids
often show a higher percentage of ants in their diet (Nentwig 1986a).
2.3 Prey Specialists
2.3.1 Specialised Webs
The ladder webs of some araneid and me tid spiders (M.H. Robinson and B.C. Robinson
1972; Eberhard 1975; Stowe 1978) (Fig. 86a) are regarded as a specialisation to catch
Lepidoptera, especially moths. They fly against the web and shed some scales that
partly render the sticky web ineffective (Eisner et al. 1964; Nentwig 1982a). However,
they fall further downward through the long ladder web until they have lost most of
their scales and finally, become stuck to its adhesive threads. Unlike most orb webs,
these ladder webs are positioned perfectly vertical, so that the moth cannot fallout.
Stowe (1978) found 68% of moths among the prey captured in the ladder web of
Scoloderus cordatus (Fig. 86a) and this seems to confirm the functional explanations
suggested by other investigators.
Among web-building spiders, prey specialisation does not occur very often since a
web is essentially a device that catches a wide prey spectrum. Consequently, specialisa-
tion has taken, in most cases, the form of a reduction in the size of the web. Eberhard
(1979) described the "web" of Argyrodes attentuatus (Theridiidae). This consists of
Fig. 86a-c. Two araneids with

specialized moth-catching behav-
iour. a The ladder web of Scolo·
derus cordatus (natural length
70 cm) (from Stowe 1978);
b hunting position of the bola spi-
der Mastophora; c swinging the
ball at a nearby moth which ap-
proaches the spider which mimics
the female sex attractant phero-
mone of the moth (b, c from Eber-
hard 1977)
only a few nonsticky threads 1 to 2 m in length and spun irregularly in the vegetation.
Ballooning spiders and a few Nematocera, which fly against the threads, seem to be the
only prey of this spider.
Dinopids catch their prey in small rectangular webs of cribellate silk held at the
corners by the first two pairs of legs. This web is suitable for catching ants (a "difficult"
prey group for many spiders) and field observations have shown that the spiders are
frequently positioned near ant trails where they can easily capture the workers. The
natural prey of Dinopis, however, is not restricted to ants and in captivity they prey
on a wide spectrum of insects (M.H. Robinson and B.C . Robinson 1971).
The bola spiders are perhaps the only web-building grou p of spiders which are truely
monophagous. The "web" of these araneids is reduced to a short thread with one big
sticky drop terminally, in Mastophora (Figs. 63 and 86b,c), or completely absent in
Celaenia. The spiders lure their prey (one or a few related species of moths) with a
substance which mimics the sex pheromone of the insects. The scent is probably emit-
ted from glands in the spider's forelegs (Lopez, this volume). The bola is probably a
derived homologue of the orb web (Peters, this volume, Fig. 63) and serves to catch
the approaching moths (Eberhard 1977, 1980b).
256 w. Nentwig
2.3.2 Social Spiders
The few species of social spiders (for a review of this group, see Buskirk 1981) can be
regarded as prey specialists, too. Social spiders share some characteristics: they are
exclusively web-building species, they live mainly in the tropics where insects are
abundant throughout the year and they build large webs up to several meters in length
or height. Social spider communities may consist of 100 to 10,000 individuals that co-
operate in prey capture. The stronger the prey and the longer it struggles, the more
spiders approach and join the attacking group. This enables these spiders to subdue the
largest insects (Fig. 85). Preying on large items has been interpreted as the ecological
niche of social spiders (Nentwig 1985c). The prey of these spiders, therefore, is highly
different from that of all other spiders. In the case of the theridiid Anelosimus eximius
it consists, of ants, wasps, beetles, lepidopterans, hemipterans, cockroaches and grass-
hoppers (85-96% of the total prey) with a mean body length of 5-20 mm (maximum
60 mm) (Nentwig 1985c, see also references in Buskirk 1981).
2.3.3 Specialised Non web-Building Spiders
Specialisation of nonweb-building spiders on one prey type could only evolve towards
a small number of taxa (Nentwig 1986a), such as ants, spiders, and perhaps termites
and bees.
Spiders: Spiders probably have a preadaptation to prey on other spiders. Most re-
cords of such specialists concern webless species which preferably prey on web-build-
ing spiders. Field observations on the mimetids Era and Mimetus show that they enter
the webs of their victims and lure the resident spider by using typical vibratory pat-
terns which may imitate potential prey or a conspecific male (Czaika 1963). The
natural diet of the salticid Portia also consists exclusively of other spiders (Jackson
and Blest 1982).
In the laboratory, however, many mimetids and Portia accept insect prey as well
(Cutler 1972). The limited data that are available for archaeids (Legendre 1961b; R.R.
Forster and Platnick 1984; KulImann, pers. commun.) indicate a similar specialisation
for spiders, however, this needs further confirmation.
Ants: These social insects are a potential prey which are relatively common every-
where, but difficult to subdue because of their defensive capabilities. It seems necessary
to specialise on this prey type in order to overcome their defences and to catch them
in high numbers. Many spiders have developed sophisticated behaviour in this respect.
Probably all Zodariidae (Schneider 1971; Harkness 1977) and Aphanthochilidae (Oli-
veira and Sazima 1984) are obligate myrmecophagous. Other examples include the
Euryopis species (Theridiidae) whose secondary weblessness is interpreted as a means
of facilitating predation on ants (Carico 1978; Porter and Eastmond 1982). Also, some
species or genera of a few more families (Thomisidae, Salticidae, Clubionidae and Gna-
phosidae) appear to be ant specialists. In contrast to widespread opinion, ant-mimick-
ing spiders generally do not prey on ants, but they only profit defensively from their
appearance (Batesian mimicry). These spiders are rather unspecialised with respect to
their food preferences. Exceptions are aphantochilids and a few other species (Oliveira
and Sazima 1984; Nentwig 1986a).
Termites: In many tropical or subtropical habitats termites are as abundant as ants.

There are a few reports on spiders associated with termites, for example, the South
African family Ammoxenidae, but nothing is known about their prey (Dippenaar and
Meyer 1980).
Spiders which hunt on flowers (e.g. some thomisids or oxyopids) often feed only a
limited range of pollinating insects (e.g. bees, bumble bees, syrphid flies or butterflies)
(Table 21). This specialisation is partly due to the specific niche where the spider hunts
(the flower), but it is known that Misumena vatia also refuses, in feeding experiments,
to take other types of prey (Nentwig 1986a). Thus, the oligophagy of spiders hunting
on flowers could be interpreted as a consequence of the narrow spectrum of the poten-
tial prey.
2.3.4 Habitat Choice
As we have seen, the width of the actual prey spectrum of most spiders depends on the
width of the potential prey spectrum. Caves and cellars or habitats under stones and
bark are examples of environments where the potential spectrum is restricted to a few
groups. Prey analyses of spiders from these habitats, therefore, can give the impression
of oligo- or even monophagous species. The prey of Uroctea durandi (Oecobiidae)
which hunt under stones or of Pholcus phalangioides living in a cellar consisted to 60%
of ants (Nentwig 1981, 1983). On sandy soil Theridion saxatile catches mainly ants
(88%) (N4)rgaard 1956). M.H. Robinson and B.C. Robinson (1970a) found that the
prey of Argiope argentata near flowering shrubs was comprised mainly of Trigona bees
(62%). Large nocturnal, synanthropic spiders (Selenopidae or Heteropodidae) are
known to eat high proportions of cockroaches. In all these cases, however, the same
spider species will accept other prey when living in another habitat or when tested in a
feeding experiment.
2.4 Scavenging Spiders?
The above descriptions may give the impression that spiders feed exclusively on living
prey which they catch themselves. There are, however, many records from field ob-
servations which indicate that at least some spiders, under natural conditions, eat other
food. For instance, particular spider species are known to feed on spider or insect eggs
(Jennings and Houseweart 1978; Jackson and Blest 1982; Nentwig unpubl.); hatching
spiderlings commonly have a first meal on the undeveloped eggs of their clutch (ValeriO
1974); and newly moulted spiders drink the liqUid from the exuvia, which contains
partly digested endocuticular proteins (Dondale 1965). Spiders may also feed on all
kind of dead arthropods (Rovner and Knost 1974; Nentwig 1985d) and could fre-
quently be observed drinking on nectar or honeydew from coccids (M. Edmunds 1978).
Observations and feeding experiments (R.B. Smith and Mornmsen 1984) have re-
vealed that the tiny orb webs of the first free instars of araneid spiders are able to catch
microscopic organic matter, such as pollen and fungus spores. This diet is eaten with
the complete web and (under laboratory conditions) doubles the life expectancy of
spiderlings and also alters their web-spinning behaviour. This nutrition probably facili-
258 w. Nentwig
tates moulting to the next instar which is able to build larger webs suitable for insect
trapping.
Under laboratory conditions many spiders accept dead insects, pieces of meat, pie,
hard-boiled eggs, marmelade, bananas, etc. (Bonnet 1924; Decae 1986). Peck and
Whitcomb (1968) maintained lycosids and clubionids successfully on a diet of milk
and egg yolk which they offered on cotton swabs. Nentwig (in preparation) modified
these experiments and developed a completely artificial diet which was accepted by
several species. At present, it is difficult to estimate the extent to which "scavenging"
supports the nonnal diet of living prey in spiders. It is especially difficult to evaluate
its necessity for the first instars of web-building spiders and many species of nonweb-
building spiders where "scavenging" may contribute essentially to the natural diet.
3 The Size of the Prey
Most spider species catch prey items smaller than themselves. In feeding experiments
the probability of a standard prey item (a cricket) being eaten by a spider was measured
by offering crickets of different lengths to various species (Nentwig and Wissel 1986).
The resulting curves (Fig. 87) show the subduing potential of spiders with respect to
the size of their prey. These results, which are comparable with the results from field
collections, show that most species overwhelm prey items not larger than 150% of the
spider's size with a preference maxima of 50-80% (among nonweb-building spiders
many lycosids, theraphosids, selenopids, pisaurids, ctenids and pilodromids). Although
some groups, such as thomisids and some salticids, can subdue animals 200- 330%
their own size. There seems to be a tendency among spiders with elongated legs to prey
on larger prey items (thomisids, pholcids, perhaps heteropodids), but this is not true in
the case of Seienops. Among the web-building spiders, the agelenids, which simply bite
their prey and hold them with their chelicerae, like a wolf spider, can kill only relatively
small prey, whereas spiders which exhibit some kind of prey-wrapping behaviour can
kill prey items of more than 200% (maximum 300- 550%) of their own size. Even the
largest arthropods can be subdued by social spiders with communal prey attacks. Since
these spiders are often small (Ane/osimus eximius is 5 mm long), the size difference be-
tween predator and prey may be 10- to 20-fold of the predator's length (Fig. 85).
100 100
50 50
a b
o+---------.-~~-----.- o+----------.------~~o--
o 100 200 o 100 200
Fig. 87a,b. Acceptance rates (% of accepted crickets in feeding experiments) versus body length of
the prey (measured as % of body length of the spider) for (a) the wolf spider Pardosa iugubris and
(b) the web-building diplurid Ischnothele guyanensis. (From Nentwig and Wissel 1986)
Fig. 88. Generalized pattern of the relative lengths of .c

predator and prey. The surrounded areas indicate size ~ parasites
relations for : a spiders; b chewing insects; c insectivo- ~ m
rous birds; d hawks, owls; e ants, dogs (pack hunters). >-
Spiders are subdivided to: 1 nonweb-building spiders; ~ dm
2 large mygalomorph spiders; 3 nonweb-building spi- 0..
ders specialized on large prey items (e.g. some salticids

or thomisids); 4 solitary web-building spiders ; 5 ocial
s
spiders. The central dotted line indicates equality of
size of predator and prey; the peripheral unbroken lines
indicate prey (predator) length one-hundredth of pre- mm
dator (prey) length ; the three broken lines indicate prey feeders
150%,300% and 500% of the length of the predator.
(From Nentwig and Wissel 1986) mm em dm m
predator length
These data can be used as a model of the size relations of predator and prey with
special reference to spiders (Fig. 88). It shows that spiders, when compared to other
predator groups, cover a much wider range , i.e. nearly two orders of magnitude, for
the lengths of predators and prey. This large predatory potential is achieved through a
variety of hunting strategies, such as specific methods of overwhelming, different web
types and a highly evolved social behaviour.
4 The Defensive Measures of Potential Prey
4.1 Behavioural Avoidance After Visual Detection of a Web
Field observations and flight experiments have shown that several insect taxa, after
visually detecting a web, display avoidance behaviour. Several brachyceran flies (e.g.
syrphid , muscid , calliphOrid and asilid flies), hymenopterans (bees, bumble bees, wasps)
dragonflies and damselflies have been frequently observed to avoid Spider webs. These
insects are good fliers , they can stop immediately or change the direction of their flight.
They also can hover in front of the web and then escape either through a large hole in
the web or by finding the borders of the snare . In flight experiments, where insects had
the opportunity of choosing between wooden frames with or without an orb web, near-
ly 70% of honeybees, syrphids and muscids flew through the empty frame, thus avoid-
ing the web. Only 50% of Chrysopa carnea, a neuropteran of poor visual abilities, flew
this way, thus indicating a random flight distribution (Nentwig 1981). There is little
known about how insects may avoid nonweb-building spiders.
4.2 Morphological Characteristics Aiding Prey Escape from a Web
Because of their large surface area, insect wings are the body structures most likely to
become attached to a sticky snare. Special surface structures of the wings, which prevent
good contact, represent therefore, the most important morphological defence against
sticky webs. Examples are the wing scales of lepidopterans, hairs of caddishflies and
260 W. Nentwig
Fig. 89. Part of an orb web

after the escape of a butterfly.
The remaining scales make this
web part ineffective in catching
prey
some midges (Psychodidae, Culicidae) or the cuticular lipids of roaches or bugs (Eisner
et al. 1964; Nentwig 1982a). Butterflies or moths which fly against an orb web, lose
some of their scales, thus rendering the sticky web ineffective and allowing the insect
to escape (Fig. 89). Comparing the resulting scale traces in the web with the frequency
of lepidopterans in the prey of Argiope argentata, M.H. Robinson and B.C. Robinson
(1970a) estimated that 54% of the lepidopterans striking the web escaped. When the
scales and hairs are cautiously removed from the insect wings with a brush, the adhesion
to a sticky surface increases 300-600% . Removing the cuticular lipids from the wings
of a cockroach increases adhesion by about 100% (Nentwig 1982a).
4.3 Activity
Insects which become entangled in a spider's web struggle to get free. Some insects
(most Brachycera, Hymenoptera and Coleoptera) are very active, and are able to free
themselves within a few seconds. Therefore, these animals constitute only a small per-
centage of the prey of most web spiders . Asecond group of prey taxa shows only weak
activity and so has a reduced chance of freeing themselves within the first minutes after
contact (e.g. Chironomidae, Aphidina, Thysanoptera). These insects ("airborne plank-
ton") constitute the majority of the prey of most web spiders.
A third group of potential prey taxa shows activity patterns consisting of escape
movements and long pauses. These pauses may provide an important rest and recovery
period for the insects themselves, but they are certainly very important for dimishing
the risk of attracting the spider's attention. Many Nematocera, Orthoptera and Heterop-
tera exhibit this intermediate type of activity (Nentwig 1982a). Web spiders orientate
towards vibrations and are attracted by struggling insects. Active insects which do not
free themselves within approximately 1-5 s, consequently, are soon detected and sub-
dued.
It is not clear whether the escape strategy is influenced by the situation of "hang-
ing in a web" or by the plucking of the spider. Field observations indicate that many
insects try to "disorientate" the resident spider by short and strong activity peaks. In
a computer simulation, Hoffmaster and Hays (1977) tested a model of spider-fly pre-
datory interactions and found that the percentage of survival time of the fly was in-
creased by decreasing the duration of struggling bouts. A fourth strategy is to remain
completely motionless (feigning death). This can frequently be observed in Bibionidae,
Rhagionidae, Coccinellidae, Chrysomelidae, Symphyta, lepidopteran larvae or Opiliones.
In most cases, the resident spider, however, will detect such a prey because of changed
thread tensions in the web (Nentwig 1982a).
4.4 Aggressivity and Cbitinisation
Insects with aggressive behaviour, chewing mandibulae or a sting, are well-protected

against spiders. A thick chitin armour is at least as effective as a defence. Most non-
web-building spiders cannot overwhelm ants or hard-chitinised beetles because they
cannot bite through the thick cuticle. Specialists, such as the agelenid, Coelotes ter-
restris, bite into the beetle's intersegmental membranes (Tretzel 1961) and the gna-
phosid Callilepis noctuma bites ants in one ofthe antennae (Heller 1976). Many web-
building spiders first wrap their victims to prevent escape. Then they search the prey
to fmd a body part suitable for the poisonous bite.
4.5 Chemical Defence
Pentatomidae are not taken as food by most nonweb-building spiders, but most web-
building spiders, on the other hand, will accept them (Nentwig 1983, 1986a). Penta-
tomids possess stink glands (Remold 1962) which are a good defence against many
predators (e.g. most nonweb-building spiders). Web spiders, like araneids, immediately
use attack wrapping on pentatomids after they have become trapped in their web. Most
pentatomids are only wrapped and not bitten. In the time that follows (minutes to
hours) the bug struggles and uses its stink glands until it is exhausted. Only then does
the spider return and bite the bug. This is a kind of mechanical detoxification through
the combination of the sticky silk of the orb web and the wrapping behaviour of the
spider.
262 w. Nentwig
The same can be achieved by the very fme silk of cribellate spiders or by the long
legs of some spiders (e.g. Pholcidae), which hold their prey items far away from their
body and wrap them carefully. Eisner and Dean (1976) described comparable adverse
chemical releases from the bombadier beetle which defends itself successfully in this
way against a wide spectrum of predators, including nonweb-building spiders. Orb
weavers, however , evade the dangerous glands of the beetles by a successive and very
effective wrapping behaviour.
Defensive chemicals, which are not stored in glands, seem to be much more difficult
to overcome. The chrysomelid Age/astica alni, which contains salicylic acid in its haemo-
lymph, was not accepted by a variety of web-building and nonweb-building spiders
(Nentwig 1986a). A large number of butterflies (among them many aposematic species),
which contain poisonous secondary plant products, are carefully cut out from the web
by many orb weavers and rejected (Marshall 1902; M.H. Robinson and B.C. Robinson
1973a; K.S. Brown 1984).
Insects containing cyanides are generally well protected by this very toxic substance.
Some spiders, however, accept cyanogenic Julidae or Zygaenidae (Nentwig 1983).
Crickets, which were rendered artifically poisonous by abdominal injections of KCN,
were offered to Cupiennius salei. Unexperienced spiders refused more crickets with in-
creasing KCN concentrations, but generally accepted items with lethal doses (Fig. 90).
Spiders which were exclusively fed with KCN-crickets for 6 months became apparently
adapted to KCN . They ate large amounts of poisoned crickets and probably changed
the composition of their digestive fluid in order to incorporate more nontoxic cricket
fractions (i .e. lipids) than toxic fractions (e .g. water) (Nentwig 1985d).
Although chemoreceptors are known in spiders (Drewes and Bernhard 1976; Harris
and Mill 1977), their specificities have not yet been investigated. Presumably, these
sensilla are able to detect a variety of chemical substances in very low concentrations,
since in most cases the first touch of the prey with the tip of one leg of the spider is
sufficient to decide whether it will be accepted or refused ("taste-by-touch", Bristowe
1941). These examples show that the "palatibility" of an insect to a spider or the
"chemical defence" of potential prey are highly complex matters which deserve much
more intensive studies.
100 lipids eaten

0/0
50
~
.... ater eaten
50
Fig. 90. Feeding experiments with KCN-poisoned crickets and
Cupiennius solei. Spiders which had been treated for months with
KCN-poisoned prey (0) eat more lipids and less water from the
crickets than control spiders (.) which received only one KCN·
o 6~--1~60--~~--~ cricket. (From Nentwig 1985d)
5 Conclusions
Complex behavioural sequences, such as prey wrapping, are found in several spider
families. Closely related species in very different families (e .g. Araneidae and Dipluridae),
however, can be completely different with respect to prey-wrapping abilities. This in-
dicates that the known patterns of the feeding physiology in spiders constitute a rather
mosaic picture instead of a straight evolutionary line. Convergent evolution in this
respect is evident. Such convergences have confused spider taxonomists, who often
have the tendency to group spiders according to simple behavioural characteristics,
such as types of web building. Morphological convergences have also had a disturbing
effect on taxonomy when, e.g. the number of claws or the cheliceral structure were
used. Considering the predator-prey dynamics, it is obvious that evolutionary con-
vergence dominates insect-spider relationships. This coevolution between spiders and
their prey has hardly been discussed and future research should focus especially on this
point.
Acknowledgments. I thank Bill Fletscher and Stefan Heimer for fruitful discussions of my ideas
and the Deutsche Forschungsgemeinschaft for financial support.
IV Feeding Ecology of Winter-Active Spiders
CASSIE W. AITCHISON 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
2 Why and How Does Winter Feeding Occur? . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
2.1 Why Does Winter Feeding Occur? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
2.2 Types of Overwintering Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
2.3 Physiological Changes Associated with Winter Feeding . . . . . . . . . . . . . . . . . . 266
3 Microclimatic Conditions During Winter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
3.1 Properties of Snow Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
3.2 Overwintering Microclimates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
4 Winter Feeding by Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
4.1 Effects of Temperature and Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
4.2 Examples of Winter Feeding by Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
4.3 Examples of Winter Feeding by Other Arthropods . . . . . . . . . . . . . . . . . . . . . 270
5 What Prey is Taken in Winter? ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
5.1 Types of Winter Prey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
5.2 Prey Activity Affecting Their Capture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
5.3 Prey Size in Relation to Predator Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
6 Implications of Winter Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272
6.1 Incompatibility of Winter Feeding and Supercooling . . . . . . . . . . . . . . . . . . . 272
6.2 Biochemical Prey Composition and Possible Effects on Supercooling . . . . . . . . . . 272
6.3 Ecological Advantage of Winter Feeding . . . . . . . . . . . . . . . . . . . . . . . . . . 272
1 Introduction
It is, indeed, an amazing fact that in the cold months of winter an ectotherrnic animal,
such as a spider, can move about and even feed. In fact, biologists usually stay indoors
at this time before their fires and rarely venture outside themselves. So only a few hardy
souls have been able to observe this phenomenon of winter activity by spiders. This
chapter discusses the ''why'', "how", ''what'' and ''where'' of feeding ecology of winter-
active spiders, together with the actual micro climatic conditions encountered and the
physiological and ecological implications of this feeding.
There is not much literature on winter feeding in the field or laboratory by spiders
(Buchar 1968; Hligvar 1973; Schaefer 1976a; Aitchison 1981, 1984a), and even then,
their prey can only be surmised. Linyphiids, on the snow surface, caught and ate winter-
active collembolans (Buchar 1968; Hligvar 1973), and other known acceptable prey in-
I 614 Patricia Avenue, Winnipeg, Manitoba R3T-3A6, Canada

Feeding Ecology of Winter-Active Spiders 265
clude the dipterans (Schaefer 1976a; Aitchison 1984a). In obscure and difficult to
observe winter habitats, the actual prey of winter-active spiders is still unknown, as is
the actual amount of feeding.
To survive under cold conditions, the spider must adapt to the low temperatures by
some means, usually by supercooling (or undercooling) its haemolymph (Salt 1961;
Kirchner 1973; Duman 1979; Kirchner, this volume). However, an incompatibility exists
between supercooling and winter feeding, and this aspect will be discussed more fully
later in this chapter.
The geographic distribution of observed instances of winter feeding by spiders re-
flects the distribution of those observing this phenomenon. Most of these observations
took place in Europe (Polenec 1962; Buchar 1968; Hftgvar 1973; Schaefer 1976a;
Aitchison 1981; Gunnarsson 1983), and one in Canada (Aitchison 1984a). Winter feed-
ing by other ectothermic arthropods was reported in Europe (Kit/>rboe et al. 1982), in
Canada (MacNamara 1924; Pritchard and Scholefield 1978; Aitchison 1983) and in
Japan (Uchida and FUjita 1968).
Winter-active is defined as locomotory movement at temperatures of 2°C or lower
during winter months; winter-inactive refers to a lack of movement during that period.
Subnivean refers to conditions under snow. Winter is defined as the period of snow
cover, Le. 5 to 6 months in south-central Canada. Cold tolerance infers some physio-
logical adaptation by an organism allowing it to survive cold temperatures.
2 Why and How Does Winter Feeding Occur?
2.1 Why Does Winter Feeding Occur?
The concept of ectothermic animals being active and feeding in subzero winter condi-
tions is not widely accepted, yet some spiders and other arthropods do maintain activity
in surprisingly cold temperatures. There must be some source of energy, either intrinsic
or extrinsic, which permits the animal to continue its activity. An intrinsic energy
source would be lipid reserves in the animal, such as in Tegenaria atrica (Collatz and
Mommsen 1974a), and an extrinsic source would be derived from feeding (Aitchison
1984a).
The families Linyphiidae (including Erigoninae), Lycosidae, Clubionidae, Thomis-
idae and Tetragnathidae are active during winter under litter and snow (Polenec 1962;
K. Thaler and Steiner 1975; Schaefer 1976a, 1977a; Aitchison 1978; Flatz 1979; Punt-
scher 1979; Aitchison 1984a,b). Feeding at low temperatures has been documented
from field observations and laboratory work for various organisms, including winter-
active spiders (Polenec 1962; Buchar 1968; Higvar 1973; Schaefer 1977a; Aitchison
1984a), insects and aquatic crustaceans (MacNamara 1924; Strobing 1958; Uchida and
Fujita 1968; Pritchard and Scholefield 1978; Leinaas 1981 ; Kit/>rboe et al. 1982; Aitchi-
son 1983) and even ectothermic amphibians (Kawakami 1966).
2.2 Types of Overwintering Spiders
There are baSically two types of overwintering behaviours (see also Schaefer, this
volume). The first type is winter inactivity (possibly diapause), without movement or
266 C.W. Aitchison
feeding at low temperatures. Such animals have low metabolic rates, possibly associated
with starvation (K. Miyashita 1969b;Dondale and Legendre 1971 ; Kirchner this volume),
often exhibit low supercooling points (Salt 1961; Kirchner 1973, this volume) and
have been found in exposed areas, such as vegetation above snow cover (Kirchner 1973;
Schaefer 1976a). The second type of behaviour is winter activity with the capability of
moving and feeding at low temperatures. It is likely that such arthropods have elevated
metabolic rates (Block and Young 1978), and seek protected microhabitats for the
winter period. Spiders of this type are freezing-susceptible and have relatively high
supercooling points of-8° to -12°C (Aitchison, unpubl. data; Kirchner, this volume).
A qualitative study showed that their haemolymph contains some cryoptrotective
polyols and free sugars (Aitchison and Hegdegar 1982); however, their mobility sug-
gests instead a reliance upon thermal-hysteretic proteins as cryoprotectants (Duman
1979; Husby and Zachariassen 1980).
2.3 Physiological Changes Associated with Winter Feeding
There must be some physiological changes occurring which allow such a complex
phenomenon as feeding at subzero temperatures. Invertebrates may exhibit a two- to
four-fold increase in metabolism seen in Antarctic mites (Block and Young 1978),
which permits activity. On the other hand, there may be fewer energy requirements
and therefore less feeding, and sometimes somewhat depressed metabolic rates, as seen
in starved spiders (K. Miyashita 1969b;Dondale and Legendre 1971). The functioning
of complex enzyme systems for digestion at low temperatures is rather surprising.
Digestion will be somewhat slow, taking about 3 weeks (Aitchison 1983), and resulting
in a very high assimilation rate (up to 99%) (Moulder and Reichle 1972). Lastly, the
animal must possess other cold tolerance mechanisms to function in the cold. Low
threshold temperatures for activity were noted in winter-active spiders below which
activity ceases (H~gvar 1973; Aitchison 1978), possibly due to congealed haemolymph.
At low temperatures active spiders are unlikely to accumulate cryoprotectants, such as
polyols and free sugars, since they increase haemolymph viscosity and therefore pro-
hibit movement. Consequently, an accumulation of thermal-hysteretic proteins seems
more likely (Duman 1979; Husby and Zachariassen 1980).
The presence of food in the gut increases the possibility of spontaneous freezing
due to ice nucleators in the food (Salt 1968). Thus, anyone species may exhibit two
ranges of supercooling points during winter months, the higher range occurring in those
individuals with food in their guts and the lower range in starved animals (S</>mme and
c,l>stbye 1969; Block and Zettel 1980; Block 1982). As a consequence it is not ad-
vantageous for the survival of winter-active spiders to feed during this period, except
in warm spells. Or is it?
3 Microclimatic Conditions During Winter
3.1 Properties of Snow Cover
Snow cover, of course, varies with geographic location. As an example, the snow cover
in south-central Canada may be present for up to 6 months of the year and forms an
insulating blanket and an ecotone between two different environments: the dry, cold
and windy atmospheric air and the moist, relatively warm and still air of the subnivean
space (Aitchison 1978). The insulating capacity of the snow cover is dependent upon
density and thickness. Less dense and friable snow contains large amounts of air, provid-
ing more insulation in a given volume than would a dense snow cover. At least 20 cm
of the less dense snow cover is required to provide a stable temperature in the subnivean
microclimate. This distinct thickness of snow cover is called the hiemal threshold, be-
low which the subnivean temperatures can fluctuate widely, accompanying those of
the ambient air temperatures (Aitchison 1978). Critical times for the survival of epigeic
fauna occur just prior to the establishment of the hiemal threshold in early winter and
in cold periods after snowmelt.
3.2 Overwintering Microclimates
During the winter months, those spiders which remain active may be found in protected
micro climates , such as litter (Edgar and Loenen 1974; Schaefer 1977a) or the sub-
nivean space (Aitchison 1978, 1984b). In relatively warm (i.e. 0 °c or higher), sunny
periods these spiders (Buchar 1968; H~gvar 1973; Huhta and Viramo 1979) and winter-
active collembolans (MacNamara 1924; Uchida and Fujita 1968; Leinaas 1981), may
be found above litter or on snow cover. In periods of adverse weather when the ambient
air temperature may drop to -35°C or lower (as in Manitoba), survival above the
protecting blanket of snow cover is only possible in efficiently supercooled and im-
mobile species of invertebrates (Kirchner 1973). In freezing-susceptible groups, such
as winter-active spiders (Aitchison unpublished data), the relatively warm, still and
moist microclimate of the subnivean space permits their survival. Subnivean tempera-
tures under 20 ern or more of snow cover in Manitoba range between _3° and -6°C,
while the ambient air temperature may range between +2° and - 35 °c (Aitchison 1978,
1984b).
4 Winter Feeding by Spiders
4.1 Effects of Temperature and Light
Temperature governs winter feeding by spiders (Edgar 1971a;Aitchison 1981, 1984a).

Below a lower threshold temperature, feeding ceases, although activity may continue;
it is not known whether this threshold temperature corresponds to the ice nucleation
temperature. In the case of juvenile Pardosa lugubris, 4 °c was the lowest temperature
at which feeding in the laboratory occurred (Edgar 1971 a), while for Coelotes atrops
the minimal temperature for laboratory feeding was 2 °c (Fig. 91) (Aitchison 1981).
Winter-active spiders could easily feed down to -2°C and rarely at -5°C in Canada
(Aitchison 1984a) and to -4°C in Sweden (Gunnarson 1983), while winter-inactive
species, often in silken chambers, do not feed below 0 °c (Aitchison 1984a).
The presence of light appears to affect feeding. An experiment, comparing C. atropos
juveniles in dark and photoperiodic incubators (Aitchison 1981, unpublished data),
however, indicated that those individuals in 8L:16D at 3.5 °c showed almost a seven-
268 C.W. Aitchison
10
'-
CII
"C
.0.
III
Q;
a.
c
2
c
CII
III
~ 5
"0
....
CII
.a
E
~
,/' ,
._~. ~~.~.~=. ~.:.: :::.~. :.~.;.z. ~:g

c
c
c
CII
~
O~~~==~~~~--~----~--~----~--~
120 160 200 240 280
Days in culture
Fig. 91. The mean number of Drosophila eaten per spider per 10 days by Coelotes atropos juveniles
at different temperatures in dark incubators (Aitchison 1981)
5o 8
a::
IJl
0::
W
Il.
IJl
W
...J 4
u..
100 140
DAYS IN CULTURE
Fig. 92 Fig. 93
Fig. 92. The mean number of flies eaten per spider per 10 days by Coelotes atropos juveniles at
3.5 °C in 8L:16D incubators
Fig. 93. Web of Lepthyphantes cristatus in a deer footprint in snow (Buchar 1968)
fold increase in feeding compared to those in the dark incubators at 4 °c (Fig. 92)
(Aitchison 1981, unpubl. data). Thus, all experiments run in the dark incubators (Ait·
chison 1981, 1984a) possibly were not representative of the actual conditions in the
field. The issue is further clouded by snow cover blocking out the light. Ught penetra-
tion into snow cover to the subnivean space and spectral changes under snow vary with
the snow thickness and density; minimal penetration occurs through thick and dense
snow (Aitchison 1979a).
4.2 Examples of Winter Feeding by Spiders
There are few articles which discuss winter feeding by spiders. The most obvious ex-
amples in nature are spider webs in animals' footprints on the snow surface (Figs. 93
and 94), where winter-active spiders may feed in winter. The primary purpose of these
webs is for copulation by winter-mature spiders (Fig. 94); some spiders mature and
mate during winter (Buchar 1968; Hftgvar 1973; Schaefer 1976a; Aitchison 1984b).
The secondary purpose of the webs is for the capture and consumption of winter-
active collembolans. Generally, the smaillinyphiids, such .as Lepthyphantes cristatus
and Bolyphantes index, move about on snow surface webs at about 0 °c (Polenec
1962; Buchar 1968; H~gvar 1973). Probably the spiders only feed occasionally during
winter. Other winter-active spiders also consume small dipterans (Schaefer 1976a).
One study specifically deals with low temperature feeding by winter-active species
from Manitoba in south-central Canada. From collections of spiders extracted from
litter and taken from bush beatings and sweepings of undergrowth, the animals were
separated into winter-active and winter-inactive species. The former group contained
30 species of linyphiids, lycosids, clubionids and thomisids, while the latter group had
18 species of theridiids, salticids, thomisids, dictynids, some lycosids and an araneid.
Maintained in separate cork-sealed vials at +2 0 , - 20 and - 5 °c in dark incubators, the
spiders stayed there for up to 5 months, the usual length of winter in Manitoba. They
were fed once a week on Drosophila melanogaster or winter-active collembolans (Ait-
chison 1984a). Winter-active spiders fed most at 2 °cand less at subzero temperatures,
Fig. 94. A pair of Bolyphantes index in a web constructed over a footprint of a fox in snow (H~gvar
1973)
270 C.W. Aitchison
Fig. 9SA-C. The mean number of prey

eaten per spider per 10-day period at
(/)0'
a 5~
~ various temperatures: A at +2 °C;
Bat -2 °C; Cat -5 0c. The winter-
52
II: active spiders are represented by a
W
~ solid line, and the winter-inactive
a: 0 spiders by a dotted line (Aitchison
~ 40 SO 120 ISO 1984a)
~
~, B '2'C
II:
~ 1.0
~
II:
~
c -S'C
120
NUMBER OF DAYS
while winter-inactive species did little feeding at 2 °c and none at 0 °c, often staying
in silken chambers (Fig. 95) (Aitchison 1984a). Thus, the winter-active species can feed
even at subzero temperatures and thereby utilise a niche with little competition (Huhta
and Viramo 1979; Aitchison 1984a; Kirchner, this volume).
4.3 Examples of Winter Feeding by Other Arthropods
Feeding by other arthropods at temperatures near freezing does happen. The orthop-
teroid Grylloblatta campodei[ormis in the Canadian Rocky Mountains lives under stones
and had full guts between 0° and 5°C, feeding on collembolans and other arthropods
(Pritchard and Scholefield 1978). Winter-active collembolans may feed on pollen at
the snow surface (MacNamara 1924) or on algae on tree trunks (Uchida and Fujita
1968). In Manitoba seven species of winter-active collembolans ate fungi down to about
-2.5 °c (Aitchison 1983). The arctic planktonic cope pod Centropages hamatus had a
low rate of food intake and very efficient digestion at near 0 °c (Ki4>rboe et al. 1982).
It seems, therefore, that some feeding and efficient digestion can occur in other arthro-
pods in winter.
5 What Prey is Taken in Winter?
5.1 Types of Winter Prey
In most habitats invertebrate predators, such as spiders, feed on invertebrate prey some-
what smaller than themselves. Under warm conditions many phytophagous, nemato-
ceran dipterans are often consumed by aerial, web spiders, and collembolans and dip-
Table 22. Prey eaten (+) and uneaten (-) by spiders of various families held at 10°C after 2-week
starvation; blanks represent untested combinations (Aitchison 1984a)
Prey Predator
Linyphiidae: Erigoninae Lycosidae Thomisidae
Collembola + + +
Aphidina +
Cicadellidae + +
Nabidae
Other hemipterans + +
Orthoptera +
Lepidoptera + +
Hymenoptera
Diptera + + +
Coleoptera +
terans form a considerable part of the diet of epigeous spiders (Moulder and Reichle
1972; Schaefer 1976a;Nyffeler and Benz 1979 ; Nentwig 1980, 1982c; Aitchison 1984a;
Nentwig, this volume). In the subnivean micro climatic and other severe environments,
both collembolans and nematoceran dipterans are winter-active, with the former insects
more abundant (Aitchison 1979a,b; Otto and Svensson 1982; Aitchison 1984a). They
thus provide a probable basis of a subnivean food chain (Aitchison 1984c). In northern
Europe winter-active collembolans constitute some of the prey of winter-active spiders
(Buchar 1968; Higvar 1973).
When individuals of three spider families had free access to prey extracted from
autumn litter, they all consumed collembolans and dipterans at 10 0 C. Larger individuals
of spiders also fed on aphids, cicadellids, other hemipterans and lepidopteran larvae
(Aitchison 1984a). Rejected prey included nabid hernipterans and unpalatable onychi-
urid collembolans (Table 22), findings corroborated by other researchers (Aitchison
1984a).
5.2 Prey Activity Affecting Their Capture
Winter activity by potential prey undoubtedly forces them to become noticeable to

predators and therefore captured. Nentwig (1982c) stated that the frequency of activity
by epigeic spiders was significantly correlated with that of their potential prey, i.e.
active dipterans and collembolans. Considering the constraints of the two-dimensional
subnivean space, winter-active collembolans appear to be the most probable prey of
winter-active spiders (Aitchison 1983, 1984a,c).
5.3 Prey Size in Relation to Predator Size
Correlations between sizes of prey and predator in warm periods have been documented
(Nentwig 1980, 1982c, this volume). Most spider prey has a mean length of 1.5 mm
272 C.W. Aitchison
(Moulder and Reichle 1972; Nentwig 1980, 1982c). The extremely numerous small
spiders, many of which are winter-active, have a readily available food source in the
epigeous populations of small collembolans (Aitchison 1983, 1984a). The ideal prey
body size for a predatoris 0.25 to 0.75 times the body length of the predator (Nentwig
1982c;Nentwig and Wissel 1986), indicating the collembolans for winter-active linyphi-
ids. Conversely, larger spiders would have fewer, larger winter-active prey available,
since most of the winter-active invertebrate fauna is small (5 mm or less) (Aitchison
1978, 1979a,b, 1983, 1984a-c).
6 Implications of Winter Feeding
6.1 Incompatability of Winter Feeding and Supercooling
Those winter-active spiders and other arthropods which feed near or below 0 °c gamble
on their chances of being able to supercool and survive temperatures of - 8°C or less
(Salt 1968; Block and Zettel 1980; Kirchner, this volume). However, subnivean tempe-
ratures rarely drop below - 8°C in cold Manitoba, except under extremely cold condi-
tions.
Nonetheless, these animals are freezing-susceptible and must somehow acquire
energy if they are to remain active. The likelihood offeeding is fairly low at -2°C (i.e.
winter-active spiders fed in the laboratory on a mean of 0.24 ± 0.34 flies/spider/lOd)
(Aitchison 1984a), and their digestion probably would be extremely efficient (about
99%) (Moulder and Reichle 1972; Kic1>rboe et al. 1982). Thus, it appears that winter
feeding is an efficient strategy, and winter-active spiders would gain some energy in
this way. However, by staying in the relatively warm, subnivean space, these animals
do not expose themselves to the climatic extremes and, therefore, do not risk being at
a disadvantage by feeding.
6.2 Biochemical Prey Composition and Possible Effects on Supercooling
Another aspect to consider is the biochemical composition of the ingested prey. Most
likely the most abundant and active prey available are winter-active collembolans (Ait-
chison 1983, 1984a,c), which also contain some cryoprotectants (Aitchison and Heg-
de gar 1982). Since spiders feed by means of external digestion and suck their prey's
juices, they probably ingest few, if any, dust particles on which spontaneous ice nu-
cleation could occur. POSSibly the cryoprotectants in the haemolymph of the prey also
confer some advantage to the predator, providing that their digestive enzymes do not
completely denature the cryoprotectants of the prey. Certainly this aspect may explain
how winter-active spiders may feed and merits further study.
6.3 Ecological Advantage of Winter Feeding
Those spiders displaying winter activity have the ecological advantage of little competi-
tion from other spiders or other predaceous invertebrates for food (Huhta and Vi ramo
1979; Aitchison 1984a,b; Kirchner, this volume; Schaefer, this volume). Probably the
niche of winter invertebrate predation is not filled fully in northern climates.
Winter-active spiders must feed occaSionally in order to obtain energy resources, espe-
cially if their lipid reserves were inadequate to sustain activity or nonexistent at the
start of winter. In order to allow winter activity, they must have undergone some
physiological changes, such as accelerated metabolic rates, cold-adapted digestive
enzymes and the acquisition of thermal-hysteretic cryoprotectants. Thick and light-
weight snow cover provides maximum protection against rigorous fluctuations of
atmospheric air temperature conditions, with the usual subnivean temperatures be-
tween _3 0 and -6 °c in southern Manitoba (January mean temperature of -19 °C),
where overwintering by winter-active invertebrates takes place in the subnivean space.
Both temperature and light affect winter feeding. Subzero temperatures usually in-
hibit feeding, while the presence of light enhances it. However, under thick snow cover
of 20 em or more, little light penetrates to the subnivean space. The most obvious ex-
amples of winter feeding by various spiders and insects are seen on the snow surface.
Laboratory experiments have also ascertained that some feeding by winter-active spi-
ders and collembolans may occur at -2 °c and some even at -S °C. The most probable
prey are the small and numerous winter-active collembolans. The consumption of
winter-active prey might possibly even confer a biochemical advantage to the winter-
active predators.
V Kleptobiosis in Spiders
FRITZ VOLLRATH 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
2 The Spider Web as a Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
3 Examples: Temporary Visitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
4 Males as Kleptos, and Kleptos Feeding on Silk . . . . . . . . . . . . . . . . . . . . . . . . . . 276
5 Examples: Spiders as Tenants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
5.1 Kleptos in Diplurid Funnel Webs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
5.2 Argyrodes in N ephilt.z Orb Webs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
6 A Host's Reaction to Its Inquilines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
7 Social Spiders as a Society of (and with) Kleptos . . . . . . . . . . . . . . . . . . . . . . . . 285
8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
1 Introduction
To a predator the pursuit of prey can be costly in time and reserves, or risky because
injuries may occur. Spiders are predators and have evolved a multitude of different
hunting methods (Bristowe 1941). We can assume that these foraging strategies, as dif-
ferent as they may be, all have in common that these costs are cut while calory intake
is optimized (Schoener 1971). Whatever the method of hunting, potential prey has to
be: awaited, spotted, identified, attacked, killed, transported to some convenient re-
treat, and, before consumption, pumped full with digestive enzymes. In addition some
spiders invest heavily in the material and construction of a web (Prestwich 1977). All
of these activities incur some costs for the predator, which can be measured in time,
risk or calories. Spiders rarely encounter conditions of prey superabundance, generally
food is a limiting resource. If it is not prey, then structural support becomes the limit-
ing resource for web spiders (Riechert and Luczack 1982; Vollrath, this volume).
Clearly a method of foraging that reduces these costs and overcomes these resource
limitations should confer large advantages to its user (Curio 1976; Barnard 1984;
Parker 1984). Stealing an insect that has been converted into prey on someone elses
account, and which for good measure may even have been predigested, is more eco-
nomical than catching this insect (Vollrath 1980). Since a resource is stolen, the prefix
"klepto" has been coined to describe such inter- or intraspecific interference competi-
tion. Stealing prey and usurping webs allows dense packing of conspecifics or coexisting
species in a habitat where the number of web sites, rather than prey, are the limiting
1 Department of Zoology, South Parks Road, Oxford OX1 3PS, Great Britain
Kleptobiosis in Spiders 275
resource. In this chapter I will discuss examples of close symbiotic interactions be-
tween spiders, mainly those where a stealthy guest benefits from its association with a
host, but also a few examples where aggressive contests occur over prey or a web.
Kleptobionts fall into a number of categories (Vollrath 1984), most notably the
pirates that take the coveted resource by force, and the pilferers that succeed owing
to their stealth. Spiders provide many examples, both as hosts and as guests; some will
be examined in this chapter. The relationship of insects with spiders is discussed in
greater detail by Nentwig and Heimer (this volume).
2 The Spider Web as a Habitat

Nearly all observations in which spiders are implicated as hosts concern web-building
spiders. This might reflect our ignorance of the interspecific interactions of hunting
spiders. However, it is much more likely that kleptobiosis is a more common occur-
rence among web-building spiders, pardy because of the potential hosts' relative im-
mobility, which makes them "sitting targets", and partly because of the constraints
and costs that always accompany a feeding strategy involving a web (Peakall and Witt
1976; Prestwich 1977). A spider web generally covers a large area in comparison to the
owner's body size. This makes the web difficult to patrol effectively against stealthy
intruders. Kleptobiotic interactions occur mainly between web-building spiders,
presumably because of the preadaptions ofthe potential guests. Specialized feet allow
them to move on threads (Foelix 1982), an idiothetic "memory" enables them to
orient in a tangle of threads (Gomer and Claas 1985; Mittelstaedt 1985; Vollrath, un-
publ.) and specialized sensory perception allows them to take full advantage of the in-
formation content of a web, its vibrations (Barth 1982, 1985; Masters et al. 1986).
However, not all kleptobiotic interactions among spiders are between web spiders. A
remarkable exception is the highly adapted thomisid Misumenops nepenticola which
lives kleptoparasitically in the carnivorous leaves of the pitcher plant Nepentes gracilis
(Bristowe 1941).
Foraging decisions guide spiders, like other animals, to remain in a profitable loca-
tion (Riechert and Luczack 1982). This often results in the accumulation of webs in a
favourable patch. Thus, the presence of one or, better still, several webs constitute a
reliable indication of site quality _A spider in search of a good feeding site could do
worse than to join other spiders, rather like birds that are attracted to a feeding flock
(Krebs 1974). Moreover, one web can also provide structural support for another web
(Buskirk 1975; Rypstra 1979). A spider web is primarily a trap to catch insects, but it
also provides an environment in which the owner/occupant can move about in relative
safety, since few enemies can approach or enter the web undetected (Bristowe 1941).
Thus, a web provides support that, in contrast to the twigs, blades of grass or leaves, is
free of ants, pompilids, mutillids, hunting spiders and other ambulatory predators. It
seems that joining an established web would be a good feeding strategy, if a safe distance
to the potentially dangerous occupant can be kept, and its attentions avoided. Taking
advantage of existing webs by adding yet another one might be beneficial to all spiders
involved by creating a "selfish herd" (Hamilton 1971), if prey is not the limiting factor.
In the peripheral strands of large orb webs one may find minute webs of first or
second instar spiderlings. These spiderlings generally rely on their own webs to catch
276 F. Vollrath
insects, but they may also venture into the orb of the host, there to remove or feed on
trapped prey. Nyffeler and Benz (1980) observed three species of juvenile Nuctenea
foraging in webs of conspecific adults. The proportion of parasitized webs differed
significantly between habitats, and ranged from 8 to 18%. Little interaction occurred
between the guests and the host, since the night-active hosts were in their retreats when
the day-active guests roamed in their webs. A number of araneid species tend to form
relatively large aggregations of webs, where the constituent spiders can belong to any
instar (Blanke 1972; Lubin 1974; Buskirk 1975). Within such clusters of webs the in-
dividual spider is frequently attracted to vibrations emanating from a neighbouring
web (Buskirk 1975). This may result in its invading the successful web, and either a
stealthy theft, or, more often, a contest which generally ends with victory of the larger
spider. Such invasion and ensuing contests are generally quite stereotyped. For an ex-
cellent review of this subject, see Riechert (1982).
3 Examples: Temporary Visitors
Inquilines can be opportunistic pilferers which may turn pirates, or prey for that mat-
ter, if detected by the host (Nentwig and Heimer, this volume). Into this category fall
a number of Panorpa species (Thornhill 1975; Nyffeler and Benz 1980), the waspMi-
schocyttarus drewseni (Jeanne 1972), and the tropical damselflies Megaloprepus and
Mecistogaster (Fincke 1984). The Costa Rica hummingbird Phaethornis longirostris
not only steals prey from the webs of NephilfI clfIvipes, but also removes web material
(Young 1971) (Fig. 97). Dipso-parasitic kleptos, like the nematoceran cecidomyiid
Didactylomyia longimana or the dipteran milichiid Paramyia nitiens, constitute no
danger to their hosts, being far too small (Sivinsky and Stowe 1980; Robinson and
Robinson 1977). They share the host's prey, a behaviour which can also be observed
in some symbiontic spiders (discussed in some detail later) and in the males of some
species.
4 Males as Kleptos, and Kleptos Feeding on Silk
The large web of the NephilfI female often contains more than one male (Robinson
and Robinson 1973a; Christenson and Goist 1979; Vollrath 1980). In this case one
male only is found on the hub, attending the female and sharing her food. The others
occupy the periphery of the barrier web, where they may feed (1) on prey stolen
from Argyrodes kleptoparasites (which have stolen it from the female); (2) on small
insects neglected by the female (which they steal from the web); or (3) on silk (Voll-
rath 1977). The silk, a protein is itself nutritious (Peakall 1971), and in addition
is often "peppered" with the aerial nekton of pollen grains and spores, which adhere
to the strands (Vollrath 1977, 1978) and provide supplementary nutrients to the spi-
ders (Smith and Mommsen 1984) (Fig. 60c).
Other inquilines may also steal just 'web and nekton'. I have seen Argyrodes elevatus
systematically take down and ingest postcard-sized areas of NephilfI's capture web. In-
deed, it is possible that the tiny, early instars of Argyrodes feed exclusively on web
material (Vollrath 1977). On several occasions I have observed an Argyrodes caudatus

female to invade the orb of a small araneid, cut the supporting strands, roll the web
into a small ball and then ingest it (Vollrath 1977). The theft of web material is not
surprising in view of the web recycling behaviour of many araneids (Peakall and Witt
1976), which take care not to discard with an obsolete web the calories it contains.
5 Examples: Spiders as Tenants
The spider web is a very special structure in terms of material, configuration and in-
formation transmission. It is not surprising that web spiders are the most successful in-
quilines in spider webs (Brignoli 1966), since they are superbly preadapted. In this
chapter I will introduce several case studies of spiders that are obligatory kleptopara-
sites, and discuss the behavioural, ecological and physiological adaptations enabling
these kleptos to be so successful. The highly specialized and obligatory kleptoparasites,
like Argyrodes argyrodes in the webs of Cyrtophora citricola (Wiehle 1928; Kullmann
1959a) or Argyrodes elevatus in the webs of Nephila clavipes (see below), have presum-
ably evolved from ancestors, like Achaearanea tepidariorum, which may pay the odd
fleeting visit to the C. citricola web, stealing prey or attacking the feeding host (Kull-
mann 1959b, 1960).
5.1 Kleptos in Diplurid Funnel Webs
A highly specialized kleptoparasite is Curimagua bayano, a small symphytognathid

that may inhabit the funnel webs of the large mygalomorph Diplura sp. in Panama
(Vollrath 1978) and the upper Amazon (unpubl. observation). Curimagua either climbs
about in the threads inside the funnel, or it mounts the host and perches on its cepha-
lothorax, riding the host even during prey-capture sorties. After a prey insect has been
thoroughly masticated and enveloped in digestive fluids by Diplura, Curimagua descends
one of the host's mighty chelicerae and partakes of the meal, very rapidly imbibing
some of the liqUidized prey (Fig. 96). Curimagua bayano, like other symphytognathids
(Forster and Platnick 1977), has chelicerae which are fused at the base and there-
fore cannot open. These chelicerae are, moreover, recessed deeply between the rela-
tively massive clypeus and the palpal coxae, the palps themselves being absent (a synapo-
morph character of all female symphytognathids). The mouth region of Curimagua
bayano is "muzzeled" by large, sensory hairs. The anatomy of the mouth makes it
highly unlikely that Curimagua bayano is able to capture, hold or masticate its own
prey. Thus, it is totally dependent on it's host, and a good example of a highly spe-
cialized kleptoparasi te.
Curimagua shares the Diplura web with other symphiles, the most conspicuous of
these being mysmenid spiders. In Panama three species of mysmenids (in order of as-
cending size: Mysmenopsis ischnamigo, M. gamboa andM. dipluroamigo) are found in
the webs of three species of diplurids (Ischnothele guyanensis and two Diplura sp.).
The largest host harbours all three mysmenid species, the medium-sized host the two
smaller species, and the smallest host harbours the smallest species (Kirkendall and
Vollrath, unpubl. observations). The mysmenids are alerted by the host's prey capture,
278 F. Vollrath
...... '
.. '
:'
\ .. ,
'. \
i ' I
'iJ..."",'"
Fig. 96. The symphytognathid Curimagua bayano is found in the webs and on the back of Diplura
sp. The kleptos climb over the mighty chelicerae of the host to partake at its meal, and they move
over its eyes without eliciting a reaction. The mouth region of C. bayano suggests a high degree of
specialization, and independent feeding is presumably no longer possible
db
Fig. 97. The web of Panamanian Nephila clavipes provides food and a habitat for a large number of
guests. The hummingbird PlU1ethornis iongirostris visits briefly and takes small prey and silk. The
theridiid Argyrodes elevatus is a permanent resident and a highly specialized thief that may share
the host's meal or steal entire prey packets. Its congeneric A. caudatus is less specialized and takes
small insects from the capture area of the host'sweb. Milichid flies may use the host's cephalothorax
as a platform to move onto its prey
280 F. Vollrath
and are attracted to the host when it feeds (Kirkendall, pers. commun.). Mysmenopsis,
similar to Curimagua, shares the host's meal. Unlike Curimagua (who even moves over
one of the diplurid's eyes without eliciting a reaction), the mysmenids regularly evoke
the host's attention, which reaction frequently rewraps its prey and moves up and
down in its tunnel. Also, unlike Curimagua, the mysmenids patrol the'capture area of
the web, and prey on small insects, which are below the host's threshold of reaction.
S.2 Argyrodes in Nephila Orb Webs
Stealthily moving about in the host web, feeding on small neglected insects or sharing
the host's prey, is a nearly universal feature of most Argyrodes species (Archer 1946;
Exline and Levi 1962). In the following paragraph I will describe the behaviour and
life history of two sympatric species of Argyrodes which I studied in Panama (Vollrath
1976, 1977, 1979a,b, 1981, 1984). Argyrodes elevatus is a medium-sized theridiid
with extremely long front legs, and a rotund or conical silver abdomen. Argyrodes cau-
datus is of equal size, but its abdomen is dull brown and irregularly cylindrical. In the
resting pOSition A. caudatus pulls its legs over its cephalothorax, and is extremely cryp-
tic. A. elevatus rests with the legs stretched out, and is highly visible, resembling a drop
of mercury. A. elevatus is a specialized klepto found rarely in the webs of Argiope
argentata, but regularly in those of Nephi/a c/avipes. A. caudatus is a generalist klepto
found in the webs of a variety of hosts. It is also a regular predator on spider egg sacs
(which it opens), early spiderlings (which it joins in their early communal web) and
small spiders (which it attacks in their own webs). A. caudatus can be found on its
own small web far from any host;A. elevatus is always found in the web of a host spi-
der.
In Panama the main host for Argyrodes is the large orb web-building spider Nephi/a
c/avipes (Fig. 97). A variety of factors make the Nephi/a web a particularly favourable
habitat for Argyrodes. The web is a relatively permanent fixture and has an extensive
three-dimensional barrier web, which provides ample shelter with a large carrying
capacity. The maximum number ofA. elevatus I observed in one web was 45 individuals,
although the average varied with season, population density and habitat type. The
mesh of the Nephi/a web is very fme, and traps even the smallest insects (Nentwig
1985b). The capture area of the orb is not only very large, but also extremely stiff,
and has diverging radii rendering it difficult to monitor. Adult Argyrodes are found
almost exclusively in the webs of sub adult and adult Nephi/a, suggesting that a positive
correlation exists between host or web size, and number and size of the guests (Vollrath,
this volume).
Both species of Argyrodes coexist in the webs of Nephi/a; in addition, I frequently
recorded A. atopus and A. dracus. The two principal species coexist, but apparently
have evolved a variety of mechanisms to avoid competition in the extremely compact
habitat provided by the web. However, it might be argued that flexibility in behaviour
as a response to encroachment by a potential competitor is indicative of past and pre-
sent selection pressures. As a generalistA. caudatus maintains a basic population density
in the biotope, and is thus able to colonize the webs of Nephi/a as they become avail-
able. In contrast, the specialist A. elevatus lags behind in its tracking of the host popu-
lation (Fig. 98), presumably because at this stage in its cycle the resident A. elevatus
# of spiders
300
100
50
10
Jan Feb Mar Apr May Jun Jut Aug Sep Oct
Fig. 98. Population profile of adult females of the host Nephiia ciavipes, and of the two klepto spe-
cies Argyrodes elevatus and A . caudatus at Cerro Galera, Panama (1978/79)
population relies on births rather than on immigration. A. elevatus builds up its popu-
lation speedily, and eventually even seems to depress the population of A. caudatus,
which has also increased considerably, starting earlier and at a higher baseline. The
faster increase of A. elevatus is probably due to its faster reproductive output, which
suggests that it is the more "r-selected" species, whereas the lower output of A. cauda-
tus suggests, together with other features, that it is more 'K-selected' (Pianka 1970).
A. elevatus's foraging behaviour predisposes it (see below) to breed rapidly: a single
meal can provide enough calories to produce an egg case, which is left in the host web.
A. caudatus has to forage longer to accumulate the necessary reserves, and it constructs
the egg case in a small web away from any host guarding it until the young hatch
after 20 days . A . elevatus lays more eggs per batch than A. caudatus (about 37 as op-
posed to 28), and can produce one batch every 5 days as opposed to every 30 days;
thus it presumably produces many more egg sacs throughout its lifetime. The depres-
sion in the population of A. caudatus when the A. elevatus population has 'caught up',
suggests that the increasing density of A. elevatus interferes with the foraging behaviour
of A. caudatus.
The foraging behaviour of the two species is quite different. In the Nephila web, the
adult females of A. caudatus feed in the capture spiral, removing small prey items
that are neglected by the host. The adult females of A . elevatus may forage similarly,
but more important, for host and guest alike, is their other foraging strategy of stealing
large prey items that have already been dealt with by the host. The success ratio of
such raids can be very high. In 200 experiments two prey items were presented to the
host IS min apart: 54% of the first prey were stolen by A. elevatus, and 32% were not
found; 12% were shared with N ephila and only in 2% of the raids was a theft prevented
by the host. Thus, the high stealing success can be attributed mainly to A. elevatus's
ability to 'read' and 'interpret' the host's behaviour, and to adjust its raids accordingly
282 F. Vollrath
Fig. 99. Recorded web vibrations generated by the activities of Nephikz: at rest, running towards
prey, catching the prey ,returning to the hub, wrapping the prey and drinking. (After Vollrath 1979b).
Argyrodes seems to perceive these vibrations, and times its raids accordingly
(Fig. 99). A. elevatus's stealing behaviour improves with experience, suggesting some
level of learning in order to adjust raiding behaviour to the host's prey capture activi-
ties.
Several ecological and behavioural measurements confirm the hypothesis of compe-
tition and character displacement between the two Argyrodes species. When they co-
habit, A. elevatus is day-active and A. caudatus is night-active (Fig. lOOa). The cryptic
colouration and posture of A. caudatus is no small advantage for this behaviour, since
daytime predation is high, especially in the periphery of the web . The silver coloura-
tion of A. elevatus is presumably an adaptation to physiological constraints associated
with its specialized life -style (as we will see later). When representatives of both species
coexist in the same web, they seem to confine themselves to different parts of it. When
alone, each species has a slightly different active radius (Fig. lOOb). This can be at-
tributed to competitive displacement in the foraging activities of both species. A. ele-
vatus keeps itself closer to the host because it largely depends on the host for informa-
tion and prey. A. caudatus forages in the web, and rests in the periphery; it does not
seek the proximity of the host. The periphery of the web is more exposed to predators,
like Trypoxylon wasps, a Phaethornis hummingbird and Mecistogaster damselflies.
Thus, when alone, A. caudatus tends to stay towards the inner regions of the web. The
centre of the web is, of course, occupied by Nephi/a, and a certain degree of activity
and danger emanates from here . Being very close to the host is presumably not without
danger, and when alone A. elevatus stays farther away from Nephi/a .
A.elevatus A.caudatus
0: 302 0; ... 17
>-
<
o
T
80
'# 80
0 A
40
>-
t: 20
>
t- 20
()
< 40
60
80
100
t-
'"
J:
~
Z
a
Oil l once 10 hos t Argyrodea caudatua
t
25
Ar gyrodes elevatus
20
15
alone
t
t.
A.cauda'us
alon8
Fig. 100. a Day-night activity of synoecious Argyrodes in Nephila webs. The observations for the
circular diagrams were made during a number of visits, when the activity for each individual was
recorded once (small dots: walking activity ; large dots: resting; empty: alert position). The histo-
grams represent 24-h observations of two adult females for each species, showing the proportion
in which the different parts of the host's web were used by the two species (T total walking activ-
ity; A Argyrodes thread; 0 orb threads; B barrier threads). (After Vollrath 1976). b Spatial displace-
ment of both Argyrodes species when cohabiting and not cohabiting. The symbols represent the
average distance to the host Nephila (filled: resting; open: active Argyrodes). The standard devia-
tion of the samples is given by the vertical bars. (After Vollrath 1984)
284 F. Vollrath
The silvery colour of the abdomen of A. elevatus is due to light reflected by lacunae
of air trapped between the inner and outer layers of its cuticle. This presumably pre-
vents overheating of the animal in direct sunlight (see also Holl, this volume). It is a
morphological adaptation that allows A. elevatus to colonize Nephila webs in exposed
locations, where A. caudatus is conspicuously absent. The observation that other spe-
cialized Argyrodes kleptoparasites are very similar in this respect supports this hypo-
thesis, A. nephila inhabits Nephila webs in southern Florida, andA. gibbosus is found
in the webs of Cyrtophora citricola in the Mediterranean. Both Argyrodes species have
bright silver abdomens and are also often found in exposed host webs. The hosts have
a considerably larger body mass, and are thus less succeptible to overheating and desic-
cation. In exposed webs during the dry season, A. elevatus becomes night-active,
presumably to reduce its metabolism and the associated water loss via its lung, rather
than tracheal, respiration (Davies and Edney 1952; Miyashita 1969b).
6 A Host's Reaction to Its Inquilines
Clearly, the many facets of kleptobiotic interactions cannot be summarized into a

simple formula describing the degree of harm done to their hosts. To justify the name
"parasitic", the symbiosis must be beneficial for the guest and harmful for the host;
the guest, moreover, must show behavioural, morphological and/or physiological adap-
tations to its life-style (Price 1977; Vollrath 1984). The harm to the host can be very
large when pirating is involved; in fact, for a "host" these interactions may have the
same effect as failed attempts of predation. Pilfering theft often involves removing
small prey items which are neglected by the host; they may have been undetected, be-
low the host's profitability threshold or too risky to be retrieved.
All three possibilities could explain the kleptobiotic interaction of the juvenile Nuc-
tenea stealing insects from the webs of mature conspecifics. The spiderlings forage in
the host webs during the day, when the night-active females sit in removed retreats
(Nyffeler 1980). Even if the inquilines are detected by the host, the energy expended
in chasing them may be higher than the caloric contents of the stolen prey.
Nephila clavipes is often alerted to the activities of the Argyrodes in its web. Some-
times the host may chase the klepto, but during many hundreds of hours of observa-
tion I have never seen a pursuit ending in capture. Indeed, my impression is that in-
experienced Nephila react more often towards the kleptos than Nephila that had been
exposed to them for some time. The activities of the thieves inevitably decrease the
host's caloric intake, and large numbers of kleptos may decrease it significantly (Voll-
rath 1980). Rypstra (1981) observed that the number of Argyrodes in the web of
Nephila clavipes is correlated with the number of prey items taken, and that hosts with
many kleptos are more likely to leave a site. Robinson and Robinson (1973a) observed
a similar phenomenon with Nephila maculata, as did Wise and Larcher with Neriene
radiata and Metepeira labyrinthea (Wise 1982; Larcher and Wise 1985). A change of
site results in a reduction (at least for awhile) in the number of associated Argyrodes
(Robinson and Robinson 1973a). One might assume that it is the decrease in the
host's feeding rate, rather than the activities of the kleptos per se, that triggers a
move (Rypstra 1981). On the other hand, evidence suggests that hosts may leave webs
Fig. 101. Effect of Argyrodes trigonum on

Metepeira labyrinth ea. The survivorship
curves show that the initial population of
o
hosts (experimental: 30/32, control: 27/31)
declines more rapidly if kleptos are intro-
o
duced. (After Wise 1982)
""o
o
~ .4
Ii o NO!::. TR IGONUM
!C
:i1 • !::. 'TR IGONUM ADDED
0.. .2
•
12 16
DAYS
simply because of the presence of kleptos. Larcher and Wise (1985) found that sup-
plementing prey did not alter the rate of host emigration.
The risks of a move can be considerable, through predation and because of the un-
certainty of finding a comparable or better web site (Vollrath, this volume). Thus, the
kleptos can be doubly harmful; on the one hand, by steady removal of small quantities
of food; on the other hand, by influencing the host to leave. Wise (1982), during an
experimental study of the effect of Argyrodes trigonum on the labyrinth spider Mete-
peira labyrinthea, observed that the activities of the kleptos had a marked effect on
the host's population density (Fig. 101). This was due to host emigration from the
site, as well as host mortality . Apart from the direct or indirect harm done by the klep-
tos' stealing behaviour, they sometimes cause immediate death to their hosts (Smith-
Trail 1980; Tanaka 1984; Larcher and Wise 1985), thus behaving like predators or
parasitoids rather than parasites.
During mOUlting, a spider is extremely vulnerable. The actual period of shedding
the old skin, inflating the new one and its preliminary hardening lasts for about 30 min
in Nephila (unpubl. observations). During this time, flying predators take full advantage
of the helpless state of the spider (unpubl. observations). Kleptos also attack the moult-
ing host. Tanaka (1984) observed that the kleptoparasite Argyrodes fissifrons preys
heavily on moulting individuals of its host Agelena lim ba ta : 75% of the studied popu-
lation of Japanese Agelena were parasitized; 32% were killed at any of the six moults
and, in all, 18% were thus killed.
7 Social Spiders as a Society of (and with) Kleptos
Some spiders are not merely gregarious, but definitely social (Buskirk 1981). Colonies
of social species may consist of very large numbers of spiders of all stages and several
generations, inhabiting a communal web which is constructed and maintained by joint
effort. Social spiders cooperate during prey capture and consume prey items together.
These seemingly cooperative societies are, however, not as egalitarian as appears at first
286 F. Vollrath: Kleptobiosis in Spiders
sight. In Anelosimus eximius many spiders may cooperate during the capture of an in-
sect, which often involves an energetic chase and risky attack (Christenson 1984). Yet
other spiders, which have not participated during the hunt, will participate in prey
consumption, sometimes even to the exclusion of those who captured the prey (Voll-
rath and Rohde-Arndt 1983). Joining a feeding aggregation at a latter stage may also
be advantageous since it allows the late arrivers to feed on the prey without the need
to inject digestive enzymes (Ward and Enders 1985). These enzymes are highly concen-
trated and accordingly quite expensive to produce (Momrnsen 1977; Collatz, this
volume).
Some females of Anelosimus eximius prevent others from accumulating enough re-
serves to produce eggs (Vollrath 1986a). The division of the females of one colony
into 'reproductives' and 'helpers' could be classified as a borderline case ofkleptobiosis,
because those who invested in the capture of prey do not benefit directly from it, and
are apparently harmed. However, the genetic complement ofthe helpers does not suf-
fer from their role, since all spiders of a colony are likely to be closely related (Smith
1986). Thus, the inclusive fitness of the apparent victims is probably enhanced rather
than diminished, because the division of labour in Anelosimus eximius is thought to
increase the overall productivitiy of a colony (Vollrath 1986a,c).
The nests of social spiders often harbour a large number of symbionts (pers. observa-
tions; Nentwig and Heimer, this volume). The exact role of any of these inquilines is
far from clear. Minute mirid hemipterans (Ranzovius) share food withA. eximius, but
I have also seen them feeding on dead A. eximius. Caterpillars of the noctuid moth
Neopaltis madata feed on the chitinous remnants of old prey insects (Robinson 1978).
Mud-dauber wasps (Sceliphron) sometimes build their brood cells inside A. eximius
nests (which they fill not with Anelosimus, but with assorted araneids). Ants may act
as pirates either evicting the spiders from their nest or stealing individual prey packets
(see Table 19). Spiders symbiotic in the colonies may catch insects which escaped
the hosts' attention or prey on their hosts. Among these spiders I found representatives
of the lycosids, salticids, oxyopids, scytodids, caponiids, dysderids, pholcids, theri-
diids, mysmenids, tetragnathids, uloborids and araneids.
8 Conclusion
Any animal that controls a resource is a potential victim for animals contesting this
resource. If the animal is per se relatively immobile, like the web spider, then pirating
and pilfering can evolve as conditional or obligatory strategies which may considerably
reduce the costs of prey acquisition for its users. The combination of behavioural,
ecological, morphological and physiological adaptations shown by Argyrodes elevatus
demonstrates the high degree of specialization that may accompany the evolution
from an opportunistic and relatively harmless guest to an obligatory parasite.
Acknowledgments. For their hospitality, generous assitance and intellectual stimulation I thank
the director and staff of the Smithsonian Tropical Research Institute, D. Windsor and M. Robinson
(who got me lust interested in kleptoparasites and spiders). C. Bromhall, M. Ridley and D. Wise
kindly read the manuscript, suggesting many improvements. David Baily did the drawings in Figs. 96
and 97.
VI Hunger and Starvation
KAZUO NAKAMURA 1
CONTENTS
1 H u n g e r . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
2 Starvation............................................... 292
3 Growth and Reproduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
1 Hunger
For animals the degree of hunger is one of the basic factors determining the feeding
occurrence and the amount of food they ingest. This is especially true for spiders, be-
cause the food ingested remains in the gut for a very long period, giving rise to a wide
range of hunger levels (Nakamura 1977).
Nakamura (1972) supplied Drosophila flies to wolf spiders Pardosa laura which were
deprived of food several days and then counted the number of flies they killed. When
the cumulative number of flies killed was plotted against the time after the start of
food supply, a saturationlike curve was obtained (Fig. 102). This suggests that the gut
of the spider was gradually filled with food until a saturation point was reached. If this
is true, we can conclude that the amount of food ingested is roughly determined by
the capacity of the gut which is not filled with food. It is then possible to use this
quantity as a representation of the degree of hunger.
Nakamura (1977), using P. laura, examined the effect of a fasting period on the
amount of food ingested. They were first supplied by unlimited food over 24 h, were
then deprived of food during varying days, and finally resupplied with unlimited food.
The amount of food ingested after deprivation was compared with the amount before
it, by determining the ratio of both amounts (Fig. 103). The ratio increased like a satu-
ration curve with an increase in fasting periods (Fig. 103). Since the food remaining in
the gut decreases due to egestion and assimilation during fasting, the available space in
the gut for food increases with an increase in the fasting period, until it reaches the
maximum, i.e. equivalent to the capacity of the gut. Therefore, the curve obtained in
Fig. 103 shows approximately the unfilled capacity of the gut expressed in terms of
the amount of food, or the degree of hunger, at certain fasting periods.
However, this curve does not exactly represent the unfilled capacity of the gut. This
is because food disappears continuously from the gut during feeding as well as during
1 Bird Control Lab., National Agriculture Research Center, Yatabe, Tsukuba Science City, Ibaraki,
305 Japan
288 K. Nakamura
Q 20
w
...J
...J
""
'"w
~ 10
u..
o
o
z: ~ - -------------~
_e--------- _.. ---...... I'IALE
°0~~~---1~0--~---2~0--~--~3~0--~---~~0~h~
Fig. 102. The mean cumulative number of flies captured by nymphs, female adults and male adults
of Pardosa laura when food was supplied after deprivation (Nakamura 1972)
fasting, and then the amount ingested after fasting is always higher than the unfilled
capacity of the gut. Therefore , we need to estimate the rate of disappearance of food,
together with the rate of feeding, in order to obtain the degree of hunger.
Writing V for the capacity of the gut expressed in terms of the weight of food and
VR for the amount of food remaining in the gut at time t, we have V- VR for the avail-
able capacity for additional food . Thus, the degree of hunger can be expressed as
(V- VR)/V(Nakamura 1972).
VR equals the difference between the amount ingested until time t, [, and the
amount of food disappearing from the gut during the same period, p . Therefore, the
FEMALE
::
Q
'" 1.0
w
Q.
'"-
Z
....
'"-< 0.5
u..
:J::
U
-<
w
0
'"....w 0 5 10
u..
-<
Q
I'IALE
....w
'" 1.0
w e
::'" e
u..
0
....
z 0.5
:0
0
:E
-<
Fig. 103. The amount ingested after differ-
w ing fasting periods in Pardo sa laura, which
:J::
...... 0 was represented by the ratio of the amount
0 5 10 after the fasting to that before it (Nakamura
FAST! NG PERIOD IN DAYS 1977)
Hunger and Starvation 289
values of f and p must be determined in order to obtain the degree of hunger. Both
quantities can be assumed as:
and
dp
-=oV,R
dt '
where {3 and 0 denote the rate of ingestion and that of food loss from the gut, respec-
tively (Nakamura 1972, 1977). Solving these equations simultaneously, we obtain the
following equation for the amount of food remaining in the gut when f = p = 0 at t = 0
(Nakamura 1977):
VR = V' (1- e-({3+6)t) ,
where V'is the amount of food remaining in the gut at satiation and equals /:0
(Nakamura 1972). This is a type of saturation equation, and VR reaches a maximum
value of V' at t = 00.
When spiders were deprived of food after the food supply to satiation, we have the
following for VR :
TT _ Vi -6 t
YR - Ve .
Thus, the equation for the degree of hunger during the fasting period is given as
V- VR = 1- V'e-ot (1)
V V·
Figure 104 shows the curve of this equation, arbitrary values for V'jV and 0 are
given. This curve corresponds to that in Fig. 103. It is apparent from Eq. (1) that the
degree of hunger during fasting depends only on the rate of food loss from the gut, o.
Thus, the greater the values of 0, the faster the maximum hunger is attained.
When spiders begin feeding after the fasting period, the degree of hunger after the
start of feeding increases according to the following equation after the start of feeding
ex: 1.0
LU
<!l
Z
:::>
:I:
u.
0
LU
0.5
UJ
ex:
<!l
UJ
'"
a
UJ
:I:
.....
TIME
Fig. 104. The change in the degree of hunger after satiated spiders were deprived of food [after
Eq. (1»)
290 K. Nakamura
(Nakamura 1977):
V- VR 1
-V-= (3 + 0 (0 + (3 e-(J3+o)t) . (2)
It is apparent from this equation that the degree of hunger in these situations is deter-
mined only by two constants, (3 and o.
Figure 105 shows the curves derived from Eq. (2), where different values of (3 and 0
were used. When 0 = 0, that is, the food ingested remains in the gut without disappear-
ing, the degree of hunger decreases with the passage of feeding time until it finally
reaches O. This is an extreme case which never occurs in nature. When (3 = 0, or the rate
of ingestion equals that of food loss in the gut, the hunger decreases until it reaches a
constant of 0.5. As the rate of food loss is considerably smaller than that of ingestion,
especially in spiders (see below), this situation is, in general, attained under the small
feeding rate due to the scarce food resources. As the feeding rate decreases less than 0,
the hunger level approaches constant values greater than 0.5.
It has been shown by many workers (e.g., K. Miyashita 1968b; Edgar 1969; J.F.
Anderson 1974) that spiders are scarcely satiated. K. Miyashita (1968b) showed that
a wolf spider, Pardosa astrigera, collected in the field, ingested a large amount of
food in the laboratory. J.F. Anderson (1974) compared weights and body dimensions
of individuals from field populations with those of known nutritional status in a wolf
spider, Lycosa lenta and a web builder, Filistata hibernalis, and showed that both spe-
cies are near starvation in nature.
Nakamura (1972) obtained the values of(3 and 0 as(3= 0.2275 h- 1 ando = 0.0392 h- 1 ,
in Pardosa laura: that is, the value of 0 is 17.5% of (3. Although both values of (3 and 0
are obtained in a few species of spiders, Nakamura (1977) concluded from the examina-
tion of the experimental results conducted for several other species that spiders have
1.0 r-------------------------------~
'"
UJ
(3 = 112 S"
'"=>z
:I:
u..
0
0.5 @= S
UJ
UJ
'"'"
UJ
Q
UJ
:I:
l-
S= a
a
TIME
Fig. 105. The change in the degree of hunger after starved spiders were supplied with food at dif-
ferent rates [after Eq. (2»)
small [j values, compared with the (3 values. This means that the hunger of spiders is
not so great under conditions of abundant food resources where the intrinsic value of
(3 is realized (Fig. 105). They ingest large amounts of food to satiate the capacity of
the gut and to store it in the gut. Since the spiders in nature are often near starvation
as mentioned above, it can be said that they live in a habitat of poor food resources,
where they can obtain food at such a small rate, i.e. less than the value of [j.
Taking this into consideration, we have assumed that the amount of food ingested
by spiders is determined only by the unfilled capacity of the gut, V - VR . However, the
initiation of feeding and successful capture of prey are also related to the hunger level.
In general, spiders begin to pursue and capture prey frequently under starvation condi-
tions. These behaviors become rare as they approach satiation.
Gardner (1964) examined the hunger effect on the hunting behavior of a jumping
spider, Phidippus clarus, which was deprived of food during different periods. He divided
the behavior into five sequences: orientation, pursuit, crouching, attaching, and jump-
ing. When the spiders deprived of food were allowed to attack prey, they captured prey
successively and approached satiation, probably according to such a curve as shown in
Fig. 105. The mean intersequence interval and the catching time increased during the
process of satiation (Fig. 106), suggesting that the hunting drive becomes less intensive
with a decrease in the degree of hunger. When satiation was approached, the initiation
of the sequences delayed and the continuation beyond orientation to the final stage,
jumping, occurred with low frequency.
Similar results were obtained in a wolf spider, Pardo sa laura, by Nakamura (1977).
Spiders were supplied with a different number of Drosophila flies. The number of flies
was always kept constant by resupplying new flies when spiders captured prey. Thus,
both times of feeding and nonfeeding were obtained during 6 h. The nonfeeding time
increased with successive captures of prey which is equivalent to the decrease in the
hunger level. Spiders at satiation ingest a small portion of prey body even if they cap-
s
, s
I
,
A. CATCHING TIME I B. INTER - SEQUENCE I
I
I
I
100 I
20 INTERVAL I
I
I I
I
I I
I I
I
I
, ,/J'I
50
oU---~--~----~---U oU---~----~--~----~
1ST 2ND 5TH 1ST 2ND 5TH
CATCHES (SUCCESSIVE FIFTHS OF TOTAL)
Fig. 106A,B. Changes in catching time and intersequence intervals against the successive fifths of
the total number of catches in Phidippus clarns. Spiders were deprived of food over 2 days (small
dashed lines), 6 days (long dashed lines) and 10 days (continuous lines). (After Gardner 1964)
292 K. Nakamura
ture prey. Thus, this proportion also changes depending on the degree of hunger (Na-
kamura 1974).
These two results suggest the existence of a threshold level which causes and drives
the feeding behavior. The sequence of behaviors in the capture of prey and feeding, in
relation to hunger, should be analyzed in more detail.
2 Starvation
It is well known that spiders have the ability to tolerate food deprivation over a long
period (Strazny and Perry, this volume). For example, adults of the wolf spider, Lycosa
lenta, survived an average of 208 days under starvation conditions and those of the
cribellate webbuilder, Filistata hibernalis, survived 305 days under the same conditions
(J.F. Anderson 1974). This survival time of L. lenta accounts for 90% of the adult
longevity of the well-fed individuals. Average survival time for starved Pardosa astrigera,
which were collected in the field in December and January, was 42-45 days, account-
ing for 41% of the longevity of well-fed individoals (K. Miyashita 1968b). However,
the same value for spiders, collected in March and April, was about 35 days for females
and 25 days for males. Adult longevity of the same species obtained by another author
(Tanaka and 1t6 1982) was 54.4 days and 28.8 days for females and males, respectively,
which accounted for 78.6% and 73.8% of the longevity of the well-fed adults. There-
fore, the starvation tolerance seems to change depending on the seasons or develop-
mental stages and sexes of spiders.
When spiders at satiation are deprived of food, body weight decreases sharply during
several days after deprivation until its decreasing rate becomes low (K. Miyashita 1969b;
Nakamura 1972; Tanaka and Ita 1982). During this period of sharp decrease in body
weight, spiders egest a considerable number of pellets (Fig. 107), suggesting that the
sharp decrease in body weight is mainly due to the decrease in food remaining in the
gut (Nakamura 1977). In the case of P. laura, a considerable number of pellets was
egested until the fifth day after the start of deprivation, which corresponded to the
time interval of sharp decrease in body weight. Therefore, the amount of food remain-
ing in the gut is considered to approach zero after the sixth day in this species.
The respiration rate also decreases sharply several days after the start of deprivation
and becomes nearly constant thereafter (K. Miyashita 1969b; Nakamura 1972; J.F. An-
derson 1974; Tanaka and Ita 1982). The reduction in metabolic rate in spiders during
fasting was first reported by Ita (1964), who obtained a 17% decrease in the rate of
starved Lycosa pseudoannulata. Similarly, females and males of P. astrigera had reduced
metabolic rates of 63.0% and 48.4%, respectively, on the 12th day after the start of
food deprivation (Tanaka and Ito 1982). In the case of the web builder, Filistata hiber-
nalis, a similar reduction of 30% to 40% was attained (J.F. Anderson 1974).
The respiration rate during fasting is generally considered to be the sum of the
energy required for the assimilation of food, for movement, and for the maintenance
of the standard metabolic rate. Of these activities, the assimilation of food ceased when
the food disappeared completely from the gut. Thus, the energy required for this
activity is regarded as zero after several days from the start of fasting, as considered
above. However, the energy required is not so small as to be neglected in the days just
Fig. l07A-C. The number of pellets

egested per day (A), change in body A. No. OF PELLETS
weight (8) and that in respiration rate 8
(CO, production in /l1/h/lO mg body
weight) (C) during fasting periods in
Pardosa laura (Nakamura 1972, 1977) 4
B. BODY WEIGHT
12
10 •
C. RESPIRATION RATE
pq
FASTING PERIOD IN DAYS
after food deprivation, because food remains in the gut and is assimilated during this
period.
The movement of spiders in the field is, in general, not very high, since they are
motionless almost at all times except for certain short periods. Web builders sit on
their web except when they move to a new site to rebuild their web (Riechert 1976;
Ford 1977b , 1978). This is also generally true for hunting spiders which are regarded
as sit-and-wait type predators (Edgar 1969; Fujii 1972 ; Ford 1977a). Ford (1978) re-
vealed that Pardosa amentata is active for only 0.0032% of the day. As the movement
of spiders is considered to decrease during a food deprivation period, the energy re-
quired for it will be so small that it can be neglected, although there is a datum con-
trary to this supposition (J.F. Anderson 1974).
Therefore, the respiration loss during long fasting is regarded mainly as the energy
for the maintenance of the standard metabolic rate. Thus, it has been considered by
many authors (Ito 1964; J.F . Anderson 1974 ; Tanaka and Ito 1982) that spiders have
the ability to reduce the metabolic rate during starvation . However, all reductions in
the respiration rate in the starved spiders are not only due to the reduction in the
metabolic rate for maintenance , but the energy for assimilation is also included. There-
fore, the reduction in the standard metabolic rate will not be as large as 4(}- 60% as ob-
tained by these authors.
294 K. Nakamura
Recently, Tanaka et al. (1985) measured the O2 consumption in P. astrigera and

found that the respiratory quotient (RQ) fell significantly during starvation. They sug-
gested from this result that starvation induces spiders to utilize fat as a metabolic fuel.
These behavioral and physiological characteristics in spiders enable them to have a
strong and long starvation tolerance.
3 Growth and Reproduction
The growth rate in invertebrate carnivores is often affected by the quantity of food
consumed. The laboratory experiment by Turnbull (1962), who reared Linyphia tri-
angularis with a different number of prey, showed that the growth rate increased with
the increasing number of prey. Kajak (1967) obtained in field populations of Araneus
quadratus an almost direct correlation between the daily food consumption and the
rate of increase in body weight. She concluded from this result that food was the limit-
ing factor in this species. Fujii (1972) observed the same phenomenon in Agelena opu-
lenta in the field, which showed a lower growth rate than the maximum one obtained
in the laboratory under conditions of unlimited food supply.
Recently, T. Miyashita (1986) analyzed the growth rate in Nephila clavata in the
field and found that the growth rate and the fecundity in adult females were greater at
sites where food was abundant. In this spider, the large individuals which might have
consumed greater amounts of food, became gravid earlier than small individuals. How-
ever, the small individuals also became gravid in late fall, producing a small number of
eggs.
As described in the preceding sections, it has been widely recognized that spiders
are in a near-starvation condition in the field and that they can be recognized baSically
as sit-and-wait type predators. These facts lead inevitably to the supposition that they
inhabit sites of scarce food resources. However, this is not always true, because spiders
inhabit various habitats with differing quantities of food and they can change their
web sites orhunting places by moving to areas providing larger amounts of food. Further-
more, they can adopt different types of prey capture tactics, such as building various
kinds of webs and hunting, both representing different efficiencies of prey capture
expending different costs for them (Nakamura 1982).
Food abundance does not only vary from habitat to habitat, but also shows strong
fluctuation. Thus, when food is abundant, spiders ingest large amounts of food, thus
satiating their gut and growing at a large rate. However, as the quantity of food de-
creases, spiders use the food stored in the gut, together with the food captured at a
low rate. In these situations they grow at a reduced rate. When food is scarce, they
endure this condition by reducing their active motion and the metabolic rate for stan-
dard maintenance and by utilizing the energy assimilated in their body. Under these
conditions, they cannot grow at all.
Since spiders have the ability to resist long starvation, the survival rate during star-
vation conditions is not considered to be significantly lower than the rate of non-
starved spiders, except in the early developmental stages (Turnbull 1962, 1973). Hence,
spiders can survive and grow at a low rate even when their chances of encountering
prey approach a very low rate, i.e. near zero. They can also reproduce even under these
conditions (J.F. Anderson 1974; T. Miyashita 1986), although the number or weight
of eggs is small, depending on the amount of food they consume (Turnbull 1962; Wise
1979). Thus, the ability of resistance to starvation, together with the behavioural and
physiological characteristics adapted to scarce food conditions, enable spiders to live
in habitats with small and strong fluctuating food resources.
D Parasites
I Nematode Parasites of Spiders
GEORGE O. POINAR, JR. 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
2 Natural Infections by Mermithid Nematodes . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
2.1 Historical Records of Spider-Mermithid Associations . . . . . . . . . . . . . . . . . . . 300
2.2 Classification of Spider Mermithids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
2.3 Bionomics of Spider Mermithids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
2.4 Effect of Mermithids on Spider Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
2.5 Host Range and Incidence of Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
3 Artificial Infections by Nematodes Belonging to the Rhabditida . . . . . . . . . . . . . . . . 307
3.1 Effect of NeoaplectaIUJ on Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
3.2 Effect of Heterorhi1bditis on Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
3.3 Other Nematodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
1 Introduction
Nematode parasites of spiders can be discussed under two aspects. The first constitutes
nematode infections which occur naturally and have been reported in the literature for
over a century. All of these reports refer to nematodes of the family Mermithidae (order
Mermithida), a large group which parasitize exclusively invertebrates, especially arthro-
pods.
The second aspect refers to certain rhabditoid nematodes which have been isolated
from the soil or insects living in the soil. Under laboratory conditions some of these
nematodes have shown an ability to infect and kill spiders. They have never been re-
corded from spiders in nature, but could, indeed, infect them since the two organisms
inhabit the same environment. They chance to be missed because, in contrast to the
mermithid nematodes, these rhabditids (e.g., Neoaplectana and Heterorhabditis sp.)
bring about host mortality within 2- 3 days after making contact with the spider.
Spider parasitism by nematodes has been a little studied subject which has casually
interested a few arachnologists and rarely any nematologists (Poinar 1985). However,
there is much of interest to be discovered regarding the behavioral and morphological
effects of parasitism on spiders, the life cycle of the parasites, and last but not least,
the effect of nematodes on spider population dynamics.
1 Division of Entomology and Parasitology, University of California, Berkeley, CA 94720, USA

300 G.O. Poinar, Jr.
2 Natural Infections by Mermitbid Nematodes
2.1 Historical Records of Spider-Mermitbid Associations
The historical information on spider-mermithid associations is important only for host

records (Table 23). Unfortunately, all of the descriptions of spider mermithids prior to
Table 23. Records of mermithid infections in spiders
Host Reference
Agelenopsis oregonensis Cl. Schlinger (pers. commun.)

Araneus ceropegius (C.L. Koch) Hoppe (1796)
Araneus diadematus (Clerck) Walckenaer (1833), Roesel (1761), von Linstow
Araneus spec. (1898)
Argyoneta aquatica (Clerck) Menge (1866)
Atypoides riversi O.P.-Cambridge Vincent (in press)
Cambridgea foliata (L. Koch) Lowe (pers. commun.)
Cesonia bilineata (Hentz) Poinar and Benton (1986)
Coelotes inermis (L. Koch) H.G. Millier (1983)
Diaea dOTIllta Fab. Rubzov (1980)
Drassodes sp. Holmberg (pers. commun.)
Geolyco!ll patellonigra Wallace Miller (pers. commun.)
Gnapho!ll Iucifuga (Walckenaer) von Siebold (1843)
Gnapho!ll sp. von Linstow (1883), Sorkin (pers. commun.)
Lyco!ll riparia sphagnicola Dahl Holm (1941)
LyCO!ll !llccata Latreille Holm (1941)
Lyco!ll scutulata Hentz Haldeman (1847, 1851)
Lyco!ll tar!lliis Thorell Holm (1941)
LyCO!ll verisimilis Montg. Montgomery (1903)
Lyco!ll vorax Walck. von Siebold (1854)
Lyco!ll spec. Leidy (1857), Kryger (1910), Bristowe (1941),
Doucet (pers. commun.)
Micryphantes bicuspidatus C.L. Koch von Siebold (1848)
Misumenops sp. Poinar and Benton (1986)
PardO!ll agrestis Rubzov (1979)
PardO!ll fureifera (Thorell) Leech (pers. corr.)
Pardo!ll glacialis (Thorell) Leech (1966)
Pardo!ll hortensis (Thorell) Parker and Roberts (1974)
Pardo!ll Iugubris (Walck.) Rubzov(1977,1979)
Pardo!ll nigropalpis Emerton Montgomery (1903)
Pardo!ll riparia (C.L. Koch) Rubzov (1977,1979)
Pardo!ll scita Montg. Montgomery (1903)
Pardo!ll vancouveri Emerton Holmberg (pers. commun.)
Pardo!ll spec. Kaston (1945), Rubzov (1978)
Peucetia viridans (Hentz) Landau (pers. commun.)
Phidippus borealis Banks Cutler (pers. commun.)
Phidippus clarus Keyserling Kaston (1945)
Phidippus johnsonii Sivertsen (pers. commun.)
Phidippus putnamii (Peckham & Peckham) Cutler (pers. commun.)
Phidippus sp. Poinar and Benton (1986)
Pseudicus sp. Sorkin (pers. commun.)
Salticus formicarius Latreille Bertkau (1888)
Nematode Parasites of Spiders 301
Table 23 (continued)
Host Reference
SchizoCOSil Silltatrix (Hentz) Montgomery (1903)

SchizoCOSil sp. Sorkin (pers. commun.)
Sitticus palustris Cutler (pers. commun.)
Sosippus florida nus Simon Kaston (1945)
Tarentula inquilina Thorell Bertkau (1888)
Tetragnatha sp. Sorkin (pers. commun.)
Theridion ovatum (Clerck) Bristowe (1931a)
Theridion redimitum Linne Bristowe (1941)
Tibellus oblongus (Walckenaer) Holmberg (pers. commun.)
Tmarus sp. Poinar and Benton (1986)
Verrucosa areanata (Wa1ckenaer) Poinar and Benton (1986)
Wadotes sp. Poinar and Benton (1986)
Wulfila alba (Hentz) Poinar and Benton (1986)
Xysticus deichmanni Soerensen Leech (1966)
Xysticus durus (Soerensen) Leech (pers. commun.)
Xysticus funestus Keyserling Kaston (1945)
Zora maculata D.P.-Cambridge Kryger (1910)
1985 can be considered species inquirendae since all descriptions were based on juvenile
material and the present systematics of the group is based on adult characters. The col-
lective genus Agamomermis was erected for such species (poinar and Welch 1981). At
the time of emergence from the spider, a mermithid is still a juvenile (postparasitic
juvenile) and it must be held at least for 3-4 weeks before molting to the adult stage.
During this period, the worm is susceptible to bacterial and fungal infection which is
the major cause of mortality under laboratory conditions.
The first report of a mermithid nematode parasitizing spiders was presented by
Roesel in 1761. This was followed by Hoppe in 1796 and Walckenaer in 1833. The lat-
ter author cited a Filaria from Aranea diadema. In 1847, Haldeman described the mer-
rnithid, Filaria lycosae from the spider Lycosa scutulata in Pennsylvania. In 1910,
Kryger also cited a Filaria from Lycosa sp. and Zora maculata. Nematology was in its
infancy at that time and Filaria was used as a collective genus for nematodes infecting
many animals, both vertebrates and invertebrates. Shortly after Haldeman's report,
Leidy (1857) collected a mermithid from a Lycosa sp. in New Jersey and called it
Mermis robusta. He considered his nematode the same as Filaria lycosae, which he con-
sidered a synonym to hisM. robusta. Later Leidy's error has been pointed out and the
name was changed to Mermis lycosae (Haldeman).
At this time the genus Mermis was used in a broad sense to represent members of
the family Mermithidae. It and the frequently used binomial, Mermis albicans, were
assigned to a range of species collected from arthropods. However, as in the case of
Filaria, these names were not accompanied by a description or the description was so
general that it was useful only to family level. Thus, the records of Mermis albicans
from Argyroneta aquatica by Menge (1866), Mermis sp. from Lycosa riparia sphagni-
cola by Holm (1941), Mermis sp. from Lycosa sp. by Bristowe (1941), Mermis sp.
from Theridion ovatum by Bristowe (1931a),Mermis sp. fromSchizocosa saltatrix and
302 G.O. Poinar, II.
Pardosa nigropa/pis by Montgomery (1903), and Mennis sp. from Tarentula inquilina
by Bertkau (1888) are all species inquirendae. The two species of Mennis that von
Linstow (1898) described from spiders, namely, Mennis drassi from Drassus (= Gna-
phosa) sp. and Mermis brasiliensis from an unidentified spider, unfortunately fall into
the same category.
The grasshopper mermithid, Agamermis decaudata was cited as a parasite of Pardosa
sp. and Phidippus clarus by Kaston (1945), but this is probably a misidentification since
juvenile material was examined and this well-studied mermithid was never previously
recorded from spiders. The same situation holds for the Hexamermis sp. recorded by
Leech (1966) for Xysticus deichmanni and Pardosa glacialis and by Kaston (1945) for
X. [unestus.
Within the past 10 years, Rubzov described four species of mermithids from spiders.
All of these are based on immature specimens and must stand as species inquirendae.
These include Pardosomermis pardosensis (1977, 1979) from Pardo sa /ugubris and
P. riparia, Arachnomermis araneosa (1978) from Pardo sa sp., Arachnomermis pardo-
sagrestis (1979) from Pardosa agrestis, and Arachnomermis dialaensis (1980) from
Diaea dorsata. The first description of a spider mermithid based on adult characters
was made by Poinar and Benton (1986) who described Aranimermis aptispicula from
a variety of spiders (Table 23).
During the present literature search, efforts were made to separate the early reports
of mermithid nematode infections of spiders from those of Gordius hairworm parasitism.
The latter group belong to a completely different phylum (Nematomorpha) yet bare a
superficial resemblance to mermithids. Both groups overlap in size and shape and it is
quite probable that some of the early reports of gordiid infections in spiders pertain to
mermithid infections. The color is the best criterion to separate members of the two
groups. Mermithid nematodes are predominately whitish or cream colored with tints
of pink, yellow, and green, whereas gordiids are usually brown or black.
2.2 Classification of Spider Mennithids
Until the present, nothing was known concerning the classification of spider mermi-
thids. Since adult stages were never collected, there was no evidence to consider mer-
mithids from spiders different from those of insects and the proposed names Mennis
albicans, Hexamermis sp., Agamennis sp., and Agamermis decaudata indicate this be-
lief. The thought was that if an insect parasitic mermithid couldn't find a suitable host,
it might attack a spider that was in the immediate vicinity. It now appears that spiders
have their own specialized mermithid fauna which represent certainly new species and
most probably a number of new genera. The new genus and species A ranim erm is apti-
spicula Poinar and Benton (1986) has specialized morphological features setting it off
from previously described insect forms. The shape of the spicules and vagina are unique
for mermithid nematodes.
2.3 Bionomics of Spider Mennithids
Aside from the fact that mermithid parasites of spiders are worldwide and that they
kill the spider upon emergence, the life history of any spider mermithid is incomplete.
There are five stages in the basic mermithid life cycle (poinar 1983). The eggs are de-
posited in the environment and the second stage juvenile (called infective juvenile or
pre parasitic juvenile) emerges. This stage makes contact with the host, enters its body
cavity, and initiates development into a third stage parasitic juvenile. When develop-
ment is completed, the parasite emerges as a mature third stage postparasitic juvenile
which then matures and molts twice to become an adult. This basic cycle may be
modified in various ways, but essentially falls into two patterns, termed direct and
indirect. The direct cycle, which is by far the commonest amongst the mermithids
studied up until the present, involves a single, developmental host. The pre parasitic
juvenile hatching from the egg deposited in the environment enters and develops inside
a susceptible host, and emerges as a postparasitic juvenile. This stage returns to the
environment to mature, mate, and oviposit. The indirect cycle involves at least two
hosts, one or possibly more paratenic hosts, and a developmental host (poinar et al.
1976). The adults deposit their eggs in an aquatic or semi-aquatic habitat. These eggs
are ingested and hatch in the gut of various invertebrates (insect larvae, turbellarians,
annelids). But instead of developing, the pre parasitic juveniles invade the tissues and
enter a period of dormancy. This dormancy is only broken when the original paratenic
host is eaten by the developmental host. The nematode then completes its develop-
ment in the predator and emerges as a postparasitic juvenile. This cycle includes two
additional variables to the normal, direct form of development. One is the chance of
the paratenic host falling prey to the right predator and the other is the need for the
parasitized host to return to an aquatic environment when the nematode is ready to
emerge.
Observations on the life history of Aranimermis aptispicula indicate that it has an
indirect life cycle. Adults were found in an aquatic habitat, whereas parasitized spiders
were found in a variety of foraging habitats (on plants, in webs, and on the ground).
Parasitized spiders were observed to enter the water and the nematodes were seen to
emerge from the hosts' bodies. These observations support an indirect type of cycle,
but further studies are necessary to substantiate this. It is likely that a number of dif-
ferent mermithid genera and species attack spiders and that some of these may also
have a direct cycle. The area is open for future studies.
2.4 Effect of Mermithids on Spider Hosts
Most parasitized spiders are killed just before, during, or just after emergence of the
postparasitic juvenile mermithid.
2.4.1 Morphological Changes
External morphological abnormalities are usually one indication of spider parasitism

by mermithid nematodes. Infected spiders often show a swollen abdomen in the later
stages of development and Leech (1966) noted that parasitizedP. glacialis had a lopsided
or greatly enlarged opisthosoma, an altered epigynum, malformed palpi, legs that were
shorter and thicker than normal, and poorly developed or absent male secondary sexual
characteristics. It is possible to see the coils of the parasite through the host's integu-
Fig. 108. a The mermithid, Aranimermis aptispicula, filling the abdominal cavity of a Tmarus (Tho-
misidae) (specimen from C. Benton). b Apostparasitic juvenile mermithid in the process of emerg-
ing from the turret spider Atypoides riversi (Antrodiaetidae) in California (specimen from L Vin-
cent; photos by the author)
Fig. l09a,b. Double infection in the hemocoel of a turret spider in California by a mermithid ne·
matode and (a) a third stage acrocerid larva and (b) a second stage tachinid larva (arrow) (specimens
from L Vincent; photos by the author)
ment since the mermithid usually occupies the entire abdomen and occasioanlly the
cephalothorax (Fig. l08a). Leech (1985, pers. corr.) also noted that the epigynum
was altered in an infected Pardosa furcifera female and that the palpal organ was slight-
ly altered in a male Xysticus durus.
Internal effects of mermithid parasitism include a reduction of the digestive gland
and parasitic castration. This was noted in spiders attacked by Aranimermis aptispicula
(Poinar and Benton 1986). In extreme examples, other organs may also be reduced.
Leech (1966) commented that parasitism of P. glacialis resulted in the loss of the main
prosomal muscles, the entire digestive system and the entire reproductive system, and
various degrees of parasitic castration in both males and females. The latter condition
was also noted in parasitized spiders by Holm (1941).
2.4.2 Beluzvioral Cluznges
It is known that nematode parasites can affect the behavior of their host. Some be-
havioral deviations are necessary for the completion of the parasite's life cycle. The
major behavioral change in spiders infected with some mature mermithids is their migra-
tion toward a source of water, which is the habitat for nematode emergence in mer-
mithids possessing an indirect cycle. This attractiveness to water was noted in spiders
parasitized by Aranimermis aptispicula as they would come out of neighboring woods
and fields to congregate at a water source. Working with the California turret spider,
Vincent (pers. commun.) noted that during the period of study, only three specimens
left their burrows and all three had a terminal mermithid infection (Fig. 108b). Leech
(1966, pers. corr.) noted that during the week before the mermithids emerged, para-
si tized spiders ceased feeding, but drank a lot of water. He also observed that infected
individuals of P. glacialis were sluggish and did not attempt to escape when approached.
2.5 Host Range and Incidence of Infection
As seen in Table 23, a wide range of spiders representing a variety of foraging habits
were subject to parasitism by mermithid nematodes. Little work has been completed
on the incidence of mermithid infection in any particular population of spiders. In a
population of P. glacialis in Canada, Leech (1966) noted that 1% of the individuals
were parasitized. However, the incidence of parasitism in select areas along stream
courses where the water ran all summer reached 5%, whereas the drier areas often did
not have any infection (Leech 1985 pers. commun.). In a population of A typo ides
riversi in California, Vincent (pers. commun.) sampled and dissected 109 individuals (all
sizes) and found nine (= 8%) to be infected with mermithid nematodes. He also found
incidences of mixed infections, in one case a mermithid nematode together with an
acrocerid fly (Fig. 109a) and another with a mermithid and tachinid fly together in
one spider (Fig. 109b).
The incidence of infection can vary over a 12-month period and mermithid nema-
todes, together with other parasites, could be decisive factors in limiting the distribu-
tion of some spiders.
3 Artificial Infections by Nematodes Belonging to the Rhabditida
Although all natural infections of spiders by nematodes involve only members of the
order Mermithida, laboratory tests show that spiders are susceptible to infection by
other soil-inhabiting nematodes belonging to the order Rhabditida.
3.1 Effect of Neoaplectana on Spiders
Members of the genusNeoaplectana belong to the family Steinernematidae of the order

Rhabditida. They comprise a half dozen or so species which are found in soil in various
parts of the world. Natural infections of these nematodes have been confined to insects
and because of their attributes, they are now being developed commercially by several
firms (Poinar 1985). The life cycle of Neoaplectana is radically different from that of
mermithids since it involves a symbiotic bacterium which kills the insect within 48 h
after contact. The infective stage of Neoaplectana is represented by a third stage juvenile
which is the only free-living stage in the life cycle of this parasite. Enclosed in the lu-
men of the intestine of the infective stages are bacterial cells of the genus Xenorhabdus
(Enterobacteriaceae). The infective stage enters a host through the natural openings,
then penetrates into the body cavity and liberates cells of Xenorhabdus. The bacteria
multiply and kill the host within 48 h. The nematodes develop into first generation
adults in the dead host, and reproduce, producing one, two, or possibly more cycles
depending on the size of the host and the initial dose of nematodes. When nutrients
are depleted, the nematodes form infective stages which leave the host and return to
the environment.
Under laboratory conditions, the spiders Pholcus phaillngioides, Latrodectus mac-
tans and Pirata sp. were challenged with infective stages of the 42 strain of Neoaplec-
tana carpocapsae (Poinar and Thomas 1985). After 20 days infection occurred in all
six of the P. phaillngiodes, all six of the L. mac trans and three of six Pirata sp. All of
the dead spiders showed the presence of Xenorhabdus bacteria in their hemocoel. The
nematodes developed to the adult stage, but did not reproduce. Although these results
show that Neoaplectana carpocapsae is capable of killing spiders, they also show that
spiders are poor hosts for these nematodes which may explain why natural infections
have never been reported.
3.2 Effect of Heterorhabditis on Spiders
Members of the genus Heterorhabditis belong to the family Heterorhabditidae of the

order Rhabditida. They comprise three or four species which are found throughout the
world, especially in warm climates. The cycle of Heterorhabditis is very similar to that
of Neoaplectana, since they also have a third stage infective juvenile which carries
Xenorhabdus bacteria. Aside from morphological characters, one distinction between
the two genera centers around the type of reproduction. The infective stages of Hetero-
rhabditis all develop into hermaphroditic females, whereas in Neoaplectana, they deve-
lop into both males and females.
Under laboratory conditions,P. phaillngioides, L. mactans, and Pirata sp. were chal-
lenged with Heterorhabditis heliothidis. After 20 days, three of six P. phaillngioides,
308 G.O. Poinar, Ir.: Nematode Parasites of Spiders
four of six Pirata sp., and all six L. mactans became infected. All dead spiders showed
the presence of Xenorhabdus bacteria in their body cavity, but as with Neoap lectana ,
the nematodes developed only to the adult stage without reproducing. Thus, H. helio-
tJddis can kill spiders, but for some unknown reason are unable to reproduce (Poinar
and Thomas 1985).
3.3 Other Nematodes
From time to time, the author has received dead spiders with nematodes crawling over
their bodies, similar to the report of Platnick and Sedgwick (1984), who found nema-
todes on live spiders and eating the remains of recently dead captive Liphistius spiders.
In most instances, these probably represent microbotrophic nematodes that are feed-
ing on living bacteria associated with decomposition of the spider. However, each spe-
cies should be examined since it is possible that spiders have their own rhabditid para-
sites that are capable of bringing about mortality and then feed on the cadaver. The
nematodes should be isolated and then placed together with healthy spiders to test
their powers of infectivity. I predict that some day natural infections of spiders by ty-
lenchid and rhabditid nematodes will be discovered.
II Spider-Hunting Sphecid Wasps
ROLLIN E. COVILLE 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
2 The Sphecid Wasps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
2.1 Geographic Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
2.2 Life History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
2.3 Nesting Habits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
2.4 Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
2.5 Food of Adult Wasps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
2.6 Oviposition and Larval Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
2.7 Survival of Paralyzed Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
2.8 Provisioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
3 Sphecidae as Spider Predators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
3.1 Impact of Sphecids on Spider Populations . . . . . . . . . . . . . . . . . . . . . . . . . 315
3.2 Hunting Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
3.3 Factors Influencing Prey Preferences of Sphecids . . . . . . . . . . . . . . . . . . . . . 317
4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318
1 Introduction
The Hymenoptera families Sphecidae and Pompilidae contain wasps that provision their
larval cells with paralyzed spiders. The predatory habits of these two groups differ in
that sphecids store several small spiders in each cell, called mass provisioning, whereas
pompilids use one spider that is as large or larger than themselves. Sphecids are general-
ly unable to pursue their small prey into confmed spaces. Hence, spiders that dwell in
burrows, in web tubes, and under bark or rocks may be heavily preyed upon by pom-
pilids, but are usually spared from sphecids.
While the use of a single spider by pompilids appears efficient, mass provisioning of
larval cells by sphecids also has advantages. First, small prey are easier to overpower
and, hence, less likely to pose a threat. Second, small prey can be carried in flight and
brought to the nest quickly, even when captured far away. This lessens the chance that
prey may be stolen from the wasp by another insect, and also means that sphecids can
establish nests in favorable sites and are not dependent on a local abundance of prey.
Although some pompilids will fly with their prey, most can only drag them about and
the majority of pompilids build their nests after, rather than before, a spider has been
Division of Entomology and Parasitology, 201 Wellman Hall, University of California, Berkeley,
CA 94720, USA
310 R.E. Coville
captured. Third, among potential prey for a wasp of a given size, small spiders may be
several times more abundant than larger spiders. In other words, mass provisioning a
cell with several small spiders may be accomplished as rapidly as provisioning a cell with
one large spider.
The family Sphecidae is a large group of wasps that prey upon many kinds of in-
sects and spiders. In this chapter I have chosen to follow the higher classification of
Sphecidae presented by Bohart and Menke (1976) in which the 7600 known species
were arranged in 11 subfamilies and 28 tribes. Nevertheless, many of these higher taxa
are so distinct that some taxonomists (see Krombein 1979) prefer to treat them as
separate families. Bohart and Menke (1976) is now the standard reference on sphecid
wasps and provides an excellent entry to the taxonomic and biological literature on all
the genera up to the time the work was published.
Two subfamilies of Sphecidae, the Sphecinae and Larrinae, contain wasps that hunt
spiders. Among Sphecinae the spider hunters are in the genera Sceliphron and Chaly-
bion in the tribe Sceliphronini. Among Larrinae spider hunting occurs in one genus,
Miscophus, in the tribe Miscophini and in three genera, Pisonopsis, Pison, and Try-
poxylon, in the tribe Trypoxylonini. In the following sections of this chapter I will use
the terms Sphecidae and sphecids only in reference to these groups.
2 The Sphecid Wasps
2.1 Geographic Distribution
Although sphecid wasps occur in all temperate and tropical regions of the world, each
genus has its particular distribution and center of diversity. Sceliphron (30 ssp.), Chaly-
bion (30), and Miscophus (150) have their greatest species diversity in the Old World.
In addition, Chalybion is unknown from South America, and Miscophus does not oc-
cur in Australia. Trypoxylon (ca. 500) andPison (150) both have widespread distribu-
tions. Pison has its greatest diversity in Australia. Trypoxylon is poorly represented in
Australia, but is abundant in most other regions.Pisonopsis (5) only occurs in the south-
western United States, Mexico, Argentina, and Chile.
2.2 Life History
Sphecids vary greatly in size. Sceliphron, Chalybion, and a few Trypoxylon have body
lengths greater than 20 mm. On the other hand, some Miscophus are less than 5 mm
long. The body length of most of the wasps range between 5 and 15 mm.
Sphecids are primarily solitary wasps, in that each female normally constructs and
provisions her own nest. Females of a few Trypoxylon may have communal nests, but
they have not been studied in detail. The general sequence of events in the construc-
tion and provisioning of nests begins with the female searching for a suitable nest site.
Once a site has been selected, she prepares a nest with a chamber or cell in which one
of her progeny will develop. Next, she hunts for spiders to be captured and stung. The
paralyzed spiders are carried to the nest and packed into the cell. After she has ovi-
posited and supplied the cell with enough spiders for one of her progeny to develop to
Spider-Hunting Sphecid Wasps 311
the adult stage, the wasp seals off the cell. She is then ready to prepare another cell or
to depart in order to found a new nest. In Pison strandi and all Trypoxylon wasps in
the subgenus Trypargi/um, a male remains in each nest as a guard while the female
searchs for building materials or hunts for spiders (Coville and Coville 1980; Coville
and Griswold 1984; Tsuneki 1970); few other wasps have similar behavior.
2.3 Nesting Habits
Nesting habits of the spider hunters include (1) excavation of burrows in the ground,
(2) modification of existing cavities, and (3) mud-daubing or the building of free mud
nests. More than one nesting habit may occur among species of a single genus. Within
Pison, for example, all three habits are known, and among Trypoxylon, the second and
third occur (Fig. 110). The remaining genera exhibit a single type of nesting behavior:
Miscophus excavate nests in the ground, Chalybion occupy preexisting cavities, and
Sce/iphron are mud-daubers (Fig. lIla).
Species that excavate burrows in the ground, includingMiscophus (Kurczewski 1969,
1982) and three Australian species of Pison (Evans 1981), prefer to nest in sandy areas
that are sparsely vegetated. Evans (1981), however, observed P. auriventre excavating
burrows at the crest of a clay bank. The nests consist of a burrow ranging in length
from 2.0 to 8.4 cm in Miscophus and 4.5 to 16.0 cm in Pison . Nests of Miscophus fre-
quently have a single oval cell at the end of the burrow, although multicellular nests
are known for a few species. Nests of Pison are multicellular.
Sphecids that establish nests by modifying existing cavities include Chalybion, Pi-
sonopsis, and most Pison and Trypoxylon (Fig. 110a). Examples of nest sites com-
Fig. 110. a Trypoxylon vagulum at the entrance to its nest in the drilled hole of a trap nest ; most
Trypoxylon prefer to nest in preexisting tubular cavities. b Mud nest of Trypoxylon aestivale; two
new adults have previously emerged through the holes in the second tube from the left
312 R.E. Coville
b
Fig. 111. a Sceliphron caementarium gathering mud to be used in building its nest. b Cell of Try-
poxylon lactitarse within a trap nest; the white, sausage-shaped wasp egg is attached to the spider
at the lower left side of the cell; the front of the cell is to the left
monly used by these wasps include hollow plant stems; abandoned beetle burrows in
logs, pillars, railings, and tree trunks; vacated burrows of bees, wasps, and tiger beetle
larvae in the ground; and cells in abandoned mud-dauber nests. In addition, many of
these wasps will r~adily nest in holes drilled in blocks of wood (Fig. llOa) or stems, a
behavior that allows them to be easily studied. Nests of Chalybion are simple struc-
tures . Generally, the nest cavity contains a single cell sealed with a thick inner layer of
mud, often deposited in chunks, and a conspicuous outer layer of uric acid from bird
or reptile droppings, dung, lime from walls, or other materials. Nests of Pisan, Pisan-
apsis, and Trypaxylan consist of a linear series of cells. Pisonopsis birkrnanni and cly-
peatum use small pieces of debris to separate the cells, whereas Pison and Trypoxylon
build thin mud partitions.
The mud-daubers, including Sceliphron, some Pison, and some Trypoxylon, build
their nests in locations that are protected from rain. Man-made structures such as
homes, sheds, barns, and bridges are especially attractive to these insects. In natural
environments they are frequently found within hollow logs, under rocks, beneath rock
overhangs, at cave entrances, attached to roots exposed beneath overhanging earth
banks, and attached to the underside of leaves.
The mud nests have diverse architecture. They generally consist of clusters of mud
cells that are placed side by side in one or more rows or layers. In completed nests, the
outline of individual cells is often obscured. The wasps frequently deposit extra quan-
tities of mud on and between the cells to provide additional strength and protection.
Nests of mud-daubing species of Trypoxylon, subgenus Trypargilum, consist of a series
of vertical mud tubes set side by side and open at the bottom (Fig. 11 Ob). Each tube is
divided into a linear series of cells separated from one another by a thin mud wall or
partition. Because of the appearance of their nests, these wasp are commonly referred
to as "pipe organ wasps".
The nests of many mud-daubers are durable structures that may last for years after
the original occupants have left. Iwata (1976) claimed that attached to the stone ruins
of Angkor Wat in Kampuchea (Cambodia) are remains of Sceliphron nests built in the
12th century. The cells in old mud nests are often repeatedly used for nests by other
wasps and bees.
2.4 Activity
All sphecids are diurnal. Foraging and nest building normally begin within 1- 3 h after
sunrise and cease within approximately the same amount of time before sunset. Activ-
ity is influenced by temperature, weather, and exposure of the nest to sunlight. For
example, along San Francisco Bay in California (USA) initiation of nesting and foraging
by M. califomicus depends upon temperature, which is influenced by fog conditions
(Powell 1967). Rain also delays or terminates foraging and nesting. In Jamaica S. as-
simile temporarily close partially provisioned cells before the onset of afternoon rain
showers and do not reopen them until the next day (Freeman and Johnston 1978a).
At night many sphecids remain in their nests, but others do not. Before sunset, some
Miscophus fill their nest entrances with sand and then dig a short sleeping burrow. Both
sexes of C. bengalense, califomicum, and zimmermani form roosting aggregations in
sheltered places (Jayakar and Mangipudi 1965;Rau 1942;G.L. Ward 1972). Likewise,
S. assimile often roost in groups on plant stems (Freeman and Johnston 1978b).
25 Food of Adult Wasps
Adult sphecids obtain carbohydrates from nectar and honeydew. Females, who re-
quire protein in order to develop their eggs, may feed on their prey. Sceliphron caemen-
314 R.E. Coville
tarium (Shafer 1949) and S. assimile (Freeman and Johnston 1978a) chew on small
spiders and drink their body fluids. Eberhard (1970) observed S. caementarium press
their mouths to the mouth region of newly stung individuals of Araneus cornutus, and
suggested that the wasps may have been ingesting fluids from the spiders' mouths.
2.6 Oviposition and Larval Growth
Females of Sceliphron usually oviposit on the first spider captured, although S. caemen-
tarium may occasionally oviposit on the second, third, or fourth spider (Shafer 1949).
There is also some question as to when Chalybion oviposit. Rau (1928) reported that
C. californicum oviposit after the cell has been provisioned, but Muma and Jeffers
(1945) claimed that they oviposit on the first spider stored. All the remaining genera
oviposit after the cell has been fully provisioned (Fig. 111 b).
The white, sausage-shaped wasp egg is always attached to the abdomen of a spider
(Fig. 111 b). Eclosion takes place after 1.5 to 3.s days, depending upon the species.
The larva punctures the spider abdomen to which it is attached and begins feeding.
Growth is rapid. After consuming the initial spider, the larva can move freely about
the cell and quickly devours the remaining provisions. Usually it eats all the prey within
5-7 days after eclosion. The larva does not defecate until after it has spun its cocoon.
In S. caementarium, Shafer (1949) found that during the feeding stage the larva's mid-
gut is a blind sac.
In temperate climates the wasps have one or two generations per year and over-
winter in the prepupal stage. In tropical climates the wasps are active all year, and
development from egg to adult can be in as little as 3 to 4 weeks depending upon the
species. Cool temperatures will delay emergence of adults.
2.7 Survival of Paralyzed Spiders
Spiders never fully recover from being stung. Any movement of the legs and chelicerae
remain more or less uncoordinated. Most prey survive just long enough to provide the
wasp larva with a fresh meal. When removed from the nest small spiders often die within
a few days, but larger well-fed spiders can survive for several weeks and in rare cases
more than a month (pers. observations).
Sphecids appear to direct their stings toward the subesophageal ganglia of their
spider prey; they probably sting through the membraneous areas at the base of the
coxae or in the mouth region. Piek et al. (1983) found that the venom of S. spirafex
contains two histaminelike components, but the mode and site of venom activity is
still poorly understood (see also Carrel, this volume).
2.8 Provisioning
Sphecids provision each cell with three or more spiders. A few examples covering the
typical range of variability are 11 to 44 in T. tenoctitlan (Coville and Coville 1980),
3 to 27 in T. collinum (Krombein 1967),8 (and possibly as few as 5) to 33 in Pison
strandi (Tsuneki 1970), 3 to 14 in M. americanus (Kurczewski 1982), and 7 to 29 in
M. kansensis (Kurczewski 1969). The number of prey per cell probably depends upon
the relative abundance of different sized spiders in the field. Cross et al. (1975) found
that T. politum in North America took more smaller spiders early in the season and
fewer larger spiders later on, but the biomass of provisions per cell was relatively con-
stant. In T. vagulum from Costa Rica, male cells were provisioned with less biomass of
spiders than female cells, but again due to the great variability, no statistical differences
were found in the number of prey (pers. observations).
The amount of time a wasp takes to leave the nest and return with a spider is also
variable, ranging from 1 to more than 70 min. The average trip usually falls between
6 and 35 min. For example, S. assimile averaged between 8.8 and 15.7 min at different
sites in Jamaica, and summer drought was positively correlated with longer times (Free-
man and Johnston 1978a). Kurczewski (1969) reported average foraging times ranging
between 8 and 35 min among different females of M. kansensis. Foraging time is in-
fluenced by abundance of prey, weather, and perhaps the condition or age of the wasp.
Irregardless, an individual sphecid can potentially capture a tremendous number of
spiders during its adult life, which ranges from 3 to 6 weeks. Consider the Costa Rican
wasp T. tenoctitlan which averaged 20 spiders per cell and provisioned cells in less than
a day (Coville and Coville 1980) or T. vagulum which usually completed 1.5 to 2 cells
per day with 19 spiders per cell (pers. observations).
3 Sphecidae as Spider Predators
3.1 Impact of Sphecids on Spider Populations
Sphecids are potentially an important mortality factor of spiders, especially the small
immature stages. The wasps would almost certainly show an immediate functional
response to an increased abundance of prey. Numerous prey would allow them to
provision cells more rapidly and perhaps complete more cells during their life than
otherwise pOSSible.
In relation to other insects, sphecids have low fecundity. Mean fecundity of each
nesting female of S. assimile in Jamaica is only about 9.3 eggs (Freeman 1980) and of
S. laetum in Australia about 8.1 eggs (A.P. Smith 1979), assuming one nest per female.
Nevertheless, mortality rates among immature stages of wasps are generally far lower
than required for replacement of the adult population. Average mortality rates for
S. assimile (Freeman 1980) and laetum (A.P. Smith 1979) were 40% and 38%, respec-
tively. Danks (1971) recorded rates of 56% to 64% in stem-nesting Trypoxylon in Eng-
land. Coville and Coville (1980) observed 44% mortality in nests of T. tenoctitlan in
Costa Rica.
Studies of Danks (1971) on stem-nesting Hymenoptera, including TrypoxyZon spp.,
Freeman (1980) on S. assimile, and A.P. Smith (1979) on S.laetum in Australia suggest
that the principal factors controlling population size act on pre-reproductive females.
In S. assimiie, for example, only about 28% of adult females establish nests. The in-
ability of many females to find suitable nest sites is probably the most important fac-
tor. For stem nesters Danks (1971) suggested that suitable nest sites have an uneven,
constantly changing distribution. For Sceliphron nest sites, such as man-made struc-
316 R.E. Coville
tures, may be widely scattered (A.P. Smith 1979). In addition, suitability of nest sites
may change with the seasons; seasonal droughts render some sites used by S. assimile
unfavorable due to a lack of mud for nest building (Freeman 1980).
3.2 Hunting Behavior
Sphecids use the following techniques to locate spiders: (1) alighting on bumps and
spots, especially those that contrast against the background; (2) tapping surface with
their antennae to detect and flush spiders that are maintaining cryptic postures or
hiding in places the wasps can reach; (3) examining webs and tearing open spider re-
treats; (4) attracting spiders by pulling and vibrating webs in the manner of an en tangled
insect. The wasps may also exploit local aggregations of spiders. For example, S. caemen-
tarium (Eberhard 1970) and C. californicum (Coville 1976) will return repeatedly to
areas where they previously captured prey. Statistical evidence, based upon the time it
took for females to return to the nest with spiders, suggest that females of S. assimile
do the same (Freeman and Johnston 1978a).
Eberhard's (1970) notes on S. caementarium preying on Araneus cornutus around
white trimmed windows of a cottage are one of the best accounts of hunting behavior.
During his observations, A. cornutus left their orb webs up during the day, while they
hid in retreats along the edges and corners of the window panes. While hunting, Sceli-
phron hovered near the windows and alighted upon objects and spots that contrasted
against the background. Upon finding an occupied retreat of A. cornutus, the wasps
used their mandibles to attempt to tear through the silk and reach the spider. The
spiders responded by dropping out of one end of the retreat. Occasional spiders escaped,
but Sceli'phron usually seized them just as they left their retreat. The wasps also cap-
tured some spiders that had dropped to the windowsill below or were hanging from
their silk thread. The wasps used their mandibles and forelegs to seize the spiders. Each
spider was stung one to four times somewhere on the ventral side of the cephalothorax.
Sometimes the prey were stung while the wasps were in flight. Sceliphron did not hunt
spiders that had webs on the shingle-covered walls of the house, because the spiders
either hid beneath the shingles or blended into the color and texture of the walls.
Coville (1976) found that Chalybion californicum uses deceit to capture the orb-
weaving araneid Zygiella x-notata. Like Araneus cornutus, Zygiella left their webs up
during the day while they hid in retreats on the underside of leaves. A Chalybion would
alight on a web. After remaining motionless for a few seconds, the wasp would begin
pulling on the web with her legs. Responding to the vibrations as it would to an entan-
gled insect, a spider often left its retreat and ran down the web to the wasp. The wasp
seized the approaching spider with her mandibles and forelegs, and then stung the spi-
der somewhere on the ventral side of the cephalothorax. The spider was carried across
the web to a leaf, where it was further rolled and manipulated and occasionally stung
again. Spiders that remained in their retreats or dropped from their webs escaped the
wasps.
Although the method of capturing Zygiella is well-suited for two-dimensional orb
webs, Chaiybion prey more heavily on theridiids, such as Latrodectus mactans (Muma
and Jeffers 1945; Rau 1935b), whose webs are a three-dimensional maze. How theridi-
ids are captured in the field is unknown.
Miscophus hunt more like pompilids than other sphecids because they will crawl
into confined spaces, and occasionally take spiders that are too large to carry in flight.
Miscophus hunt on or near the ground. They also hunt close to the nest; Kurczewski
(1969, 1982) observed Miscophus capture spiders between 1.4 and 2.4 m from their
nest entrances. Hunting females move in short rapid flights between locations that the
wasps investigate with nervous searching movements accompanied by flicking the wings
and bobbing the abdomen (powell 1967; Kurczewski 1969). They generally search for
spiders beneath stones, sticks and leaves; in cracks and insect burrows; and in spider webs
found in axils oflow-growing plant stems. Mter flushing a spider, the wasp pounces on
the prey, and while holding the prey with its legs, stings it on the underside of the body.
Trypoxylon are agile fliers, but slow and awkward when walking or running. For
this reason, these wasps probably hunt while in flight. Trypoxylon hover close to the
ends of branches, leaves, and along surfaces of plants and buildings. During these flights
they tap the surface with their antennae and investigate small protuberances and small
spots that contrast against the background. Probably the tapping and air currents will
disturb spiders that are resting or maintaining cryptic postures outside of their webs.
C. Griswold (pers. commun.) has found that many such spiders will move when lightly
touched or exposed to a slight puff of air. The movement may be detected by the wasp
which then can pounce on the spider and sting it. Trypoxylon politum may use the
same methods as Chalybion when hunting orb weavers. Rau (1944) observed one in-
dividualland in two webs. In each web it waited quietly for 0.5-4 min before pulling
the web with its legs. As with Chalybion, this species never became entangled.
3.3 Factors Influencing Prey Preferences of Sphecids
Most sphecids will capture many kinds of spiders, including those that use webs to
capture prey, as well as those that pursue or ambush their prey. Even wasps that prey
predominantly on one family of spider occasionally capture other spiders that have
totally different habits. In most cases sphecids probably take any spider of appropriate
size that they encounter. Among exceptions is S. assimile; these wasps seem to ignore
orb-weaving araneids of the genus Gasteracantha while preying heavily on several other
orb-weaving genera (Freeman and Johnston 1978a). Gasteracantha possess a spiny ab-
domen. Certain types of spiders are seldom or never taken because they dwell in micro-
habitats that the wasps do not frequent. For example, there are no records of mygalo-
morphs ever being among prey of sphecids. Likewise, ground-inhabiting spiders, such
as Lycosidae, are seldom preyed upon, except by some Miscophus and Pison.
One of the most important factors influencing prey preferences is the relative
abundance of different kinds of spiders. The abundance of each prey species changes
throughout the season in temperate climates and varies from one region to another.
Muma and Jeffers (1945) observed seasonal changes in the prey of three spider hunters.
Among prey of S. caementarium, Araneidae predominated when the wasps first began
nesting, but declined later on. Anyphaenidae and Oxyopidae also decreased in abun-
dance, while Thomisidae and Clubionidae increased. Chalybion californicum captured
about the same number of Theridiidae as other families early in its nesting season, but
later it took mostly theridiids. Trypoxylon politum took the same kinds of spiders,
principally Neoscona and Araneus, throughout its nesting period, but each cell was
318 R.E. Coville: Spider-Hunting Sphecid Wasps
provisioned with more smaller spiders early in the season and fewer larger ones later
on.
The kinds of spiders captured also depends upon where the wasps hunt. Based upon
its exclusive use of Senoculidae, Trypoxylon xanthandrum probably hunts along the
underside of branches where the spiders flatten themselves during the day (Coville and
Griswold 1983)_ Sceliphron prey heavily on spiders such as Araneidae, Clubionidae,
Salticidae, and Thomisidae, and occasionally take other spiders as well (Freeman and
Johnston 1978a; Grandi 1961; Muma and Jeffers 1945). Based upon the habits of their
prey and observations of Freeman and Johnston (1978a), Sceliphron probably hunt
among flowers, open vegetation, and the outside of human habitations. Many Try-
poxylon probably do the same thing, since their prey preferences are similar to those
of Sceliphron. Other species such as T. superbum (Coville and Griswold 1984) and
T. bicolor (Yoshimoto 1964) specialize on Salticidae. These wasps probably search the
upper surface of foliage for moving spiders. AmongPison a few species such as P. strandi
(Masuda 1939; Tsuneki 1970) prey heavily on Thomisidae, suggesting that these wasps
hunt primarily around flowers. Ground-nesting wasps such as M cali/ornicus (Powell
1967) and Pison auriventre (Evans 1981) appear to search on the ground because their
prey consist of Lycosidae. M. cali/ornicus also preys heavily upon the ground-dwelling
salticid Pellenes.
Individual wasps of the same species may exhibit different prey preferences because
they hunt in different areas, exploit aggregations of spiders, or become conditioned to
certain types of spiders or hunting behavior. For example, of two female T. tenoctitlan
nesting next to each other, one preyed heavily on Araneidae and the other took main-
ly Salticidae (Coville and Coville 1980). In the case of C cali/ornicum, Muma and Jef-
fers (1945) found nests usually contained theridiids and other spiders found in webs
near the ground and around human habitations, but in some cases only foliage-inhabit-
ing araneids were used.
Prey preferences and patterns of predatory behavior other than the examples men-
tioned above exist. Iwata (1976) contains several tables that attempt to summarize the
prey preferences of sphecids. Publications providing extensive data and reviews on prey
preferences of Trypoxylon include Krombein (1967), Matthews and Matthews (1968),
and Coville (1982). Kurczewski (1969, 1982) has summarized most of what is known
of Miscophus. Evans (1969,1970) has recorded the only prey records for Pisanapsis,
and Evans et al. (1980) and Evans (1981) provide additional information on Pisan.
Prey records consist of samples from only one or a few cells for most sphecids that
have been studied. In most cases small samples are insufficient to form a basis for
generalizations.
4 Conclusion
Much remains to be learned about the ecology and behavior of spider-hunting sphecid
wasps. The habits of more than 90% of the species are unknown, yet many of these
are potentially quite easy to study. The mud-daubing species commonly occur around
human habitations and species that nest in preexisting cavities will readily nest in trap
nests. Nests of these wasps can provide vast collections of spiders, including species
that are rarely collected. Study of these wasps should not be confined to hymenopterists.
Only arachnologists can provide the knowledge of spider ecology that is required for
an understanding of the wasp's predatory behavior.
III The Biology of Acroceridae (Diptera):
True Endoparasitoids of Spiders
EVERT I. SCHLINGER 1
CONTENTS
1 Dipterous Parasitoids of Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
2 Acroceridae.............................................. 320
3 Adult Flies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
4 Larvae................................................. 323
4.1 Host Finding .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
4.2 Second to Fourth Instar Larvae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324
5 Host-Parasitoid Relationship .... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
1 Dipterous Parasitoids of Spiders
The dipterous parasitoids and predators of spiders are restricted to a few families of
Diptera and vary considerably in their mode of action. Members of the families Droso-
philidae, Phoridae, Chloropidae, and Sarcophagidae are known to predate on eggs of
several families of spiders (Clausen 1940; Eason et al. 1967). Recently, the first endo-
parasitic member of the family Tachinidae, Lypha sp., was reported to attack the fos-
sorial spider A typo ides riversi in California (Vincent 1985). The only truly coevolved,
host-restricted, parasitoid family of Diptera known to attack spiders is the endoparasitic
family Acroceridae (Cyrtidae of authors). Although I am convinced that acrocerid
larvae attack and develop successfully only in true spiders (Araneae), I should note the
case cited by Sferra (1986) where the first instar larvae of Pterodontia [lavipes were
found internally in the bodies of several types of Acari. However, because of the small
size of the hosts, there is no possibility that this parasitoid could develop further. The
only real competitor to acrocerids for this endoparasitic mode of life appears with cer-
tain nematodes (Poinar, this volume). In fact, Vincent and Schlinger (unpublished notes)
recently found a case where the spider, Atypoides riversi was serving as a host for both
a nematode and an acrocerid (Eulonchus) larva simultaneously. Which one of these
parasitoids (if any) would have been the successful competitor was not determined due
to the early dissection of the host.
1 Department of Entomological Sciences, University of California, Berkeley, CA 94720, USA

320 E.l. Schlinger
2 Acroceridae
The most recent overall review of the family Acroceridae was presented in the Manual
of Nearctic Diptera (Schlinger 1981). In order to conserve space, I have chosen to cite
references mostly published since 1938 when Millot carefully cited and reviewed all
the limited, earlier works. Plomley (1947) again reviewed the older references, and all
host and biology references to the most common genus Ogcodes were cited in Schlinger
(1960). A summary to the more recent and important bio-ecology references to Acro-
ceridae is as follows: King 1916 (Pterodontia); Vellard 1934 (Exetasis); Millot 1938
(Acrocera and Ogcodes); Clausen 1940 (Ogcodes); Jenks 1940 (Ocnaea); Dumbleton
1941 (Ogcodes); Plomley·1947 (Ogcodes); Schlinger 1952 (Opsebius = Turbopsebius);
Baerg 1958 (Lasia); Lamore 1960 (Ogcodes); Schlinger 1960 (Ogcodes); Capelle 1966
(Ogcodes); Eason et a1. 1967 (Ogcodes, Lasia, and Acrocera); Schlinger 1969 (Acro-
cera, Ogcodes and Pterodontia); Coyle 1971 (Eulonchus); von Eickstedt 1971 (Exe-
tasis); Schlinger 1972 (Exetasis); Humphreys 1976 (Ogcodes and Pterodon tia); Pichka
1977 (Ogcodes); McQueen 1978 (Pterodontia); Schlinger 1981 (general review); Holl
et al. 1982 (Ogcodes); Vincent 1986 (Eulonchus).
The family Acroceridae is cosmopolitan except for oceanic (volcanic) islands and
polar regions. It contains 380 species in 55 genera, with more than 200 additional new
species to be described. The family Acroceridae contains three subfamilies: Acroceri-
nae, Panopinae, and Phllopotinae. A world list of known hosts for acrocerid flies is
given in Table 24.
Table 24. World list of known host spider taxa of families with associated reared acrocerid parasito-
ids (new records are indicated by an asterisk)
Spider family Spider species Parasitoid taxa
Mygalomorphae
Ctenizidae Cteniza ariana Astomella hispaniae
Aptostichus standfordianus* Eulonchus s. smaragdinus
Bothriocyrtum californicum Ocnaea smithi
Antrodiaetidae A typo ides riversi Eulonchus sp.
Antrodiaetus unicolor Eulonchus marialiciae
Antrodiaetus sp.* Eulonchus sp.
Aliatypus sp. * Eulonchus tristis
Migidae Moggridgea crudeni Astomella capen sis
Dipluridae* Diplura cousini Lasia ecuadorensis
Theraphosidae Phrixotrichus roseus Arrhynchus maculatus
Chaetopelma sp.* Astomella gravis
Lasiodora klugi Exetasis eickstedtae
Grammostola actaeon Exetasis sp.
Dugesiella hentzi Lasia purpurata
Aphonopelma sp. * Ocnaea sp.
Araneomorphe
Plectreuridae * Kibramoa sp. Acrocera arizonensis
Segestriidae * Ariadna maxima Sphaerops appendiculata
Araneidae Araneus sericatus Pterodontia [lavipes
Mecynogea lemniscata Ogcodes dispar
Zygiella x-notata Ogcodes fumatus
The Biology of Acroceridae (Diptera): True Endoparasitoids of Spiders 321
Theridiidae Steatoda p%~~mara Ogcodes pallidipennis

Theridion ovatum v. lineatus Ogcodes gibbosus
Lycosidae Geolycosa domifex Pterodontio [lavipes
Geolycosa godeffroyi Ogcodes basalis
Geolycosa godeffroyi Pterodontia melli
Lycosa "helluo" Ogcodes pallidipennis
Lycosa ocreata Acrocera fasdata
Lycosa pretensis Pterodontia [lavipes
Lycosa pullata Ogcodes pallipes
Lycosa sp. Acrocera sp.
Lycosa sp. Acrocera orbicula
Lycosa sp. Ogcodes pallidipennis
Pardosa banksi Ogcodes eugonatus
Pardo sa distincta Ogcodes eugonatus
Pardosa lapidicina Ogcodes pallidipennis
Pardosa milvina Ogcodes eugonatus
Pardosa saxatilis Ogcodes pallidipennis
Pardosa sternalis'" Acrocera convexa
Pardosa sternalis Ogcodes adaptatus
Pardosa sternalis Ogcodes eugonatus
Pardosa sternalis Pterodontia misella
Pardosa tuoba Ogcodes adaptatus
Pardosa utahensis Ogcodes rufoabdominalis
Pardosa vancouveri'" Acrocera convexa
Pardosa sp. Acrocera convexa
Pirata sedentarius Ogcodes dispar
Tarentula barbipes Ogcodes pallipes
Tarentula kochi Ogcodes melampus
Trochosa sp. Ogcodes gibbosus
Agelenidae Agelenopsis naevia Turbopsebius sulphuripes
Agelenopsis oregonensis'" Acrocera bakeri
Agelenopsis sp. '" Ogcodes dispar
Barronopsis sp.'" Acrocera bimaculata
Hololena curta Ogcodes adaptatus
Hololena curta Ogcodes pallidipennis
Hololena curta Turbopsebius diligens
Hololena sp. Acrocera subfasciata
Rualena sp. '" Turbopsebius diligens
Tegenario agi/is Acrocera sanguinea
Tegenario agilis Acrocera trigramma
Amaurobiidae Amaurobius sylvestris Acrocera fasciata
Amaurobius sp. Ogcodes pallidipennis
Exlinea sp. '" Holops cyaneus
New genus'" Megalybus pictus
New genus'" Megalybus n.sp.
Psechridae Matachia ramulicola Ogcodes brunneus
Oxyopidae Oxyopes salficus Ogcodes dispar
Oxyopes salticus Ogcodes eugonatus
Oxyopes lineatus Ogcodes fumatus
Anyphaenidae Wulfia saltabunda Ogcodes borealis
Miturgidae'" Machadonia sp.? Thyllis cra ssa
322 E.I. Schlinger
Clubionidae Clubiona putris Ogcodes pal/ipes

Clubiona sp. Ogcodes pal/ipes
Clubiona sp. Agrocera orbicula
Clubionidae Trachelas mexicanus* Ogcodes pallidipennis
Gnaphosidae Herpyllus sp. Ogcodes pallidipenn is
Zelotes sp. Ogcodes gibbosus
Zelotes sp. * Acrocera melanderi
Philodromidae Philodromus sp. Ogcodes adaptatus
Thomisidae Diaea sp.* Ogcodes nitens
Xysticus cunctator Ogcodes melampus
Xysticus luctuosus Ogcodes pal/ipes
Xysticus montanensis Ogcodes borealis
Xysticus sp. Ogcodes eugonatus
Salticidae Aelurillus insignitus Ogcodes pal/ipes
Aelurillus insignitus Ogcodes varius
Cobanus sp. * gen.sp.nov.near Terphis
Cosmophasis bitaeniata Ogcodes basalis
Heliophanus sp. Ogcodes zonatus
Heliophanus sp. Ogcodes pal/ipes
Metaphidippus aeneolus* Ogcodes borealis
Metaphidippus manni* Pterodontia vix
Metaphidippus sp. * Acrocera bulla
Metaphidippus sp. * Ogcodes boharti
Pellenes hallani* Ogcodes eugonatus
Phlegra fasciata Ogcod es pallipes
Phidippus ardens* Ogcodes boharti
Phidippus femoratus* Ogcodes boharti
Phidippus formosus* Ogcodes adaptatus
Phidippus johnsoni* Ogcodes boharti
Phidippus johnsoni* Ogcodes eugonatus
Phidippus principes* Ogcodes eugonatus
Phidippus rimator* Ogcodeseugonatus
Phidippus sp. * Acrocera sp.
Sassacus sp. * Ogcodeseugonatus
Lyssomaninae gen.sp. * Ogcodes guttatus
3 Adult Flies
Adult acrocerid flies (Fig. 112) are morphologically diverse, sexually dimorphic, small
(2 mm) to large (30 mm), often mimetic to bees or wasps, and may be dull, spotted, or
brilliantly colored in blue, green, red, purple, or shiny black. They often have long
mouth parts extending twice the length of the body (e.g., Lasia and Euionchus) , and
are often very specialized pollinators; or they may appear to have no mouth parts at
all (Acrocera, Ogcodes). Mating usually takes place in flight, and being pro-ovigenic,
they begin to lay up to 5000 eggs soon after mating and may continue during the fol-
lowing 2 to 10 days. The tiny, pear-shaped, black, microtype eggs are deposited either
in flight upon the ground (Euionchus), upon dead branches (Ogcodes) , upon tree trunks
The Biology of Acroceridae (Diptera) : True Endoparasitoids of Spiders 323
Fig. 112. Adult fly , Ocnaea boharti

Schlinger, (Acroceridae), a parasitoid of
mygalomorph spiders
(Pterodontiaj, or upon grass stems (Acroceraj . Adult longevity is usually rather short,
lasting from 3 days to about 1 month. There is only one generation per year (rarely 1.5)
with the acrocerines on their araneomorph hosts; but it appears that in the case of
many pan opines , there may be only one generation every 5 to 10 years due to the
longer immature stages of their mygalomorph hosts.
4 Larvae
4.1 Host Finding
Acrocerid eggs hatch in 3 to 6 weeks giving rise to small planidiallarvae. These larvae
are most often not directly associated with their intended host spider and must actively
seek out and locate their hosts. Most species of acrocerids have planidia with well-
developed hairs and spines, and a caudal suction disk which allows the larvae to crawl,
jump, attach to a substrate , or move about in geometridlike, larval movements. An
exception to this type of pia nidi urn occurs in the genus Acrocera, where the larvae are
nearly hairless, lack the heavy sclerotized plates, and are able to move only in a crawling
fashion (Schlinger, unpubl.) . The proper host spider either has to come in contact with
the planidium, in which case the planidium attaches to the leg of the spider, or the
larvae must seek out the host. This latter case occurs when a fossorial spider is involved.
In the case of Ogcodes adaptatus (Schlinger 1960), a common parasitoid of Pardosa,
once the planidium attaches to the spider's leg it becomes motionless until the spider
moves. Once the spider moves, the planidium moves up the leg slowly and surely as
long as the spider moves, but it stops abruptly when the spider halts. This movement
continues until the larva moves off onto the cephalothorax or the dorsal base of the
spider's abdomen . The larva usually then moves to the most anterodorsal part of the
abdomen (conveniently out of reach of the spider's legs), and proceeds to cut a tiny
hole in the abdomen for its entrance. This may take as long as 24 h to complete, during
which period the spider appears quite agitated. Larvae have been observed also to enter
324 E.I. Schlinger
the spider through leg joints, but eventually all larvae end up in the book lung area of
the spider and attach themselves in the air pockets between the lung lamellae, and
breathe outside air. The first instar planidium then apparently goes into a state of
physiological diapause. In araneomorph hosts this diapause is usually broken in the
spider's penultimate stage. Thus, this diapause lasts from 6 to 9 months. In mygalo-
morph hosts this diapause may be much longer, possibly as long as 10 years! It should
be noted that while the endoparasitoid phase of these hypermetamorphic larvae may
be oflong duration, the free-living phase of these larvae may last only 1 to 3 weeks.
4.2 Second to Fourth Instar Larvae
The simple linear second and the more rounded, undifferentiated third instar larvae
of acrocerine parasitoids in araneomorph hosts are not easily observed in the spider's
abdomen. The mouth parts and head capsules easily distinguish these stages, if they
can be extracted from the dead spider, but they are often lost in the ensuing develop-
ment of the mature feeding larvae. The linear second instar larvae of panopine para-
sitoids (Eulonchus spp.) in mygalomorph hosts (Atypoides and Antrodiaetus) have
recently been observed (Schlinger, unpubl.), and appear to be a stage of short dura-
tion. The third instar panopine larvae are more easily seen through the integument of
the living hosts. In mygalomorph hosts this larva may be clearly seen for up to two or
more years. These larvae often attach themselves to the book lung for breathing pur-
poses as in the first instar, but occasionally they may be found attached to the integ-
ument of the abdomen at other areas of its surface, leaving behind a spiracular plate
remnant upon emergence from the spider (Schlinger 1972 and unpubl.).
The fourth instar, or mature larva, is the active feeding and destructive stage. In
acrocerine larvae this stage lasts from 24 to 48 h, while in panopine larvae this stage
may last for a week or more. The host spider in any case does not appear to die until
about 12 h before the emergence of the parasitoid. Feeding by this larvae is active and
often all of the spider's tissues from the legs, cephalothorax, and abdomen are con-
sumed. The last several hours of feeding is accomplished while the posterior end of the
parasitoid's abdomen is extruded from the spider. Larval emergence usually occurs at
the epigastric furrow area of the spider, e.g. Ogcodes, Turbopsebius, and Pterodontia),
but in the case of Acrocera, larvae always emerge through the dorsum of the spider's
abdomen. About 24 h prior to the active fourth instar's lethal attack on the host spi-
der, the spider most often spins a "protective" molting web as if it was going to molt.
It is this web which the naked, mature, acrocerid larva has to locate and affix itself to
as it exists from the spider to successfully pupate. The ventral side of the parasitoid's
abdomen contains a series of transverse rows of minute, hooked spines. These spines
facilitate the attachment of the larva's heavy body to the "premolting" web. The larva,
after attaching itself to this silk, ventral side up, remains in this stage for 1 or 2 weeks
before passing a large meconial pellet and becoming a pupa. The head of the pupa con-
tains a set of clublike devices, structurally distinct for each genus. These clublike knobs
hold the pupa firmly in the web during pupation, but more importantly, they securely
fasten the pupal skin to the web during ecdysis. Adult flies usually emerge about 2 to
3 weeks after pupation.
The Biology of Acroceridae (Diptera) : True Endoparasitoids of Spiders 325
Fig. 113. Recently killed mature theraphosid Grammostola sp., showing the many emerged mature
larvae of it's parasitoid, Exetasis eickstedtae, from Brazil. (Original photography by Lou Sorkin)
Fig. 114. Dying segestriid spider, Ariadna maxima, with mature, externally feeding acrocerid larva
(a) Sphaerops appendiculata, on the spider's dorsum, collected in Santiago, Chile. Both are inside
the silken tube prepared by Ariadna . (Original sketch by M.E. Irwin; redrawn by M.E. BuegJer)
326 E.!. Schlinger
All reported acrocerid larvae are endoparasitic and while they are usually solitary,
up to two, three, four, or even as many as fourteen larvae may emerge from a single
host spider (Fig. 113). The only exception to this total endoparasitic trait is the newly
reported parasitoid, Sphaerops appendiculata, which I have reared many times from its
host,Ariadna maxima, from Chile. Figure 114 shows the mature larva feeding externally
on the living spider in its tubelike retreat. These larvae may feed up to 3 weeks in this
ectoparasitic mode before killing their host. At first glance, these larvae appeared similar
to certain ectoparasitic polysphinctine ichneumonids, which are well-known spider
parasi toids.
5 Host-Parasitoid Relationship
The complete world list of known acrocerid-spider relationships now consists of 111 spe-
cimen rearings representing 54 species of Acroceridae. Of these, 68 specimens have
been recorded from the first rearing (Menge 1866) to the present. Included here are
43 new rearings representing 22 new species records which I have gathered mostly since
the publication on Ogcodes (Schlinger 1960). For the first time rearings of Megalybus
(from Chile), Thyllis (from South Africa), and a new species near Terphis (from Costa
Rica), all of the subfamily Philopotinae, are recorded. Also the first records of the
Chilean genera Holops and Sphaerops are given. Further new records are cited for the
genera Acrocera, Astomella, Eulonchus, Exetasis, Lasia, Ocnaea, Ogcodes, Pterodontia,
and Turbopsebius (Table 24).
While it may seem premature to make too strong a case out of the host-parasitoid
relationships, since there are records for only 22 families of spiders of the 96 listed by
Brignoli (l983), certain observations need to be made at this time.
It is clearly evident that the most primitive subfamily, Panopinae, are parasitoids
of the Mygalomorphae (Schlinger 1957, 1972, 1982). All parasitoids of the more ad-
vanced subfamily, Acrocerinae, are clearly associated with the more advanced Araneo-
morphae {loco cit.). Evidence is now available for the first time for the subfamily Philo-
potinae, showing some intermediate relationships, i.e., Megalybus and Thyllis, and one
unexpected an "off-line" relationship for a new genus near Terphis. The Amaurobiidae
are hosts for the Chilean Megalybus, and the Miturgidae serve as hosts for the South
African Thyllis. The peculiar, yet seemingly typical salticid, Cobanus, serves as a host
for a new genus near Terphis.
There is considerable host cross-over with a number of parasitoids of the genera
Acrocera, Ogcodes, and Pterodontia, all of the subfamily Acrocerinae, for they attack
spiders in a variety of families within the Araneomorphae. This has been known for
some time (Schlinger 1957, 1960), yet at the same time there appears to be consider-
able host restriction to members of the family Agelenidae (Rualena, Hololena, and
Agelenopsis) by the parasitoids of the North American genus Turbopsebius, also a
member ofthe Acrocerinae. Major cross-overs occur for such species as: Ogcodes adap-
tus (using species in Philodromidae, Lycosidae, and Agelenidae); Ogcodes dispar (using
species in Araneidae, Oxyopidae, and Lycosidae); Ogcodes eugonatus (using species in
Lycosidae, Oxyopidae, Salticidae, and Thomisidae), and Ogcodes pallidipennis (using
species of Lycosidae, Gnaphosidae, Agelenidae, Theridiidae, Amaurobiidae, and Club-
The Biology of Acroceridae (Diptera): True Endoparasitoids of Spiders 327
ionidae). In these latter cross-over cases it appears as though habitat has a clear pre-
ference of host selection, and/or host suitability has a much wider base for species of
certain parasitoid genera than for others. By way of example I would like to point out
that I was able to rear seven specimens of Sphaerops appendiculata from Ariadna
maxima in Chile (Fig. 114), but even though I reared many other acrocerid parasitoids
from more than 5000 spiders, I was unable to find this parasitoid in any other host.
Thus, there appear to be considerable host restrictions here and there throughout
the family Acroceridae even though this trait is often obscured by the most common
acrocerids, which seem to have almost unlimited host spider species. I am continuously
studying the more interesting and important host-parasitoid relationships with the hope
of using this coevolved biological system to elucidate and predict old, vicariant distribu-
tional problems. I would greatly appreciate any records or information concerning this
subject from anyone at anytime.
Acknowledgments. Many arachnologists have always cheerfully aided me in obtaining my spider

identifications for the many acrocerid rearings I have made during the past 35 years. Without their
considerable interest and perseverance in identifying those immature, dried, dirty little spider
"skins", we would not be in a position to begin to understand the true host relationships of these
unique spider parasitoids. Space does not permit my listing all the individuals and institutions in-
volved in helping me over the years, but I sincerely wish to thank the following taxonomists for the
many answered requests: J.A. Beatty, B. Cutler, G.B. Edwards, W.J. Gertsch, C.E. Griswold, H.W.
Levi, W.P. Maddison, N.I. Platnick, and V.D. Roth. I also wish to thank several of my graduate
students, who since 1963 have greatly assisted me in collecting and/or caring for the spiders (often
for several years in order to obtain either mature spiders or parasitoids), and for their consistent
intellectual stimulation and assistance: M.E. Bentzien, M.E. Buegler, C.E. Griswold, M.E. Irwin,
P.A. Rauch, and 1.S. Vincent.
E Environment and Life History
I Life Cycles and Diapause
MATTHIAS SCHAEFER 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
2 Ontogeny: Components of Life Cycles and Diversity of Patterns . . . . . . . . . . . . . . . 331
2.1 Embryogenesis and Larval Period . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
2.2 Juvenile Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
2.3 Adulthood and Maturity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
2.4 Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
3 Life Cycles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
3.1 Classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
3.2 Life Cycles in Seasonal Climates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
3.3 Hibernation and Aestivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
4 Mechanisms of Dormancy: The Control of Life Cycles . . . . . . . . . . . . . . . . . . . . . 340
4.1 The Role of Diapause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
4.2 Types of Diapause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
5 Life Cycle Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
1 Introduction
The developmental cycle of spiders concerns many aspects of morphology, physiology

and ecology: embryogenesis and postembryonic growth, life cycle patterns, adaptation
to seasonally occurring adverse environments, forms of dormancy, population dynamics
and the contribution of life cycle patterns to the structuring of animal communities.
In this short chapter I do not intend to give an exhaustive review of the extensive litera-
ture on life histories of spiders. Instead, I will concentrate on the following questions:
Which types of life cycles occur? How are spider life cycles adjusted to seasonal clima-
tes? How are life cycles controlled? What is the adaptive Significance of specific life
cycles?
2 Ontogeny: Components of Life Cycles and Diversity of Patterns
The complete development of an individual spider comprises four phases (Vachon

1957; Foelix 1982): (1) embryogenesis from the fertilized egg to the final inversion of
the germinal band; (2) a larval period with prelarva and larva as "incomplete stages"
1 Abteilung 6kologie, II. Zoologisches Institut der Universitiit, Berliner Str. 28, D-3400 Gottingen,
FRG
332 M. Schaefer
Egg
Development of gonads Ra te of development

Diapause Diapause
Reproductive ef for t
1st nymphal
Adul t ---------- instar
Juvenile period
Growth rate
Size increment
Number of instars
Diapause
Fig. 115. Influence of environmental factors on life-cycle phases. In rectangles: important life-cycle
characteristics
(Holm 1940); (3) "complete" juvenile or nymphal stages; (4) adulthood and maturity.
Environmental factors, such as temperature, photoperiod, humidity or availability of
food resources, may influence spider life cycles during these different phases of develop-
ment (Fig. 115).
2.1 Embryogenesis and Larval Period
Temperature influences the duration of embryonic development (Deevey 1949; Schaefer

1976a). The process of development of the fertilized egg to hatching of the spiderlings
occurs above a threshold of 0 0 C. Within certain limits the relationship between rate of
embryogenesis and temperature is linear (see Fig. 118b). Dependence on temperature
may be modified by diapause phenomena. However, embryogenic development is
generally independent from environmental variables other than temperature. The num-
ber of prelarval and larval stages may differ from species to species. Many spiders have
two prelarval and one larval stage (Holm 1940; Vachon 1957). The egg shell is left
during the prelarval or larval phase. The prelarval and larval periods do not contribute
much to the total time of development, and like embryonic development, are mainly
influenced by temperature.
2.2 Juvenile Phase
Further postembryonic development (summaries in Bonnet 1930; Deevey 1949; Levy

1970) is a rather variable period during a spider's life and is influenced by a series of
environmental events. Spiders grow by successive moults, although most spider species
Life Cycles and Diapause 333
do not moult after having attained adulthood. Exceptions are females of orthognath
species and the Filistatidae among the true spiders (Juberthie 1954).
Different size parameters have been used for describing growth: total length (e.g.
Vollrath 1983); length of the carapace (e.g. Jackson 1978a); carapace width (e.g. K.
Miyashita 1968a; Hagstrum 1971; Seligy 1971; van Dyke and Lowrie 1975; Schaefer
1976a); distance between the eyes (e.g. Jackson 1978a); fang length (e.g. Stradling
1978); ratio tarsal width/width of the pedipalp patella (e.g. Seligy 1971); length of
tibia I (Seligy 1971; Toft 1978b; Coyle 1985b) and other segments of the first leg
(Vollrath 1983). The latter author found an allometric correlation between several
morphological measurements in the spider Nephila clavipes. The parameter most widely
used as an index of the size of spiders is the length or width of the carapace.
The growth ratio (growth increment) is a species-specific character, which may be
influenced by diet, e.g. in Nephila clavipes (Vollrath 1983). Van Dyke and Lowrie
(1975) observed a linear increase of carapace width with instar number. The growth
ratio of Lycosa T-insignita did not obey Dyar's rule (K. Miyashita 1968a), which
postulates a regular geometric progression through a series of juvenile instars. It is a
well-known fact that spiders generally exhibit large intrapopulation size variations as
adults (Levy 1970). In some species this variation is large: e.g. Hypochilus sp. (Coyle
1985b) and Nephila clavipes (Vollrath 1983). Size is positively correlated with abund-
ance of food and the number of instars (e.g. in Lycosa T-insignita, K. Miyashita 1968a).
However, it cannot be ruled out that some of the size variation in adult spiders may be
genetically determined.
The number of moults depends upon the size of the spider, environmental influences
and the state of dormancy (see also Vollrath, this volume). Large spiders tend to have
more juvenile stages (Bonnet 1930; Levy 1970). However, there is no clear overall cor-
relation between spider size and number of moults. In the linyphiids, for instance, large
species (e.g. Linyphia triangularis) and small species (e.g. Saloca diceros) do not differ
in their number of nymphal instars (normally four to five). Generally, the number of
moults is variable (Deevey 1949; van Dyke and Lowrie 1975 ; Schaefer 1976a)(Fig. 116a).
On the other hand, there are remarkable cases of constancy in the number of stages.
Seligy (1971) observed a constant number of five immature instars (including the larva)
in the theridiid Enoplognathtl ovata. The me tid Metellina segmentata goes through for
juvenile instars (Toft 1983). Many linyphiids have four complete juvenile stages, e.g.
Linyphia triangularis (Schaefer 1976a) and Macrargus rufUS (Buche 1966; Christoophe
1977) (Fig. 116a). Apparently many species of the related families Araneidae, Metidae
and Theridiidae have this number of instars.
Temperature and humidity may influence the number of juvenile stages under sub-
optimal conditions. Diversity in food supply may contribute to variation in the number
of instars. This has been already observed by Bonnet (1930) in Dolomedes. In Latro-
dectus mactans spiderlings that were fed maximally required fewer instars to complete
development (Deevey 1949). The number ofinstars and its variation increased in Lycosa
T-insignita with decreasing food supply (K. Miyashita 1968a). Female Nephila clavipes
require an additional moult and are smaller at maturity when their food supply is re-
duced (Vollrath 1983).
The length of the intermoult phases and of total postembryonic development vary
considerably (van Dyke and Lowrie 1975; Schaefer 1976a; Coyle 1985b). During early
334 M. Schaefer
- - ----<, Ph.ph .
Ph. ph. J - - _ - C ! . phr.
t----+-----<---+---+-- ,2
CI. phr.
t___-+-----<--+__ 14 - 18 - 4
B
U1
J
_ _ _ _ _ _ _ - - ---I'Th .st r.
T
Th. ~tr .
, --o_ _--+--+---+--
..... 8 - 3
c
- J ~g~~~==-------I' P.p.
(!:l P p.
"0 10- 10-11-4 J =- 'T.m.
I
I I
. ~ A. p. J -.....:--_ - - - - I P. ct.
~I 9 - '0-2
....J A. a .
_--.--+--+---+--+_-+-- 6 - 8
J--
T. m. '0 _ -----1 B.g.
_--.---+-_--+-- 3 - " - ,3 S J- - __ ' F. b.
P. el.
O--~--+--_ 2 - 29 - 3
, 1-,-, i ,
M . ~. 20 100 200
T .....
, ~-_5 -2'-' b Time ( days)
Oe.r.
O--~--+-_+_ ,8
C. ~ .
0----+---+-_ '8 300
C. b.
III
~ 8 . g. ~
'C ' '2 o~200
:.§. D. ~ . CC
04> 1.0 :::;0
>-' .;: E 0-
." F . b. 00. 0
....J _-+---+_+-- 2 - ,2
c3~ 100 a.
st. !. 0.5 Q
_ _ _-+-_ '5
'C
~ '"!.
L. tr.
t---+-- - + _ _ 14
A . ~e . c o 5 10 15 20
J. ~::::::::::::::=_!8~__<__+~--<
t-- I I Temperature ( 0 C )
2 3 4 5 6 7 8 9 10
a Postembryonic ins tar
Fig. 116a-c. Postembryonic growth of spiders. a Number of instars. The numbers in the lines in-
dicate the number of adult individuals at that stage, from results of laboratory experiments (at
23°C and 16L:8D). b Length of inter moult phases. Experiments at 23°C and 16L:8D. c Duration
(e) and rate (x) of total postembryonic development. Experiments at 0 °e, 5°C, 10°C, 16 °e,
20.2 °C, 23°C and mostly at 16L:8D. (All data from Schaefer 1976a, 1977c).
Abbreviiltions: Ph.ph. = Pho[cus pha[angioides; C!.phr. = Clubiona phragmitis; Th.str. = Thanatus
striatus; P.p. = Pirata piratieus; A.p. = Aretosa perita; A.a. =Alopecosa aeeentuata; T.m. = Tetrag-
natha montana; P.c!. =Paehygnatha clereki; M.s. =Metellina segmentata; Oe.r. = Oedothorax retus·
us; e.S. = Centromerus silvaticus; e.b . = Centromerita bieolor; B.g. = Bathyphantesgraeilis; D.s. =
Drapetisea soeiillis; F.b. = Floroniil bueeulenta; St.!. = Stemonyphantes lineatus; L.tr. = Linyphia
trianguiaris ; A.sc. =Allomengea scopigera
nymphal development, moults succeed one another rapidly, but the time elapsing be-
tween moults and its variability increase with later stages (Fig. 116b). In Latrodectus
mactans longer instars resulted in a smaller number of instars and vice versa (Deevey
1949). The developmental rate is not directly size-dependent. Smalilinyphiids often
need 2 years for maturation (Toft 1976; Albert 1982), whereas some large species re-
quire only 1 year. Species with brood care tend to have a longer development time
(Toft 1978b). Thus, the length of the intermoult cycle depends on endogenous factors
and is modulated by temperature, state of hunger and other environmental influences.
The duration of total postembryonic development obeys the temperature summa-
tion rule (Fig. 116c). However, for postembryonic growth in Gongylidium rufipes
(duration of the ins tars 1- III) Baert (1980) found differen t mathematical descriptions:
the equation of Blunck (hyperbola), the equation of Janisch (exponential relation) and
linear regression. These equations were valid only within a certain temperature range.
The temperature threshold depends on the preferred temperature range of the spider.
For example, the littoral species Allomengea scopigera has a lower value than the dune·
inhabiting Tlumatus striatus (Fig. 116c). In the field both species developed more rapid-
ly than in the laboratory at comparable temperatures because fast-growing individuals
had a higher mean survival rate than slow growing individuals (Schaefer 1977c). Low
humidity prolonged instar length in Tegenaria atrica (Browning 1941).
The feeding regime has an important influence (e.g. Turnbull 1962 for Linyphia
triangularis; Vollrath 1983 for Nephila clavipes, Edmunds 1982 for Argiope spp.). The
influence of the rate of food consumption on the rate of development declines with
each consecutive instar (Turnbull 1962). For example, in Latrodectus mactans with
maximum food intake, the penultimate instar was the longest; at lower feeding rates,
the longest instar occurred earlier in development (Deevey 1949). A similar tendency
was observed for Lycosa T-insignita by Miyashita (1968a).
2.3 Adulthood and Maturity
Spiders usually become sexually mature soon after moulting to the adult stage. This is
true especially for males, which undoubtedly are under strong compulsion to transfer
sperm as soon as possible. Sexual maturity of the females depends much more on
environmental conditions. There may be a delay between final moult and sexual ma-
turity in diapausing females.
In most cases, males, which tend to have fewer moults due their small body size
(Foelix 1982), reach adulthood before females. However, there are exceptions. In
the winter-active species Macrargus rufUS and Centromerus silvaticus females mature
before the males (Buche 1966). If the male life span is much shorter, matings between
siblings are improbable because males meet females which originated from earlier co-
coons [found for Latrodectusmactans by Deevy (1949), for Thomisusonustus by Levy
(1970), for Nephila by M.H. Robinson and B.C. Robinson (1978)). Generally, however,
dispersal of males and females after having left the cocoon prevents sib-sib matings to
a large degree. Males mature earlier than females because they compete for matings.
Additionally, sexual dimorphism with small-sized males may lead to higher survival of
the males until they reach adulthood (B.C. Robinson and M.H. Robinson 1978).
Sex ratio of spiders in the field is normally characterized by a smaller number of
males. Austin (1984) found a sexual index of 1 :2.3 for Clubiona robusta. Huhta (1965)
observed an average sex ratio of 1 :3.5 in Finnish forests. The surplus of females can be
explained by the fact that males regularly have a shorter life span and die very soon
after mating.
The number of eggs per batch varies and is positively correlated with body size (e.g.
Jackson 1978, Phidippus; Austin 1984, Clubiona). Workman (1979) found a positive
correlation between biomass (production) of eggs and biomass (production) of females
in Lycosidae. If a female of Achaearanea lunata or Theridion varians produces a large
(small) number of eggs in her first cocoon, she will also have a large (or small) number
in the following ones (Toft 1978b). The number of egg sacs ranges from one to three
for many species (Schaefer 1976a). However, it differs from individual to individual.
336 M. Schaefer
Many species show a reduction in the number of eggs in consecutive egg masses. The
number of egg sacs reached six in Achaearanea lunata (Toft 1978b) with a reduction in
egg numbers in later cocoons. Austin (l984) found for Clubiona that the number of
eggs per cocoon does not decrease if the spiders are fed between consecutive oviposi-
tions.
2.4 Variability
The outstanding feature of spider development is its variability. For instance, Turnbull
(l962) found that the age at maturity of female Linyphia triangularis varied from 70
to about 200 days, and the size at maturity ranged from 2 to 25 mg, depending on the
feeding rate. In a Hypochilus population some individuals grew five to six times (in-
crease in linear dimenSions!) faster than their slowest growing contemporaries during
the first full year of growth following spiderling emergence (Coyle 1985b). Some spi-
ders have both early and late maturing adults (Ramousse 1973; Wise 1976, 1984a;
Jackson 1978a). Wise (l976, 1984a) observed variable rates of development in the
filmy dome spider Neriene radiata. Some spiderlings, emerging from eggs laid in the
spring, develop rapidly and mature within a few months; others, however, do not mature
until the next spring. Spring females are almost twice as heavy as summer females be-
fore eggs are laid and may lay more than twice as many eggs.
3 Life Cycles
3.1 Classification
Studies on spider phenology have led many authors to attempt a classification of spi-
der life cycles (Bonnet 1930; Tretzel1954; Juberthie 1954; Huhta 1965; Levy 1970;
Schaefer 1976a; Toft 1976; Albert 1982; Aitchison 1984a). It is not easy to reconcile
the different interpretations and classifications; perhaps an attempt to do so is not
even fruitful. Assigning a species to a certain life-history pattern is only a very crude
characterization (Toft 1978a). The main problem is the variability of life history events.
For instance, in the stenochronous, spring- and summer-maturing Achaearanea lunata,
which normally spends the winter in the juvenile stage, some adults are hibernating
(Toft 1978b). Additionally, classification of life cycles depends upon the method of
collection, definition of seasons and the latitude of the study area (Aitchison 1984a).
Three major components of a life cycle can be recognized: (l) duration of develop-
ment; (2) longevity and seasonal occurrence of adults; (3) period of reproduction (i.e.
mating and deposition of egg sacs) (Table 25).
Duration of Development. Mainly orthognath spiders in warmer climates need more

than 2 years for development. Most species in temperate climates are annual, annual-
biennial or biennial (Almquist 1969; Toft 1976; Aitchison 1984b). Hunting spiders in
Canada (Dondale 1977) and linyphiids in Denmark (Toft 1976) comprise annuals and
biennials (in many cases these are the larger species). In tropical and subtropical climates
many and even large species have two or more generations a year (Nentwig 1985b).
Table 25. Components of life cycles
Duration of development Longevity of adults Occurrence of adults Period of reproduction
Long High Spring Eurychrony

Short Low Summer Diplochrony
Asynchronization Autumn Stenochrony
of dd and 99 Winter
Even in temperate latitudes some species may have at least two generations a year, e.g.
Bathyphantes gracilis (Schaefer 1976a) or Pardosa pseudoannulata (Kawahara et al.
1974). Some individuals of the filmy dome spider Neriene radiata complete a genera·
tion in 3 months (Wise 1976, 1984a). As the duration of development depends upon
environmental conditions, especially upon temperature, life cycles vary geographically,
as many studies have demonstrated.
The wolf spider Pardo sa lugubris is annual-biennial, with a bimodal distribution of
female weights during the summer in Holland, but it is biennial in Scotland and Den·
mark (Edgar 1971a; Toft 1976). The metid Metellina mengei is annual-biennial in Den-
mark and annual further south. Biennial individuals are larger and have one additional
moult. There is a smooth transitional zone between annual-biennials and annuals (Toft
1983). Many linyphiids (including the erigonids) are biennial in the north (e.g. Denmark,
Toft 1976), but tend to be annuals in more southern regions (Schaefer 1976a). Saltic-
idae and Clubionidae are often biennials in Sweden and annuals in France (Almquist
1969). Achaearanea lunata and Theridion varians have two generations per year in
southern France (Juberthie 1954) and are biennial in Denmark; they must have annual
cycles in some areas of Central Europe (Toft 1978b). A similar trend is observed for
altitudinal gradients. At higher altitudes annual species change into biennial ones, e.g.
linyphiids (Albert 1982).
Lack of synchronization between males and females in the timing of maturation
(Levy 1970) occurs only in few species of spiders and is viewed as a basic category of
life-cycle types by Levy (1970).
Longevity and Seasonal Occurrence of Adults. In warmer environments and latitudes

spider species tend to mature more rapidly and to be more long-lived than in cooler
climates. This contrasts with the trend of shortened developmental time at higher am-
bient temperatures. Primitive (orthognath) spiders can survive up to 20 years. Depend-
ing on the type of life cycle, adults may occur in every season of the year. In temperate
climates most spiders are adult in spring (Schaefer 1976a; Toft 1976). Tretzel (1954)
and Huhta (1965) observed two maxima of adults, one in spring and a second in
autumn.
Period of Reproduction. I propose to use the terms "eurychronism", "diplochronism"

and "stenochronism" for the reproductive phase, and not for the seasonal occurrence
of adults, in contrast to other authors (e.g. Tretzel1954; Aitchison 1984b). Eurychro-
nous species have an extended period of reproduction. The prolongation of the repro-
ductive period is achieved either by females developing from consecutively laid egg
sacs, or by variations in the timing of sexual maturity. Stenochronous spiders reproduce
338 M. Schaefer
during a definite season of the year. A special category of stenochrony is winter activity
(in the strict sense as winter maturity with the period of reproduction during the winter
months). Diplochronous species are those having two periods of reproduction during
the year. This category includes species with autumn-breeding, long-lived females that
are in dormancy during winter and lay eggs again in spring, and species with two activ-
ity periods of males during the year (Aitchison 1984b). Therefore, this author suggests
including diplochronism with eurychronism.
Life cycles are represented by two extremes. On the one hand, long-lived eurychro-
nous species require a long period of development, whereas short-lived stenochronous
species with rapid growth and maturation represent the other extreme. Most life-cycle
patterns lie in between this continuum and show a high diversity (Levy 1970), making
classifications difficult.
3.2 Life Cycles in Seasonal Climates
The most comprehensive studies have been performed in temperate latitudes. For the
temperate climate of Central Europe, Schaefer (1976a, 1977a) differentiated between
five types of annual cycles, following and modifying the classification of Tretzel (1954).
The length of the reproductive period and the seasonal occurrence of stenochrony and
the overwintering stages are the most important distinguishing features.
Eurychronous Species. They do not reproduce at a defmite period and overwinter in

different stages. Developmental time may be short or long. Eurychronous species oc-
cur in a broad range of families, e.g. Linyphiidae (including Erigoninae) and Clubion-
idae. Maritime climates (e.g. in salt marshes, Heydemann 1961) favour eurychronism
in comparison to more continental habitats.
Diplochronous Species. These are mainly adult in spring and autumn and predominantly
overwinter in this stage. In autumn a part of the population can mate, but the true
reproductive season is confined to spring, when the eggs are laid (e.g. some Lycosidae).
Stenochronous Species Reproducing in Spring and Summer. These hibernate as imma-

ture spiders in different instars. Almost all families contain stenochronous species repro-
ducing in spring and summer. Typical families are Lycosidae, Thomisidae or Salticidae
(Dondale 1977). Apparently, this type of life cycle contains many biennial species
(Toft 1976).
Stenochronous Species Reproducing in Autumn. These hibernate in the egg stage (e.g.
some Linyphiidae and Araneidae). In some species (e.g. Argiope bruennichi) the spider-
lings hatch in late fall and remain inside the cocoon until spring.
Stenochronous, Winter-Mature Species. They reproduce in winter and are active during
this season. These species are mainly linyphiids.
The relative proportion of the types within the total spider fauna of a given area is
different. Stenochronous species reproducing in spring and early summer dominate
(Toft 1976; Dondale 1977; Stippich 1981; Albert 1982). In the maritime climate of
northern Germany (Schleswig-Holstein), Schaefer (1976a, 1977a) found that among a

total of 277 spider species 22.7% were eurychronous, 9.0% diplochronous, 13.4% either
eurychronous or diplochronous, 44.5% stenochronous spring-maturing, 6.7% steno-
chronous autumn-maturing and 9.0% winter-mature. Thus, only a low nwnber of spe-
cies hibernates in the safe, cold-resistant egg stage. The percentage of winter-mature
species is high, however, only 2- 3% of the spiders in the more continental climate of
Erlangen (Germany) are winter-mature (Buche 1966).
3.3 Hibernation and Aestivation
Summer, with periods of drought, and winter, with low ambient temperatures, are the
most adverse seasons in temperate regions. The tropics are characterized by the regular-
ity of wet and dry seasons. Mygalomorph species survive dry summer periods by aesti-
vating in the soil (Buchli 1970). Spiders prepare to hibernate in a different way. The
preferred hibernation site is the protecting litter layer (Schaefer 1976a, 1977b). Lipids,
carbohydrates and proteins are stored in autumn (Collatz and Mommsen 1974a; Collatz,
this volume). Life-cycle strategies are related to patterns of cold resistance and their
annual variation. In some species the supercooling points are lowered in autumn. Other
species hibernate in safe, resistant stages (mainly the egg stage). The degree of cold
resistance is related to the hibernation site (Kirchner, this volume).
Winter Activity. Some spider species are active during winter, moving, feeding and re-
producing in the coldest season of the year (Schaefer 1976a, 1977a;Aitchison 1984a,b).
Some spiders are even known to be active on and under snow at winter temperatures
below freezing (H~gvar 1973; Aitchison 1978; Huhta and Viramo 1979).
The term "winter-active" is ambiguous, e.g. some authors use this term for spiders
that can stay mobile (= "active") during winter (Huhta and Viramo 1979; Aitchison
1984a,b). Hence, ''winter-active'' species (in this sense) can occur in all classes of
phenology except the strictly annual, autumn-stenochronous species with overwinter-
ing eggs. A stricter definition includes only spider populations with their period of
adulthood and reproduction during winter, the so-called winter-mature spiders (Tretzel
1954; Schaefer 1976a, 1977a). The latter category contains mainly linyphiids (Tretzel
1954; Schaefer 1976a; Aitchison 1984a) (Figs. 93 and 94). However, juveniles from
many other spider families may be active (mobile) during winter (e.g. Lycosidae,
Clubionidae, Thomisidae, Tetragnathidae, Linyphiidae). Such winter-active spiders
may attain a high diversity. Huhta and Viramo (1979) found 102 species that were
active on snow and had come from almost any species of the community.
Winter-active and/or winter-mature spiders move, produce silk, feed, lay eggs and
moult during the annual cold season. Obviously the mobility of spiders being active
down to -8°C is not impeded by cryoprotectants, which may be due to the presence
of thermal-hysteretic proteins (Kirchner this volume). Winter-mature spiders build
webs in winter and spiders active in the vegetation or litter layer produce draglines
(Schaefer, unpublished observation). Winter-active spiders may feed at 2 °c and even
at -2°C or - 5 °c (Aitchison 1984a; this volume), since feeding may be necessary for
spider metabolism, although generally energy is derived from stored lipids (Collatz and
340 M. Schaefer
Mommsen 1974a; Collatz, this volume). The main food of winter-active spiders are col-
lembolans (Aitchison 1984a). Aitchison (1984a) observed infrequent moulting near
zero temperatures; one juvenile moulted at 2 0 C.
4 Mechanisms of Dormancy: The Control of Life Cycles
Principal controlling factors of spider life cycles are endogenous mechanisms, tempera-
ture and photoperiod. Mi.iller (1970) differentiated between quiescence and diapause
as a dormancy phenomenon, depending upon the rapidity of its termination in response
to favourable changes in the environment. Quiescence is a mere retardation or arresting
of development, which can be resumed without delay after reoccurrence of favourable
conditions. Diapause is defmed as a state of suppressed development, the manifestation
of which may be induced by environmental factors (Beck 1980). Its termination re-
quires a period of diapause development.
4.1 The Role of Diapause
Only a few authors have studied the occurrence of diapause in spiders (e.g. Dondale
and Legendre 1971; Baert 1978; Kurihara 1979; Bonaric 1980). Schaefer (1976a,
1977a) analyzed the influence of temperature and photoperiod on the development
of spider species with different life cycles. In the following sections some results of
these studies are given.
Eurychronous Species. Pachygnatha clercki develops rapidly at long-day and short-day

photo periods without significantly different rates (Fig. 117a). Other spider species
have a similar life cycle: e.g. Pholcus phalangioides, Bathyphantes gracilis and Clubiona
phragmitis (Schaefer 1976a). Embryogenesis, hatching, posthatching growth and re-
production are possible under differen t temperatures and photoperiod conditions. States
of diapause do not occur.
Diplochronous Species. Postembryonic growth of Arctosa perita does not depend on

photoperiod (Fig. 117b). At short day lengths, however, vitellogenesis and production
of egg batches is not possible. For short-day females oviposition occurs after a period
of chilling or transfer to long-day conditions. The life cycle of A/opecosa accentuata is
controlled by a similar mechanism (Schaefer 1976a): i.e. prevention of oviposition by
short-day photo periods. Short day lengths in autumn inhibit oviposition, but do not
prevent copulation. Often the females hibernate, preserving the sperm mass in their
receptacula seminis, and produce viable eggs in spring without further copulation. This
diapause mechanism secures the timing of egg laying at the beginning of the vegetation
period. In Tegenaria africa ovaries of young females start to grow as early as November
(Collatz and Mommsen 1974a). According to Dondale (1977), in autumn-maturing
species of Lycosidae and Thomisidae mating may occur in autumn or spring, but ovi-
position always occurs in spring.
Stenochronous Species Maturing in Spring and Summer. The wolf spider Pirata pira-
ticus develops more slowly at short day lengths (Fig. 117c). The retardation of post-
EurycNonism Diplochronism
.. .
1
P.clerck i Aper i ta
Oviposi tion
Adult
Ul timate maul t
stage
-----+I
16L :8D //
23° •.
/1 16L: 80
23" •.~
Postembryonic / ~Diap
growth / 8~ : 160 ~? · SL.,:~~O
Ha tch ot spiderl ings Ii; 23°
•.
1
23°
Egg and III
larval per iod
Oviposition ~~--~-----r----'
a 100 200 b 100 200 300
Time (days)
Stt>nochronism (spring) Stenochronism (autumn) StenocNonism (winter)
Ii 1/·
C.bicolor
II
P. piraticus F. bucculenta
8L:160 SL : 160
16L : 80
A :so
23° • • 23° .• 10° . •
~ ~80
•. 10·
/
23°
Oiap. (otter chill ing)
"'-----Oiap
c 100 200 300 d 100 200 300 e 100 200 300

Time (days)
Fig. 117a-e. Types of life cycles and diapause events for a eurychronous species (example: Pachy-
gnatha clercki); b diplochronous species (Arctosa perita); c stenochronous spring-maturing species
(Pirata piraticus); d stenochronous autumn-maturing species (Flaronia bucculenta); e stenochro-
nous winter-maturing species (Centromerita bicalor). Heavy line: development at long-day photo-
periods; thin line: at short-day photoperiods; Diap diapause. (Data from Schaefer 1976a)
embryonic growth is more pronounced at low temperatures (lO °c,Schaefer 1976a).

After a period of chilling, development to maturity is rapid, due to a shortening of
intermoult phases and a reduction in the number of juvenile instars. For other spiders
a similar retardation of postembryonic development at short photophases has been ob-
served: e.g. Tetragnatha montana, Oedothorax retusus and Thanatus striatus (Schaefer
1976a), Pisaura mirabilis (Dondale and Legendre 1971). Predominant mechanisms are
prolongation of intermoult cycles, increase in instar numbers (probably not for all spe-
cies), high developmental rate and reduction of the number of immature instars after a
period of chilling (not for all species). Sexual maturation, egg laying and embryogenesis
are not dependent on photoperiod. Temperature changes are influential by inducing
quiescence. Spiderlings emerge from the eggs in late spring or summer and grow rapidly
at high ambient temperatures. Further moults are retarded by short days in autumn,
accompanied by falling temperatures. Consequently, the individuals do not attain
maturity and hibernate as immatures. In spring longer days and rising temperatures
allow the completion of development. After the ultimate moult spiders mature rapid-
ly. Gongy/idium rufipes needs a photoperiod of 11 -14 h for the normal development
of ovaries (Baert 1978); this species is mature in April and produces egg sacs in May.
342 M. Schaefer
Overwintering nymphs of Lycosa T-insignita were retarded in their development when

they were collected in the field in November- December and kept at high temperatures
in the laboratory (K. Miyashita 1969a). The inhibition of development began to dis-
appear in individuals collected in January.
Stenochronous Species Reproducing in Autumn. The linyphiid Floronkl bucculenta

grows after hatching of spiderlings at long and short day lengths (Fig. 117d). Develop-
ment is somewhat retarded at long photophases by prolonged intermoult phases and
addition of another instar. Eggs do not hatch at high temperatures without a previous
period of chilling. A similar mechanism - retardation of egg development at high tem-
peratures - has been found in other autumn-mature spider species: Linyphkl trklngularis,
Drapetisca sockllis, Allomengea scopigera, Metellina segmentata (Schaefer 1976a).
However, an influence of photoperiod on postembryonic growth of these species has
not been observed. This mechanism prevents premature hatching of spiderlings in
autumn. For spiderlings of Agelena limbata, which overwinter in the egg sac, short-
day photoperiods induce diapause. Long day lengths and/or a period of chilling termi-
nate it (Kurihara 1979). This mechanisms is similar to that in stenochronous spring-
and summer-maturing species.
Stenochronous, Winter-Mature Species. Oviposition and complete embryogenesis of

Centromerita bicolor are possible at a wide range of temperatures (23 0 C to 0 0 C)
(Fig. 117e). Photoperiod does not influence the rate of postembryonic development.
The control of the development cycle of a closely related species, Centromerus silva-
ticus, is similar: slow postembryonic growth at high temperatures and a low tempera-
ture optimum for mating, oviposition and postembryogenesis, lead to maturation late
in the autumn or early winter. Stemonyphantes lineatus is endowed with some well-
defined mechanisms of environmental responsiveness that postpone maturity to autumn
and winter: (1) The moulting intervals and the number of moults are reduced by short
days in autumn; (2) the males are ready to mate only after having experienced low
temperatures for some weeks; (3) under short-day conditions the ovaries of the females
only develop after exposure to cold. Even at long days the percentage of ovipositing
females is raised by incubating them at low temperatures (Schaefer 1976a).
4.2 Types of Diapause
In general, spider diapause mechanisms appear to be more primitive or imperfect than

corresponding phenomena in insects or mites (cf. H.J. Miiller 1970; Tauber and Tauber
1976; Beck 1980; V.K. Brown and Hodek 1983). However, these ontogenetic events
are effective in controlling life cycles of noneurychronous spider populations.
4.2.1 Egg Dklpause
At low temperatures normal embryogenesis without dormancy is possible from the

fertilized egg to the eclosion of spiderlings. At high temperatures the growth of the
embryo is completely arrested at an early stage of development (usually after the forma-
F. bucculenta -.
...c
250 250
II
II>
01
01
"10
o
o
c:
.2
c..
~ 5 25°C ~./
.- .... .-.-.
~
o
..
u
c:
Oiapause >/< Postdiapause
g
01
25"C,..
~.-.-.-.-. -.
:-.-.-.-.--'-' 5°C
.-.-.- .-.- ......
,,/
b 5 10 15 20 25 cS 0 N 0 J F M A
Temperature ( 'C ) Time (months)
Fig. llSa-c. Egg diapause (species studied: Floronia bucculenta). a Influence of temperature on
embryonic development (numbers = stages of embryogenesis). b Development after a period of
chilling and termination of diapause. c Oxygen consumption of eggs collected in the field. (Data
from Schaefer 1976b)
tion of the genu band). This inhibitory effect can be annihilated by chilling at low
temperatures (usually ranging from - 5 DC to +10 dc). The diapause does not occur in
response to photoperiods experienced by the female parent.
The egg diapause of F/oronia bucculenta was analyzed in detail by Schaefer (1976b)
(Fig. 118a). Embryogenesis proceeded with a maximum rate at about +5 DC. Complete
inhibition of development occurred at temperatures above 10 DC . Diapause develop-
ment was most rapidly completed at 0 DC, after about 8-10 weeks. Physiogenesis is
not confined to a certain morphogenetic stage. Thus, physiogenesis and morphogenesis
are independent processes. During postdiapause, eggs develop rapidly (Fig. 118b). Egg
diapause is accompanied by low metabolic rates (Fig. 118c).
The suppression of egg development in spiders may be regarded as a rather primitive
diapause mechanism. The egg diapause is neither a genuine obligate diapause nor a
"parapause" (in the sense of J J. Miiller 1970), two fonus of embryonic donuancy
which are common among insects or mites (cf. HJ. MUller 1970; Jungreis 1978). Per-
haps opilionids have an egg diapause similar to that of the spiders studied (Bachmann
and Schaefer 1983).
4.2.2 Nymphal (Juvenile) Diapause
During late, postembryonic development, the intervals between moults are prolonged
and the number of instars increased by short days (stenochronous spiders maturing in
spring or summer) or long days (some spiders maturing in autumn or winter). However,
the inhibitory effect on the growth rate is highly variable. In some species the photo-
period cannot induce diapause after the juvenile spiders have been exposed to low tem-
344 M. Schaefer
P.piraticus
. P.mirabilis
Adult
'O
9
.", ... ~ .-. Ptaoperiod I5L :8D
{ ./ .-1"
stage : \ Prevailing
16L : 8~fo~/ ._. photoperiod
IS
/s •• ~
8L:16D
Hatch of ~~--r------'--""
......
Chilling
\\ .--Field c~itions
~~
.,f::::-- ...
,,0'"
!!:t
.. ....-.-.-::...-
spiderlings 100 200 300 N 0 J F M
a Time (days) b
Fig. 119a,b. Nymphal diapause. a Influence of temperature on the development of Pirata piraticus.
Further explanation see Fig. 117 . (Data from Schaefer 1976a). b Oxygen consumption of juvenile
individuals of PiSllura mirabilis. (Data from Dondale and Legendre 1971)
peratures for some time. In addition, cold may reduce the number of moults. In all
cases the diapause is terminated almost immediately after transferring the spiders to
long-day or short-day photoperiods, respectively , and thus may be classified as a facul-
tative dormancy (Fig. 119a). Apparently similar mechanisms operate when spiderlings
hibernate in the egg sac (Kurihara 1979). Food advances the development of larger
juveniles of Lycosa T-insignita during the incubation period (K. Miyashita 1969a).
Winter diapause is associated with low levels of ecdysteroid hormones (Bonaric , this
volume). In Pisaura mirabilis the incidence of diapause appears to be controlled by the
neurosecretory cells of the central nervous system (Bonaric 1980). During diapause,
metabolism is reduced and respiration rates are low, as demonstrated for Pisaura mi-
rabiUs by Dondale and Legendre (1971) (Fig. 119b). For first ins tar spiderlings of
Argiope aurantia in the egg sac, respiration rates remained constant at single test tem-
peratures during the overwintering period from 24 November to 11 May. Respiration
rates increased sharply prior to emergence in late May at the termination of diapause
(Riddle and Markezich 1981).
A photoperiodically induced retardation of development is regarded as quiescence
by some authors (Beck 1980; H.J. Muller 1970). Lees (1955), however, argues that it
may be classified as true diapause, because such inhibition of growth must be effected
by neurohormonal processes (see Bonaric, this volume). Larval or pupal diapause in in-
sects is most frequently confined to a definite ins tar and leads to a complete arrest of
morphogenetic developmen t. In many cases diapause is induced by certain photoperiods
and terminated by other photoperiods or by cold (Beck 1980). There are only few
examples in insects where diapause consists of a photoperiodically induced retardation
of development.
4.2.3 Reproductive Diapause
Some spiders do not reproduce at short-day cycles. The development of the female
reproductive system and the deposition of yolk is inhibited by this photoperiod. Prob-
ably this adult diapause is induced by photoperiods experienced by the spiders during
their juvenile stages. The diapause is terminated by transferring the females to long-day
Fig. 120. Reproductive diapause: influence of photoperiod on ovi- A. peri ta
,L
position of Arctosa per ita . Further explanation see Fig. 117. (Data Ovi-
from Schaefer 1976a) posit ion
Chil,ling 'Diap
Mating
100 200
Timeldays)
conditions or exposing them to cold . When the spiders develop at long-day photoperi-
ods, adult diapause is prevented (facultative diapause) (Fig. 120). Nothing is known
about eventual diapause events in males. Data on metabolism of diapausing females are
not available. Similar cases of photoperiodic control of adult diapause have been de-
monstrated for many insects and mites (Lees 1955; Beck 1968).
5 Life Cycle Strategy
A spider species or individual can allocate resources to growth, longevity and/or egg
production. A general theory of spider life-cyde strategy does not exist. Only specula-
tions on ultimate factors determining life cycles are possible. However, some general
patterns emerge. Rapid growth of individuals may occur at the expense of longevity
(Reed and Witt 1972). Spiderlings that postpone maturation tend to mature to a larger
size, which leads to higher female fecundity (Workman 1979; Wise 1984a). Brood care
increases the rate of development (Toft 1978b). Most spider species extend their re-
productive period by producing several egg sacs. Investment in reproduction is positively
correlated with longevity.
Life-cycle strategy should be related to the pattern of mortality. Generally, the
highest mortality occurs during growth of the freshly hatched spiderlings to larger im-
mature stages (Schaefer 1977b). Key factors for mortality may be, among other in-
fluences, low winter temperatures (Schaefer 1977b), predation pressure (e.g. McQueen
1983; Austin 1984) and low availability of food (van Wingerden 1977).
We may hypothesize that a short life cycle and early maturity are advantageous.
Thus, if conditions are favourable, many species exhibit rapid development (Vollrath
1983; Nentwig 1985b). On the other hand, stable and favourable conditions with high
survival rates allow slow development, high longevity of adults and iteroparity of fe-
males, as exemplified by mygalomorph spiders. In risky microhabitats with adverse
abiotic conditions and/or high predation pressure, spider species should have a short
life cycle and mature early. In some habitats stenochrony may be advantageous, be-
cause the seasonal occurrence of sexes is synchronized and location of mates is facili-
tated.
Several hypotheses exist to explain variability in developmental rate, size and/or
reproductive output. Most typically, phenotypic differences reflect the effects of varia-
tion in environmental factors. Differences in the rate of food supply can produce a
346 M. Schaefer
wide range in age and size at maturity (Deevey 1949; Bristowe 1958; Coyle 1985b).
Local variations in temperature lead to differences in developmental rate. However, it
is difficult to imagine that temperature conditions are sufficiently different within one
spider population to cause significant variation in development. Varying temperature
regimes may be experienced by populations in contrasting habitats or web spiders oc-
cupying different heights of vegetation. Variability might also reflect a genetic poly-
morphism, for example in the rate of development. However, the existence of such
genetic polymorphisms in developmental rate has not yet been convincingly demon-
strated for spiders (Wise 1986).
A crucial point is the proportion of the variability that is genetically determined
versus variation that results from a plastic phenotype responding to differences in
proximate limiting factors (Smith-Gill 1983; Wise 1984a). Wise (1986) found no
evidence of substantial genetic variation in developmental rate in Neriene radiata.
Variation in rates of growth and maturation may lead to increased chances of survival
in spatially and temperally heterogeneous environments. However, a passive response
of individuals to environmental variation constitutes phenotypic modulation which is
not necessarily adaptive (Smith-Gill1983). On the other hand, developmental "conver-
sion" (Smith-Gill 1983), as a switch in the pattern of development (in many cases as
diapause, Schaefer 1976a, 1977a) in response to environmental cues may be regarded
as an adaptation to a variable environment. Examples of the latter are mixed life-
history patterns of spider populations in seasonal climates (Wise 1984a). Size variations
in males may be determined by the degree of competition for females (B.C. Robinson
and M.H. Robinson 1978): small males maturing early are less likely to evoke predatory
attacks by the female; larger males may be superior competitors for mates, but may
suffer a lower mating frequency, if they mate too late. The smaller size of males may
be related to the need for lower investment in reproductive tissue, but higher allocation
of resources to searching behaviour.
Constraints on the life-history pattern include features of the abiotic habitat, oc-
currence of adverse seasons, food availability, predation pressure and interspecific
competition. These relationships allow predictions which can be tested in future studies
on spider ecology and life-cycle characteristics.
Relation to Features of the Abiotic Habitat. It is not clear to what extent spider life
cycles depend on habitat characteristics. Spiders - mainly Mygalomorphae - that in-
habit the comparably safe, favourable and predictable soil environment are long-lived.
Spiders of the vegetation layer tend to develop rapidly (Toft 1976, 1978a), because
they have to endure more pronounced fluctuations in microclimate and probably higher
predation pressure.
Adaptation to Adverse Seasons. Spiders have to be adapted to survive winter condi-

tions or drought, periods with physiological constraints and low prey abundance. Spe-
cies passing winter only in one stage are an exception (Almquist 1969). Some species
have a flexible cycle, others have a relatively fixed developmental pattern. In many
cases only certain developmental stages are adapted to survive winter conditions
(Schaefer 1976a). There are safe species with low winter mortality and unsafe species
with high winter mortality (Schaefer 1977b), perhaps reflecting different degrees of
adaptation to northern latitudes. Timing of oviposition in spring in stenochronous spe-

cies (and in diplochronous species; cf. Dondale 1977) is advantageous, because initial
development occurs in the climatically favourable seasons with high food resources.
Food Availability. Low metabolic rates of spiders allow them to live in unpredictable
environments. The size increment and time intervals between moults are determined
by the amount of food consumed during the intermoult interval (Vollrath 1983), thus
allowing a flexible response of the spider to varying degrees of food availability. Large
orb weavers tend to mature in autumn because of high availability of large prey organ-
isms during that season (Olive 1981a). In habitats with more food, annual instead of
biennial cycles are favoured.
Interspecific Competition. Spider communities contain a high number of species with

different life cycles. The temporally isolated populations use different food resources
and the idea is appealing that competition may be a structural feature of these com-
munities. However, they may result from independently evolved adaptations to seasonal
changes in limiting factors. A central hypothesis by several workers (Tretzel 1955;
Dondale 1977) is that the diversity of spider life cycles reduces competition within the
spider and arthropod community, as interactions between coexisting populations would
decrease survival. Winter activity is interpreted as the utilization of niches with relatively
little competition. However, there is no convincing evidence for an influence of inter-
specific competition on the formation of life cycles (Connell 1980; Simberloff 1982).
The overall importance of competition in structuring spider communities is question-
able (Wise 1984b). A crucial point in determining spider life-cycle strategy is the avail-
ability of resources, mainly food.
Acknowledgments. I thank Prof. D.H. Wise for fruitful discussions on spider life-cycle strategy.
II Dispersal: Ballooning and Other Mechanisms
ARTHUR E. DECAE 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
2 Why Disperse? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
3 Dispersal in Primitive Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
4 Ballooning...........................·.................... 351
5 Distribution Patterns in Spiders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
6 Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
1 Introduction
The meaning of the word dispersal both in a general and in a biological context, is some-
what equivocal. Confusion originates from the fact that dispersal may indicate either
the event of being dispersed or the act of dispersing (Longman 1978). In order to ob-
tain useful definitions for various migrational processes, Pielou (1979) distinguished
three kinds of dispersal. Her two classes of diffusion and secular migration, both con-
cerned with slow and gradual shifts in the distribution range of a species, are not relevant
here. Only the events falling into her classification of jump-dispersal are relevant to the
type of dispersal considered here. Quoting Pielou's definition: "Jump-dispersal is the
movement of individual organisms across great distances, followed by successful estab-
lishment of a population of the original dispersers' descendants at the destination. The
salient points are that the whole journey is completed in a short period of time, usually
very much shorter than the life-span of an individual; and that the journey usually takes
the dispersers across totally inhospitable terrain as, for example, when spiders are car-
ried by air currents far across the open sea." As interpreted here, dispersal does not
necessarily involve an extension of the original distribution range of a species. This is
in contrast to Platnick's (1976b) view, in which the use of the term dispersal is spe-
cifically restricted to indicate events of range extension.
Zoologisch Laboratorium Rijksuniversiteit Groningen, Postbus 14, 9750 AA Haren (Gn),

The Netherlands
Dispersal: Ballooning and Other Mechanisms 349
2 Why Disperse?
Travel, to some extent, is part of every individual organism's life even if this is only
due to the fact that offspring cannot occupy the exact location of the organism from
which it sprang. Generally, however, an individual will be "born" in an environment
favourable to its life requirements. What then motivates so many animals to leave their
native region and travel across inhospitable terrain in order to find new places to live?
After all, these pioneers take severe risks of failure during the course of their journey.
Basically, two hypotheses have been put foreward to explain dispersal functionally:
1. The founder hypothesis which is probably best defmed in the following quote
from CJ. Krebs (1978:34): "Natural selection will clearly favour an individual that
leaves a relatively crowded habitat and colonizes an empty one in which it can leave
many descendants." In this hypothesis emphasis is placed on the adaptive advantage
of colonization. In sexually reproducing species, one might particularly expect to see
this mechanism at work in fertilized females as these represent the optimal life stage to
initiate new populations. A prerequisite for this is that new suitable habitats become
regularly available at distances that can be crossed with a reasonable chance of success
by the dispersing animal. Dispersal behaviour in these females might be controlled by
some endogenous mechanism, probably associated with the reproductive system, and
this may occur irrespective of environmental circumstances.
2. The escape hypothesis sees dispersal as a reaction of animals to signals originating
in their environment indicating deterioration of living conditions. Under this hypo-
thesis, all life stages are expected to show dispersal behaviour in response to some
identifiable environmental change. This could occur irrespective of the overall stability
of the habitat or the proximity of other suitable living spaces.
Focussing on terrestrial arthropods, some groups, insects, mites and spiders, have
evolved highly effective means of dispersal by aerial travel. Conspicuously, in insects
(C.G. Johnson 1960) and mites (Brader 1972) aerial dispersal seems largely restricted
to adult forms, while the aerial spider fauna mainly consists of juvenile animals. Excep-
tions, however, occur widely, such as with ballooning Lepidoptera larvae (Brader 1972)
and adult linyphiid spiders (e.g. Duffey 1956). Superficially, it follows that dispersal
in insects and mites has largely evolved under conditions that act in accordance with
the founder hypothesis, while in spiders escape from unfavourable conditions might
have been the principal dispersal dqve, the real situation appears to be more complex.
Although some workers show a realistic influence of the founder principle (Duffey
1956; Southwood 1962; Greenstone 1982), others present evidence of modulating
environmental effects on dispersal behaviour (Richter 1970; Meyer 1977; van Winger-
den 1977, 1980; Legel and van Wingerden 1980).
Ultimately, two strategies can be followed to combat high environmental pressure
levels. One is to have sufficient dispersal capacity to avoid areas of excessive environ-
mental stress, the other being an increased habitat specialization in order to better with-
stand adverse conditions. The second strategy, however, through the necessity of in-
creasingly narrow specialization holds a high risk of extinction in an ever changing
world. The ability to retain a considerable degree of flexibility coupled to sufficient
powers of dispersal are fundamental properties of organisms in order to continue the
flow of life.
350 A.E. Decae
3 Dispersal in Primitive Spiders
All spiders package their eggs in silken "egg sacs". The number of eggs contained in
one egg sac, according to Bristowe (1958), may vary between 2 and 1000, with ap-
proximately 100 eggs per egg sac being probably average. The emergence of spiderlings
from the egg sac is rather simultaneous, temporarily resulting in a severe situation of
overcrowding on or very near the empty egg sac. As a consequence , activities of young
spiderlings are migratory in character and apparently aimed at finding sufficient in-
dividual "living space" (exceptions being the few species that exhibit forms of social
interactions, such as brood care). The most primitive method of achieving this goal for
spiders is probably to walk the distance it takes to find a suitable site.
Characteristically, the great majority of species contained in the primitive groups of
the Liphistiomorphae and the Mygalomorphae (collectively known as the Orthognatha)
use this method of dispersal. Indeed, these displacements usually involve the crossing
of totally inhospitable terrain and, therefore, may be termed "dispersal" under the de-
finition given in the introduction with one possible objection to the fact that, generally,
no great distances are covered. The majority of orthognath spider species are fossorial
in habit and spend their entire lives in places where conditions are very stable (Main
1976 :56). Nevertheless, due to behavioural adaptation, the orthognaths have colonized
a wide variety of general habitat types from rain forests to deserts. Activities, such as
dispersal, which takes them outside the protection of their dwellings, must be restricted
o
c
5cm
o
o
Fig. 121. Cluster of the (orthognath) trapdoor spider Bolostromus spec. as observed on a creek bank
in a tropical lowland forest in the Republic of Panama. Circles: spider burrows; hatched areas: rocks
to periods of favourable conditions (Main 1976:53). Due to the low speed of displace-
ment of the walking spiderlings and the limited time in which favourable conditions
prevail, many Orthognatha occur in so-<:alled "clusters" (Fig. 121). Typically, acluster
consists of a central, large spider nest inhabited by an adult female around which
numerous small nests, occupied by juveniles, radiate.
4 Ballooning
The limited dispersal potential, just described as characteristic for most orthognath
spiders, contrasts strongly to the great aerial dispersal powers sometimes credited to
the Labidognatha or "true spiders". As spiders have no wings, aeronautic displace-
ments in these animals cannot comfortably be termed flying, but instead are common-
ly known as ballooning (Fig. 122).
Many romantic, as well as scientific, descriptions of spiders floating through the air
on long buoyant strands of silk have been published (e.g. Comstock 1948; Bristowe
1958; Gertsch 1979). Adaptation to a more exposed way of life, the increased and
more versatile application of silk and a reduction in size as compared to the primitive
orthognath spiders have provided the conditions suitable for aerial dispersal in Labido-
gnatha. This does not mean, however, that ballooning is totally unknown in Ortho-
gnatha. Coyle (1983 , 1985a) has recently published reports on ballooning observed in
two North American orthognath species and a few Australian and one European species
Fig. 122A,B. Two methods observed to accomplish ballooning. A The spider stands on "tiptoes"
facing the wind and squeezes out a drop of silk which is expanded further by the wind. When the
pull on the threads from the air currents is strong enough, the spider will float off into the air. B The
spider drops from an elevated position on a dragline. The breeze will blow the spider further and
further away from its original stand until the dragline, after reaching a certain length, will break at
its attachment point and the spider drifts away with the wind. See Coyle (1983) for a more elaborate
description of both methods of ballooning
352 A.E. Decae
are also said to disperse in this way (Enock 1885; Main 1976). On the whole, however,
ballooning in Orthognatha must be regarded as exceptional. In Labidognatha, on the
other hand, ballooning is rather common. A literature survey carried out by Nentwig
(pers. commun.) mentions numerous species belonging to 22 families of labidognaths
that have been reported to balloon. Dean and Sterling (1985) took representatives of
no less than 19 families from aerial samples collected with the aid of suction traps in
Texas.
Apart from the questions regarding the adaptive function of dispersal, in general,
and the stimuli inducing dispersal in animals that were briefly discussed earlier, several
workers have discussed the power of ballooning as a dispersal mechanism. Observations
of ballooning spiders landing on ships more than 300 km from the nearest land (Gertsch
] 979) and the presence of spiders in samples of aerial plankton taken at heights up to
approximately 5 km (Glick 1939; Gressitt 1961; Gertsch 1979), inspired beliefs that
few geographical barriers can possibly obstruct the spread of animals possessing such
powerful means of dispersal (Platnick 1976a). Undoubtedly, ballooning is a very ef-
ficient dispersal mechanism and many remote places, such as isolated valleys and oceanic
islands, may have received their spider populations largely through this means of travel.
To regard ballooning as virtually unlimited in its ability to carry individuals success-
fully around the world must be thought of as rather unrealistic. To traverse long, con-
tinuous distances of uninhabitable terrains, such as oceans, spiders would have to float
at great heights for a considerable length of time. To find air currents sufficiently per-
sistent in one particular direction at a global scale, as would be required for intercon-
tinental spider ballooning, the animals would have to ascend into tropopausic altitudes.
This would be at elevations of approximately 8 km above the poles up until approxi-
mately 15 km across the equator at temperatures of around - 60 0 C (Rumney 1969: 16).
To have an idea of the minimum time involved in such travels, one may consider that
the jet route New York/Amsterdam is 5850 km (KLM distance tables) and that winds
at the acquired jet stream altitudes average 130 km/h (Rumney 1969). On this basis,
the journey would take 45 h. I don't know of any experiments that test spider en-
durance under tropopausic conditions, but given the extreme circumstances at these
altitudes, the chances of survival for any length of time are very small indeed. More-
over, if the particular spider that survives such a journey is not a fertilized female,
colonizing a new habitat can only occur after it has found a conspecific of the op-
posite sex that has survived the same journey, mated successfully, etc. Taking into ac-
count that most ballooning goes on at low altitudes (Glick 1939) and the fact that the
density of the aerial plankton decreases rapidly with increasing height (Fig. 123), there
is probably much realism in Platnick's (1976a) statement when he discussed the same
topic: "The point is not that such a series of events is impossible, but that even given
enormous spans of geologic time such an occurrence must be unlikely." However, the
maximum heights at which spiders have been collected, i.e. ±5 km and distances of
300 km away from the nearest land, although impressive in themselves, are far below
the requirements needed for intercontinental travel as described above.
The very fact that ballooning spiders were observed to arrive on ships, occasionally
300 km from the coast, but undoubtedly very common at shorter distances or in
harbours, gives us a good idea of how some species may become cosmopolitan in their
distribution. Ballooning spiders seem particularly suitable of becoming dispersed by
Fig. 123. Density decrease of aero-

height 1m
nautic arthropods with increasing
height based on data collected with 5000 I
I
the aid of airplane traps by Glick I
I
(1939) above Louisiana I
I
I
I
II
4000 I
I
I
I
II
I
I
I
I
3000 I
I
I
I
I
,
I
I
I
I
2000 I
I
\
I
I
I
,
\
\
\
1000 \
"
" .................
............ _------
------ --------
10 30 50 70
arthropOds collected ( NIh )
mediation of man, although, the huge spiders that sometimes stir up commotion among
workers unloading banana boats, indicate that ballooning is not necessarily involved.
5 Distribution Patterns in Spiders
That man may act as a very powerful dispersal agent for many species of animals and
plants, hardly needs to be elucidated. As far as spiders are concerned, the typical en-
demic faunae of areas that have long been isolated from northern human influences
are often "contaminated" by species with an apparent northern temporate origin
(Levi 1967a; R.R. Forster and L.M. Forster 1973; Main 1976; van Helsdingen 1977;
Platnick 1977; Unzicker 1977; Saaristo 1978; Holm 1980). Most probably such spe-
cies have reached these places traveling as "stowaways" on ships and planes. In fact,
it would be rather interesting to study the faunae aboard such carriers in relation to
species distribution.
Although it may be true that widely distributed species are generally of northern
temporate origin and that man has acted as the prime dispersal agent, some aspects
remain to be explained. For instance, human travels usually involve a return trip. Despite
this few species from the south have apparently settled in the northern countries
(for an exception, see R.R. Forster and L.M. Forster 1973 :178). If the phenomenon
is real, it may be partly explained by the much more restricted tolerance or greater
degree of specialization of inhabitants of the warmer parts of the globe. Another ex-
354 A.E.Decae
planation might be sought in the much more limited dispersal capacity of tropical and
southern species and consequently, the lesser susceptibility of being transported by
man. R.R. Forster and L.M. Forster (1973) claimed that few New Zealand species bal-
loon because it would be disadvantageous for island inhabitants having a high risk of
being blown into the sea. Due to the fact that land masses in the southern hemisphere
are all comparatively narrow, this might hold in general for southern species.
M.H. Robinson (l982a) discussed the possibility that aerial dispersal in tropical spi-
ders is very rare or even absent. This idea is inspired by the belief that tropical organ-
isms, in particular those inhabiting forests, show an exceptionally high degree of habitat
specialization, reflected, for instance, in complex interspecific or symbiotic inter-
actions (M.H. Robinson 1980). For discussions of latitudinal diversity gradients and
causal explanations, see Dobzhansky (1950); Simpson (1964); Pianka (1966); Stehli
et al. (l969); Pielou (1975) and the many references listed in these papers. Emphasis
on biological investment in habitat specialization and competitive ability in tropical
species may have evolved at the expense of dispersal capacity and, hence, would ex-
plain the rarity of tropical ballooning. However, as M.H. Robinson (1980, 1982a)
Table 26. Summary of results of a preliminary study of aeronautic dispersal in some tropical spider
species. Hatchlings from spider egg sacs collected in the tropical forest area around Gamboa (Pana-
ma) were released on a vertical branch that was placed in an exposed position. During conditions of
clear sky and light winds, these spiders were observed to either descend along the branch to the
ground or ascend and then mostly become airborne by either of the two methods illustrated in
Fig. 122
Family Species Pre ballooning behaviour a Actual

ballooning
2 3
Dipluridae Ischnothele guyanensis

Filistatidae Filistata insidiatrix + + + +
Scytodidae Scytodes longipes
Pholcidae gen.spec. + + +
Gnaphosidae gen.spec.
Selenopidae Selenops spec. + + +
Pisauridae Thaumasia uncata + + +
Lycosidae gen.spec. + + + +
Ctenidae Cupiennius foliatus + + +
Salticidae Phiale guttata + + + +
Salticidae gen. spec. (1)
Theridiidae Theridion rufipes + + +
Araneidae Argiope argentata + + + +
Araneidae Nephilll clavipes + +
Araneidae Eriophora fuliginea + +
Araneidae Cyclosa spec. + +
Araneidae Mecynogea lemniscata + + +
a Preballooning behaviour: 1 = ascending to the top of the experimental setup; 2 = tiptoe behaviour
(Bristowe 1958; Richter 1970); 3 =dropping on dragline (Coyle 1983).
pointed out, our knowledge of tropical biology, in general, is very limited and the
reality of aerial dispersal in tropical spiders being scarce remains to be investigated.
A preliminary and far from conclusive study of aerial dispersal in tropical spiders is
reported on here (Table 26). This study was carried out in the Republic of Panama
(9°N/79°W) and shows that at least a substantial proportion of the species occurring
in this particular tropical area are capable of becoming airborne. Here, however, as in
temperate species (Meyer 1977; van Wingerden 1980) ballooning probably mainly
serves as an (effective) means of short-range travel through which the established spe-
cies distribution area is not normally substantially extended.
Successful long-range dispersal byway of ballooning, as pointed out before, involves
considerable climatolOgical tolerance. Discussing topographical barriers with respect to
the dispersal of tropical species, Janzen (1966) suggested that tropical species may be
less tolerant to low temperatures, a condition that would effect dispersers particularly
at high altitudes. Nentwig investigated cold resistance in many of the same tropical
species that were studied for their ballooning tendencies as mentioned in Table 26.
Also some additional species from the same environment and one Palaearctic species
(Argiope bruennichi) have been tested. The results of this study, shown in Table 27,
clearly indicate a comparatively low cold resistance in tropical species.
Table 27. In Gamboa (Panama), Nentwig tested the cold resistance of hatchlings from spider egg
sacs he had collected from the surrounding tropical lowland forest and one batch of Argiope bruen·
nichi sent over from Germanya
Family Species N A B
Dipluridae lschnothele guyanensis 30

Scytodidae Scytodes longipes 143
Selenopidae Selenops spec. 199
Pisauridae Thaumasia uneota 81
Lycosidae gen spec. 295
Ctenidae Cupiennius foliatu s 322
Salticidae Phiale guttata 229 +
gen. spec. (1) 71
gen. spec. (2) 45
gen. spec. (3) 40 0
Theridiidae Theridion rUfipes 165
Araneidae Argiope argentata 5432 0 +
Nephila clavipes 1221 0 +
Eriophora fuliginea 723 +
Cyclosa spec. 128
Mecynogea lemniscata 32
Argiope bruennichi 132 + +
a Groups of spiderlings were kept for 24 h at (A) 0 °c ± 3° and (B) 4 °c ± 10.

+ indicates that all spider lings survived the treatment; 0 few survived; - none survived. N = total
number of spiders tested.
356 A.E. Decae: Dispersal: Ballooning and Other Mechanisms
6 Epilogue
Few arachnids are great travelers (Savory 1977). Undoubtedly this accounts for the
very local distribution of many, in particular the more primitive, species. The fact that
in spiders it is mainly the juveniles that engage in aerial dispersal, suggests that this has
evolved as an adaptation to short-term fluctuations in the living conditions of the
environment. Van Wingerden's (1980) hypothesis: "that the main function of aero-
nautic activity of the immatures of the species under concern (two linyphiid species)
is the ability to cover distances within the population in order to reach adulthood in
more favourable sites" might contain a great deal of realism as well as his further con-
siderations that this type of dispersal may incidently contribute to the founding of
new populations. Cosmopolitan distributions of some species are probably largely
achieved through the mediation of man. The fact that tropical spiders have comparative-
ly restricted species distributions is probably an effect of their narrower environmental
tolerance rather than of a limited power of dispersal.
Acknowledgments. I want to thank Dr. G. Thomas and Dr. A. Hallensleben for their help during
the preparation of the manuscript and Dr. W. Nentwig who allowed the use of his unpublished
results on cold resistance in tropical spiders. The work in Panama was carried out under the sup-
port of a research grant from the Netherlands Foundation for the Advancement of Tropical Re-
search (WOTRO).
III Growth, Foraging and Reproductive Success
FRITZ VOLLRATH 1
CONTENTS
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357
2 Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
2.1 Mechanism of Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
2.2 Growth Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
3 Foraging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
3.1 Environmental Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
3.2 Assessment of a Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364
4 Foraging and Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366
5 Growth and Reproductive Success . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
6 Conclusions.............................................. 369
1 Introduction
The life history of any animal is greatly influenced by three factors (Stearns 1976):
the need to grow and reach maturity, the need to feed in order to accumulate the
energy necessary for growth, and fmally, the need to reproduce. Reproduction is often
seasonal, and often the reproductive success depends greatly on body size and physical
fitness (Thornhill and Alcock 1983). Hence, the effectiveness of foraging, the efficiency
of growth and the timing of maturation all directly affect reproductive fitness. Ac-
cordingly, we would expect natural selection to have sharpened the animal's process of
making optimal decisions regarding foraging and growth.
In spiders, foraging and growth are closely interrelated, i.e. foraging decisions in-
fluence growth and vice versa (Fig. 124). Growth determines body size and consequent-
ly prey size, as well as web size (should the spider build one) (Enders 1975; Murakami
1983). Constraints on web size and variations in prey size may in turn compel a grow-
ing spider to move from one site to another, and thus growth may influence foraging
decisions (Enders 1975). Foraging, in turn, influences growth decisions through the ef-
fect it has on the accumulation of reserves, which is a necessary precondition for growth.
Spiders, especially web spiders,are notoriously immobile predators. Immobility hinders
an animal in sampling the environment, and thus impedes the assessment of environ-
mental quality at large. The uncertainty about the quality of other sites, coupled with
high costs of moving, seems to have resulted in a life history in which growth is general-
ly adjusted to the web site. Since the reproductive success of a male spider can greatly
1 Department of Zoology, South Parks Road, Oxford OXI 3PS, Great Britain
358 F. Vollrath
Fig. 124. The interaction of foraging, growth

and reproductive success. The quality of a feed-
do
ing site affects the rate and ratio of a spider's
u\
growth. Timing of maturation and the size of
the adult greatly influences reproductive fitness.
we8-f
Body size also affects mortality and web param-
Mov;..3tay eters in immatures and adult females. Thus, size
~'
lB
has a direct effect on foraging decisions concern-
ing the risk of moving, the filter action of the
{Size Timing
Survival , J web and the efficiency of prey capture. These
Adult V foraging decisions, of course, largely affect the
profitability of a web site
depend on the body size of conspecific competitors, the interaction of growth decisions
and foraging decisions can be viewed as a problem in game theory (see Maynard-Smith
1982, for review).
In this chapter I will firstly consider the "rules" of growth in a web-building spider,
secondly outline the problems which face a foraging spider, thirdly evaluate the inter-
action of foraging success and moulting decisions and fmally examine the importance
of body size for reproductive fitness. Rather than giving exhaustive reviews on these
subjects, I will attempt to point out, in the form of an essay, the importance of each
of these parameters for one facet of the spider's life history, whilst illustrating the great
degree of their interaction.
2 Growth
In spiders the commitment of body reserves to increase in body size is an irreversible

decision. The body which has grown during a moult needs larger muscles, more blood,
longer nerves. Moreover, with a given number of moults (which, like steps leading to a
landing, lead to the final size) any "step" whose height is below maximum precludes
the upper limit of the final size. In growth not only size, but also time is an important
parameter, especially in spiders where mating occurs at a certain period, when all moults
must have been shed. Spiders incorporate time and size into their moulting decisions
which are based on the rate of food intake.
2.1 Mechanism of Growth
The exoskeleton of the spider's cephalothorax grows only during a moult, not between
moults, with the possible exception of the first ins tars (Homann 1949). The abdomen
stays soft throughout the life of the spider and expands with the nutrients imbibed by
the animal and stored in the interstitial tissue of the abdomen as lipids, and to a lesser
degree carbohydrates and proteins (Collatz and Mommsen 1974a). Just before a moult
these are liquified and transferred into the leberidiocytes which are specialized blood
cells (Browning 1942; Eckert 1967). Haemolymph volume is augmented by water up-
take (Homann 1949). In a series of elegant experiments Homann (1949) analyzed the
Growth, Foraging and Reproductive Success 359
physical changes which accompany ecdysis in Tegenaria agrestis. During ecdysis the
abdomen contracts and haemolymph pressure in the cephalothorax nearly doubles,
from 1.2 atm to 1.35 atm. Prior to moulting the firmness of the old cuticle decreases
by one-third from its intermoult value of 270 kg cm- 2 (Fig. 33). Contraction of the
abdomen forces the haemolymph into the cephalothorax, thus inflating it. Homann
found that a decrease in abdominal weight of ca. 70% corresponds to an increase in
cephalothoracical weight of 180%, suggesting that during ecdysis a large proportion of
body mass is shifted from the abdomen into the cephalothorax of Tegenaria agrestis.
In Nephila clavipes this relationship varies between 30- 50% and 170- 300%, and de-
pends on the diet (Vollrath, unpubl.).
It emerges from these observations that the growth of the cephalothorax depends
on several factors. It is limited by the softness of the new and qUickly hardening cuticle,
by the pressure which can be generated and by the amount of reserves which are avail-
able for liquification. The physical properties of the young cuticle are presumably
genetically determined. The amount of reserves in the abdomen is determined by the
foraging success of the animal.
The three components of spider growth (weight, size, time) can be combined into
one figure (Fig. 125), showing the influence which different volumes of reserves can
have on growth. The diagrams show that the timing of ecdysis affects body size as
much as the amount of available reserves. Well-fed Nephila clavipes grow rapidly and in
large increments, poorly-fed individuals grow slowly and in small increments (Fig. 126a).
In Nephila females (but not in males) very slow growth produces an additional instar
(Bonnet 1930; Vollrath 1983).
Size
(b)
iQ)
If)
o
....
Q)
(.)
....c
n m Instar n m Instar
Time
Fig. 12Sa,b. Hypothetical growth of a spider. During a moult, cephalothorax size increases step-
wise. Weight increases continuously, the gain is stored in the abdomen. During a moult, these ab-
dominal reserves are funneled into the cephalothorax, to fuel the growth. Spiders on a poor diet
use up all reserves during a moult (a), spiders on a rich diet retain a proportion of their reserves in
the abdomen (b)
360 F. Vollrath
20 (a) Rich
E
E
+-
c
a..
(I)
N
(f)
140 160
Time, days
6 (b)
)(
5
0
l5
c ....0
.c
'"0E 0
4
"C
.c
Q.
LJ
'"
.
<l u
3
VI
(I)
>
~ 2
(I)
VI
!
(I)
0: 1
T ime, days
Fig. 126. a The growth of Nephiltz cltzvipes varies greatly, depending on the diet. Body size as well
as the intervals between moults, and consequently also total developmental time, are strongly af-
fected by the feeding rate . Starved females undergo an additional 8th moult. Independently of the
diet, males mature either in the 4th or 5th instar. (After Vollrath, in prep.). b Growth of 4th instar
Nephiltz cltzvipes. Each point represents a decision to moult. The curve indicates average timing of
ecdysis, the shaded area shows the "region" in which Nephiltz may moult. The spiders were raised
under diets ranging from rich to poor. At the beginning of the instar all animals had nearly the same
relative amount of reserves, at ecdysis these differed greatly. The available reserves, adjusted for
body size (abdomen weight divided by cephalothorax weight), are shown on the y-axis; the x-axis
shows the time interval between moults. (Redrawn from Vollrath, in prep.)
2.2 Growth Decisions
The decision to moult (as an evolved rule of behaviour) depends largely on the rate of
reserve accumulation , i.e. the amount of reserves accumulated in one instar as a func-
tion of the time needed (Fig. 126b). Males, even as early immatures, may have dif-
ferent decision rules from the females . In all instars the males of N. clavipes had sig-
nificantly longer intervals between moults than the females; this trend was independent
of the diet (Vollrath unpubl.). Turnbull (1965) observed in Ageienopsis potteri a good
correlation between feeding and growth rates. Browning (1941) found that in Tegenaria
atrica any variation in diet that was above minimum (defined by 50% mortality) had
no significant effect on instar length. Parentage , however, did have a measurable effect,
suggesting that some of the variation of growth within a species can be attributed to
differences in the inherited decision rules of the timing of ecdysis. Temperature and
humidity might also affect the timing of growth (Jones 1941; Browning 1941; Schaefer
1976a, 1977c, and this volume), as may photoperiod (Schaefer 1976c), but not il-
lumination (Browning 1941). The spectacular effect which temperature and photo-
period can have on growth is shown in Figs. 115 and 127.
The growth of a spider is greatly influenced by food intake and temperature . Spi-
der growth is not a function of time or body weight, but a function of time and body
weight. Using the rate of feeding and metabolism as the measure by which to determine
the timing of ecdysis allows the animal to take advantage of a rich diet, but ensures at
the same time that the deadline of the mating period is met (when the other sex is avail-
able). The slow growth of Nephila's males (which have fewer instars) can be explained
by this crucial requirement. The additional moult of the very slow growing female may
allow maturation in the following generation, rather than inbetween generations. Over-
lap of generations would result from a division in fast and slow growing genotypes, as
they occur in Araneus diadematus (Ramousse 1973) and Linyphia (Schaefer 1977c).
Similarly, genetic variation in instar numbers, as apparently occurs in the males of
Nephila (Vollrath unpubl.) would result in overlapping generations. Rather complex
life histories could easily occur in the same population (Fig. 128), if these genotypes
show the same phenotypic response to the environment, especially if the number of
23° LO 16 : 8
!
i
E
::;) 16° LO 16:8
"0
o
+-
(f)
Fig. 127. The effect of both tem-

perature and photoperiod on the
t_~-":;i:=::!I~',~=:r' ~O
rate of development in Tetragnatha
montana. The marks on each line
indicate the mean (up : males; ',---23o 8 :16 I
down : females). (After Schaefer
1976a) o 50 100 150 200 250 300
Days
362 F. Vollrath
Fig. 128. The development of

Meta mengei. I Four different
life histories in a schematic de-
scription; the different growth
rates are interrupted by one or
two diapauses (shaded bars).
II The change in relative im-
portance of developmental
types along the gradient of
environmental severity (lati-
tude, altitude, vegetation den-
sity): I Southern Europe (inter-
% % % polated); 2 Northern France;
3 Holland; 4 Southern Jut-
land; 5 Northern J utiand;
6 Scandinavia. The letters in
the two graphs oorrespond to
each other, indicating the pro-
portion of the different life
histories (A -D) in the popula-
tions /-6. (After Toft 1983)
o
2 3
instars vary depending on the feeding conditions (Levy 1970; Workman 1979; Toft
1983; Wise 1984a). But of greater immediate importance is the observation that the
resilience towards environmental conditions would allow a life style which does not
require active foraging because maturation can be reached at the specified time even
with a minimum supply of food. The effect of season and the importance of timing in
spider populations is discussed by Schaefer (this volume).
3 Foraging
An excellent and in-depth analysis of spider foraging can be found in Riechert and
Luczak (1982), who reviewed the literature on the feeding ecology of hunting spiders
and web spiders. In this section I will only give a brief introduction to those aspects of
foraging theory which are immediately relevant to an understanding of the interaction
of foraging and growth.
3.1 Environmental Uncertainty
The foraging behaviour of web spiders (I will concentrate on web spiders) has to cope
with some problems rarely encountered by other animals. Because of the web, the
costs of extensively sampling the environment can be prohibitively high. The web spi-
der's foraging decisions are accordingly rarely based on extensive personal experience.
Yet large variations and fluctuations in prey availability often encountered by succes-
sive generations (Vollrath 1980) and/or in different environments (Jocque 1981;Toft
1983; Vollrath 1986b) make "phylogenetic experiences" a less than accurate basis for
foraging decisions. The spider's foraging dilemma can be visualized in a 2 X 2 table.
Both web site and environment at large can be assigned a value for profitability. These
values are independent of one another, and they are additive. The location of the spi-
der's web can be: in a good site in a good season (++), in a good site in a poor season
(+-), in a poor site in a good season (- +) or in a poor site in a poor season (--). Both,
very good (++) and very poor (--) profitability are presumably easily recognized by
"preset" thresholds. The other two options, however, are problematic and potentially
difficult to solve: if the animal fmds itself in a poor site in a good season (- +) it should
move on rapidly, if it is in a good site in a poor season (+-) it should stay. The profit-
ability of the two sites would be very similar. The question arises: can the spider dis-
tinguish between the two? To anticipate the answer: we do not know. However, random
departure from web sites (see below) suggests a feeding strategy that takes no account
of these differences.
The theory of optimal foraging has amply demonstrated that animals respond to
changes in food abundance in a manner which is well adapted (Schoener 1971; Krebs
1978; Pyke 1984). When prey declines in one location, the animals move to another,
where they decide to stay if it is profitable. Profitability is ,measured by the feed-
ing rate (the amount of food ingested during a certain time span). The profitability
of a feeding site is often measured in relative terms (marginal value) by comparing it
with other potential sites in the habitat (Charnov 1976). Food is rarely distributed
evenly in the environment, but is generally concentrated in patches (Hassell and May
1974). A good patch maybe distinguished by certain proximate features of the environ-
ment, or it may be indistinguishable from poor patches, and the predator may have to
invest time and energy in sampling site quality. The predator'S action may deplete
the quality of the patch by removing prey items. Alternatively predation may have no
effect on site quality, this should occur when the prey are highly mobile and the pre-
dator is not. The predator may either search out individual prey (the feeding strategy
of an active hunter), or it may capture prey when they themselves venture close to the
stationary predator (sit-and-wait strategy). Frantically searching for prey, or sitting
rooted to one spot and waiting for prey are strategies at the two extremes of a con-
tinuous scale of foraging options (Enders 1975). Most spiders seem to employ a mix-
ture of both strategies, but tend to spend more time waiting than searching (Riechert
and Luczak 1982). The salticids can be described as very actively searching hunters
(Hill 1979), the linyphiids as extreme sit-and-wait predators (Ford 1977b; Janetos
1982).
Suggesting lack of mobility, the term sit-and-wait encompasses the two scales of
space and time. Such foraging behaviour can be measured either (1) by examining the
frequency of site changes, or (2) by comparing the amount of time spent waiting for
prey with the time spent in transit and searching for a new site. Orb spiders move rarely,
but each move presumably takes a long time. Hunters in comparison move frequently,
but in short bursts of activity, interspersed with periods of waiting. In spiders the term
'sit-and-wait strategy' should be used when changes of a feeding site are rare in relation
364 F. Vollrath
to the animal's life span. For example, web-building spiders do not change web sites
more frequently than once adayjnight (Turnbull 1964; Le Berre et al. 1981; Ramousse
et al. 1981). With an average life span of 6 months this gives most orb spiders a theo-
retical maximum of only 180 opportunities to change sites, in practice it is only about
one-third, i.e. ca. 60 opportunities (Vollrath 1986b; Houston and Vollrath, in prep.).
Webless spiders, on the other hand, are much more mobile and presumably change
position more frequently.
The main currencies of foraging are time and calories (Schoener 1971): the foraging
animal strives to acquire a maximum of food in a minimum of time with a minimum of
expense, at the same time minimizing the risks to its health and life. Thus, predation
risk is one constraint which prevents maximal exploitation of these two currencies. We
can assume that the risk of predation is higher for a webless web spider moving be-
tween sites than for the same spider in, and protected by, its web, which acts not only
as a trap or snare, but also as an 'early warning' device against predators (Robinson
and Mirick 1971).
In Panama the predation on a cohort of immatures of Nephila cavipes was about
90% in 90 days (Vollrath 1986b). In the same habitat the mortality of travelling Nephila
males, of sizes similar to the immatures, was assumed to be around 83 to 96% in 6 days,
depending somewhat on the size of the male (Vollrath 1980). During a study in North
America lasting 3 months, Wise (1979) observed that survival (the net effect of mortality
and emigration) was 40% in Metepeira labyrinthea, and under 10% in Mecynogea [em-
niscata; supplementary feeding did not increase survival. Rypstra's (1984) figures of
survival of tropical web spiders without a web add up to about 40% diurnal and 10%
nocturnal "predation" during a 6-h period; in the temperate zone it was about 6%,
during both day and night (Rypstra 1984).
3.2 Assessment of a Web Site
A foraging spider is confronted with two important decisions: firstly, when to abandon
an unprofitable site,and secondly, where to settle next. A site may be left either (1) in-
dependently of profitability, (2) immediately if it is recognized as unprofitable, or
(3) at random, after recognition as being poor. A new site can be found either (1) by
trial and error, or (2) it can be identified as rich by proximate factors before settling.
An animal may be able to return to a site, when no better site is found, or it may be
unable to do so. The expenses of active foraging are presumably higher in web spiders
than in hunting spiders, owing to the extended investment of time as well as the ex-
penditure for the web. I will only discuss the foraging decisions of web-building spiders.
Foraging theory makes it unlikely that sites are abandoned independently of their
profitability (Krebs 1978). Janetos (1982) observed non-random departure in a
collection of orb spiders (Fig. 129). Vollrath (1986b), on the other hand, found that
immatures of the orb spider, Nephila clavipes, move at random when the environment
is very poor (Le. unprofitable), but stay when food is abundant. When feeding condi-
tions suddenly worsen, for Nephila used to a rich diet, then these spiders all move
within a few days. The spiders may be responding to the sudden decrease in their feed-
ing rate, or they may be responding to the lower feeding rate. In a set of laboratory ex-
periments Vollrath and Houston (1986) found that previous experience in other sites
Fig. 129. Actual distribution of re- 300 80

sidence times at web sites compared
with expected distributions, should Sheet web
movements occur at random. (After 270 60 Weavers N= 220
Janetos 1982)
'X..2 :
20.41
240 40 -- ,
~
.01 < p< .02
210
180
0 1 2345677
Inter - movement Times
150
Observed
- -- Random Hypothesis
120
-,,
90 Orbweavers N = 458
'X..2 : 340.33
60 -- ., p«.oOI
I
I
30
3
0 1 2345677
Inter-movement Times
does not seem to affect Nephi/a's tenacity to a web site. This suggests that Nephi/a,
and possibly other orb spiders as well, either respond rapidly with a move when the
profitability of a site declines significantly, or that they move at random when the
profitability of a site is consistently very low. In the same habitat, where he observed
orb spiders to move non-randomly, Janetos (1982) found that the linyphiids moved at
random. This suggests that linyphiids, compared with the araneids, either (1) had a
lower threshold of profitability (because of the higher initial expenses for the web, and
lower web-maintenance costs), and/or (2) that they had a lower rate of prey capture
(because of a lower effectiveness of their webs) and/or (3) that they had a worse ratio
of nutrient extraction. Rypstra (1983) observed that araneids respond to changes in
prey density or structural support more rapidly than linyphiids.
Like the question of the previous paragraph: "When and how does a spider decide
to abandon a web site?", the question: "Is the quality of a site judged before settling?",
cannot be answered clearly (Reichert and Gillespie 1986). Turnbull (1964) observed in
the laboratory that Achaearanea tepidariorum finds a profitable site by trial and error.
He observed the spiders moving about in his laboratory, building a web at each stopping
point, only to abandon it after a few days, till they arrived at the windowsill which was
a profitable site. Enders (1975), Brown (1981) and Riechert (1985) observed in the
field that Argiope and AgeZenopsis apparently choose profitable sites before settling by
proximate cues like vegetation, light, shadows, etc. Riechert and Luczak (1982) even
366 F. Vollrath
speculated that some spiders might be able to predict prey abundance by sampling air-
borne vibrations in potential web sites. Clearly it would be of great advantage for a web
spider if site quality could be assessed before settlement.
The presence of an occupied web suggests that this site is likely to be profitable.
Thus, a foraging spider might decide to settle in the vicinity of a web. Indeed, some
species form large aggregations of webs in favourable sites (Lubin 1974; Rypstra 1979).
Alternatively the foraging animal could invade and usurp the existing structure, and
incidentally also gain the protein of the web's silk. This seems to be a widespread
method (or strategy) of sampling web sites (Blanke 1972; Riechert 1982). The main
precondition for success is for the foraging/contesting animal to be larger than the
owner/occupant (Riechert 1982). However, if both are of roughly equal size, the de-
fender has a home advantage. It appears that in Age/enopsis aperta the contestant can
gauge the occupant's determination to defend a site (Riechert 1982). This determina-
tion is generally based on the occupants 'knowledge' about the quality of the site. It
is a puzzle how the contestant can prize this information from the occupant, who
should be little inclined to part with it.
4 Foraging and Growth
Active foraging is costly for several reasons: firstly, it requires that energy is spent to
fuel locomotion; secondly, it incurs some risks concerning the energy budget and the
developmental time by exchanging a site of known quality for a site of unknown quality;
and finally, foraging is risky for a web spider because of the considerable risks of preda-
tion. The expenses of foraging subtract energy from the accumulated stock of reserves
which could otherwise be spent on growth.
A spider, which finds itself in a suboptimal site and cannot grow maximally, has
two options. It may stay and grow at the rate possible in this site. This strategy has two
advantages: (1) from experience the animal can assess the quality of the site and it can
"predict" its rate and ratio of growth, (2) the animal avoids the risks of predation as-
sociated with moving. On the other hand, the animal can decide that its present site is
not profitable, and leave, incurring the costs and risks. A move always affects the rate
and ratio of growth: if a more profitable site is found quickly, growth is enhanced; if
one is not found, growth suffers. The speed with which a foraging animal finds a new
site of better or equal profitability is an important variable that ultimately affects the
animal's adult size and its time of maturation.
Food intake can be envisaged as a steadily climbing "path" of reserve accumulation
in a graph depicting the moult decisions of a spider (Fig. l30a). The basic metabolic
costs and the daily costs of web maintenance must be covered by the nutrients ingested.
The discrete capture of prey items results in a discontinuous intake of nutrients. This
path of reserve accumulation (an expression of the feeding rate and the rate of energy
expenditure) leads towards a set of possible points of ecdysis. From its trajectory the
animal might be able to predict timing and growth at the next moult, and it might be
able to compare its trajectory with the best and the worst conditions possible. When
feeding conditions suddenly worsen, the animal's actual trajectory deviates from the
prOjected path (Fig. l30b). At this point the animal must decide if it should move,
Fig. 130a,b. The interaction of

growth and foraging decisions.
a The curve for moulting decisions
extracted from Fig. 126a. The
field of scatter is indicated by the
shaded area. Minimum and maxi-
..
0
mum feeding rates would lead to ~
minimum and maximum growth .r:
.
+-
rates and ratios. Actually, reserves ~
0
are accumulated rapidly during (!)
the consumption of prey, but
diminish again slowly due to the Time
costs of basic metabolism and
maintenance (silk production and Growth rate
locomotion). Therefore, the
animal approaches a projected
point of ecdysis, not in a direct
(b)
line, but oscillating in a 'path' of
reserve a accumulation.
b Growth depends on site tenacity
and site profitability. When the III
profitability of a site drops (*), '"~
the spider may leave it at any of '"
II>
the following nights (t), thus site a::'"
tenacity affects growth. A new site
may be better, similar or worse,
accordingly the spider's feeding
rate will improve, remain un-
.-
Min.
RIP
changed or decline, thus the B
choice of site affects growth
Time, days
rather than stay and 'accept' the new projection of prolonged and reduced growth. A
move costs energy. If no better site is encountered, the animal may lose all reserves
and eventually starve. If the animal ends up in a similar site, it will have lost time and
size. Only if a better site is found speedily does the animal 'gain' time and size, com-
pared with its projected performance in the previous site. For the web spider each move
is a risky gamble, because of the limited opportunities of moving, the inability to predict
the quality of other sites and the high risks of predation (J anetos 1986).
To be able to make the right decision about the profitability of a site compared with
other sites, the animal needs to 'know' the variance in quality between sites (which it
can estimate accurately only by extensive and expensive sampling). The variance (fluc-
tuation of prey) within a site can only be used as a very rough estimate of environ-
mental quality: it may be a poor site in a good season or vice versa (see above). The in·
accuracy of site quality as a measure of the quality of the environment is shown by the
observation that the variance of the capture rates of adjacent sites is often very large,
as is the variance in spider growth (Jocque 1981; Miyashita 1968a; Toft 1983; Vollrath
1986b).
Very little is known about the interaction of foraging and growth decisions in spi-
ders. The evidence (Vollrath 1986b) suggests that orb spiders minimize risks and adapt
368 F. Vollrath
their growth to a web site, rather than forage very actively in order to find better sites
which would allow faster and larger growth. In a rich site Nephila clavipes moved very
little, growing quickly and in large steps. But even in a poor site the animals moved in-
frequently, adjusting their growth to the site. On average they grew in poor sites with
a growth rate that was twice as long, and had a growth ratio that was one-third smaller
than those in a rich site (Vollrath 1986b). The large size differences, which can be ob-
served among orb spiders from adjacent sites that presumably have different prey
densities (Toft 1983), suggest that many web spiders largely adjust their growth to their
feeding site rather than attempt to maximize feeding and growth rates by active forag-
ing. However, among the web spiders large differences exist in foraging behaviour
(Janetos 1986; Riechert and Gillespie 1986; Rypstra 1983).
5 Growth and Reproductive Success
In spiders, like in many other animals, adult size may greatly influence reproductive
success (Kessler 1971). Larger males presumably have a higher probability of inseminat-
ing a female, and fertilizing a larger number of eggs (if the females mate repeatedly),
than smaller males (Vollrath 1980; Austad 1984). As important as adult size is the
timing of maturation. Early maturation allows males to find immature females and
thus secure first matings (Robinson and Robinson 1973a; Lubin 1986). In spiders
first matings secure the male a larger number of fertilized eggs than later matings
(Austad 1984). In addition, early maturation facilitates an increased number of matings.
Accordingly most spider species appear to be protandrous (unpubl.literature survey).
The problem of foraging and growth facing the web spider can be seem as a problem
in game theory, i.e. the animals actions must be tailored to the actions of conspecifics.
It is not absolute size which gives a particular male the edge over his competitors, but
his relative size. In Nephila clavipes the largest male in the female's web occupies the
position on the hub where he has the opportunity to copulate, feed (sharing the fe-
male's prey) and "protect" the female from the advances of other males, thus guarding
his reproductive investment (Christenson and Goist 1979; Vollrath 1980). Other males
are delegated to the periphery of the female web. The frequency spectrum of male size
changes with the season and differs between habitats (Vollrath 1980, 1986b). To out-
do his competitors a male has to be '1ust" larger than them. In Nephila there appears
to be no benefit attached to a particular male size, with the exception of travel survivor-
ship, in which the medium-sized males did best, the large males second best and the
small males worst (Vollrath 1980). Apart from this, small males had some advantages
over medium-sized males: they seemed to be tolerated by immature females which
gave them the opportunity to wait for their maturation, and thus mate with a virgin
(Robinson and Robinson 1973a). In spiders, where the sperm of the first mating
fertilizes most of the eggs, this strategy has obvious advantages (Jackson 1980; Vollrath
1980; Austad 1984). Small males also could sneak copulations when the large guarding
male was fighting off contestants (Christenson and Goist 1979; Vollrath 1980).
Females generally live much longer than males, accumulating after mating the re-
serves for the eggs, which they lay towards the end of the season. In some species the
females die after having laid a single egg sac (like Araneus diadematus or Amaurobius),
whilst the females of other species may resume feeding after egg laying, and produce
up to five batches (like Nephila clavipes, Zygiella x-notata, Mecynogea lemniscata)
(unpubl. observations). Thus, whereas the adult males of web spiders no longer forage
actively, the adult females do (Turnbull 1962). This suggests that the decisions about
growth, and especially the timing of maturity, are different for the sexes. For a female
it is important to mature early before the end of the season in order to be able to ac-
cumulate enough reserves for one or several egg batches. For females (as for the males)
timing of maturity can be as important as body size (Olive 1981 b). A large female is
able to lay a large egg batch (Kessler 1973), and an early female is able to lay several
(Kessler 1971; Jackson 1978a). Depending on the species, several combinations of
fixed or variable clutch sizes and fixed or variable egg sizes seem to be possible (Tum-
bull 1962; Kessler 1971).
Spider males are generally smaller than the females, often they are much smaller.
The differences in size are due to the fact that males have fewer moults than the fe-
males (e.g. Schaefer 1976). Thus, they are able to mature before the females. How-
ever, in some species the males are very much smaller and reach maturity after half as
many moults as the females, yet they mature on average only a short period before the
females. This is especially obvious in Nephila clavipes, where the females mature after
seven or eight moults, and the males after four or five moults, yet where the males
(under the same, optimal diet) proceed the females by little more than a week (Vollrath,
in prep.). The growth decisions of the Nephila males already differ in the first post-
eclosion instars, when no other differences between the sexes are apparent, neither in
morphology, foraging biology nor web parameters (unpubl. observations).
6 Conclusions
It emerges that little is known about the interaction of growth and foraging decisions
in spiders. Food is hardly ever superabundant, and consequently growth is generally
submaximal. It appears that a spider's growth is greatly affected by its feeding rate.
The feeding rate depends on parameters of the environment which are very "private",
rarely is the feeding rate identical for any two individuals of a population; environ-
mental parameters, like temperature and photoperiod, in contrast, are shared by all.
Spiders generally have a brief mating period, and they must adjust their growth to
mature at the correct time, having completed a fixed number of instars. Large size at
maturation conveys benefits. Thus, the average spider, fmding itself in a suboptimal
site, has to balance the costs of staying and growing sub optimally against the costs of
moving.
Since web spiders cannot easily and inexpensively sample the qUality of the environ-
ment, foraging decisions should present a particularly "dicey" problem for them. They
seem to solve it by staying in a very good site, and by leaving poor sites at random.
This is a good strategy if one assumes that the animal has no information about the
probability of fmding a better site.
Foraging and growth are closely interlinked. Foraging greatly influences growth
through its costs and payoffs for the animal's energy budget. Growth affects foraging
though its influence on the potential prey spectrum and choice of web site. Growth
370 F. Volhath: Growth, Foraging and Reproductive Success
influences reproductive success by determining the timing of maturation and the adult
size. The number of eggs a female can lay increases with her size; timing affects the
number of batches she can produce. In a male as well, timing of maturation may be
as important as size, since early maturation allows him to mate with virgins, which in-
creases his chances for fertilization. Absolute size may influence a males's travel surviv-
orship, but his chances of inseminating a female and fertilizing her eggs depends on his
relative size (relative to his competitors).
Acknowledgments. I thank C. Bromhall, A. Houston, J. Endler and M. Ridley who kindly read
and criticized various drafts of the manuscript. M. Amph1ett prepared the figures.
IV The Biorhythms of Spiders
JOHN L. CWUDSLEY -THOMPSON 1
CONTENTS
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
2 Short-Period Rhythms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
3 Circadian Rhythms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 372
3.1 Circadian Rhythms in Sensitivity and Metabolism . . . . . . . . . . . . . . . . . . . . . 372
3.2 Diel Rhythms of Locomotory Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
3.3 Ecological Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
3.4 Diel Rhythms and Seasonal Activity in Populations . . . . . . . . . . . . . . . . . . . . 377
4 Photoperiodic Timing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
1 Introduction
Changes in the physiology and behaviour of living organisms usually take place cyclically
or rhythmically. This is as true of spiders as it is of all other animals and plants. Some
innate rhythms, such as that of the beating of the heart, have short periods. At the op-
posite extreme, endogenous or persistent rhythms with a periodicity of about a year
have been shown experimentally to exist in a number of long-lived plant and animal
species. Rhythmic functions of ecological significance in animals are mainly of three
kinds: endogenous daily, lunar and seasonal or annual rhythms; photoperiodic pheno-
mena; and the sense of time used in celestial navigation. Endogenous or 'persistent'
daily biorhythms, which free-run with a circadian period are widespread among spi-
ders. Circalunar activity rhythms do not appear to have been much investigated, since
most species do not live long enough to exhibit endogenous circannual rhythms. Al-
though these may occur among Mygalomorphae, and especially in Theraphosidae, the
possibility of their existence has not yet been tested. Until now, although the autecology
of phenology of many orders of arachnids have been investigated, no evidence of a
circannual clock has ever been reported, even in the most long-lived species (Cloudsley-
Thompson 1978).
1 Department of Zoology, University College, University of London, Grower Street, London,

WClE 6BT, England
372 J.1. Cloudsley-Thompson
2 Short-Period Rhythms
The spider's heartbeat, a typical short-period innate rhythm, has been described and
analyzed among others by R.S. Wilson (1967), who suggested that the heart rate may
be influenced by back pressure from the prosoma, as well as being controlled by the
central nervous system. After severance from prosomal control by ligature, the isolated
heart of Heteropoda venatoria shows co~rdinated beating for a long period. Its rate is
controlled by a free-running pacemaker in the anterior half of the dorsal cardiac nerve,
whose cells may form part of the pacemaker system. For interactions of heart rate and
physiological processes, see Carrel (this volume).
3 Circadian Rhythms
3.1 Circadian Rhythms in Sensitivity and Metabolism
Circadian rhythms may appear at various levels of physiological and behavioural organ-
ization. For instance, the posterior median eyes of the net-casting spider Dinopis sub-
rufUS are highly specialized for night vision. During the day, the receptive segments of
the rhabdomeres are short, and the membrane occupies only a small proportion of their
volumes. At nightfall, however, the segments lengthen, and new membrane is added to
them in a rapid burst of synthesis. The sequence is reversed at dawn when the mem-
brane is removed. Synthesis and destruction of membrane are controlled, in part, by
the immediate state of retinal illumination, on which a daily physiological rhythm is
superimposed which is probably endogenous (Blest 1978). Similar mechanisms have
been described in the eyes of other arthropods. This type of dual control of a cycle by
both exogenous and endogenous factors is by no means unusual, and such rhythms have
been deSignated as 'composite'.
Because of the circadian rhythm, metabolism may vary in the same animal, even
under controlled constant conditions, depending upon the time of day. Thus, the
metabolic rates of lycosid spiders (Schizocosa sp.), measured in respirometers at dif-
ferent temperatures, were not only higher at the higher temperatures, but were also
higher at night than during the day, thereby reflecting the nocturnal habit of the ani-
mals (Moulder and Reichle 1972). Hourly rates for single individuals revealed a major
peak of activity between 20.00 and 23.00 h, with a secondary peak at 06.00-07.00 h
(Moulder et al. 1970).
Nycthemeral or 'dieI' respiratory cycles in spiders have also been reported by Hag-
strum (1970a), J.F. Anderson (1970) and Humphreys (1975b, 1977b), who suggested
that these patterns decayed if the animals were kept in captivity for a long time.
McQuenn et al. (1979), however, found that while some individual wolf spiders (Geoly-
cosa domifex) in Canada showed diel patterns of respiration while resting, others did
not, and that high rates of respiration occurred significantly more often in the dark,
when the spiders are normally most active, than during the day, despite a high degree
of variation between individuals. Humphreys (1975b, 1977b) showed that Geolycosa
godeffroyi had definite peaks in respiration, separated by a mean interval time of 23.3 h,
but that these peaks were not associated with any particular time of day. He suggested
The Biorhythms of Spiders 373
that this might have been because the spiders had been maintained under continuous
light and constant temperature for some days before the experiments. Such an expla-
nation does not seems unreasonable, since J.F. Anderson (1970) found that the diel
pattern produced by L. lenta and Phidippus regius decayed within 3 weeks when the
animals were kept under constant environmental conditions, while Granstrom (1977)
noted that Pardosa spp. kept at constant temperature tended to become arrhythmic
during midsummer and midwinter in northem Sweden. Respiration cycles must be
linked with rhythms of locomotion and, therefore, may be exogenous, endogenous or
composite.
When the number of consecutive responses elicited at night from Amaurobius ferox
to the vibrations of a tuning fork was compared with the number of consecutive times
that the spiders could be enticed from their holes during the daytime, the following
results were obtained: of 20 tests made in darkness, 10 were positive in that the spiders
rushed from their holes. In the 10 negative tests, the spiders did not show a response
to any of three applications of the vibrating tuning forks to their webs. In comparison,
only 4 positive responses were obtained in over 30 tests made during the daylight hours
(Cloudsley-Thompson 1953).
3.2 Diel Rhythms of Locomotory Activity
A considerable amount of information is available about rhythms of locomotory activ-

ity in Arachnida, especially scorpions. Somewhat less is known about spiders in this
context, but probably most species would display endogenous periodicities of move-
ment and rest when main tained under constan t conditions in the laboratory. By means
of an aktograph apparatus, Buchli (1961, 1965a,b) showed that locomotory activity
takes place in Nemesia caementaria when the level of illumination falls below 100 lx.
Daily patterns of movement and rest have been described in females of Aphonopelma
seemanni during the rainy season in Costa Rica. These spiders are active during the
rainy season between 18.00 and 06.00 h, but show little movement during the day-
light hours. The difference observed between the number of tarantulas active in the
same area during the day (6.64 ± 1.30) and at night (26.25 ± 4.48) was highly Significant
(Herrero et al. 1983). In contrast, Sphaerobothria hoffmanni has a biorhythm which
includes a period of inactivity at the entrance to the burrow (Valerio 1980). Likewise,
a study of daily activity patterns in the North American desert tarantula Aphonopelma
chalcodes confirmed that these animals, too, are most active at night, although there is
not a total cessation of activity during the daylight hours (Minch 1978). The biological
"clock" is probably timed to coincide with the beginning of nocturnal activity, which
takes place at sunset, when surface and overhead light intensity also playa role in
initiating activity. Tarantulas tend to be more active on cool afternoons, even with the
overhead light intensity at 344,000 lx, indicating that favourable surface temperature
alone might bring about some increase in activity. The diurnal rhythm in this case is
evidently composite.
Many of the animals that inhabit arid environments survive by being active at night
and remaining hidden in sheltered retreats during the day. In the scrubby flatwoods of
the central highlands of southern California, the nocturnal emergence of Lycosa ceratiola
is correlated with ambient temperature, and there is a complex relationship between
374 I.L. Cloudsley-Thompson
body size and emergence behaviour. Since the animals are arrhythmic in the absence of
thermal cues, however, it seems probable that this rhythm is largely exogenous (Carrel
1980).
In its natural habitat in the highlands of Central America, Cupiennius salei (Ctenidae)
is , likewise, markedly nocturnal and can seldom be seen outside its retreat except at
night. The onset of locomotory activity is abrupt: after sunset, at light intensities of
about 20 lx, C. salei turns round so that its prosoma points upwards , but remains mo-
tionless until the night is completely dark (below 0.1 lx).1t then starts moving slowly
up a leaf to wait for its prey (Barth and Seyfarth 1979). Later, Seyfarth (1980) showed
by means of an aktograph apparatus that the rhythm of C. salei is endogenous, with a
mean period of 24 .9 h . During the first four to six cycles in constant darkness , the
rhythm usually 'splits' into two components running at different frequencies: in con-
stant light (26 Ix), locomotory activity becomes arrhythmic (Fig. 131) .
--- .
:
--
---
--
--
- ~
u
---
~
~ 10
-
:!
~
10
- -- - - 2
<(>
-- .-
CI
- .
~IO
- --
~
•-
a b ~
.. 11 :II
'IN( I~ I
11
• .
---------- -- ----
---
-
- :!
~
5
,
..S"
~ 1
)
c d
" " • "ICI~I
1.4 11
• l'
O ~_-_ _-~~-~~-~~-~
10 12 " 16 " :lO n 24 16 28 )0
pERICO Ihl
----
=- .---- - -- r l:':@b : ! \ d
~ I,S
------- ..!
;!
J 1,0
I
.DO
-- - -
- - - - - - -- --
O,S
1\ f 0'0 11 " 16 ,. 10 11 l' 16 18 )0

e PERICO Ihl
11
• ..TIM[ I~I
11
• l'
Fig. 13la-f. Locomotor activity of Cupiennius salei. a Entrained activity rhythm during LD 12:12,
lights off at 18.00 h, note lights-on activity at 6.00 h. b Mean hourly distribution of activity amounts
(± SE) for same data. c Free-running activity in LL. d Periodogram for same animal; no periodic
component is discernable over the range from 10 to 30 h. e Free-running rhythm in DO. f Periodo-
gram for same data. Main peak is at 25 .7 h and indicates period of endogenous rhythm. The acto-
gram records in a, c, e are drawn on a double 24 h scale' to aid visual inspection. (From Seyfarth
1980)
A 5
>-
I-
:>
~
uct
B
21°C
18
15
60 %
50 RH
C 40
Fig. I32A-C. Analysis of aktographic records showing diurnal rhythm of movement and rest in
spiders of the genus Amaurobius in natural daylight and darkness at room temperature and humid-
ity. Ordinates: Activity on the left , temperature and relative humidity on the right. Abscissa: Time
in days. The bTilck stripes represent 12-h periods from 18.00 to 06.00 h. A Periodicity in A. ferox;
B periodicity in A. similis; C periodicity in A. fen estra lis . (From Cloudsley-Thompson 1957)
Aktograph experiments have shown that spiders of the genus Amaurobius are
strictly nocturnal, and the rhythm persists under constant laboratory conditions . Al-
though A. ferox loses water by transpiration much more rapidly than A. similis or
A. fenestra lis, over 90% of the activity of all three species takes place in darkness
(Fig. 132). It seems probable , therefore, that nocturnal behaviour may be related more
to biotic than to physical factors of the environment (Cloudsley-Thompson 1957).
The same may well be true of the North American desert mygalomorph spiders Apho-
nope/ma spp., which are strictly night-active (Cloudsley-Thompson 1968) and spend
the days in burrows at a depth sufficient to reduce temperature variations (Seymour
and Vinegar 1973).
Many studies on the nycthemeral rhythms of araneid spiders have been carried out,
but in relatively few cases has persistence of the rhythm under constant conditions
been investigated. Le Guelte and Ramousse (1979), however, showed that the rhythm
of web building in Araneus diadematus , recorded with a time-lapse camera under a
cycle of 12 h light (100 Ix) followed by 12 h darkness (0.5 Ix) (L:Dj12:12), was weak
under constant temperature and humidity compared with the rhythm exhibited when
these parameters fluctuated naturally. Furthermore, under constant dim illumination
(10 Ix) the rhythm of web destruction was initiated first by the initial reduction in
light intensity, followed by the rhythm of web construction, which was inhibited by
constant illumination (100 Ix).
Most orb-web spiders build their webs early after the onset of darkness (Eberhard
1976a; Foelix 1982; M.H. Robinson et al. 1971 ; Stowe 1978). Ramousse and co-wor-
kers made detailed investigations on a number of species. All those studied were claim-
ed to be nocturnal. Araneus umbracticus, A. quadratus, Argiope obscuripes and Zygie/la
376 J.1. Cloudsley-Thompson
x-notata build only one web at the end of the night. A. diadematus, A. cucurbitinus,
A. serieatus (= A. sc/opetarius) and A. daimaticus, on the other hand, build webs at the
beginning of the night. They are able to construct several webs within 24 h (Ramousse
1980a,b; Ramousse and Davis 1976; Ramousse and Le Guelte 1979,1984; Ramousse
et al. 1981).
The development of nycthemeral rhythmicity during the ontogeny of web-building
spiders has been investigated by a number of authors. Araneus eornutus hatchlings are
initially active in daylight, but become more nocturnal after their second ecdysis
(Le Berre 1979a,b), while synchronization of moulting decreases from hatchling to the
second moult in A. suspieax, according to Ramousse and Wurdak (1984). The spider-
lings are day-active at the beginning of the first nymphal instar, but become nocturnal
towards the end. Livecchi (1979) reared A. diadematus from the moment of dispersal
from the cocoon under environmental conditions (L:D/12 :12) differing only in the
times of extinction of the light, the principal synchronizer of the circadian rhythms.
Most of the web-building activity of the spiders plunged into darkness (0.5 Ix) at
19.00 h occurred at the onset of the light period at 07.00 h. Spiders which were in
darkness from 01.00 h to 13.00 h, on the other hand, were nocturnally active and
responded more rapidly to their synchronizer during the course of development.
3.3 Ecological Considerations
It is often extremely difficult to prove the adaptive significance of behavioural traits,

and this is particularly true of biorhythms, as will be seen from the following example.
Zodarion frenatum (Zodariidae) is a predator of the ant Cataglyphis bieolor in
Greece. Harkness (1977) noticed that spiders appeared outside the nests of ants from
dusk onwards, when they attacked and killed the guard ants. Their numbers rose to a
maximum of about one per nest between 21.00 and 22.00 h, when the ants were most
active, and fell around 05.00 h, about 0.5 h before sunrise.
Two alternative predatory strategies are available to large burrowing and secretive
arthropods, such as tropical mygalomorph spiders, scorpions and scolopendromorph
centipedes. They may either remain most of the time inactive in their retreats, or else
they can venture forth in search of prey. This increases expenditure of energy and the
risk of predation, but may also increase the chances of obtaining food. By means of
aktographic experiments, it has been shown that desert species tend to be nocturnal
and active at night, while tropical forest and woodland forms invoke the first of these
two alternative modes of life. The species of Mygalomorphae investigated included
Aphonopelma sp. which is typically rather inactive and lacks any marked locomotoric
rhythms (Fig. 133) (Cloudsley-Thompson 1981). Unlike most other Theraphosidae,
the arboreal Aviculilria aviculilria shows a weak circadian rhythm of activity and rest.
It is crepuscular, with a bimodal peak of locomotoric activity which is more pronounced
at subjective dusk (lights off) than at subjective dawn (lights on), which is in agreement
with observations made in the field (Cloudsley-Thompson and Constantinou 1985).
Studies on 24·h cycles of activity in Tetragnatha montana and Dolomedes fimbriatus,
based on field observations and on counts made in the Kampinos Forest near Warsaw,
showed that both species moved about and fed most during the evening and at night.
During cloudy and wet weather, however, the period of activity was extended into the
Fig. 133a-c. Activity of Time(h)

Aphonopelma sp. a At first 3456789
under room conditions I
(20° ± 1 0c) (days 1-15), 2
then bin L:D/ l :23 at 23°C 3 :
4 :
(days 16-21) and finally
5 :
c in darkness at 30°C 6 :
alternating with 20 °c. 7 :;
(From Cloudsley-Thomp- 8 "
9 :
son 1981)
10
II
12
13
14
15
a
..
>-
0
0
. '•
21
b
22 :'
23 ""'"
': - : =:::::,;:
24 8;:;;-: ::
25 :;'11:: - ::
26
27
28
29
C
hours of daylight, while the rest period would be interrupted by the presence of large
numbers of the mosquitoes on which these spiders normally prey (Horn 1969).
The colonial orb weavers Eriophora bistriata of Paraguay are markedly nocturnal in
habit, leaving their daytime retreats at dusk to spin their horizontal webs and return-
ing to shelter at the approach of dawn . Here they clump, one upon the other, grasping
each others' bodies (Fowler and Diehl 1978). The adaptive functions of this activity
rhythm may reside in the avoidance of harsh physical factors. The social Agelena con-
sociata is also strictly nocturnal. Indeed, there seems to be a general tendency for social
spiders to be nocturnal.
3.4 Diel Rhythms and Seasonal Activity in Populations
Diel and seasonal activity in spiders and other arthropods have been investigated on a
number of occasions by means of pitfall traps. Williams (1962) established the fact
that cursorial spiders exhibit ecological similarity between daytime and summer activity
at night and activity during the winter. He suggested that when a species occurs over
a wide range of habitats, its diel periodicity is most marked in more favourable environ-
ments. Diel activity and season abundance in Pardosa ramulosa and its prey species
were investigated by Yeargan (1975), using time-sorting pitfall traps similar to those
378 J.L. Cloudsley-Thompson
employed for a similar purpose by Dondale et al. (1972). The correlations between
diel activity patterns and seasonal abundance of insect species such as the leaf-hopper
Aceratagallia obscura and Geocoris spp. (Heteroptera) with those of P. ramulosa in-
dicate that the former could fulfil the temporal requirements of suitable prey species.
Diel trends in a population of prairie spiders, based on the number of specimens
taken in 50 sweep samples, were analyzed by Muma and Muma (1949). On the high
prairie, numbers were lowest during the late hours of the morning, increasing gradually
to a peak in the evening and decreasing again abruptly in the early hours of the morning.
The same trend was noticeable in post-climax tall grasses, but was modified in de-
ciduous shrubs, where a peak in numbers was obtained in the afternoon and early even-
ing. Oxyopes salticus, a night-hunting spider, reached a distinct peak in the late evening
in high prairie and post-climax tall grasses, and a light peak in the afternoon and early
evening in deciduous shrubs.
Daily vertical migrations have been recorded by Le Sar and Unzicker (1978) in a
number of species of spider that live on soybean plants. Although about two-thirds of
the entire spider population was located on the lower halves of the plants, regardless
of the time of day, the others showed noticeable changes in location, apparently in
response to heat and humidity stress. Spatial and temporal patterns in a North American
sagebrush stepper spider community were studied by Abraham (1983), using pitfall
traps, sweep-netting and beating techniques. He found that seasonal patterns were in-
fluenced by climate and by the migration of the spiders between the different vegeta-
tional strata. Diel activity patterns were examined only in the herb and shrub strata:
spider activity was found to be strongly influenced by light intensity and relative
humidity, especially in the herb layer.
In an investigation of the locomotory activity of Pardosa lugubris andP. amentata
in northern Sweden by means of automatic pitfall traps (with 2-hintervals) (Granstrom
1973), and under experimental conditions (Granstrom 1977), it was found that both
species are day-active under natural and experimental conditions. Granstrom concluded
that the position of the active phase was decided mainly by the light-dark cycle, al-
though the pattern was modified by tern perature.
4 Photoperiodic Timing Mechanisms
There are other biological phenomena which depend upon measurement of time. These
include the photoperiodic regulation of diapause, seasonal timing of reproduction,
development and hibernation (Schaefer, this volume), seasonal resistance to cold
(Kirchner, this volume) and time-compensated astronomical navigation (Garner and
Claas 1985).
Photoperiodic regulation of winter diapause in the univoltine grass spider Agelena
limbata has been described by Kurihara (1979). Of egg sacs collected in the field in
late September, 95% emerged in 26-39 (average 31) days when kept under long-day
conditions (L:Dj13.5:10.5) and 5-31% in 140-165 days under short-day conditions
(L:D /13: 11). There is clear evidence here of the operation of a circadian "clock" that
measures day length. Although comparatively little studied, photoperiodic timing of
reproduction and development is probably widespread among the spiders of the tem-
perate and boreal biomes as it is in insects.
According to Minch (1979), temperature and photoperiod are probably the major
factors controlling timing of the events of the annual cycle of tarantulas. This con-
clusion was based upon observations of Aphonope/ma chalcodes in the field and labora-
tory which indicated, but did not prove, the presence of an internal biological "clock"
with temperature involved as a modifying factor as the period for unplugging the bur-
row approached. Some plasticity was, therefore, allowed in the event of unseasonable
weather, such as a cold snap in mid-April.
5 Conclusion
Although many different kinds of rhythmic events have been described in animals, the
ones that are generated by endogenous biological "clocks" tend to have periods con-
siderably longer than those which control heartbeat and the biochemical rhythmicity
of intermediate metabolism. Endogenous rhythms are probably driven by endogenous
pacemakers contained within specialized groups of neurones, although it is generally
believed that every living cell is equipped with a biological "clock" or "clocks" (see
discussion in Cloudsley-Thompson 1961a). All the more important known kinds of
biological time-keeping, apart from circannual rhythms, have been demonstrated among
spiders. Nevertheless, predatory animals are more difficult to work with than vege-
tarians, and the study of spiders has not contributed anything like so much to the
understanding of animal biorhythms as has the study of insects. What has been ascer-
tained, however, indicates that spiders do not differ from other arthropod taxa in re-
spect of their biological "clock" systems.
Appendix
Commented Checklist of Spider Families
The following list gives a short overlook on the taxonomic system of the Araneae. Any
classification, however, is controversial, reflects the actual stand of knowledge and al-
ways includes personal assumptions. Therefore, the order in which the families are
listed here or the number and limits of the families may differ from other sources.
I would like to stress the point that this list does not represent a new classification
based on my own taxonomic work. The main intention of this list is to provide non-
taxonomists with an idea of the system of spiders and to offer taxonomic help to the
reader of this book. I want to thank PM. Brignoli, S. Heimer, H.W. Levi, R.R. Forster,
N.1. Platnick and RJ. Raven for comments on earlier drafts of this list. I could not fol-
low all their (often controversial) suggestions and deviations from published opinions
should be considered as my personal points of view.
For more information on spider taxonomy the reader is referred to a few publica-
tions which cover a wide range of the Systema Araneorum (Lehtinen 1967; R.R. Forster
1970; R.R. Forster and Wilton 1973; Levi 1982; Brignoli 1983; Raven 1985) and to a
wide range of special papers some of which are mentioned in the following list.
This list gives the valid family name, common English names (if existing), approxi-
mate number of species, dominant type of web (but only if it is used to capture prey),
geographic distribution, some typical genera (including all genera mentioned in this
book or otherwise important) and a few references (limited to important papers and
with special reference to this book).
Appendix 381
Order Araneae
1. Suborder: Uphistiomorphae (Mesothelae)
Uphistiidae
20, burrow with trapdoor, signal threads, SE Asia: Liphistius, Heptathela, Ryuthela (Bristowe
1932; Haupt 1979; Murphy and Platnick 1981; Platnick and Sedgwick 1984)
2. Suborder: Mygalomorphae (Orthognathae)
Microstigmatidae
10, ? web, S. Africa, S. America: Micromygale, Microstigmata, Ministigmata
Mecicobothriidae
10, funnel webs, America: Hexura, Mecicobothrium
Dipluridae (including Hexathelidae)

250, funnel web, worldwide (tropics): Atrax, Diplura, Euagrus, Hexathele, Ischnothele, Macrothele,
Po"hothele (Hickman 1964; Palmer 1985)
N emesiidae (including Pycnothelidae) (also considered a subfamily of the Ctenizidae)

Trapdoor spiders, 300, burrow with trapdoor, worldwide: Nemesill, Pycnothele (Buchli 1961,
1965a,b,1970)
Barychelidae
200, burrow with trapdoor, worldwide (tropics): Barychelus
Theraphosidae
Tarantulas, 600, no web, worldwide (mainly tropical): Aphonopelma, Avicularill, Brachypelma,
Chaetopelma, Dugesiella, Eurypelma, Grammostola, Lasiodora, Melopoeus, Phormictopus, Phrixo-
trichus, Pterinochilus, Sericopelma, Sphaerobothrill (Baerg 1958; Melchers 1964; Stewart and Martin
1970,1974,1982; Minch 1978, 1979; Stradling 1978; Valerio 1980)
Paratropidae
10, funnel web, Amazon basin: Paratropis
Atypidae
30, purse web, holarctis, Africa, S.E. Asia: Atypus, Calommata, Sphodros
Antrodiaetidae (also considered a subfamily of the Ctenizidae)

Turret spiders, 25, burrow with trapdoor, N. America, Japan: Alilltypus, Antrodiaetus, Atypoides
(Coyle 1983)
Ctenizidae (including Cyrtaucheniidae, Idiopidae)

Trapdoor spiders, 400, burrow with trapdoor, worldwide: Aptostichus, Bothriocyrtum, Cteniza, Cyc-
locosmill, Cyrtocarenum, Idiops, Neocteniza, Ummidill (R.R. Forster and Wilton 1968;Gray 1968)
382 Appendix
Actinopodidae (also considered a subfamily of the Ctenizidae)

Trapdoor spiders, 50, burrow with trapdoor, C. and S. America, Australia: Actinopus
Migidae
75, funnel web mostly with a trapdoor, Southern Hemisphere: Heteromigas, Migas, Moggridgea
(R.R. Forster and Wilton 1968)
3. Suborder Araneomorphae (Labidognathae)
Hypochilidae
10, cribellate, lampshade-shaped webs, N. America, China: Ectatosticta, Hypochilus (Marples 1968;
Fergusson 1972; Coyle 1985b)
Hickmaniidae
1, cribellate, sheet web, Tasmania: Hickmania (Marples 1968)
Thaididae (= Austrochilidae) (also considered a subfamily of the Hypochilidae)

1, cribellate, funnel web, Argentina, Chile: Thaida (= Austrochi/Us) (Marples 1968)
Gradungulidae
10, cribellate, irregular webs (or reduced), Australia, New Zealand: Gradungu/a, Progradungu/a
(Marples 1968; R.R. Forster and Gray 1979)
Filistatidae
50, cribellate, irregular web, worldwide (tropics, subtropics): Filistata
Caponiidae
50, no webs, America, Africa: Caponia, Nops
Sicariidae
25, no webs, S. America, S. Africa (arid areas): Sicarius
Scytodidae (also considered a subfamily of the Sicariidae)

Spitting spiders, 130, irregular webs, worldwide: Scytodes (Millot 1930b; Bristowe 1931b; Dabe-
low 1958; C. Gilbert and Rayor 1985; Nentwig 1985a)
Loxoscelidae (also considered a subfamily of Scytodidae, Sicariidae)

Brown spiders, 100, irregular web, worldwide: Loxosceles (Rekow et al. 1983)
Diguetidae
10, space web, America (arid areas): Diguetia (Bentzien 1973; Lopez 1983b)
Plectreuridae
20, funnel web, N. America: Kibramoa. Plectreurys
Appendix 383
Tetrablemmidae
100, no web, worldwide (tropics): Ablemma, Tetrablemma
Pacullidae (also considered a subfamily of the Tetrablemmidae)

20, no web, S.E. Asia: Paculla
Dysderidae
200, no webs, worldwide: Dysdera, Harpactea (Cooke 1965; Jackson and Pollard 1982; Pollard in
press)
Orsolobidae
170, no webs, S. Hemisphere: Orsolobus
Segestriidae
100, burrow with signal threads, worldwide: Ariadna, Segestria
Oonopidae
250, no web, worldwide (tropics): Dysderina, [schnothyreus, Oonops
Leptonetidae
100, space web, worldwide (caves): Leptoneta (Lopez and Emerit 1978)
Telemidae
20, space web, worldwide (caves), Telema (Lopez 1977b; Legendre and Lopez 1978; Juberthie
et al. 1981)
Ochyroceratidae
50, irregular webs, worldwide (tropics): Speocera, Theotima
Pholcidae
Daddy long legs spiders, 350, space webs, worldwide: Holocnemus, Modisimus, Pholcus (Bricefio
1985; Eberhard and Bricefto 1985; Kirchner 1986)
Archaeidae
17 (including some fossile amber species), no web, Madagascar, S. Africa, Australia: Archaea (Le-
gendre 1961b; R.R. Forster and Platnick 1984)
Mecysmaucheniidae (including Pararchaeidae, Holarchaeidae) (often considered a sub-

family of the Archaeidae)
30, no web, S. Hemisphere: Holarchaea, Mecysmauchenius, Pararchaea, Zearchaea (R.R. Forster
and Platnick 1984)
Stenochilidae
5, no web, S.E. Ashl: Stenochilus
Palpimanidae
100, no web, worldwide (except Australia): Palpimanus
384 Appendix
Huttoniidae
10, no web, New Zealand: Huttonia (R.R. Forster and Platnick 1984)
Eresidae
100, cribellate, space webs, Africa, Eurasia, Pacific area: Dresserus, Eresus, Stegodyphus (NI/Jrgaard
1941; Millot and Bourgin 1942; Nawabi 1974; P. Ward and Enders 1985)
Zodariidae
350, no web, worldwide: Storena, Zodarion (Schneider 1971; Harkness 1977)
Homalonychidae
2, no webs, N. America (arid): Homalonychus
Oecobiidae
50, cribellate, space webs, worldwide: Oecobius, Uroctea, Uroecobius (Zimmermann 1975)
Hersiliidae
75, no web, worldwide: Hersilia, Tama
Uloboridae
200, cribellate, orb webs, worldwide: Hyptiotes, Miagrammopes, Philoponella, Polenecia, Sybota,
Uloborus, Zosis (Wiehle 1927, 1931; Reukauf 1931; Eberhard 1973; Opell 1979, 1982; Peters
1984)
Dinopidae
Ogre-faced spiders, 50, cribellate, web held by the anterior legs, worldwide (tropics): Dinopis, Men-
neus (M.H. Robinson and B.C. Robinson 1971)
Theridiidae
Comb-footed spiders, 2000, different forms of space webs, often modified or reduced, worldwide:
Achaearanea, Anelosimus, Argyrodes, ChrysSQ, Enoplognatha, Episinus, Latrodectus, Lithyphantes,
Pholcomma, Steatoda, Teutana, Theridion (Bristowe 1931a; D'Amour et aI. 1936; Wiehle 1937;
Archer 1946; R. Braun 1956; NI/Jrgaard 1956; Exline and Levi 1962; Turnbull 1964; Oxford 1976,
1983; Vollrath 1976, 1979a,b, 1985a; Carico 1978; Lubin et al. 1978; Toft 1978b; Eberhard 1979;
Christenson 1984; Nentwig 1985c)
Nesticidae
200, space webs, worldwide (caves): Nesticus (Kirchner and Kullmann 1972; Gertsch 1984)
Hadrotarsidae (often considered a part of the Theridiidae)

20, web ?, Australian region: Hadrotarsus
Tetragnathidae
600, orb webs, worldwide: Eugnatha, Leucauge, Pachygnatha, Tetragnatha (Wiehle 1927; Levi
1980)
Metidae (often considered a subfamily of the Tetragnathidae)

100, orb webs, worldwide: Meta, Metellina (Potzsch 1966; Blanke 1974; Levi 1980; Toft 1983)
Appendix 385
Cyatholipidae (often considered a part of other araneoid families)

20, sheet web, probably worldwide: Cyatholipus, Tekella
Araneidae (= Argiopidae, Epeiridae)

Orb weavers, 4000, mainly orb webs, worldwide: Aculepeira, Alpaida, Arachnura, Araneus (= Epei-
ra), Araniella, Argiope, Celaenia, Cyciosa, Cyrtarachne, Cyrtophora, Dicrostichus, Eriophora,
Gasteracantha, Gea, Kaira, Mastophora, Mecynogea, Metepeira, Micrathena, Neoscona, Nephila,
Nephilengys, Nuctenea, Paraplectana, Pasilobus, Phonognatha, Poecilopachys, Sallassina, Scolo-
derus, Singa, Verrucosa, Zygiella (Wiehle 1927,1928, 1931; Peters 1955;Kajak 1967 ;M.H. Robinson
and B.C. Robinson 1970a,b, 1973a, 1974, 1976; PeakaIl1971; Blanke 1972, 1975a; B.C. Krakauer
1972; Clyne 1973; Buskirk 1975; Levi 1980, 1975b; Tolbert 1975; Eberhard 1977, 1980b; Wise
1979; Rypstra 1981; Murakami 1983; Nentwig 1985b; T. Miyashita 1986)
Mirnetidae
Pirate spiders, 150, no webs, worldwide: Arcys, Australomimetus, Ero, Gelanor, Mimetus (Czaika
1963; Cutler 1972; Heimer 1984, 1986a)
Linyphiidae (including Erigoninae = Micryphantidae)

Sheet web spiders and dwarf spiders, 4000, different forms of sheet webs, worldwide: Allomengea,
Bathyphantes, Bolyphantes, Centromerita, Centromerus, Ceratinella, Drapetisca, Erigone, Erigoni-
dium, Floronia, Frontinella, Gnathonarium, Gongylidium, Hypomma, Lepthyphantes, Linyphia,
Macrargus, Microneta, Micryphantes, Mynoglenes, Neriene, Oedothorax, Pityohyphantes, Saloca,
Stemonyphantes (Turnbull 1962; Wiehle 1956; Buche 1966; Wise 1975, 1976; Schaefer 1976b;
Blest and Pomeroy 1978; Toft 1980; Suter 1981)
Theridiosomatidae
Ray spiders, 100, different web types, worldwide: Wendilgarda, Theridiosoma
Symphytognathidae
50, sheet webs and orb webs, probably worldwide (tropics): Curimagua, Patu, Symphytognatha
(R.R. Forster and Platnick 1977)
Anapidae (often considered a subfamily of the Symphytognathidae)

65, orb webs, worldwide: Anapis, Anapisona, Pseudanapis (R.R. Forster and Platnick 1977)
Micropholcommatidae (often considered a part of the Symphytognathidae)

30, web ?, Australia, New Zealand: Micropholcomma (R.R. Forster and Platnick 1977)
Textricellidae (often considered a part ofthe Symphytognathidae)

30, sheet web, Chile, Australia to New Zealand: Textricella (R.R. Forster and Platnick 1977)
Mysmenidae (often considered a subfamily of the Symphytognathidae or Theridiidae)

50, space web, worldwide: Maymena, Mysmena, Mysmenopsis (R.R. Forster and Platnick 1977;
Platnick and Shadab 1978; Vollrath 1978)
Nicodamidae
10, irregular web, S.E. Asia to New Zealand: Megadictyna, Nicodamus (R.R. Forster 1970)
Lycosidae (including Bradystichidae)

Wolf spiders, 3000, no webs (except Hippassinae with funnel webs), worldwide: Alopecosa (= Ta-
rentula), Arctosa, Geolycosa, Hippassa, Lycosa, Pardosa, Pirata, Schizocosa, Sosippus, Trochosa
386 Appendix
(Nct>rgaard 1951; W. Hackman 1957; Bucherl 1960; Engelhardt 1964; Hallander 1967, 1970; K.
Miyashita 1968a,b, 1969a,b; Edgar 1971a,b; Schaefer 1974; van Dyke and Lowrie 1975; Ford
1978; Humphreys 1978a; McQueen 1978, 1983)
Pisauridae (including Dolomedidae, Trechaleidae)

Nursery-web spiders, 600, probably all species with funnel webs (sometimes only juveniles), world-
wide: Dolomedes, Pisaura, Thaumasia (Palmgren 1939; D.S. Williams 1979)
Oxyopidae
Lynx spiders, 450, space web often reduced, worldwide: Oxyopes, Peucetia (Turner 1979)
Senoculidae
35, no webs, tropical America: Senoculus
Cycloctenidae
No web: Cycioctenus
Ctenidae
600, no webs, worldwide (tropics, subtropics): Cupiennius, Ctenus, Phoneutria (Melchers 1963;
Seyfarth 1980)
Acan thoctenidae
25, cribellate, no web, tropical America: Acanthoctenus
Zoropsidae
20, cribellate, worldwide: Zoropsis
Zoridae (also considered a subfamily of the Ctenidae)

50, no web, worldwide: Zora
Psechridae
30, cribellate, funnel webs or sheet webs, S.E. Asia to New Zealand: Fecenia, Matachia, Psechrus
Selenopidae
200, no web, worldwide (tropics): Selenops
Heteropodidae (= Eusparassidae, Sparassidae)

Giant crab spiders, huntsman spiders, 900, no webs, worldwide: Heteropoda, Micromata, Olios
(Homann 1946)
Philodromidae
500, no webs, worldwide: Philodromus, Thanatus, Tibellus
Amaqrobiidae
300, cribellate, irregular web, worldwide: Amaurobius, Exlinea, Metaltella (Forster and Wilton
1973)
Titanoecidae (also considered a subfamily of the Amaurobiidae)

50, cribellate, irregular web, worldwide: Titanoeca (Szlep 1966)
Appendix 387
Stiphidiidae (also considered a subfamily of the Amaurobiidae)

30, cribellate, sheet web, Australia to New Zealand: Stiphidius. Cambridgea (R.R. Forster and Wil-
ton 1973)
Neolanidae
3, crlbellate, sheet web, New Zealand: Neolana (R.R. Forster and Wilton 1973)
Am phinectidae
100, cribellate, irregular webs, Australia, New Zealand: Amphinecta. Mamosa (R.R. Forster and
Wilton 1973)
Tengellidae
3, cribellate, funnel web, C. America: Tengella
Agelenidae
Funnel weavers, 600, funnel webs, worldwide: Agelena. Agelenopsis. Barronopsis, Cicurina, Coelo-
tes, Histopona, Hololena, Rualena. Tegenaria (Schlott 1931; S.E. Jones 1941; Tretzel 1961;Tum-
bull 1965 ; Krafft 1967,1970; Riechert 1976, 1985)
Dictynidae
500, cribellate, irregular web, worldwide: Dictyna. Heterodictyna, Mallos, Nigma (Jackson 1982b,c)
Hahniidae
150, sheet web, worldwide: Antistea, Hahnia
Cybaeidae (also considered a subfamily of Agelenidae, Dictynidae)

100, sheet or funnel web, worldwide: Cybaeus (Lehtinen 1967; R.R. Forster 1970)
Anyphaenidae
500, no webs, worldwide: Anyphaena, Aysha, Wulfila
Amaurobioididae (also considered a part of Anyphaenidae)

10, no webs, S. Hemisphere: Amaurobioides (Lamora11968; R.R. Forster 1970)
Miturgidae
Worldwide: Machadonia, Miturga
Desidae (including Toxopidae) (also considered a part of Agelenidae)

100, no web, worldwide: Desis, /xeuticus, Toxopus (Lamora11968; R.R. Forster 1970)
Argyronetidae (also considered a subfamily of Agelenidae)

Water spider, 1, dome-shaped web, Eurasia: Argyroneta (F. Braun 1931; Crome 1952/53)
Clubionidae (including Corinnidae)

Sac spiders, 1300, no webs, worldwide: Castianeira, Chiracanthium, Clubiona, Corinna, Trachelas
(Orl 1975; Austin 1984; Pollard 1984; Pollard and Jackson 1984)
388 Appendix
Liocranidae (also considered a subfamily of Clubionidae)

300, no webs, worldwide: Agroeca, Liocranum
Gnaphosidae
2000, no webs, worldwide: Call1iepis, Cesonia, Drassodes, Gnaphosa, Haplodrassus, Herpyllus,
Prodidomus, Zelotes (Heller 1976)
Trochanteriidae (also considered a part of the Gnaphosidae)

20, no web, Africa, Argentina: Trochanteria
Platoridae
10, no webs, tropical America and Asia: Plator
Ci thaeronidae
10, no webs, Africa to India: Cithaeron
Gallieniellidae
8, no webs, Madagascar: Gallieniella, Legendrena
Ammoxenidae
10, no webs, S. Africa (arid): Ammoxenus (Dippenaar and Meyer 1980)
Thomisidae
Crab spiders, 1500, no web, worldwide: Diaea, Heriaeus, Misumena, Misumenoides, Misumenops,
Thomisus, Tmarus, Xysticus (Packard 1905; Gabritschevsky 1927; Weigel 1941 ; Levy 1970; Morse
1981)
Aphantochilidae (also considered a subfamily of the Thornisidae)

10, no webs, tropical S. America: Aphantochilus (Oliveira and Sazima 1984)
Salticidae (including Lyssomanidae)

Jumping spiders, 5000, worldwide: Aelurillus, Brettus, Cobanus, Corythalia, Cosmophasis, Cyllo·
belus, Cyrba, Euophrys, Euryattus, Evarcha, Heliophanus, Lyssomanes, Marpissa, Metacybra,
Metaphidippus, Pellenes, Phiale, Phidippus, Philaeus, Phlegra, Portia, Pseudicius, Salficus, Sassacus,
Sitticus, Trite (Bonnet 1933; Crane 1949; Jackson 1978a,b, 1980, 1982b; Hill 1979; L.M. Forster
1982; Wanless 1980, 1984a,b, 1985)
References
Important papers and review articles are marked with an asterisk (*)
Abraham BJ (1983) Spatial and temporal patterns in a sagebrush steppe spider community (Arach-
nida:Araneae). J ArachnoI11:31-50
Adams PA, Heath JE (1964) An evaporative cooling mechanism in Pholus achemon (Sphingidae). J
Res Lepid 3 :69-72
Ahearn GA (1970) Water balance in the whip scorpion Mastigoproctus giganteus (LUCAS) (Uropygi).
Comp Biochem Physiol35 :339-353
Ahearn GA, Hadley NF (1976) Water transport in desert scorpion ileum: functional roles of luminal
sodium and potassium. Nature (London) 261 :66-68
Ahearn GA, Hadley NF (1977) Water transport in perfused scorpion ileum. Am J Physiol Regul
Integr Comp PhysioI2:198-207
Aitchison CW (1978) Spiders active under snow in southern Canada. Symp Zool Soc London 42:
139-148
Aitchison CW (1979a) Winter-active invertebrates in southern Canada I Co llembola. Pedobiologia
19:113--120
Aitchison CW (1979b) Winter-active invertebrates in southern Canada IV Diptera and Hymenoptera.
Pedobiologia 19:176-182
Aitchison CW (1981) Feeding and growth of Coelotes atropos (Agelenidae, Araneae) at low tem-
peratures. J Arachnol 9 :327 -330
*Aitchison CW (1983) Low temperature and preferred feeding by winter-active Collembola (Insecta,
Apterygota). Pedobiologia 25 :27 -36
*Aitchison CW (1984a) Low temperature feeding by winter-active spiders. J Arachnol12:297 -305
Aitchison CW (1984b) The phenology of winter-active spiders. J ArachnoI12:249-271
Aitchison CW (1984c) A possible subnivean food chain. In: Merritt JF (ed) Winter ecology of small
mammals. Carnegie Mus Nat Hist Spec PubllO:363-372
Aitchison CW, Hegdegar B (1982) Cryoprotectants in winter-active collembolans and spiders. Cryo
Letters. Abstr Pap Symp Invert Cold Hardiness, Oslo
Akai H (1971) Ultrastructure of fibroin in the silk gland of larval Bombyx mori. Exp Cell Res 69:
219-223
Albert R (1982) Untersuchungen zur Struktur und Dynamik von Spinnengesellschaften verschiede-
ner Vegetationstypen im Hoch-Solling. Thesis, Univ Bremen
Alberti G (1973) Ernlihrungsbiolcigie und Spinnverm6gen der Schnabelmilben (Bdellidae, Trombi-
diformes). Z Morphol Tiere 76:285-338
Alberti G (1974) Fortpflanzungsverhalten und Fortpflanzungsorgane der Schnabelmilben (Bdelli-
dae, Trombidiformes). Z Morphol Tiere 79:133-153
Alberti G, Ehrnsberger R (1977) Rasterelektronenmikroskopische Untersuchungen zum Spinnvermo-
gen der Bdelliden und Cunaxiden (Acari, Prostigmata). Acarologia 19:56-61
Alberti G, Storch V (1973) Zur Feinstruktur der "Munddriisen" von Schnabelmilben (Bdellidae,
Trombidiformes). Z Wiss ZooI186:149-160
Alberti G, Storch V (1974) tiber Bau und Funktion der Prosoma-Driisen von Spinnmilben (Tetra-
nychidae, Trombidiformes). Z Morphol Tiere 79:133-154
Alberti G, Storch V (1983) Zur Ultrastruktur der Mitteldarmdriisen von Spinnentieren (Scorpiones,
Aranae, Acari) unter verschiedenen Ernlihrungsbedingungen. Zool Anz 211 :145 -160
390 References
Alexander AJ, Ewer DW (1958) Temperature adaptive behaviour in the scorpion Opisthophthalmus
latimanus Koch. J Exp Bioi 35 :349-359
Alexander CP (1936) Crane-flies (Tipulidae, Diptera) from the Malayan Islands and Northern
Australia. Rev Suisse Zool43 :73
Almquist S (1969) Seasonal growth in some dune-living spiders. Oikos 20 :392-408
Almquist S (1970) Thermal tolerances and preferences of some dune-living spiders. Oikos 21:
230-236
Almquist S (1971) Resistance to dessication in some dune-living spiders. Oikos 22:225-229
Altner H, Prillinger L (1980) Ultrastructure of invertebrate chemo- thermo- and hygroreceptors
and its functional significance. Int Rev CytoI67:69-139
Anatasiu-Dumitresco M (1937) Les glandes sericigenes tubulaires de Teutana triangulosa Walck.
Arch Zool exp gen 79:58-68
Anatasiu-Dumitresco M (1942) Contributions a l'etude anatomique et cytologique de l'appareil
sericigene des araignees. 2e Partie. Anal Acad Rom Mem Sect stiin Ser III 17 :263-349
*Andersen SO (1970) Amino acid composition of spider silks. Comp Biochem Physiol 35 :705 -711
Anderson CM, Tillinghast EK (1980) GABA and taurine derivatives on the adhesive spiral of the
orb web of Argiope spiders, and their possible behavioural significance. Physiol En tomol 5:
101-106
Anderson JF (1966) The excreta of spiders. Comp Biochem PhysioI17:973-982
*Anderson JF (1970) Metabolic rates of spiders. Comp Biochem Physiol 33 :51-72
*Anderson JF (1974) Responses to starvation in the spiders Lycosa lenta Hentz and Filistata hiber-
naZis (Hentz). Ecology 55 :576-585
Anderson JF (1978) Energy content of spider eggs. Oecologia 37:41-57
Anderson JF, Prestwich KN (1975) The fluid pressure pumps of spidera (Chelicerata, Araneae). Z
Morphol Tiere 81 :257 -277
Anderson JF, Prestwich KN (1980) Scaling of subunit structures in book lungs of spiders (Araneae).
J MorphoI165:167-174
*Anderson JF, Prestwich KN (1982) Respiratory gas exchange in spiders. Physiol ZooI55:72-90
Anderson JF, Prestwich KN (1985) The physiology of exercise at and above maximal aerobic ca-
pacity in a theraphosid (tarantula) spider, Brachypelma smithi (F.O. Pickard-Cambridge). J
Comp Physiol B 155 :529-539
Andre M (1932) La secretion de la soie chez les acariens. Livre du Centenaire Soc Zool Fr 457 -473
*Angersbach D (197 Sa) Oxygen pressures in haemolymph and various tissues of the tarantula, Eury-
pelma helluo. J Comp PhysioI98:133-145
Angersbach D (1975b) Die Sauerstoffversorgung der Gewebe zweier Spinnen, Eurypelma helluo
und Cupiennius salei: Lokale P0 2 Messungen in der Hiimolymphe und in verschiedenen Mus-
keln. Verh Dtsch Zool Ges Bochum:277-280
*Angersbach D (1978) Oxygen transport in the blood of the tarantula Eurypelma californicum: P0 2
and pH during rest, activity and recovery. J Comp PhysioI123:113-125
*Apstein C (1889) Bau und Funktion der Spinndriisen der Araneida. Arch Naturgesch 55 :29-74
Archer AF (1946) The Theridiidae or comb-footed spiders of Alabama. Ala Mus Nat Hist 22:5-67
Aspey W, Lent C, Meeker M (1972) Effect of humidity on dessication by living and dead wolf
spiders (Araneae:Lycosidae). Experientia 28:1249-1250
Austad SN (1982) First male sperm priority in the bowl and doily spider, Frontinella pyramitela.
Evolution 36 :777 -785
Austad SN (1984) Evolution of sperm priority patterns in spiders. In: Smith RL (ed) Sperm com-
petition and the evolution of animal mating systems. Academic Press, London New York
Austin AD (1984) Life history of Club iona robusta L. Koch and related species (Araneae, Clubioni-
dae) in South Australia. J Arachnol12 :87 -104
Bachmann E, Schaefer S (1983) Notes on the life cycle of Phalangium opilio (Arachnida:Opilioni-
da). Verh Naturwiss Ver Hamburg NF 26:255-263
Baehr B, Eisenbeis G (1985) Comparative investigations on the resistance to dessication in Lycosi-
dae, Hahniidae, Linyphiidae and Micryphantidae (Arachnida, Araneae). Zool Jahrb Abt Syst
112:225-234
Baerg WJ (1958) The tarantula. Univ Kansas Press, Lawrence
References 391
Baert L (1978) Influence de la photoperiodicite sur la maturation ovarienne chez Gongylidium

rufipes (Sundevall) (Araneae, Linyphiidae). Rev ArachnoI2:23-27
Baert L (1980) Autoecologie de Gongylidium rUfipes (Sundevall, 1829) (Araneae, Linyphiidae).
1. Influence de temperatures constantes sur la duree de developpement postembryonnaire. Bull
Inst R Sci Nat Belg 52; 19:1-14
Bakken GS (1976) An improved method for determining thermal conductance and equilibrium
body temperature with cooling curve experiments. J Therm Bioi 1 :169-175
Barnard 0 (1984) When cheats may prosper. In: Barnard CJ (ed) Producers and scroungers. Chrom
Helm, Beckenham, pp 6-33
Barnard EA (1973) Comparative biochemistry of digestive enzymes. In: Prosser CL (ed) Comparative
animal physiology, 3rd edn. Saunders, Philadelphia
Bartels M (1930) tiber den Frefl>mechanismus und chemischen Sinn einiger Netzspinnen. Rev Suisse
ZooI37:1-42
Barth FG (1969) Die Feinstruktur des Spinneninteguments. I. Die Cuticula des Laufbeins adulter
hautungsferner Tiere (Cupiennius salei Keys). Z Zellforsch Microsk Anat 97: 137 -159
Barth FG (1970) Die Feinstruktur des Spinneninteguments. II. Die raumliche Anordnung der Mikro-
fasern in der lamellierten Cuticula und ihre Beziehung zur Gestalt der PorenkanaIe (Cupiennius
salei Keys., adult, hautungsfern, Tarsus). Z Zellforsch Microsk Anat 104:87 -106
*Barth FG (1973) Microfiber reinforcement of an arthropod cuticle. Laminated composite material
in biology. Z Zellforsch Microsk Anat 144:409-433
Barth FG (1982) Spiders and vibratory signals: sensory reception and behavioural significance. In:
Witt PN, Rovner JS (eds) Spider communication. Princeton Univ Press, Princeton, pp 67-120
*Barth FG (1985) Neuroethology of the spider vibration sense. In: Barth F (ed) Neurobiology of
arachnids. Springer, Berlin Heidelberg New York, pp 162-189
*Barth FG, Blickham R (1984) Arthropoda. Mechanoreception. In: Bereiter-Hahn J, Matoltsy AG,
Richards KS (eds) Biology of the integument, vol 1. Invertebrates. Springer, Berlin Heidelberg
New York, pp 554-582
Barth FG, Seyfarth E-A (1979) Cupiennius salei Keys. (Araneae) in the highlands of central Guate-
mala. J Arachnol 7:255-263
Bartholomew GA (1981) A matter of size: an examination of endothermy in insects and terrestrial
vertebrates. In: Heinrich B (ed) Insect thermoregulation. Wiley, New York, pp 45-78
Barton C (1938) How it feels to be bitten by a black widow. Nat Hist 42:43-44
Baust JG, Edwards JS (1979) Mechanisms of freeZing tolerance in an Antarctic midge, Be/gica
antarctica. Physiol EntomoI4:1--5
Beasley BI (1942) Arachnidism; case report of arachnidism. With death and autopsy finding. South
Surg 11:737-741
Beck SD (1980) Insect photoperiodism, 2nd edn. Academic Press, London New York
Behrens W, BUckmann D (1983) Ecdysteroids in the pycnogonid Pycnogonum litorale (Strom)
(Arthropoda, Pantopoda). Gen Comp Endocrinol51 :8-14
Beittinger TL, Fitzpatrick LC (1979) Physiological and ecological correlates of preferred tempera-
ture in fish. Am ZooI19:319-329
Bell AL, Peakall DB (1969) Changes in fine structure during silk protein production in the ampul-
late gland of the spider Araneus sericatus. J Cell Bioi 42:284-295
Bentzien M (1973) Biology of the spider Diguetia imperiosa. Pan-Pac EntomoI49:110-123
Bertkau Ph (1882) tiber das Cribellum und Calamistrum. Ein Beitrag zur Histologie, Biologie und
Systematik der Spinnen. Arch Naturgesch 48:316-362
Bertkau PH (1884) Verdauungsapparat der Spinnen. Correspondenzbl Naturhist Ver Preuss Rheinl
Westfalen 2:1-11
Bertkau PH (1885) tiber den Verdauungsapparat der Spinnen. Arch Mikrosk Anat 24:398-451
Bertkau PH (1887) tiber den Bau der Chernetiden oder Pseudoscorpione. Sitzungsber Niederrhein
Ges Naturwiss 44:112-117
Bertkau PH (1888) Ueber Mermis in Tarentula inquilina und die durch den Parasiten bedingte Ste-
rilitat des Wirthes. Verh Naturhist Ver Rheinl Westfalen 45 :91-92
Bethge W (1973) C>kologisch-physiologische Untersuchungen Uber die Bindung von Erigone longi-
palpis (Araneae, Micryphantidae) an das Litoral. Faun-6kol Mitt 4:223-240
Bettini S (ed) (1978) Arthropod venoms. Springer, Berlin Heidelberg New York, p 977
392 References
Bettini S, Cantore G (1959) Quadro clinico dellatrodectismo. Riv Parassit 20:49-72

Bettini S, Maroli M (1978) Venoms of Theridiidae, genus Latrodectus. In: Bettini S (ed) Arthropod
venoms. Springer, Berlin Heidelberg New York, pp 149-185
Bettini S, Toschi-Frontali N (1960) Biochemical and toxicological aspects of Latrodectus tredecim-
guttatus venom. In: Bettini S (ed) Proc XI Int Congr Entomol. 1st Entomol Agric Univ Pavia,
pp 115-121
Bettini S, Antonini E, Cantore G (1953) La terapia del morso di Latrodectus 13-guttatus Rossi.
Nota II. - Cinque novi casi di latrodectismo trattati con gluconata di calcio endovena. Arch
Ital Soc Med Trop Parassit 34:579-587
Biere 1M, Uetz GW (1981) Web orientation in the spider Micrathena gracilis (Araneae:Araneidae).
Ecology 62 :336 -344
Biggemann U (1971) Estudios bioquimicos del veleno de la arafia L.laeta y su mecanismo de accion.
Thesis, Concepcion, Chile
Blackwall I (1841) On the number and structure of the mammulae employed by spiders in the pro-
cess of spinning. Trans Linn Soc London 18:219-224
Blair AW (1934) Spider poisoning, experimental study of the effect of the bite of the female Latro-
dectus mactans in man. Arch Intern Med 54:831-834
Blake I (1926) A comparison of the animal communities of coniferous and deciduous forests. III.
BioI Monogr 10:1-148
*Blanke R (1972) Untersuchungen zur Okophysiologie und Okoethologie von Cyrtophora citricola
ForskYil (Araneae, Araneidae) in Andalusien. Forma Functio 5: 125 -206
Blanke R (1973a) Nachweis von Pheromonen bei Netzspinnen. Naturwissenschaften 60:481
Blanke R (1973b) Neue Ergebnisse zum Sexualverhalten von Araneus cucurbitinus Cl (Araneae,
Araneidae). Forma Functio 6:279-290
Blanke R (1974) Rolle der Beute beim Kopulationsverhalten von Meta segmentata (C1.). Forma
Functio 7:83-94
Blanke R (1975a) Untersuchungen zum Sexualverhalten von Cyrtophora cicatrosa (Stoliczka) (Ara-
neae, Araneidae). Z TierpsychoI37:62-74
Blanke R (1975b) Die Bedeutung der Guanocyten fUr den physiologischen Farbwechsel bei Cyrto-
phora cicatrosa (Arachnida: Araneidae). Entomol Germ 2: 1-6
Blanke R (1982) Die Variabilitat von Zeichnungsmuster und Helligkeit des Abdomens beiAraneus
diadematus Clerck und Araneus marmoreus Clerck (Arachnida:Araneae). Z Zool Syst Evolu-
tionsforsch 20 :63 -75
Blest AD (1978) The rapid synthesis and destruction of photoreceptor membrane by a dinopid
spider: a daily cycle. Proc Roy Soc London Ser B 200:463-483
Blest AD (1985) The fine structure of spider photoreceptors in relation to function. In: Barth FG
(ed) Neurobiology of arachnids. Springer, Berlin Heidelberg New York, pp 79-102
Blest AD, Pomeroy G (1978) The sexual behaviour and genital mechanics of three species of Myno-
glenes (Araneae: Linyphiidae). I Zool London 185 :319 -340
Blest AD, Taylor HH (1977) The clypeal glands of Mynoglenes and of some other linyphiid spiders.
I Zool (London) 183:473-493
*Blickhan R,Barth FG (1985) Strains in the exoskeleton of spiders. I Comp Physiol A 157:115-147
Bligh 1, Cloudsley-Thompson lL, Macdonald AG (1976) Environmental physiology of animals.
Blackwell, Oxford
Block W (1979) Cold tolerance of micro-arthropods from Alaskan taiga. Ecol EntomoI4:103-110
Block W (1980) Survival strategies in polar terrestrial arthropods. Bioi I Linn Soc 14:29-38
Block W (1982) Supercooling points of insects and mites on the Antarctic Peninsula. Ecol Ento-
moI7:1-8
Block W, SI/lmme L (1982) Cold hardiness of terrestrial mites at Signy Island, maritime Antarctic.
Oikos 38:157 -167
Block W, Young SR (1978) Metabolic adaptations of Antarctic terrestrial micro-arthropods. Comp
Biochem PhysioI61A:363-368
Block W, Zettel I (1980) Cold hardiness of some alpine Collembola. Ecol EntomoI5:1-9
Block W, Young SR, Conradi-Larsen EM, SI/lmme L (1978) Cold tolerance of two Antarctic ter-
restrial arthropods. Experientia 34:1166-1167
References 393
*Blumenthal H (1935) Untersuchungen tiber das "Tarsalorgan" der Spinnen. Z Morphol Oekol Tiere
29:667-719
Bodnaryk RP (1977) Stages of diapause development in the pupa of Mamestra configurata based
on the l3-ecdysone sensitivity index. J Insect Physiol 23:537 -542
Bogen E (1926) Arachnidism. A study in spider poisoning. Arch Intern Med 38:334-623
Bogen E (1932) Poisonous spider bites. Ann Intern Med 6:375-388
*Bohart RM, Menke AS (1976) Sphecid wasps of the World. Univ California Press, Berkeley
Boissin L, Bouix G, Maurand J (1970) Recherches histologiques et histochimiques sur Ie tractus
genital male du pseudoscorpion, Hysterochelifer meridianus (1. Koch). Bull Mus Hist Nat 2nd
Ser 42:491-501
Bonaric JC (1976) Effects of ecdysterone on the molting mechanisms and duration of the inter-
molt period inPisaura mirabilis Cl. Gen Comp EndocrinoI30:267-272
Bonaric JC (1977a) Action de l'ecdysterone sur la reduction de duree de la periode d'intermue
chez Pisaura mirabilis Cl. (Araneae, Pisauridae): determination de l'effet optimal par inocula-
tion de concentrations croissantes. Arch Zool Exp Gen 118:43-51
Bonaric JC (1977b) Action de l'ecdysterone sur Ie cycle de mue de l'araigm!ePisaura mirabilis Cl.:
variation des reponses en fonction de la date d'intervention au cours de l'intermue nymphal.
Arch Zoo I Exp Gen 118:409-421
Bonaric JC (1977c) Existence d'une diapause hivernale revelee par l'action de traitements hormon-
aux chez l'araignee Pisaura mirabilis Cl. (Pisauridae). CR Acad Sci 284:1297 -1300
Bonaric JC (1978) Action de traitements hormonaux sur Ie deroulement des phenomenes de mue
chez l'araignee Pisaura mirabilis Cl. (Araneae, Pisauridae). Symp Zoo I Soc London 42 :469 -475
Bonaric JC (1980) Contribution a l'etude de la biologie du developpement chez l'araigneePisaura
mirabilis (Clerck, 1758). Approche physiologique des phenomenes de mue et de diapause hiver-
nale. Thesis, Univ Montpellier
Bonaric JC (1982) Cycle tegumentaire et hormone de mue chez les nymphes de Pisaura mirabilis
(Araneae, Pisauridae). Bull Soc Zoo I Fr 107:449-452
Bonaric JC, De Reggi M (1977) Changes in ecdysone levels in the spider Pisaura mirabilis nymphs
(Araneae, Pisauridae). Experientia 33:1664-1665
Bonnet P (1924) Sur la nature des aliments que les araignees peuvent absorber et sur Ie sens du gout
chez les animaux. C R Soc Bioi 91:1194-1196
Bonnet P (1930) La mue, l'autotomie et la regeneration chez les araignees avec une etude des Do·
lomedes d'Europe. Bull Soc Hist Nat Toulouse 65 :237-700
Bonnet P (1933) Cycle vital dePhilaeus chrysops Poda. Arch Zool Exp Gen 71 :129-144
Bosenberg W, Strand E (1906) Japanische Spinnen. Abh Senckenberg Nat Ges 30:98-422
*Bouligand Y (1965) Sur une architecture torsadee repandue dans de nombreuses cuticles d'arthro-
podes. C R Acad Sci 261:3665-3668
Brader L (1972) Verspreiding van insekten door de lucht. Aerobiologie. Cent Landbouwpubl
Landbouwdok, Wageningen, pp 58-77
Bradley R (1982) Digestion time and reemergence in the desert grassland scorpion Paruroctonus
utahensis (Williams) (Scorpionida, Vaejovidae). Oecologia 55:316-318
Braun.F (1931) Beitriige zur Biologie und Atmungsphysiologie der Argyroneta aquatica Cl. Zool
Jahrb Syst 62:175-262
Braun R (1956) Zur Biologie von Teutana triangulosa (Walck) (Araneae, Theridiidae, Asageneae).
Z Wiss ZooI159:255-318
Braun R (1969) Zur Autokologie und Phanologie der Spinnen (Araneida) des Naturschutzgebietes
"Mainzer Sand". Mainzer Naturwiss Arch 8:193-288
Brazil OV, Vellard J (1925) Contribucao as estudo de veneno das aranhas. Mem Inst Butantan 2:
1-70
Breed AL, Levine VD, Peakall DB, Witt PN (1964) The fate of the intact orb web of the spider
Araneus diadematus Cl. Behavior 23:43-60
Briceflo RD (1985) Sticky balls in webs of the spider Modisimus sp. (Araneae:Pholcidae). J Arach-
nol13:267-269
Bridges CR, Kestler P, Scheid P (1980) Tracheal volume in the pupa of the saturniid moth Hyalo-
phora cecropia determined with inert gases. Respir PhysioI40:281-291
394 References
Brignoli PM (1966) Le societa eterotipiche degli Araneidi (I). Rend Accad Naz 16:219-246
Brignoli PM (1974) Ober die Gruppe der Haplogynae (Araneae). Proc 6th Int Arachnol Congr,
Amsterdam, pp 33-38
*Brignoli PM (1983) A catalogue of the Araneae described between 1940 and 1981. Manchester
Univ Press, Manchester
Bristowe WS (1923) A British semi-marine spider. Ann Mag Nat Hist 12:154-156
Bristowe WS (1931a) Notes on the biology of spiders V Theridion ovatum Clerck, its habits and
varieties. Ann Mag Nat Hist 8:466-469
Bristowe WS (1931b) Notes on the biology of spiders VI Spitting as a mean of capturing prey by
spiders. Ann Mag Nat Hist 10th Ser 8:469-471
Bristowe WS (1932) The liphistiid spiders. With an appendix on their internal anatomy by J. Millot.
Proc Zool Soc 103 :1015 -1057
*Bristowe WS (1941) The comity of spiders. Ray Soc, London
*Bristowe WS (1958) The world of spiders. Collins, London
Brown KM (1981) Foraging ecology and niche partitioning in orb-weaving spiders. Oecologia Berlin
50:380-385 .
Brown KS (1984) Adult-<lbtained pyrrolizidine alkaloids defend ithomiine butterflies against a
spider predator. Nature (London) 309:707-709
Brown VK, Hodek I (1983) Diapause and life cycle strategies in insects. Junk, Boston London
Browning HC (1941) The relation of instar length to the external and internal environment in Te-
genaria atrica (Arachnida). Proc Zool Soc London Ser Alii :303-317
Browning HC (1942) The integument and moult cycle of Tegenaria atrica (Araneae). Proc R Soc
London Ser B 131 :65 -86
Buatois A, Croze JP (1978) Thermal responses of an insect subjected to temperature variations. J
Therm Bioi 3:51-56
Buchar J (1968) A winter trip among spiders. Ziva 16:24-25
Buche We (1966) Beitrage zur Okologie und Biologie winterreifer Kleinspinnen mit besonderer
Beriicksichtigung der Linyphiiden Macrargus rufus rufUS (Wider), Macrargus rufUS carpenteri
(Cambridge) und Centromerus silvaticus (Blackwall). Z Morphol Okol Tiere 57 :329-448
Biicher! W (1956) Studies on dried venom of Phoneutria fera Perty 1833. Adv Sci 44:95-97
Biicher! W (1960) Die Tarantelspinne Lycosa erythrognatha Lucas 1836. Arzneimittel Forsch 10:
774-777
Biicherl W (1969) Biology and venoms of the most important south american spiders of the genera
Phoneutria, Loxosceles, Lycosa and Latrodectus. Am Zool 9 :157 -159
Buchli H (1961) Observations preliminaires sur Ie rythme d 'activite et la biologie de Nemesia caemen-
taria. Vie Milieu 12:297 -304
Buchli H (1965a) Notes preliminaires concernenant Ie comportement de chasse et Ie rythme d'ac-
tivite de la mygale maconne Nemesia caementaria. Rev Ecol BioI 2 :403-438
Buchli H (1965b) Le rythme d'activite chez la mygale ma!tonneNemesia caementaria (Ctenizidae).
Ann Epiphytes 9:195-196
Buchli H (1970) Notes sur Ie cycle de reproduction, la ponte et Ie developpement post-embryon-
naire de Nemesia caementaria (Latr.) (Ctenizidae, Mygalomorphae). Rev Ecol BioI Sol 7 :95-
143
Biickmann D (1974) Die hormonale Steuerung der Pigmentierung und des morphologischen Farb-
wechsels bei den Insekten. Fortschr ZooI22:1-22
Burri P, Weibel ER (1971) Morphometric estimation of pulmonary diffusion capacity. II. Effect
of P0 2 on the growing lung. Adaptation of the growing rat lung to hypoxia and hyperoxia.
Respir Physiol 11: 24 7 - 264
Burton RF (1984) Hemolymph composition in spiders and scorpions. Comp Biochem Physiol A
78:613-616
Buskirk RE (1975) Aggressive display and orb defence in a colonial spider, Metabus gravidus. Anim
Behav 23:560-567
*Buskirk RE (1981) Sociality in the Arachnida. In: Herman H (ed) Social insects. Academic Press,
London New York
Cabanac M, LeGuelte L (1980) Temperature regulation and prostaglandin EI fever in scorpions. J
Physiol (London) 303:365-370
References 395
Candelas GC, Cintron J (1981) A spider fibroin and its synthesis. J Exp Zoo 1 216:1-6
Candelas GC, Lopez F (1983) Synthesis of fibroin in the cultured glands of Nephila clavipes. Comp
Biochem Physiol 74B:637 -641
Cantore GP, Bettini S (1958) Contributo alIo studio del veleno di Latrodectus tredecimguttatus
Rossi. Rend 1st Sup San 21 :794-805
*Capelle TO (1966) Observations on the life history of Ogcodes rufoabominalis in northern Utah
(Diptera:Acroceridae). J Kans Entomol Soc 39:641--649
Carico JE (1978) Predatory behavior in Euryopis funebris (Hentz) (Araneae:Theridiidae) and the
evolution significance of web reduction. Symp Zool Soc London 42:51-58
Carrel JE (1978) Behavioural thermoregulation during winter in an orb-weaving spider. Symp Zool
Soc London 42:41-50
Carrel JE (1980) Determinants of nocturnal emergence patterns in a wolf spider. Proc 8th Int Congr
Arachnology, Vienna, pp 41-46
Carrel JE, Eisner TE (1984) Spider sedation induced by defensive chemicals of milliped prey. Proc
Nat! Acad Sci USA 81 :806-810
*Carre1JE, Heathcote RD (1976) Heart rate in spiders: influence of body size and foraging energetics.
Science 193:148-150
Carrel JE, Doom JP, McCormick JP (1985) Arborine and methaqualone are not sedative in the wolf
spider Lycosa ceratiola Gertsch and Wallace. J Arachnol 13 :269-271
Casey TM (1981) Behavioural mechanisms of thermoregulation. In: Heinrich B (ed) Insect thermo-
regulation. Wiley, New York, pp 79-114
*Chamberlin JC (1931) The arachnid order Chelonethida. Stanford Univ Publ Bioi Sci 7 :1-284
Charnov EL (1976) Optimal foraging: the marginal value theorem. Theor Popul Bioi 9 :129-136
Chippendale GM (1977) Hormonal regulation of larval diapause. Annu Rev EntomoI22:121-138
Christenson TE (1984) Behaviour of colonial and solitary spiders of the theridiid speciesAnelosi-
mus eximius. Anim Behav 32:725-734
Christenson TE, Goist KC (1979) Costs and benefits of male-male competition in the orb-weaving
spider Nephila clavipes. Behav Ecol Sociobiol 5 :87 -92
Christophe T (1977) Etude demographique d 'une popUlation de l'araignee Macrargus rufus (Wider)
(Linyphiidae). Bull Soc Zoo1102:187 -196
Church NS (1960a) Heat loss and the body temperatures of flying insects. I. Heat loss by evapora-
tion of water from the body. J Exp Bioi 37 :171-185
Church NS (1960b) Heat loss and the body temperatures of flying insects. II. Heat conduction
within the body and its loss by radiation and convection. J Exp Bioi 37:186-212
Claret J, Porcheron P, Dray F (1978) La teneur en ecdysones circulantes au cours du dernier stade
larvaire de l'hymenoptere endoparaiste Pimpla instigator et l'entree en diapause. CR Acad Sci
286:639-641
Clark AW, Mauro A, Longenecker HE Jr, Hurlbut WP (1970) Effects of black widow spider venom
on the frog neuromuscular junction. II Effects on the fine structure of the frog neuromuscular
junction. Nature (London) 225 :703-705
Clausen CP (1940) Entomophagous insects. McGraw-Hill, New York
Cloudsley-Thompson JL (1953) Nocturnal responses of spiders to the vibrations of a tuning fork.
In: Cloudsley-Thompson JL, Sankey JHP (eds) Some aspects of the fauna of the district around
Juniper Hall, Mickleham, Surrey - II. Entomol Mon Mag 89:284-286
Cloudsley-Thompson JL (1957) Studies in diurnal rhythms. - V. Nocturnal ecology and water-
relations of the British cribellate spiders of the genus Cinijlo (BI.). J Linn Soc (Zoo!) 43:134-
152
*Cloudsley-Thompson JL (1961a) Rhythmic activity in animal physiology and behaviour. Academic
Press, London New York
Cloudsley-Thompson JL (1961b) Some aspects of the physiology and behaviour of Galeodes arabs.
Entomol Exp App14:257 -263
Cloudsley-Thompson JL (1962) Lethal temperatures of some desert arthropods and the mechanism
of heat death. Entomol Exp App15 :278-280
Cloudsley-Thompson JL (1964) Terrestrial animals in dry heat: arthropods. In: Dill DB, Adolph EF,
Gilber CG (eds) Handbook of physiology, sect 4. Am Physiol Soc, Washington, pp 451-465
396 References
Cloudsley-Thompson IL (1968) The water-relations of scorpions and tarantulas from the Sonoran
desert. Entomol Mon Mag 103 :217 -220
Cloudsley-Thompson IL (1970) Terrestrial invertebrates. In: Whittow GG (ed) Comparative phy-
siology of thermoregulation, vol I. Academic Press, London New York, pp 15-77
Cloudsley-Thompson IL (1978) Biological clocks in Arachnida. Bull Br Arachnol Soc 4:184-191
Cloudsley-Thompson IL (1981) A comparison of rhythmic locomotory activity in tropical forest
Arthropoda with that in desert species. I Arid Environ 4:327-334
Cloudsley-Thompson IL (1983) Desert adaptations in spiders. I Arid Environ 6:307 -317
Cloudsley-Thompson IL, Constantinou C (1983) Transpiration from forest dwelling and woodland
Mygalomorphae (Araneae). Int I Biometeorol 27 :69-74
Cloudsley-Thompson IL, Constantinou C (1985) Diurnal rhythms of activity in the arboreal taran-
tula Avicularia avicularia (1.) (Mygalomorphae:Theraphosidae). I Interdiscipl Cycle Res 16:
113-116
Clyne D (1973) Notes on the web of Poecilopachys australasia (Griffith and Pidgeon, 1833) (Ara-
neida:Argiopidae). Aust Entomol Mag 1 :23-29
Coddington I (1986) The monophyletic origin of the orb web. In: Shear WA (ed) Spider webs
and spider behavior. Stanford Univ Press, Stanford
Cohen AC (1980) Hemolymph chemistry of two species of araneid spiders. Comp Biochem Physiol
66A:715-717
Collatz KG (1973) Temperature-dependent storage processes in crayfish and spiders. In: Wieser W
(ed) Effects of temperature on ectothermic organisms. Springer, Berlin Heidelberg New York,
pp 195-199
*Collatz KG, Mommsen T (1974a) Lebensweise und jahreszyklische Veranderungen des Stoffbe-
standes der Spinne Tegenaria atrica CL Koch (Agelenidae). J Comp PhysioI91:91-109
Collatz KG, Mommsen T (1974b) Die Struktur der emulgierenden Substanzen verschiedener Inver-
tebraten. I Comp PhysioI94:339-352
Collatz KG, Mommsen T (1975) Veranderung der Korperzusammensetzung und der Stoffwechsel-
intensitat der Spinne Tegenaria atrica CL Koch (Agelenidae) nach kurzem und langem Hunger.
I Comp PhysioI98:205-212
Comstock IH (1948) The spider book. Rev edn by WI Gertsch. Comstock Publ Comp Inc, Ithaca,
New York
Connell JH (1980) Diversity and the evolution of competitors, or the ghost of competition past.
Oikos 35:131-138
Constantinou C, Cloudsley-Thompson IL (1983) Uptake of water by tarantulas (Araneae, Mygalo-
morphae). Ent Month Mag 119:39-41
Cooke IAL (1965) A contribution to the biology of the British spiders belonging to the genus
Dysdera. Oikos 16 :20-25
Cooper PD (1983) Components of evaporative water loss in the desert tenebrionid beetles Eleodes
armata and Cryptoglossa ve"ucosa. Physiol ZooI56:47-55
Cooper PD, Schaffer WM, Buchmann SL (1985) Temperature regulation of honey bees (Apis mel-
li/era) foraging in the sonoran desert. I Exp Bioi 114:1-15
Costa FG, Sotelo IR (1984) Influence of temperature on the copulation duration of Lycosa maliti-
osa Tullgren (Araneae, Lycosidae). I ArachnoI12:273-277
Coville RE (1976) The predatory behavior of the spider wasp Chalybion cali/ornicum (Hymen-
optera:Sphecidae). Pan-Pac EntomoI52:229-233
*Coville RE (1982) Wasps of the genus Trypoxylon subgenus Trypargilum in North America (Hy-
menoptera: Sphecidae). Univ Calif Publ Entomo197:1-147
Coville RE. Coville PL (1980) Nesting biology and male behavior of Trypoxylon (Trypargilum)
tenoctitlan in Costa Rica (Hymenoptera:Sphecidae). Ann Entomol Soc Am 73:110-119
Coville RE. Griswold C (1983) Nesting biology of Trypoxylon xanthandrum in Costa Rica with
observations on its spider prey (Hymenoptera:Sphecidae; Araneae:Senoculidae). J Kans Ento-
mol Soc 56 :205 -216
Coville RE, Griswold C (1984) Biology of Trypoxylon (Trypargilum) superbum (Hymenoptera:
Sphecidae), a spider-hunting wasp with extended guarding of the brood by males. J Kans
Entomol Soc 57 :365-376
References 397
Coyle FA (1971) Systematics and natural history of the mygalomorph spider genusAntrodiaetus
and related genera. Bull Mus Comp ZooI141:269-402
Coyle FA (1983) Aerial dispersal by mygalomorph spiderlings (Araneae:Mygalomorphae). 1 Arach-
nol11:283-286
Coyle FA (1985a) Ballooning behaviour of Ummidia spiderlings (Araneae, Ctenizidae). 1 Arachnol
13:137-139
Coyle FA (1985b) Two-year life cycle and low palpal character variance in a Great Smoky moun-
tain population of the lamp-shade spider (Araneae, Hypochilidae, Hypochilus). 1 Arachnol13:
211-218
Coyle FA, Harrison FW, Gimsey WC, Palmer 1M (1983) Observations on the structure and func-
tion of spermathecae in haplogyne spiders. Trans Am Microsc Soc 102:272-280
Crane 1 (1949) Comparative biology of salticid spiders at Rancho Grande, Venezuela, part IV. An
analysis of display. Zoologica 34:159-215
Crawford CS (1981) Biology of desert invertebrates. Springer, Berlin Heidelberg New York
Crawford CS, Cloudsley-Thompson lL (1971) Water relations and dessication-avoiding behavior
in the vinegaroon Mastigoproctus giganteus (Arachnida:Uropygi). Entomol Exp ApplI4:99-
106
Crawford CS, Riddle WA (1974) Cold hardiness in centipedes and scorpions in New Mexico. Oikos
25:86-92
Crome W (1952/1953) Die Respirations- und Circulationsorgane der Argyroneta aquatica Cl.
(Araneae). Wiss Z Humboldt-Univ Berlin 2, II Math Nat R 3/4:53-83
Crome W (1955) Die Beziehungen zwischen dem dorsalen Zeichnungsmuster und der Metamerie
des Spinnenabdomens (II). ZoolJahrb Syst 83:541-638
Cross EA, Stith MG, Bauman TR (1975) Bionomics of the organ-pipe mud-dauber Trypoxy[on
politum (Hymenoptera:Sphecoidea). Ann Entomol Soc Am 68:901-916
Curio E (1976) The ethology of predation. Springer, Berlin Heidelberg New York
Cutler B (1972) Notes on the biology of Mimetus puritanus Chamberlin (Araneae: Mimetidae). Am
Midi Nat 87 :554
Cutler B, Richards AG (1972) Sclerotization and localization of brown and black colors in cheli-
cerates (Arthropoda). Zool lahrb Anat 89:404-421
Czaika M (1963) Unknown facts of the biology of the spider Ero furcata (Villers) (Mimetidae,
Araneae). Pol Pismo Entomol 33:229-231
Dabelow S (1958) Zur Biologie der Leimschleuderspinne Scytodes thoracica (Latreille). Zool lahrb
Syst 86 :85 -126
Dainton BH (1954) The activity of slugs. I. The induction of activity by changing temperatures. 1
Exp Bioi 31:165-187
Dainton BH, Wright J (1985) Falling temperature stimulates activity in the slug Arion ater. J Exp
Bioi 118:439-444
Dalingwater JE (1975) Further observations on eurypterid cuticles. Fossils Strata 4 :271-279
Dalingwater JE (1980) SEM observations on the cuticles of some chelicerates. 8th Int Congr Arach-
nol, Egermann, Vienna, pp 285-289
Dalingwater JE (1985) Biomechanical approaches to eurypterid cuticles and chelicerate exoskele-
tons. Trans R Soc Edinburgh Earth Sci 76:359-364
Dalwigk HB von (1973) Uber die Temperaturabhiingigkeit von Instinkthandlungen bei Libellenlar-
ven (Aeschna cyanea MOLL.) und Springspinnen (Salticus scenicus CL.). Zool Anz 190:361--
380
D'Amour FE, Becker FE, Riper W van (1936) The black widow spider. Q Rev Bioi 11 :123-160
Danks HV (1971) Populations and nesting-sites of some aculeate Hymenoptera nesting in Rubus. 1
Anim EcoI40:63-77
*Danks HV (1978) Modes of seasonal adaption in the insects. I. Winter survival. Can Entomol110:
1167-1205
*Davies ME, Edney EB (1952) The evaporation of water from spiders. J Exp Bioi 29:571-582
*Dean DA, Sterling WL (1985) Size and phenology of ballooning spiders in Texas. 1 Arachnol13:
111-120
Decae AE (1986) Popularly reports on field experiments with meat and banana eating spiders.
Newslett Br Arachnol Soc (in press)
398 References
De Carlo JM (1973) Anatomia microscopica de las espermatecas de los generos Grammostola y

Acanthoscurria (Araneae, Theraphosidae). Physis 32 :343-350
Deevey GB (1949) The developmental history of Latrodectus mactans (Fabr.) at different rates
of feeding. Am Midi Nat 42:189-219
Deevey GB, Deevey ES (1945) A life table for the black widow. Trans Conn Acad Arts Sci 1930:
115-l34
Delbecque JP, Diehl PA, O'Connor JD (1978a) Presence of ecdysone and ecdysterone in the tick
Amblyomma hebraeum Koch. Experientia 34:l379-l380
Delbecque JP, Hirn M, Delachambre J, DeReggi M (1978b) Cuticular cycle and molting hormone
levels during the metamorphosis of Tenebrio moUtor (Insecta, Coleoptera). Dev Bioi 64:11-30
Dennell R (1975) The structure of the cuticle of the scorpion Pandinus imperator (Koch). Zool J
Linn Soc 56 :249-254
*Dennel1 R (1976) The structure and lamination of some arthropod cuticles. Zool J Linn Soc 58:
159-164
Dennell R (1978) The cuticle of the hoplocarid crustacean Squilla desmaresti Risso. Zool J Linn
Soc 62:309-316
Denny M (1976) The physical properties of spider's silk and their role in the design of orb-webs.
J Exp Bioi 65 :483 -506
DenOtter CJ (1974) Setiform sensilla and prey detection in the bird-spider Sericopelma rubroni-
tens AUSSERER (Araneae, Theraphosidae). Neth J ZooI24:219-235
DeReggi M, Hirn M, Delaage M (1975) Radioimmunoassay of ecdysone an application to Droso-
phila larvae and pupae. Biochem Biophys Res Commun 66:1307-1315
DeVoe RD (1975) Ultraviolet and green receptors in principal eyes of jumping spiders. J Gen
PhysioI66:193--207
DeWilde J (1943) Some physical properties of the spinning threads of Aranea diadema L. Arach
Neerl Physiol27 :118-132
Diehl PA, Germond JE, Morici M (1982) Correlations between ecdysteroid titers and integument
structure in nymphs of the tick, Amblyomma hebraeum Koch (Acarina:Ixodidae). Rev Suisse
Zool 89 :859 -868
Digby PSB (1955) Factors affecting the temperature excess of insects in sunshine. J Exp Bioi 32:
279-298
Diniz CR (1963) Separayao de proteinas e caracteriza<;a:ode substancias ativas en venenos de aranhas
do Brasil. Anais Acad Bras Cienc 35:283-291
Dippenaar AS, Meyer MKP (1980) On the species of the African genus Ammoxenus (Araneae:
Ammoxenidae), with descriptions of two new species. J Entomol Soc South Afr 43 :41-49
Dobzhansky T (1950) Evolution in the tropics. Am Sci 38:209-221
Dondale CD (1965) A spider's first meal after molting. Can Entomol 97 :446
Dondale CD (1977) Life histories and distribution patterns of hunting spiders (Araneida) in an
Ontario meadow. J ArachnoI4:73-93
Dondale CD, Legendre R (1971) Winter diapause in a Mediterranean population of Pisaura mirabilis
(Clerck). Bull Br Arachnol Soc 2:6-10
Dondale CD, Redner JH, Semple RB (1972) Diel activity periodicities in meadow arthropods. Can
J Zool50 :1155 -1163
Dreesbach K, Uhlenbruck G, Tillinghast EK (1983) Carbohydrates of the trypsin soluble fraction
of the orb web of Argiope trifasciata. Insect Biochem 13 :627 -631
Dresco-Derouet L (1960) Etude biologique comparee de quelques especes d'araigm!es lucicoles et
troglophiles. Archs Zool Exp Gen 98 :271-354
Dresco-Derouet L (1964) Le metabolisme respiratoire des scorpions II. Mesures de I'intensite res-
piratoires chez quelques especes adifferentes temperatures. Bull Mus Hist Nat Paris 36 :97 -99
Drewes CD, Bernard RA (1976) Electrophysiological responses of chemosensitive sensilla in the
wolf spider. J Exp Zool198 :423-
Dubach M (1981) Quantitative analysis of the respiratory system of the house sparrow, budgerigar
and violet-eared humming bird. Respir Physiol 46 :43 -60
Duffey E (1956) Aerial dispersal in a known spider population. J Anim Ecol 25 :85 -111
Duffey E (1962) A population study of spiders in limestone grassland. The field layer fauna. Oikos
13:15-34
References 399
*Duman JG (1979) Subzero temperature tolerance in spiders: the role of thermal-hysteresis-factors.

J Comp PhysioI131:347-352
*Dumbleton LJ (1941) Oncodes brunneus Hutton: A dipterous spider parasite. NZ J Sci Tech
XXII (2A):97 A-102A
Dyke D van, Lowrie DC (1975) Comparative life histories of the wolf spidersPardosa ramulosa and
P. sierra (Araneae:Lycosidae). Southwest Nat 20:29-44
Eason FR, Peck WB, Whitcomb WH (1967) Notes on spider parasites, including a reference list. J
Kans Entomol Soc 40:422-434
Eberhard WG (1967) Attack behavior of diguetid spiders and the origin of prey wrapping in spiders.
Psyche 74:173-181
Eberhard WG (1970) The predatory behavior of two wasps, Agenoideus humilis and Sceliphron
caementarium on the orb weaving spider, Araneus comutus. Psyche 77 :243 -251
Eberhard WG (1973) Stabilimenta on the webs of Uloborus diversus (Araneae:Uloboridae) and
other spiders. J Zooll71 :367 -384
Eberhard WG (1975) The inverted ladder orb web of Scoloderus sp. and the intermediate orb of
Eustala(?) sp. (Araneae:Araneidae). J Nat Hist 9:93-106
Eberhard WG (1976a) Photography of orb webs in the field. Bull Br Arachnol Soc 3:200-204
Eberhard WG (1976b) Physical properties of sticky spirals and their connections: sliding connec-
tions in orb webs. J Nat Hist 10:481-488
*Eberhard WG (1977) Aggressive chemical mimicry by a bolas spider. Science 198:1173-1175
Eberhard WG (1979) Argyrodes attentuatus (Theridiidae): a web that is not a snare. Psyche 86:
407 -413
Eberhard WG (1980a) Persistant stickiness of cribellum silk. J Arachnol8:283
Eberhard WG (1980b) The natural history and behaviour of the bola spider Mastophora dizzyde·
ani sp.n. (Araneidae). Psyche 87 :143-169
Eberhard WG, Bricefio RD (1985) Behavior and ecology of four species of Modisimus and Blech·
roscelis (Araneae, Pho1cidae). Rev Arachnol6:29-36
Eckert M (1967) Experimentelle Untersuchungen zur Hiiutungsphysiologie bei Spinnen. Zool Jahrb
Physiol73:1-38
Edgar WD (1969) Prey and predators of the wolf spiders Lycosa lugubris. J ZoolI59:405-411
Edgar WD (1970) Prey and feeding behaviour of adult females of the wolf spider Pardosa amentata
(Clerck). Neth J Zool 20:487 -491
Edgar WD (1971a) Seasonal weight changes, age structure, natality and mortality in the wolf spider
Pardosa lugubris Walck. in Central Scotland. Oikos 22:84-92
*Edgar WD (1971 b) Aspects of the ecological energetics of the wolf spider Pardosa (Lycosa) lugubris
(Walckenaer). Oecologia (Berlin) 7 :136-154
Edgar WD (1971c) The life-cycle, abundance and seasonal movement of the wolf spider, Lycosa
(Pirata) lugubris, in central Scotland. J Anim Ecol40:303-322
Edgar WD, Loenen M (1974) Aspects of the overwintering habitat of the wolf spider Pardosa lu-
gubris. J Zool (London) 172:383-388
Edmunds M (1978) On the association between Myrmarachne ssp. (Salticidae) and ants. Bull Br
Arachnol Soc 4:149-160
Edmunds J (1982) The species Argiope (Araneae:Araneidae) found in Ghana, West Africa, with a
description of the male of Argiope jlavipalpis (Lucas). Bull Br Arachnol Soc 5 :285 -293
Edney EB (1977) Water balance in land arthropods. Springer, Berlin Heidelberg New York
Edney EB (1982) The truth about saturation deficiency: an historical perspective. J Exp Zool
222:205-214
Efrati P (1969) Bites by Loxosceles spiders in Israel. Toxicon 6:239-241
Ehrnsberger R (1977) Fortpflanzungsverhalten der Rhagidiidae Acarina, Trombidiformes). Acaro-
logia 19:67-73
*Ehrnsberger R (1979) Spinnvermogen bei Rhagidiidae (Acari, Prostigmata). Osnabr Naturwiss Mitt
6:45-72
*Eickstedt VRD von (1971) Three cases of parasitism in the mygalomorph spider, Lasiodora klugi
(C.L. KOCh), by a fly of the genus Exetasis (Diptera:Acroceridae) in Brazil. Mem Inst Butantan
35:139-146
400 References
Eisenack A (1956) Beobachtungen an Fragmenten von Eurypteriden-Panzern. Neues Jahrb Geol

Paleontol Abh 104:119-128
Eisenbeis G (1982) Physiological absorption of liquid water by Collembola: absorption by the
ventral tube at different salinities. J Ins PhysioI28:11--20
Eisner T, Dean J (1976) Ploy and counterploy in predator-prey interactions: Orb-weaving spiders
versus bombardier beetles. Proc Nat! Acad Sci USA 73:1365-1367
Eisner T, Nowicki S (1983) Spider web protection through visual advertisement: Role of the stabi-
limentum. Science 219:185-187
Eisner T, Alsop R, Ettershank G (1964) Adhesiveness of spider silk. Science 146:1058-1061
EI Banhawy EM (1977) Growth inhibition of the predacious mite Amblyseius brazilli (Mesostig-
mata:Phytoseiidae) by a synthetic juvenile hormone under laboratory conditions. Entomophaga
22:429-434
Ellis CH (1943) The mechanism of extension in the legs of spiders. BioI Bull 86 :41-50
Enders F (1975) Change in web site in Argiope spiders (Araneidae). Am MidI Nat 94:484 - 490
*EngeJhardt W (1964) Die mitteleuropiiischen Arten der Gattung Trochosa c.L. Koch, 1848 (Ara-
neae, Lycosidae). Morphologie, Chemotaxonomie, Biologie, AutCikologie. Z Morphol Okol
Tiere 54:219-392
Enock F (1885) The life-history of Atypus piceus. Trans Entomol Soc London
Esch H (1976) Body temperature and flight performance of honey bees in a servo-mechanically
controlled wind tunnel. J Comp Physiol A 109:265-277
Eskafi FM (1974) Venom histochemistry and water balance of Loxosceles reclusa (GERTSCH
and MULAIK). Thesis, Mississippi State Univ
Evans HE (1969) Notes on the nesting behavior of Pisonopsis clypeata and Belomicrus forbesii
(Hymenoptera, Sphecidae). J Kans Entomol Soc 42:117 -125
Evans HE (1970) Ecological-behavioral studies of the wasps of Jackson Hole, Wyoming. Bull Mus
Comp ZooI140:451--511
Evans HE (1981) Biosystematics of ground-nesting species of Pison in Australia (Hymenoptera:
Sphecidae:Trypoxylonini). Proc Entomol Soc Wash 83:421-427
Evans HE, Matthews JR, Hook A (1980) Notes on the nests and prey of six species of Pison in
Australia (Hymenoptera:Sphecidae). Psyche 87 :221-230
Ewers R (1972) The devices in the web of the West African spider Argiope [lavipes. J Nat Hist 6:
159-167
Exline H, Levi HW (1962) American spiders of the genus Argyrodes (Araneae, Theridiidae). Bull
Mus Comp Zool Harvard 127:75-204
Fabre JH (1912) The life of the spider. Hodder and Stoughton, London (translated by de Mattos
AT)
Feinstein R, Aspey WP, Schmale MC (1976) A miniaturized photocell assembly for monitoring
heart rate in freely behaving Aplysia (Moliusca:Gastropoda). Behav BioI 17 :271-278
Fergusson I (1972) Natural history of the spider Hypochilus thorelli Marx (Hypochilidae). Psyche
86:179-199
*Filshie BK, Hadley NF (1979) Fine structure of the cuticle of the desert scorpion, Hadrurus ari-
zonensis. Tissue Cell 11 :249-262
Fincke 0 (1984) Giant damselflies in a tropical forst: Reproductive biology of Megaloprepus coeru-
latus with notes on M ecistogaster (Zygoptera :Pseudostigmatidae). Adv Odonatol 2: 13 - 27
Fincke T, Tiling K, Paul R, Linzen B (1986) Relations between respiration and circulation in the
tarentula, Eurypelma californicum. In: Linzen B (ed) Invertebrate oxygen carriers. Springer,
Berlin Heidelberg New York, pp 327-331
Finlayson MH (1936) "Knoppie-Spider" antivenene. S Afr Med J 10:735 -736
Finlayson MH (1956) "Knoppie-Spider" bite in Southern Africa. Med Proc 2:634-638
Fischer E (1907) Uber Spinnenseide. Hoppe-Seyler's Z Physiol Chern 53:126-139
Fischer FG, Brander J (1960) Eine Analyse der Gespinste der Kreuzspinne. Hoppe-Seyler's Z Physiol
Chern 320:92-102
Flatz S (1979) Winteraktivitiit epigiiischer Arthropoden (ibs. Aranei, Carabidae) im Bereich der
Landesanstalt fiir Pflanzenzucht und Samenpriifung Rinn (Nordtirol 900 mNN). Mag Thesis,
Univ Innsbruck, p 75
References 401
Fleschner CA, Badgley MA, Ricker DW, Hall JC (1956) Air drift of spider mites. J Econ Entomol
49:624-627
Florey E (1975) Lehrbuch der Tierphysiologie. Eine Einflihrung in die allgemeine und vergleichende
Physiologie der Tiere, 2nd edn. Thieme, Stuttgart
Focke P (1981) Zur Ventilation von Spinnen: Morphologische und physiologische Untersuchungen
an der Vogelspinne Eurypelma californicum. Diplomarbeit, Univ Miinchen
Foelix RF (1971) Some morphological aspects of the innervation of a spider's silk gland. Tissue Cell
3:423-432
*Foelix RF (1979) Biologie der Spinnen. Thieme, Stuttgart
*Foelix RF (1982) Biology of spiders. Harvard Univ Press, Cambridge Mass
Ford MJ (1977a) Metabolic costs of the predation strategy of the spider Pardosa amentata (Clerck)
(Lycosidae). Oecologia (Berlin) 28 :333-340
*Ford MJ (1977b) Energy costs of the predation strategy of the web-spinning spider Lepthyphantes
zimmermanni Bertkau (Linyphiidae). Oecologia (Berlin) 28:341-349
Ford MJ (1978) Locomotory activity and the predation strategy of the wolf-spider Pardo sa amen-
tata (Clerck) (Lycosidae). Anim Behav 26:31-35
Fornes RE, Work RW, Morosoff N (1983) Molecular orientation of spider silks in the natural and
supercontracted states. J Poly Sci Poly Phys Ed 21 :1163-1172
Forster LM (1982) Visual communication in jumping spiders (Salticidae). In: Witt PN, Rovner JS
(eds) Spider communication. Mechanisms and ecological significance. Princeton Univ Press,
Princeton, pp 161-212
*Forster RR (1970) The spiders of New Zealand. Part 3. Otago Mus Bull 3 :1-184
Forster RR, Forster LM (1973) New Zealand spiders: an introduction. Collins, Auckland
Forster RR, Gray M (1979) Progradungula, a new cribellate genus of the spider family Gradungu-
lidae (Araneae). Aust J ZooI27:1051--1071
Forster RR, Platnick N (1977) A review of the spider family Symphytognathidae (Arachnida,
Araneae). Am Mus Nov 2619:1-74
Forster RR, Platnick N (1984) A review of the archaeid spiders and their relatives, with notes on
the limits of the superfamily Palpimanoidea (Arachnida, Araneae). Bull Am Mus Nat Hist 178:
1-106
*Forster RR, Wilton CL (1968) The spiders of New Zealand. Part II. Otago Mus Bull 2:1-169
*Forster RR, Wilton CL (1973) The spiders of New Zealand. Part IV. Otago Mus Bull 4: 1-309
Fowler HG, Diehl J (1978) Biology of a Paraguayan colonial orb-weaver, Eriophora bistriata (Reng-
ger) (Araneae, Araneidae). Bull Br Arachnol Soc 4:241-250
Freeman BE (1980) A population study in Jamaica on adult Sceliphron assimile (Dahlbom) (Hy-
menoptera: Sphecidae). Ecol Entomol 5 :19-30
Freeman BE, Johnston B (1978a) The biology in Jamaica of the adults of the sphecid wasp Sceli-
phron assimile Dahlbom. Ecol Entomol 3 :39--52
Freeman BE, Johnston B (1978b) Gregarious roosting in the sphecid wasp, Sceliphron assimile.
Ann Entomol Soc Am 71 :435-441
Friedrich VL, Langer RM (1969) Fine structure of cribellate silk. Am Zool 9 :91-96
Fritz LC, Atwood HL, Jahromi LL (1980) Lobster neuromuscular junctions treated with black
widow spider venom - correlation between ultrastructure and physiology. J Neurocytol9 :699
Frontali N (1972) Catecholamine-depleting effect of black widow spider venom on iris nerve fibers.
Brain Res 37 :146-148
Frontali N, Grasso A (1964) Separation of three toxicologically different protein components of
the venom of the spider Latrodectus tredecimguttatus. Arch Biochem Biophys 106 :213-218
Frontali N, Granata F, Parisi P (1972) Effects of black widow spider venom on acetylcholine release
from rat cerebral cortex slices in vitro. Biochem Pharmacol 21 :969-974
Frontali N, Ceccarelli B, Gorio A, Mauro A, Siekevitz P, Tzeng M, Hurlbut WP (1976) Purification
from black widow spider venom of a protein factor causing the depletion of synaptic vesicles
at neuromuscular junctions. J Cell Bioi 68 :462-479
Fujii Y (1972) Study of energy utility efficiency in Agelena opulenta L. Koch (Araneae:Ageleni-
dae). Bull Nippon Dental Coll, Gen Educ No 1 :79-95
Furlanetto RS (1961) Estudos sobre preparayao do soro antiloxoscelico. Thesis, Fac Farm Odontol,
Univ Sao Paulo
402 References
Gabritschevsky E (1927) Experiments on the color changes and regeneration in the crab spider,
Misumena vatia (CI.). J Exp ZooI47:251-267
Gadeau de Kerville H (1907) Sur l'homochromie protectrice des femelles de Misumena vatia Clerck.
Bull Soc Entomol Fr 1907 :145-146
Gajardo-Tobar R (1941) Ellatrodectismo. Prensa Med 6 :3-18
Gajardo-Tobar R (1950) Algo mas sobre latrodectismo. Rev Med Valparaiso 3:150-159
Gajardo-Tobar R (1966) El araneismo en el mundo tropical y sUbtropical. Mem Inst Butantan 33:
45-54
Gajardo-Tobar R, Vildosola E (1944) Anotaciones acerca del tratamiento dellatrodectismo. Los
seis primeros casos curados con suero anti-Latrodectus. Riv Soc Arg Bioi 20:12-20
Gans C, Pough FH (1982) Biology of reptilia, vol XII. Physiology C: physiological ecology. Aca-
demic Press, London New York
Gardefors D (1964) The innuence of rapid temperature changes on the activity of Chorthippus
albomarginatus DE GEER (Acrididae, Orthoptera). Entomol Exp Appl 7 :71-84
Gardner BT (1964) Hunger and sequential responses in the hunting behavior of salticid spiders. J
Comp Physiol PsycoI58:167-173
Gaubert P (1892) Recherches sur les organes des sens et sur les systemes tegumentaire, glandulaire
et musculaire des appendices des arachnides. Ann Sci Nat Zool Paleontol13 :31-184
Gaud J, Delesalle D (1949) Araneisme du aux morsures de Latrodectes du Maroc. Bull Inst Hyg
Maroc 9:233-237
Geren CR, Chan TK, Ward BC, Howell DE, Pinkston K, Odell GV (1973) Composition and proper-
ties of extract of fiddleback (Loxosceles rec1usa) spider venom apparatus. Toxicon 11:471-479
Gerhardt U (1923) Araneina. In: Schulze P (ed) Biologie der Tiere Deutschlands, part 20. Borntra-
ger, Berlin, pp 1-37
Gerhardt U (1925) Neue sexualbiologische Spinnenstudien. Z MorpholOkol Tiere 3:567 -618
Gerhardt U (1928a) Biologische Studien an griechischen, korsischen und deutschen Spinnen. Z
MorpholOkol Tiere 10:576-675
Gerhardt U (1928b) Uber Vorkommen und Lebensweise einiger griechischer Spinnen. Zool Jahrb
Abt Allg Zoo I Physiol 45 :31-44
Gerhardt U, Kaestner A (1937) Araneae. In: Kiikenthal W, Krumbach T (eds) Handbuch der Zoolo-
gie, vol III. De Gruyter, Berlin Leipzig, pp 395-656
Germond JE, Diehl PA, Morici M (1982) Correlation between integument structure and ecdystero-
id titers in fifth-stage nymphs of the tick, Ornithodoros moubata (Murray, 1877; sensu Walton,
1962). Gen Comp Endocrinol46 :255 -266
Gertsch WJ (1947) Spiders that lasso their prey. Nat Hist 56:152-158,189
*Gertsch WJ (1979) American spiders, 2nd edn. Van Nostrand Reinhold, New York
Gertsch WJ (1984) The spider family Nesticidae (Araneae) in North America, Central America and
the West Indies. Bull Texas Mem Mus, Austin 31: 1-91
Gilbert C, Rayor L (1985) Predatory behavior of spitting spiders (Araneae, Scytodidae) and the
evolution of prey wrapping. J Arachnol13 :231-242
Gilbert EW, Stewart CM (1935) Effective treatment of arachnidism with calcium salts; preliminary
report. Am J Med Sci 189 :532-536
*Giraud-Guille M-M (1984) Fine structure of the chitin-protein system in the crab cuticle. Tissue Cell
16:75-92
Glatz L (1972) Der Spinnapparat haplogyner Spinnen (Arachnida, Araneae). Z Morphol Tiere
72:1-26
Glatz L (1973) Der Spinnapparat der Orthognatha (Arachnida, Araneae). Z Morphol Tiere 75:
1-51
Glick PA (1939) The distribution of insects, spiders and mites in the air. Tech Bull US Dep Agric
673
Gomer P, Claas B (1985) Homing behavior and orientation in the funnel-web spider, Agelena laby-
rinthica Clerck. In: Barth FG (ed) Neurobiology of archnids. Springer, Berlin Heidelberg New
York, pp 275-297
Gonzalez-Fernandez F, Sherman RG (1984) Cardioregulatory nerves in the spider Eurypelma marxi
SIMON. J Exp ZooI231:27 -37
References 403
Gordon CJ, Heath JE (1983) Reassessment of the neural control of body temperature: importance
of oscillating neural and motor components. Comp Biochem Physiol A 74:479-489
Gorham JR (1968) The brown recluse spider Loxosceles rec1usa and necrotic spiderbite. A new
public health problem in United States. J Environ Health 31 :138-145
Gosline JM, Denny MW, DeMont ME (1984) Spider silk as rubber. Nature (London) 309:551-552
Gothe R, Morawietz M (1979) Zum Wirkungsspektrum von Juvenoiden in Postembryonalphasen
von Argas (Persiargas) walkerae Kaiser und Hoogstraal, 1969. Zentralbl Vet Med B 26:779-797
Goyffon M, Martoja R (1983) Cytophysiological aspects of digestion and storage in the liver of a
scorpion, Androctonus australis (Arachnida). Cell Tissue Res 228 :661-77 5
Grainge CA, Pearson RG (1966) Cuticular structure in the Phalangida. Nature (London) 211 :866
Grandi G (1961) Studi di un entomologo sugli imenotteri superiori. Boll 1st Entomol Univ Studi
Bologna 25 :1-659
Granstrom U (1973) Pitfall traps for studying the activity of groundliving spiders (Araneida). Aquilo
(Zoo!) 14:93-98
Granstrom U (1977) Tages- und Jahresperiodik vonPardosa lugubris (Walck.) undP. amentata (C1.)
(Lycosidae, Araneida) unter natUrlichen Lichtbedingungen in Nordschweden. Aquilo (Zoo!) 17:
42-48
Grasso A (1976) Preparation and properties of a neurotoxin purified from the venom of black
widow spider (Latrodectus mactans tredecimguttatus). Biochirn Biophys Acta 439:406-412
Grasso A, Toschi-Frontali N (1962) Studi sui veleno del ragno Latrodectus tredecimguttatus. Bol
Soc Ital Bioi Sper 38:1814-1816
Gray MR (1968) Comparison of three genera of trapdoor spiders (Ctenizidae, Aganippini) with
respect to survival under arid conditions. Thesis, Univ Western Australia, Perth
*Greenstone MH (1982) Ballooning frequency and habitat predictability in two wolf spider species
(Lycosidae, Pardosa). Fla Entomol65 :83-89
Greenstone MH, Bennett AF (1980) Foraging strategy and metabolic rate in spiders. Ecology 61:
1255-1259
Greenstone MH, Morgan CE, Hultsch A-L (1985) Ballooning methodology: equations for estimating
masses of sticky-trapped spiders. J Arachnol 13 :225 - 230
Greer WER (1949) Arachnidism. N Engl J Med 240:5-8
Gressitt JL (1961) Problems in the zoogeography of pacific and antarctic insects. Pac Insects Monogr
2:1-94
Gunnarsson B (1983) Winter mortality in spruce-living spiders: effect of spider interaction and bird
predation. Oikos 40:226-233
Gunnarsson B (1985) Phenotypic variation in dark coloration in Pityohyphantes phrygianus (C.L.
Koch) (Araneae:Linyphiidae). Bull Br Arachnol Soc 6:369-374
Habermehl G (1974) Vergiftungen durch Dornfingerspinne. Naturwissenschaften 61:1
Hackman RH (1982) Structure and function in tick cuticle. Annu Rev Entomol 27 :75 -95
*Hackman RH (1984) Arthropoda. Cuticle: biochemistry. In: Bereiter-Hahn J, Matolsky AG, Richards
KS (eds) Biology of the integument, vol 1. Invertebrates. Springer, Berlin Heidelberg New York,
pp 583-610
Hackman W (1957) Studies on the ecology of the wolf spider Trochosa ruricola Deg. Commentat
Bioi 16:1-34
Hadley NF (1970a) Micrometeorology and energy exchange in two desert arthropods. Ecology 51:
434-444
Hadley NF (1970b) Water relations of the desert scorpion, Hadrurus arizonensis. J Exp Bioi 53:
547-558
Hadley NF (1971) Water uptake by drinking in the scorpion Centruroides sculpturatus (Buthidae).
Southwest Nat 15 :495 -505
Hadley NF (1978) Cuticular permeability and lipid composition of the black widow spider, Latro·
dectus hesperus. Symp Zool Soc London 42:429-438
*Hadley NF (1981) Fine structure of the cuticle of the black widow spider with reference to surface
lipids. Tissue Cell 13 :805 -817
*Hadley NF (1984) Cuticle: ecological significance. In: Bereiter-Hahn J, Matoltsy AG, Richards KS
(eds) Biology of the integument, vol 1. Invertebrates. Springer, Berlin Heidelberg New York,
pp 685-694
404 References
*Hadley NF, Fi1~hie BK (1979) Fine structure of the epicuticle of the desert scorpion, Hadrurus
arizonensis, with reference to location of lipids. Tissue Cell 11 :263-275
Hadley NF, Ahearn GA, Howarth FG (1981) Water and metabolic relations of cave-adapted and
epigean Iycosid spiders in Hawai. J Arachnol 9 :215 - 222
Hagstrum DW (1970a) Ecological energetics of the spider Tarantula koehi (Araneae:Lycosidae).
Ann Entomol Soc Am 63:1297-1304
Hagstrum DW (1970b) Physiology of food utilisation by the spider Tarentula koehi (Araneae :Lyco-
sidae). Ann Entomol Soc Am 63 :1305 -1308
Hagstrum DW (1971) Carapace width as a tool for evaluating the rate of development of spiders in
the laboratory and the field. Ann Entomol Soc Am 64:757 -760
Hll.gvar S (1973) Ecological studies on a winter-active spider, Bolyphantes index (Thorell) (Araneida,
Linyphiidae). Nor Entomol Tidsskr 20 :309--314
Haldeman SS (1847) Lyeosa seutulata and its parasite, Filaria. Proc Am Philos Soc 4:356
Haldeman SS (1851) Invertebrates. In: Baird S (ed) Outlines of general zoology. R Garrigue, New
York, pp 48-196
Hallander H (1967) Courtship display and habitat selection in the wolf spider Pardosa ehelata
(O.F. MULLER). Oikos 18:145-150
Hallander H (1970) Environments of the wolf spider Pardosa ehelata (O.F. Miiller) and P. pullata
(Clerck). Ekol Pol 18A:I-72
Hallas SEA, Jackson RR (in press) Comparative study of Old and New World 'Iyssomanines' (Ara-
neae, Salticidae): utilisation of silk and predatory behaviour of Asemonea tenuipes and Lysso-
manes viridis. NZ J Zool
Hamilton WD (1971) Geometry for the selfish herd. J Theor Bioi 31 :295 -311
Harkness RD (1977) Further observations on the relation between an ant, Cataglyphis bieolor (F.)
(Hym., Formicidae) and a spider, Zodarium frenatum (Simon) (Araneae, Zodariidae). Entomol
Mon Mag 112:111-121
Harris DJ, Mill PJ (1977) Observations on the leg receptors of Ciniflo (Araneida :Dictynidae). II.
Chemoreceptors. J Comp Physiol119 :55 -62
Hassell MP,May RM (1974) Aggregation of predators and insect parasites and its effect on stability.
J Anim Ecol 43 :567 -594
Haupt J (1979) Lebensweise und Sexualverhalten der mesothelen Spinne Heptathela nishihirai n.sp.
(Araneae, Liphistiidae). Zool Anz 202 :348-3 74
Hazan A, Gerson U, Tahori AS (1974) Spider mite webbing I. The production of webbing under
various environmental conditions. Acarologia 16:68-84
*Hazan A, Gertler A, Tahori AS, Gerson U (1975) Spider mite webbing III. Solubilization and
amino acid composition of the silk protein. Comp Biochem Physio151B:457 -462
Heath IE, Hanegan JL, Wilkin PJ, Heath MS (1971) Adaptation of the thermal responses of insects.
Am Zool11:147-158
Heathcote RD, Carrell IE (1979) Heart rate in wolf spiders: Effect of temperature on resting values.
Am Zoo119:919
Heatwole H (1970) Thermal ecology of the desert dragon Amphibolurus inermis. Ecol Monogr 40:
425-457
Heckel E (1891) Sur Ie mimetisme de Thomisus onustus. Bull Sci Fr Belg 23 :347 -354
Heidger C, Nentwig W (1985) The prey of Dietyna arundinacea (Araneae:Dictynidae). Zool Beitr
NF 29:185-192
Heimer S (1984) Remarks on the spider genus Arcys WALCKENAER, 1837, with description of
new species (Araneae, Mimetidae). Entomol Abh Mus Tierk Dresden 47:155-178
Heimer S (1986a) Notes on the spider family Mimetidae with description of a new genus from
Australia. Entomol Abh Mus Tierk Dresden 49:(in press)
Heimer S (1986b) Untersuchungen zur Evolution der Kopulationsorgane bei Spinnen. Thesis (in
press)
Heimer S, Nentwig W (1982) Thoughts on the phylogeny of the Araneoidea Latreille, 1806 (Arach-
nida, Araneae). Z Zoo I Syst Evolutionsforsch 20 :284-295
Heinrich B (1976) Heat exchange in relation to blood flow between thorax and abdomen in bumble-
bees. J Exp Bioi 64 :561-585
Heinrich B (1979) Bumblebee economics. Harvard Univ Press, Cambridge, Mass
References 405
Heinrich B (ed) (1981) Insect thermoregulation. Wiley, New York

Heinrich B, Casey TM (1978) Heat transfer in dragonflies: 'fliers' and 'perchers'. J Eltp Bioi 74:
17-36
Heller G (1976) Zum Beutefangv.erhalten der ameisenfressenden Spinne Cal/ilepis noctuma (Arach-
nida:Araneae:Drassodidae). Entomol Germ 3 :100-103
Helsdingen PJ van (1965) Sexual behaviour of Lepthyphantes /eprosus (Ohl) (Araneida, Linyphi-
idae), with notes on the function of the genital organs. Zool Meded 41 :15 -42
Helsdingen PJ van (1977) La faune terrestre de l'tle de Sainte Helene. Ann Mus R Mr Cent, Ser 8,
Sci Zool220
Helversen 0 von (1976) Gedanken zur Evolution der Paarungsstellung bei den Spinnen (Arachnida:
Araneae). Entomol Germ 3:13-28
Hemmingsen AM (1960) Metabolism in relation to body size. Rep Steno Mem Hosp Nord Insulinlab
9:1-110
Henking H (1882) Beitriige ZUI Anatomie, Entwicklungsgeschichte und Biologie von Trombidium
fuliginosum Herm. Z Wiss Zool 37:553-663
Herman WS (1972) Molt initiation in response to phytoecdysones and low doses of animal ecdysones
in the Chelicerate Arthropod, Limulus polyphemus. Gen Comp EndocrinoI18:301-305
Herreid CF (1969) Water loss of crabs from different habitats. Comp Biochem PhysioI28:829-839
Herreid CF, Lee LW, Spampata R (1981) How do spiders breathe? Am Zool 21 :917
Herrero MV, Morales A, Vargas R (1983) Patrones de actividad diaria de las hembras de Aphono-
pelma seemanni (Araneae:Theraphosidae) durante la epoca lluviosa en Guanacaste, Costa Rica.
Rev Bioi Trap 31 :161-162
Heydemann B (1961) Die biozonotische Entwicklung vom Vorland zum Koog. I. Tei1: Spinnen
(Araneae). Abh Math-Naturwiss KI Akad Wiss Lit 1960:745-913
Hickman VV (1964) On Atrax infensus sp.n. (Araneida :Dipluridae) its habits and a method of trap-
ping the males. Pap Proc R Soc Tasmania 98:107 -112
Hieber CS (1985) The "insulation" layer in the cocoons of Argiope aurantia (Araneae:Araneidae).
J Therm Bioi 10:171-175
Hill DE (1979) Orientation by jumping spiders of the genusPhidippus (Araneae:Salticidae) during
the pursuit of prey. Behav Ecol Sociobiol5 :301-322
Hingston RWG (1927) Protective devices in spiders' snares, with a description of seven new species
of orb-weaving spiders. Proc Zool Soc London 18 :259-293
Hinton HE (1976) Possible significance of the red patches of the female crab-spider, Misumena
vatia. J Zool (London) 180:35-39
Hippa H, Oksala I (1979) Color polymorphism of Enoplognatha ovata (Clerck) (Araneae:Theridi-
idae) in western Europe. Hereditas 90:203-212
Hochachka PW, Somero GN (1980) Strategien biochemischer Anpassung. Thieme, Stuttgart New
York
Hochrainer H (1942) Der Wassergehalt der Insekten und die Faktoren, die denselben bestimmen.
Zool Jahrb Abt Allg Zool Physio160:387 -436
Hoffmann JA (ed) (1980) Progress in ecdysone research. Elsevier/North Holland, Amsterdam New
York
Hoffmann KH (ed) (198Sa) Environmental physiology and biochemistry of insects. Springer,
Hoffmann KH (ed) (1985b) Metabolic and enzyme adaptation to temperature. In: Environmental
physiology and biochemistry of insects. Springer, Berlin Heidelberg New York, pp 1-32
Hoffmaster DK, Hays HE (1977) Simulation of predatory interactions: spiders and flies. Proc Pa
Acad Sci 51 :131-133
Holl A (1982) Temperaturabhiingiger Farbwechsel bei Larven der griinen Huschspinne Micromata
rosea (Sparassidae). Z Naturforsch 37c:l040-1041
Holl A (1986) Reifefiirbung und Hypodermispigmente miinnlicher Micromata rosea Clerck (Arach-
nida:Sparassidae). In press
*Holl A, Riidiger W (1975) Micromatabilin, a new biliverdin conjugate in the spider Micromata rosea
(Sparassidae). J Comp PhysioI98:189-191
Holl A, Klett T, Kracht M (1982) Ogcodes fumatus (Erichson) (Diptera:Acroceridae) als Parasit der
Radnetzspinne Zygie/Ia x-notata (Clerck). Hess Faun Briefe 3:26-30
406 References
Holm A (1940) Studien tiber die Entwicklung und Entwicklungsbiologie der Spinnen. Zool Bidr
Upps 19:1-210
Holm A (1941) tiber Gynandromorphismus und Intersexualitat bei den Spinnen. Zool Bidr Uppsala
20:397-416
Holm A (1950) Studien tiber die Spinnenfauna des Tornetraskgebietes. Zool Bidr Uppsala 29:
103-213
Holm A (1980) Spiders (Araneae) from Faroes. Bull Br Arachnol Soc 5 :108-114
Holmquist AM (1931) Studies in arthropod hibernation. III. Temperatures in forest hibernacula.
Ecology 12:387-400
Homann H (1946) tiber die Jugendform von Micromata viridissima (Deg.) (Araneae). BioI Zentralbl
65:82-83
*Homann H (1949) Uber das Wachstum und andere mechanische Vorgange bei der Hautung von
Tegenaria agrestis. Z Vergl Physiol 31 :413
Hopfmann W (1935) Bau und Leistung des Spinnapparates einiger Netzspinnen. Jena Z Naturwiss
70:65-111
Hoppe DH (1796) Entomologisches Taschenbuch fiir Anfanger und Liebhaber dieser Wissenschaft
auf das Jahr 1796. Regensburg, 16 pp
Horn E (1969) 24-hour cycles of locomotor and food activity of Tetragnatha montana Simon
(Araneae, Tetragnathidae) and Dolomedes fimbriatus (Clerck) (Araneae, Pisauridae). Ekol Pol A
17:533-549
Horton C (1980) A defensive function for the stabilimenta of two orb weaving spiders (Araneae:
Araneidae). Psyche 87 :13-20
Howell F, Ellender R (1984) Observations on growth and diet of Argiope aurantia Lucas (Arane-
idae) in a succession habitat. J ArachnoI12:29-36
Huey RB, Stevenson RD (1979) Integrating thermal physiology and ecology of ectotherms. Am
Zool19 :35 7 -366
Huhta V (1965) Ecology of spiders in the soil and litter of Finnish forests. Ann Zool Fenn 2:260-
308
Huhta V, Viramo J (1979) Spiders active on snow in northern Finland. Ann Zool Fenn 16:169-176
Humphreys WF (1974) Behavioural thermoregulation in a wolf spider. Nature (London) 251 :502--
503
Humphreys WF (1975a) The influence of burrowing and thermoregulatory behaviour on the water
relations of GeolycoSIJ godeffroyi (Araneae:Lycosidae), an Australian wolf spider. Oecologia
21:291-311
Humphreys WF (1975b) The respiration of GeolycoSIJ godeffroyi (Araneae:Lycosidae) under con-
ditions of constant and cycling temperature. Physiol ZooI48:269-281
Humphreys WF (1976) The population dynamics of an Australian wolf spider, Geolycosa godeffroyi
(1. Koch 1865) (Araneae:Lycoside). J Anim Ecol45 :59-80
*Humphreys WF (1977a) Respiration studies on Geolycosa godeffroyi (Araneae:Lycosidae) and their
relationship to field estimates of metabolic heat loss. Comp Biochem Physiol 57 A:255 -263
Humphreys WF (1977b) Variables influencing laboratory energy budgets of Geolycosa godeffroyi
(Araneae). Oikos 28 :225 -233
*Humphreys WF (1978a) The thermal biology of GeolycoSIJ godeffroyi and other burrow inhabiting
Lycosidae (Araneae) in Australia. Oecologia (Berlin) 31 :319-347
Humphreys WF (1978b) Ecological energetics of GeolycoSIJ godeffroyi (Araneae:Lycosidae) with
an appraisal of production efficiency in ectothermic animals. J Anim EcoI47:627-652
Humphreys WF (1983) The surface of spiders' eggs. J Zoo I 200:303-316
*Hunt S (1970) Amino acid composition of silk from the pseudoscorpion Neobisium maritimum
(Leach): a possible link between the silk fibroins and the keratins. Comp Biochem Physiol 34:
773-777
*Hunt S, Legg G (1971) Characterization of the structural protein component in the spermatophore
of the pseudoscorpion, Chthonius ischnocheles (Hermann). Comp Biochem Physioi40B:475--
479
Husby JA, Zachariassen KE (1980) Antifreeze agents in the body fluid of winter-active insects and
spiders. Experentia 36 :963-964
Iizuka E (1983) III. The physico-chemicai properties of silk fibers and the fiber spinning process.
Experientia 39 :449 -454
References 407
Ismail S, Porcheron P, Fuzeau-Braesh S (1979) L 'hormone de mue et la diapause chez Gryllus cam·
pestris L (Orthopteres): dosage, injections et metabolisme. CR Soc Bioi 107 :37 -40
Ito Y (1964) Preliminary studies on the respiratory energy loss of a spider, Lycosa pseudoannulata.
Res Popul EcoI6:13-21
Iwata K (1964) Ethological notes on four Japanese species of Pison. Mushi 38:1-6
*Iwata K (1976) Evolution of instinct: Comparative ethology of Hymenoptera. Amerind, New Dehli
Jackson RR (1971) Fine structure of thread connections in the orb web of Araneus diadematus.
Psyche 78:12-31
Jackson RR (1977) Prey of the jumping spider Phidippusjohnsoni (Araneae:Salticidae). J Arachnol
5:145-149
Jackson RR (1978a) Life history of Phidippus johnsoni (Araneae, Salticidae). J Arachnol 6 :1-29
Jackson RR (1978b) The mating strategy of Phidippus johnsoni (Araneae:Salticidae) 1. Pursuit
time and persistence. Behav Ecol Sociobiol4 :123-132
Jackson RR (1980) The mating strategy of Phidippus johnsoni (Araneae, Salticidae) II. Sperm
competition and the function of copulation. J Arachnol 8 :217 -240
*Jackson RR (1982a) The behavior of communicating in jumping spiders (Salticidae). In: Witt PN,
Rovner JS (eds) Spider communication: mechanisms and ecological significance. Princeton
University Press, Princeton, pp 213-247
*Jackson RR (1982b) Comparative studies of Dictyna and Mallos (Araneae, Dictynidae): IV. Silk
mediated interattraction. Insectes Soc 29: 15 - 24
Jackson RR (1982c) Particularites des filieres de l'araignee cribellate Mallos gregalis (Dictynidae).
Rev Arachnol 3 :141-145
Jackson RR (1985) The biology of Euryattus sp. indet., a web-building jumping spider (Araneae,
Salticidae) from Queensland: utilization of silk, predatory behaviour and intraspecific inter-
actions. J Zool (London) 13 :145 -173
Jackson RR (1986a) Cohabitation of males and juvenile females: a prevalent mating tactic of spiders.
J Nat Hist
Jackson RR (1986b) Use of pheromones by males of Phidippus johnsoni (Araneae, Salticidae) to
detect subadult females that are about to molt. J Arachnol
Jackson RR (in press) Comparative study of releaser pheromone associated with the silk of jump·
ing spiders (Araneae, Salticidae)
Jackson RR, Blest AD (1982) The biology of Portia fimbriata, a web-building jumping spider
(Araneae, Salticidae) from Queensland: utilization of webs and predatory versatility. J Zool
(London) 196:255-293
Jackson RR, Hallas SEA (1986a) Comparative studies on Portia, araneophagic web-building jump-
ing spiders (Araneae, Salticidae): predatory versatility, utilisation of silk and intraspecific inter-
actions of P. africana, P. albimana, P. fimbriata, P. labiata and P. schultzi. NZ J Zool13
Jackson RR, Hallas SEA (1986b) Predatory versatility and intraspecific interactions of spartaeine
jumping spiders (Araneae, Salticidae): Brettus adonis, B. cingulatus, Cyrba algerina, and Phaeacius
sp.n. NZ J Zool13
Jackson RR, Pollard SD (1982) The biology of Dysdera crocata (Araneae, Dysderidae): intraspecific
interactions. J Zool (London) 198:197-214
Janetos AC (1982) Foraging tactics of two guilds of web-spinning spiders. Behav Ecol Sociobiol
10:19-27
Janetos AG (1986) Web site selection: are we asking the right questions? In: Shear WA (ed) Spiders:
webs, behavior and evolution. Stanford Univ Press, Stanford CA, pp 9-22
Janzen DH (1966) Why mountain passes are higher in the tropics. Am Nat 101 :203-249
Jaw.orowski A (1894) Die Entwicklung der sogenannten Lungen bei den Arachnoiden. Z Wiss Zool
58:54-78
Jayakar SD, Mangipudi RS (1965) Dormitories of Chalybion bengalense Dahlb. J Bombay Nat Hist
Soc 61 :708-711
Jeanne RL (1972) Social biology of the neotropical wasp Mischocyttarus drewseni. Bull Mus Comp
ZooI144:63-150
Jegla TC (1974) Ecdysone activity in Limulus polyphemus. Am Zoo114:1288
*Jegla TC (1982) A review of the molting physiology of the trilobite larva of Limulus. In: Alan RL
(ed) Physiology and biology of horseshoe crabs: studies on normal and environmentally stressed
animals. Liss, New York, pp 83-101
408 References
Jegla TC, Costlow JD (1970) Induction of molting in horseshoe crab larvae by polyhydroxy stero-
ids. Gen Comp EndocrinoI14:295-302
Jegla TC, Costlow 10 (1979a) The Limulus bioassays for ecdysteroids. Bioi Bu11156:103-114
Jegla TC, Costlow 10 (1979b) Ecdysteroids in Limulus larvae. Experientia 35 :554-555
Jegla TC, Costlow JD, Alspaugh J (1972) Effects of ecdysones and some synthetic analogs on horse-
shoe crab larvae. Gen Comp Endocrinol19 :159-166
Jenks GE (1940) The spider's "uninvited" fly brings doom. Nat Hist 45:157 -161
Jennings DT (1972) An overwintering aggregation of spiders (Araneae) on cottonwood in New
Mexico. Entomol News 83:61-67
Jennings DT, Houseweart MW (1978) Spider preys on spruce budworm egg mass. Entomol News
89:183-186
Jensen M, Weis-Fogh T (1962) Biology and physics of locust flight. V. Strength and elasticity of
locust cuticle. Philos Trans R Soc London Ser B 245:137-169
Jocque R (1981) On reduced size in spiders from marginal habitats. Oecologia (Berlin) 49:404-408
Johnson CG (1960) A basis for a general system of insect migration by flight. Nature (London)
186 :348 -350
Johnson SF, Gessaman JA (1973) An evaluation of heart rate as an indirect monitor of free-living
energy metabolism. In: Gessaman JA (ed) Ecological energetics of homeotherms. Monogr
Series, vol XX. Utah St Univ Press, Logan, pp 44-54
Joly R, Porcher on P, Dray F (1979) Etude des variations du taux d'ecdysteroides au cours du
cycle de mue dans l'hemolymphe de Lithobius forficatus (Myriapode, Chilopode). CR Acad
Sci 288 :243 -246
Jones JC, Smith LS, Larimer IL (1971) Part X. Invertebrate circulation. Section 201. Heart rate.
In: Altman PL, Dittmer DS (eds) Respiration and circulation. A biological handbook. Fed Am
Soc Exp Bioi, Bethesda, pp 600-606
Jones SE (1941) Influence of temperature and humidity on the life history of the spider Agelena
naevia WALCKENAER. Ann Entomol Soc Am 34:557-570
Juberthie C (1954) Sur les cycles biologiques des araignees. Bull Soc Hist Nat Toulouse 89:299-
318
Juberthie C, Lopez A (1980) La glande clypeale d'Argyrodes argyrodes (Walck): nouvelles pn!ci-
sions sur son ultrastructure. Rev Arachnol 3 :1-11
Juberthie C, Lopez A, Kovoor J (1981) Spermiogenesis and spermatophore in Telema tene11a
Simon (Araneae:Telemidae). An ultrastructural study. Int J Invert Reprod 3:181-191
Jungreis AM (1978) Insect dormancy. In: Clutter ME (ed) Dormancy and developmental arrest.
Academic Press, London New York, pp 47 -112
Kaestner A (1927) Pseudoscorpiones. Bioi Tiere Dtsch Part 18:1-68
Kaestner A (1929) Bau und Funktion der Fachertracheen einiger Spinnen. Z Morphol Okol Tiere
13:463-558
Kaestner A (1952) Die Mundwerkzeuge der Spinnen, ihr Bau, ihre Funktion und ihre Bedeutung
fdr das System. 1. Teil Orthognatha, Palaeocribellata. Zool Jahrb Abt Anat Ontog Tiere 7·2:
101-146
Kaestner A (1953) Die Mundwerkzeuge der Spinnen, ihr Bau, ihre Funktion und ihre Bedeutung
fdr das System. 2. Teil Herleitung und biologische Bedeutung der Labidognathie. Zool Jahrb
Abt Anat Ontog Tiere 73 :47 -6 8
Kaestner A (1969) Lehrbuch der speziellen Zoologie. I. 1. Teil, Fischer, Stuttgart
Kaiser E (1953) The enzymatic activity of spider venom. Mem Inst Butantan 25 :35 -39
Kaiser E, Michl H (1958) Biochemie der tierischen Gifte. Deuticke, Wien
Kajak A (1965) An analysis of food relations between the spiders Araneus cornutus and A. quadra-
tus and their prey in meadows. Ekol Pol A 13 :717 -768
Kajak A (1967) Productivity of some populations of web spiders. In: Petrusewicz K (ed) Secondary
productivity of terrestrial ecosystems. Warsaw, pp 807 -820
Kammer AE (1981) Physiological mechanisms of thermoregulation. In: Heinrich B (ed) Insect
thermoregulation. Wiley, New York, pp 115-158
Karlson P, Sekeri KE, Richards AG, Richards PA (1969) The amino acid com position of the various
types of cuticle of Limulus polyphemus. J Insect Physiol15 :495-507
Kaston BJ (1945) Notes on nematode parasites of spiders. Trans Conn Acad Sci 36 :241-244
References 409
Kaston BJ (1948) Spiders of Connecticut. Peiper Press, Wallingford (State Geol Nat Hist Surv Bul1
70)
Kaston BJ (1964) The evolution of spider webs. Am Zool 4 :191-207
Kaufman WR, Sauer JR (1982) Ion and water balance in feeding ticks: mechanisms of tick excre-
tion. In: Obenchain FD, Galun R (eds) Physiology of ticks. Current themes in tropical Science,
vol!. Pergamon Press, Oxford New York, pp 213-244
Kavanagh EJ, Til1inghast EK (1979) Fibrous and adhesive components of the orb webs of Araneus
trifolium and Argiope trifasciata.J MorphoI160:17-32
Kavanagh EJ, Tillinghast EK (1983) The alkaline proteases of Argiope. II. Fractionation of protease
activity and isolation of a silk fibroin digesting protease. Comp Biochem Physiol 74:365-372
Kawahara S, Kiritani K, Kakiya N (1974) Population biology of Lycosa pseudoannulata (Boes. et
Str.). Bun Kochi Inst Agric For Sci 6:7 -22
Kawakami M (1966) [The feeding habits of alpine salamanders.] Proc Hokkaido Gov High School
Teachers Assoc Sci Res:I-15
Kennaugh JH (1959) An examination of the cuticles of two scorpions, Pandinus imperator and
Scorpiops hardwicki. Q J Microsc Sci 100:41-50
Kennaugh JH (1968) An examination of the cuticles of three species of Ricinulei (Arachnida). J
Zoo I (London) 56 :393-404
Kerkut GA, Taylor BJR (1958) The effect of temperature changes on the activity of poikilotherms.
Behaviour 13:259-279
Kessler A (1971) Relation between egg production and food consumption in species of the genus
Pardosa (Lycosidae, Araneae). Oecologia (Berlin) 8:93-109
Kessler A (1973) A comparative study of the production of eggs of eight Pardosa species in the field.
Tijdschr EntomoI1l6:23-41
Kessler A, Vermeulen JWC, Wapenaar P (1984) Partitioning of the space in tussocks of the sedge,
Carex distans, during winter, by a spider community. J Zool (London) 204:259-269
Kestler P (1985) Respiration and respiratory water loss. In: Hoffmann KH (ed) Environmental
physiology and biochemistry of insects. Springer, Berlin Heidelberg New York, pp 137 -183
Kew HW (1930) On the spermatophores of pseudo scorpions Chthonius and Obisium. Proc Zool
Soc London 1930:253-256
*King JL (1916) Observations on the life history of Pterodontia j7avipes Gray. Ann Entomol Soc
Am 9:309-321
King WW, Hadley NF (1979) Water flux and metabolic rates of free-roaming scorpions using the
doubly label1ed water technique. Physiol ZooI52:176-189
Kingsley JS (1885) Notes on the embryology of Limulus. Q Microsc Sci 25
Kiquboe T, Mct>hlenberg F, Nicolajesn H (1982) Ingestion rate and gut clearance in the planktonic
copepod Centropages hamatus (Lil1jeborg) in relation to food consumption and temperature.
Ophelia 21:181-194
Kirchner W (1965) Wie iiberwintert die Schilfradspinne Araneus cornutus? Nat Mus 95: 163 -170
Kirchner W (1973) Ecological aspects of cold resistance in spiders (A comparative study). In: Wieser
W (ed) Effects of temperature on ectothermic organisms. Springer, Berlin Heidelberg New York,
pp 271-279
Kirchner W (1986) Das Netz der Zitterspinne (Pholcus phalangioides Fuesslin) (Araneae:Pholcidae).
Zool Anz 216:151-169
Kirchner W, Kestler P (1969) Untersuchungen zur Kiilteresistenz der Schilfradspinne Araneus
cornutus (Araneidae). J Insect PhysioI15:41-53
Kirchner W, Kullmann E (1972) 6kologische Untersuchungen an einer Freilandpopulation von
Nesticus cellulanus irn Siebengebirge unter besonderer Beriicksichtigung der Kiilteresistenz
(Araneae, Nesticidae). Decheniana 125 :219-227
Kirchner W, Kullmann E (1975) Oberwinterung und Kiilteresistenz der Haubennetzspinnenarten
Theridion impressum (1. Koch) und Theridion sisyphium (Clerck) (Araneae, Theridiidae).
Decheniana 127 :241-250
Klein-Koch C (1976) Die Wirkung von Juvenoiden und lihnlichen Substanzen auf Spinnmilben.
Z Angew EntomoI82:192--199
Kloppenburg P, Pulz R (in press) Das gerippte Haarsensil1um auf dem Vogelspinnentarsus ist kein
Thermorezeptor. Verh Dtsch Zool Ges 79
410 References
Knight J, Sutton L (1982) Successful treatment of Atrax formidabilis envenomation. Med J Aust
1:434-435
Kniker WT, Morgan PN, Flanigan WJ, Reagan PW, Dillaha CJ (1969) An inhibitor of complement
in the venom of the brown recluse spider, Loxosceles reclusa. Proc Soc Exp Bioi Med 131:
1432-1439
Kniille W, Rudolph D (1982) Humidity relationships and water balance of ticks. In: Obenchain
FD, Galun R (eds) Physiology of ticks. Current themes in tropical science, vol 1. Pergamon
Press, Oxford New York, pp 43-70
Kobayashi Y, Fujiwara T, Kyogoku Y, Kataoka K (1980) Abstracts of the 19th Meeting of Nuclear
Resonance in Japan, pp 149-152
Kobert R (1901) Beitrage zur Kenntnis der Giftspinnen. Enke, Stuttgart
Kolosvary G von (1939) Neue Angaben tiber das Uberwintern einiger Spinnenarten. Veroeff Dtsch
Kolonial-Obersee-Mus 2 :268-271
Kolosvary G von (1940) Weitere Angaben tiber das Uberwintern einiger Spinnenarten. Veroeff
Dtsch Kolonial-Ubersee-Mus 3:56--57
Konishi T, Kurokawa M (1968) The structure of silk fibroin-Q!. Sen-i Gakkaishi 24:550-554
Kovoor J (1972) Etude histochimique et cytologique des glandes sericigenes de quelques Argiop-
idae. Ann Sci Nat Zool Bioi Anim 14: 1-40
Kovoor J (1976) Caracteres adaptatifs et caracteres familiaux des glandes sericigenes dans Ie genre
Hippasa E. Simon (Araneae, Lycosidae). CR 3e Coli Arachnol Exp Fr, Les Eyzies·83-96
*Kovoor J (1977a) La soie et les glandes sericigenes des arachnides. Ann Bioi 16 :97 -171
Kovoor J (1977b) Donnees histochimiques sur les glandes sericigenes de la veuve noire Latrodec-
tus mactans Fabr. (Araneae, Theridiidae). Ann Sci Nat Zool12 Ser 19:63 -87
Kovoor J (1977c) L'appareil sericigene dans Ie genre Uloborus Latr. (Araneae, Uloboridae). I.
Anatomie. Rev Arachnoll:89-102
Kovoor J (1978) L 'appareil sericigene dans Ie genre Uloborus Latreille (Araneae, Uloboridae). II.
Donnees histochimiques. Ann Sci Nat Zoo 1 Bioi 12 Ser 20:3-25
Kovoor J (1979) Les glandes sericigenes d 'Uroctea durandi (Latreille) (Araneae:Oecobiidae) Revi-
sion, histochimie, affinites_ Ann Sci Nat Zool13 Ser 1 :187 -203
Kovoor J (1981) Une source probable de pheromones sexuelles: les glandes tegumentaires de la
region genitale des femelles d 'araignees. Atti Soc Toscana Sci Nat Pisa Mem 88:1-15
Kovoor J (1984) Anatomie, histologie et affinites de l'appareil sericigene des Hersilia Sav. & Aud.
(Araneae, Hersiliidae). Can J Zoo I 62:97-106
Kovoor J (1986) L'appareil sericigene dans les genres Nephila Leach etNephilengys Koch: anatomie
microscopique, histochimie affinites avec d'autres Araneidae. Rev Arachnol 7 :15-34
Kovoor J (in press) Affinites des Pholcidae (Araneae) decelables a partir de caracteres des glandes
sericigenes. Bull Inst Roy Sci Nat Belg
Kovoor J, Lopez A (1980) Variations de l'appareil sericigene dans la famille des Araneidae: cas des
genres Cyclosa Menge etNemoscolus Simon. CR5th Coli ArachnolExp Fr, Barcelone, pp 119-
127
Kovoor J, Lopez A (1982) Anatomie et histologie des glandes sericigenes des Cyrtophora (Araneae,
Araneidae): affinittls et correlations avec la structure et la composition de la toile. Rev Arachnol
4:1-21
Kovoor J, Lopez A (1983a) Structure et ultrastructure de l'appareil sericigene chez Telema tenella
Simon (Araneae, Telemidae). Mem Biospeoll0:419-426
Kovoor J, Lopez A (1983b) Composition et histologie de l'appareil sericigene des Argyrodes Simon
(Theridiidae), relations avec Ie comportement de ces araignees. Rev Arachnol5 :29-43
*Kovoor J, Zylberberg L (1972) Histologie et infrastructure de la glande chl!licerienne de Scytodes
delicatula Sim. (Araneida, Scytodidae). Ann Sci Nat Zool12 Ser 14:333-388
*Kovoor J, Zylberberg L (1979) Ultrastructure du canal des glandes agregees et flagelliformes d'Ara-
neus diadematus Clerck (Araneae, Araneidae). Zoomorphologie 92:217-239
*Kovoor J, Zylberberg L (1980) Fine structural aspects of silk secretion in a spider (Araneus dia-
dematus) I. Elaboration in the pyriform glands. Tissue Cell 12:547 -556
Kovoor J, Lopez A, Emerit M (1981) Des glandes tegumentaires particulieres aux femelles chez
Lepthyphantes sanctivincentii et Linyphia triangularis (Araneae, Linyphiidae). Atti Soc Toscana
Sci Nat Pisa Mem 88:53-60
Krafft B (1967) Thermopreferendum de I'araignee saciale Agelena consociata DENIS. Insectes
Soc 14:161-182
References 411
Krafft B (1970) Les rythmes d'activite d'Agelena consocillta Denis, activite de tissage et activite
locomotrice. Rev Bioi Gabonica 6 :99 -130
*Krafft B (1980) Les systemes de communication chez les araignees. In: Proc 8th Int Congr Arach-
nol, Wien, pp 197-213
Krafft B (1982) The significance and complexity of communication in spiders. In: Witt PN, Rovner
JS (eds) Spider communication. Princeton Univ Press, Princeton
*Krafft B, Roland Ch (1980) Quelques remarques au sujet de la communication chimique chez les
araignees.In: CR 5th Coli Arachnol, Barcelona, pp 129-135
Krakauer T (1972) Thermal responses of the orb-weaving spider Nephila clavipes (Araneae:Argio-
pidae). Am Midi Nat 88:245-250
Krebs CJ(197 8) Evolutionary advantages of dispersal. Ecology 2nd edn. Harper & Row, New York,
pp 34-36
Krebs JR (1973) Colonial nesting and social feeding as strategies for exploiting food resources in
the Great Heron (Ardea herodias). Behaviour 51 :99-134
Krebs JR (1978) Optimal foraging: decision rules for predators. In: Krebs JR, Davies NB (eds) Be-
havioural ecology. Blackwell, Oxford
Krebs JR, Dawkins R (1984) Animal signals: mind-i:eading and manipulation. In: Krebs JR, Davies
NB (eds) Behavioural ecology: an evolutionary approach. Blackwell, Oxford
Krishnakumuran A (1962) A comparative study of the cuticle in Arachnida. 1. Structure and stain-
ing properties. Zool Jahrb Anat 80:49-64
Krishnakumuran A, Schneiderman HA (1968) Chemical control of moulting in arthropods. Nature
(London) 220 :601-603
Krishnakumuran A, Schneiderman HA (1970) Control of molting in mandibulate and chelicerate
arthropods by ecdysones. Bioi Bul1139:520-538
*Krombein KV (1967) Trap-nesting wasps and bees: life histories, nests, and associates. Smithsonian
Inst Press, Washington DC
Krombein KV (1979) Superfamily Sphecoidea. In: Krombein KV, Hurd PD, Smith DR, Burks BD
(eds) Catalogue of Hymenoptera in America. North of Mexico, vol 2, Apocrita (Aculeata). Smith-
sonian Inst Press, Washington DC, pp 1573 -1740
Kryger JP (1970) Snyltere i Edderkoppeaeg. Entomol Medd 3 :257 -285
Ku1lmann E (1959a) Beobachtungen und Betrachtungen zum Verhalten der Theridiide Conopista
argyrodes Walck. Mitt Zool Mus Berlin 35 :275 -292
Kullmann E (1959b) Ober parasitiires Verhalten der Spinne Theridium tepidariorum Koch. Verh
Deut Zool Ges 332--342
Kullmann E (1959c) Beobachtungen an der Raumnetzspinne Cyrtophora citricola Forsk?il auf
Sardinien (Araneae, Araneidae). Dtsch Entomol Z NS 6:65-81
Kullmann E (1960) Beobachtungen an Theridium tepidariorum Koch als Mitbewohner von Cyr-
thophora Netzen. Dtsch Entomol Z 7:146-164
Kullmann E (1968a) Soziale Phiinomene bei Spinnen. Insectes Soc 15 :289-298
Kullmann E (1968b) Das Cribellum zweier Stegodyphus-Atten im elektronenoptischen Bild (Arach-
nida:Araneae :Eresidae). Senckenbergiana Bioi 49 :451-460
Ku1lmann E (1971) Bemerkenswerte Konvergenzen im Verhalten cribellater und ecribellater Spin-
nen. Freunde Koln Zoo 13 :123-150
Kullmann E (1974) Theridion impressum (Theridiidae), Brutfiirsorge und periodisch-soziales Ver-
halten. Encycl Cinemat E 1864, Gottingen
Kullmann E (1975) Die Produktion und Funktion von Spinnenfiiden und Spinnengeweben. In:
Inst Leichte Fliichentragwerke (ed) Netze in Natur und Technik. Stuttgart-Vaihingen, pp 318-
378
Kumar R (1980) Effect of administration of synthetic ecdysone on the moulting of Palamnaeus
bengalensis. Experientia 36 :411-412
*Kurczewski FE (1969) Comparative ethology of female digger wasps in the genera Migcophus and
Nitelopterus (Hymenoptera:Sphecidae, Larrinae). J Kans Entomol Soc 42:470-509
*Kurczewski FE (1982) An additional study on the nesting behaviors of species of Miscophus
(Hymenoptera:Sphecidae). Proc Entomol Soc Wash 84:67-80
Kurihara K (1979) Photoperiodic regulation of winter diapause in the grass spider. Experientia
35:1479-1480
Lagerspetz K, Jayniis E (1959) The behavioural regulation of the water content in Linyphill mon-
tana (Araneae, Linyphiidae) and some other spiders. Ann Entomol Fenn 25 :210-223
412 References
Lagueux M, Hirn M, Hoffmann lA (1977) Ecdysone during ovarian development in Locusta migra-
toria. 1 Insect Physiol23 :109-119
Lahmann El, Zuniga CM (1981) Use of spider threads as resting places by tropical insects. 1 Arach-
noI9:339-341
Lamoral BH (1968) On the ecology and habitat adaptations of two intertidal spiders, Desis formid-
abilis (O.P. Cambridge) and Amaurobioides africanus Hewitt, at "The Island" (Kommetjie, Cape
Peninsula), with notes on the occurrence of two other spiders. Ann Natal Mus 20:151-193
*Lamore DH (1960) Cases of parasitism of the basilica spider, Allepeira lemniscata (Wa1ckenaer), by
the dipteran endoparasite, Ogcodes dispar (Macquart). Proc Entomol Soc Wash 62:65 -85
*Lamy 1, Truchot 1-P, Gilles R (eds) (1985a) Respiratory pigments in animals. Relation structure-
function. Springer, Berlin Heidelberg New York, 175 p
Lamy J, Sizaret P-Y, Billard P, Motha G (1985b) Quaternary structure of arthropod hemocyanins.
In: Lamy 1, Truchot 1-P, Gilles R (eds) Respiratory pigments in animals. Springer, Heidelberg
New York, pp 73-87
*Land MF (1985) The morphology and optics of spider eyes. In: Barth FG (ed) Neurobiology of
arachnids. Springer, Berlin Heidelberg New York, pp 53-78
Lane WI (1965) The biology of, and some ecological aspects of some Lycosa species. Thesis, Univ
West Aust, Perth
Larcher SF, Wise D (1985) Experimental studies of the interactions between a web-invading spider
and two host species. 1 Arachnol13:43-59
Le Berre M (1979a) Mise en place du rythme d'activite chez Araneus cornutus. C R Acad Sci 288:
839-842
Le Berre M (1979b) Modification temporelles de I'activite chez de jeunes argiopides. Rev Arachnol
2:193-200
Le Berre M, Ramousse R, LeGuelte L (1981) Eco-Ethologie des argiopides: 1. Evolution temporel-
Ie d 'une popUlation d 'Araneus diadematus Clerck dans son milieu nature!. Atti Soc Toscana Sci
Nat Pisa Mem (B) 88:71-83
Lebez D (1953) Prispevki k studiju strupa tarantele. Bioi Vestn 2:27---35
Lebez D (1954) Beitriige zum Studium des Giftes von Latrodectus tredecimguttatus Rossi. Hoppe-
Seiler's Z Physiol Chern 298:74-76
*Leborgne R (1981) Soie et communication chez les araignees (Le rapprochement des sexes). Atti
Soc Toscana Sci Nat Pisa Mem 88:132-142
Leborgne R, Roland Ch, Horel A (1980) Quelques aspects de la communication chimique et vibra-
toire chez certaines Agelenidae. In: Proc 8th Int Congr Arachnol, Wien, pp 215-220
Leech RE (1966) The spiders (Araneida) of Hazen Camp 81°49'N, 71°18'W. Quaest Entomol2:
153-212
Lees AD (1955) The physiology of diapause in arthropods. Cambridge Univ Press, Cambridge
Legel GJ, Wingerden WKRE van (1980) Experiments on the influence of food and crowding on
aeronautic dispersal of Erigone arctica (White, 1852) (Araneae, Linyphiidae). Proc 8th Int
Arachnol Congr, Vienna, pp 97-102
Legendre R (1961a) Le mechanism de la prise de nourriture chez les araigm!es. CR Acad Sci 252:
321
Legendre R (1961b) Etudes sur les Archaea (Araneides). II. La capture des proies et la prise de
nourriture. Bull Soc Zool Fr 86 :316-319
Legendre R (1972) Les glandes epigastriques de la mygale Scodra calceata Fabr. (Orthognatha,
Theraphosidae). C R Acad Sci D 274:542-545
Legendre R (1978) Quelques prognls recent concernant l'anatomie des araignees (systeme nerveux
sympathique et appareil digestif). Symp Zool Soc London 42:379-388
Legendre R, Lopez A (1978) Presence d 'un spermatophore dans Ie genre ApneumoneUa (Araneae:
Telemidae): valeur taxonomique et probhlmes de biologie sexuelle. Bull Soc Zool Fr 103 :35 -41
Legg G (1971) The comparative and functional morphology of the genitalia of the British Pseudo-
scorpions. PhD Thesis, Univ Manchester
Legg G (1973) Spermatophore formation in the pseudoscorpion Chthonius ischnocheles (Chthoni-
idae).l Zool (London) 170 :367 -394
Legg G (1974a) The genitalia and accessory glands of the pseudoscorpion Cheiridium museorum
(Cheiridiidae). J Zoll (London) 173 :323-339
References 413
Legg G (1974b) A generalized account of the male genitalia and associated glands of pseudoscorpions
(Arachnida). Bull Br Arachnol Soc 3 :66-74
LeGuelte L, Ramousse R (1979) Effets de facteurs environnementaux constructeur chez l'araignt!e
Araneus diadematus CI. BioI Behav 4:289-302
Lehmensick R, Kullmann E (1956a) tiber den Feinbau der Faden einiger Spinnen. Vergleich des
Aufbaues der Fangfiiden cribellater und ecribellater Spinnen. Zool Anz Supp1123-129
Lehmensick R, Kullmann E (1956b) tiber den Feinbau der Spinnenf:i.den. Proc Stockholm Conf
Electron Microsc, pp 307 -309
Lehtinen P (1967) Classification of the cribellate spiders and some allied families, with notes on
the evolution of the suborder Araneomorpha. Ann Zool Fenn 4:199-468
Leidy J (1857) A synopsis of Entozoa and some of their ectoconginers. Proc Acad Nat Sci Phila-
delphia 8:42-58
Leinaas HP (1981) Activity of Arthropoda in snow within a coniferous forest with special reference
to Collembola. Holarct EcoI4:127-138
Le Sar CD, Unzicker JD (1978) Soybean spiders: species composition, population densities and
vertical distribution. BioI Not III Nat Hist Surv No 107:3-14
Levi HW (1967a) The theridiid spider fauna of Chile. Bull Mus Comp Zool136:1-20
*Levi HW (1967b) Adaptations of respiratory systems of spiders. Evolution 21 :571-583
Levi HW (1975a) Mating behavior and presence of embolus cap in male Araneidae. In: Proc 6th
Int Congr Arachnol, Amsterdam, pp 49-50
Levi HW (1975b) The orb-weaver genera Verrucosa, Acanthepeira, Wagneria, Acacesia, Wixia,
Scoloderus and Alpaida north of Mexico. Bull Mus Comp Zool Harvard 147:351-391
Levi HW (1980) The orb-weaver genus Mecynogea, the subfamily Metinae and the generaPachygna-
tha, Glenognatha and Azilia of the subfamily. Tetragnathinae north of Mexico. Bull Mus Comp
ZooI149:1-74
*Levi HW (1982) Chelicerata. In: Synopsis and classification of living organisms. McGraw-Hill, New
York
Levi HW, Coddington J (1983) Progress report on the phylogeny of the orb-weaving family Arane-
idae and the superfamily Araneoidea (Arachnida:Araneae). Verh Naturwiss Ver Hamburg 26:
151-154
*Levy G (1970) The life cycle of Thomisus onustus (Thomisidae:Araneae) and outlines for the clas-
sification of the life histories of spiders. J Zoo I (London) 160 :523 -526
Lhotsky S, Phaneuf, Langlois JM, Gagne S, Poussart D (1975) Optocardiography: in vivo measure-
ments of the insect cardiac activity with a new optical method. J Insect Physiol 21:237 -248
Ligon PJB, Greenberg MJ (1971) Part X. Invertebrate circulation. Section 202. Response of heart
rate to temperature: arthropods, annelids, and mollusks. In: Altmann PL, Dittmer DS (eds)
Respiration and circulation. A biological handbook. Fed Am Soc Exp BioI, Bethesda, pp 607-
608
Linstow 0 von (1883) Nematoden, Trematoden und Acanthocephalen, gesammelt von Prof. Fedt-
schenko in Turkestan. Arch Naturgesch 49:274-314
Linstow 0 von (1898) Das Genus Mermis. Arch Mikrosk Anat 53:149--168
Linzen B (1967) Zur Biochemie der Ommochrome. Naturwissenschaften 54:259-267
Linzen B, Gallowitz P (1975) Enzyme activity patterns in muscles of the lycosid spider, Cupien-
nius salei. J Comp PhysioI69:101-109
*Linzen B, Schartau W, Schneider H-J (1985a) Primary structure of arthropod hemocyanins. In:
Lamy J, Truchot J-P, Gilles R (eds) Respiratory pigments in animals. Springer, Berlin Heidelberg
New York, pp 59-71
*Linzen B, Soeter NM, Riggs AF, Schneider H-J, Schartau W, Moore MD, Yokota E, Behrens PQ,
Nakashima H, Takagi T, Nemoto T, Vereijken JM, Bak HI, Beintema JJ, Volbeda A, Gaykema
WPJ, HoI WGI (1985b) The structure of arthropod hemocyanins. Science 229:519-524
Livecchi G (1978) Ontogenese du comportement constructeur chez deux especes d'araignt'!es (Ara-
neus diadematus; Zygiella x-notata). These 3. Cycle, Lyon
Livecchi G (1979) Ontogent!se du rythme du comportement constructeur chez Araneus diadematus.
Rev ArachnoI2:201-203
Locket GH, Millidge AF (1951) British spiders, voll. Ray Soc, London
Locket GH, Millidge AF (1953) British spiders, vol 2. Ray Soc, London
414 References
Low AM (1983) Untersuchungen zur cuticularen und pulmonaren Transpiration der Vogelspinne
Eurypelma californicum. Dipl Thesis, Univ Miinchen
Loewe R, Brauer de Eggert H (1979) Blood gas analysis and acid-base status in the hemolymph of a
spider (Eurypelma californicum) ~ influence of temperature. J Comp Physiol134:331 ~338
Loewe R, Linzen B (1973) Haemocyanin in spiders. I. Subunits and stability region of Dugesiel!a
californicum haemocyanin. Hoppe Seyler's Z Physiol Chern 354:182~188
Loewe R, Linzen B (1975) Haemocyanins in spiders II. Automatic recording of oxygen binding
curves, and the effect of Mg++ on oxygen affinity, cooperativity, and subunit association of
Cupiennius sa lei haemocyanin. J Comp Physiol 98:14 7~156
*Loewe R, Linzen B, Stackelberg W von (1970) Die geliisten Stoffe in der Hiimolymphe einer Spin-
ne, Cupiennius salei Keyserling. Z Vergl Physiol66:27 ~34
Longenecker HE, Hurlbut WP, Mauro A, Clark AW (1970) Effects of black widow spider venom
on the frog neuromuscular junction. I. Effects on end-plate potentials, miniature end-plate
potential and nerve terminal spike. Nature (London) 225 :70 1 ~ 703
Longman (1978) Longmans dictionary of contemporary English. Longman, Harlow, London
Lopez A (1974) Glandes epigastriques et classification des araneides. Bull Assoc Anat 58:899~904
Lopez A (1977a) Contribution a l'etude des caracteres sexuels somatiques chez les males des ara-
neides. Thesis, Univ Montpellier
Lopez A (1977b) Sur un nouveau mode de reproduction chez les araignees: existence de spermato-
phores chez Telema tene1la Simon (Telemidae). Bull Soc Zool Fr 102:261~266
Lopez A (1983a) Coxal glands of the genus Metepeira (Araneae, Araneidae). J Arachnoll1 :97 ~98
Lopez A (1983b) Some observations on the internal anatomy of Diguetia canities (Mc Cook, 1890).
J Arachnolll :377 --384
Lopez A, Emerit M (1978) Le dimorphisme sexuel gnathocoxal de Leptoneta microphthalma Simon
(Araneae, Leptonetidae). Rev ArachnoI2:1~15
Lopez A, Juberthie-Jupeau L (1983) Structure et ultrastructure de la spermatheque chez Telema
tene1la Simon (Araneae, Telemidae). Mem Biospeoll0:413~418
Lopez A, Juberthie-Jupeau L (1985) Ultrastructure comparee du tube seminifere chez les males
d 'araignees. Mem Biospeol12 :97 ~ 109
Lopez A, Stowe MK (1985) Observations sur quelques araignees "excentriques" du Nouveau
Monde et leurs glandes a soie. Bull Soc Sci Nat Beziers 10:16~23
Lopez A, Juberthie-Jupeau L, Ribera C (1980) Les glandes gnathocoxales des Leptoneta (Araneae:
Leptonetidae): structure, ultrastructure et interet systematique. In: CR 5th Coli Arachnol,
Barcelona, pp 163 ~178
Lopez A, Stowe MK, Bonaric J-C (1985) Anatomie interne de l' "araignee a bolas" nord-ameri-
caine Mastophora cornigera (Hentz, 1850) (Araneae:Araneidae) apres sa sortie du cocon. Publ
Sci Accel No 8:1 ~8
Loveridge JP (1980) Cuticular water relations teChniques. In: Miller TA (ed) Cuticle technique in
arthropods. Springer, Berlin Heidelberg New York, pp 301~366
Lubin YD (1974) Adaptive advantages and the evolution of colony formation in Cyrtophora. Zool
J Linn Soc 54:321~339
Lubin YD (1975) Stabilimenta and barrier webs in the orb webs of Argiope argentata (Araneae,
Araneidae) on Daphne and Santa Cruz Islands, Galapagos. J ArachnoI2:119~126
Lubin YD (1986) Courtship and alternative mating tactics in a social spider. J Arachno114;239~25 7
*Lubin YD, Eberhard WG, Montgomery GG (1978) Webs of Miagrammopes (Araneae:Uloboridae)
in the Neotropics. Psyche 85:1 ~23
Lucas F (1964) Spiders and their silks. Discovery 25 :20~26
Lucas F, Shaw JTB, Smith SG (1958) The silk fibroins. Adv Protein Chern 13:107~242
Luczak J, Dabrowska-Prot E (1969) Preliminary observations on the food of the spider Theridion
pictum and its predators. Bull Br Arachnol Soc 1 :109~111
Macchiavello A (1937) La Loxosceles !aeta, causa del aracnoidismo cutaneo 0 mancha gangrenosa
de Chile. Rev Chil Hist Nat 41 :11 ~19
Machin J (1979) Atmospheric water absorption in arthropods. Adv Insect Physiol14: 1 ~48
MacNamara C (1924) The food of Collembola. Can Entomol56 :99~105
MacPhee AW (1961) Mortality of winter eggs of the European red mite Panonychus ulmi (Koch),
at low temperatures, and its ecological significance. Can J ZooI39:229~243
MacPhee AW (1964) Cold hardiness, habitat and winter survival of some orchard arthropods in Nova
Scotia. Can Entomol 96:617 ~625
References 415
Main BY (1976) Spiders. Aust Nat Library, Collins Sydney London

Main BY (1982) Adaptations to arid habitats by mygalomorph spiders. In: Barker WR, Greenslade
PJM (eds) Evolution of the flora and fauna of arid Australia. Peacock, Frewville S Aust, pp 273-
283
Mango C, Odhiambo TR, Galun R (1976) Ecdysone and the super tick. Nature (London) 260:
318-319
Mansingh A, Rawlins SC (1977) Antigonadotropic action of insect hormone analogues on the cattle
tick Boophilus microplus. Naturwissenschaften 64 :41
Maretic Z (1951) Beobachtungen liber Pathologie und Klinik des Latrodektismus in Istrien (1948-
1950). Acta Trop 8:136-144
Maretic Z (1953) Arachnidism treated with cortisone. JAMA 152:609
Maretic Z (1955) Medicinsko znaeenje ujeda paukova u naSim krajevima. Med Glas 9:159-163
Maretic Z (1959) Araneizam - s osobitim osvrtom na Istru. Thesis, Med Fac Zagreb
Maretic Z (1963) Electrocardiographic changes in man and experimental animals provoked by the
venom of Latrodectus tredecimguttatus. TOlticon 1: 127 -130
Maretic Z (1965) Latrodectus und Latrodectismus. Nat Mus 95 :124-132
Maretic Z (1967) Venom of an east african orthognath spider. In: Russel FE, Saunders P (eds)
Animal toxins. Pergamon Press, Oxford New York, pp 23-28
Maretic Z, Jela~ic F (1953) Uber den Einflu1l. des Toxins der Spinne Latrodectus tredecimguttatus
auf das Nervensystem. Acta Trop 10:203-221
*Maretic Z, Lebez D (1979) Araneism. With special reference to Europe. Natl Library Med, Bethesda
- NOLIT, Beograd
*Maretic Z, Lebez D (1985) Otrovni pauci. Araneizam s posebnim osvrtom na Istru. 0.0. Zbora
lije<!nika Hrvatske. Med Cent, Pula. (rransl. from English and 2nd edn.)
Maretic Z, Russell FE (1979) A case of necrotic arachnidism in Yugoslavia. Toxicon 17 :412-413
Maretic Z, Levi HW, Levi LR (1964) The Theridiid spider Steatoda paykulliana poisonous to mam-
mals. Toxicon 2:149-154
Markl J, Schmid R, Czichos-Tiedt, Linzen B (1976) Haemocyanins in spiders. III. Chemical and
physical properties of the proteins in Dugesiella and Cupiennius blood. Hoppe Seyler's Z Physiol
Chern 357:1713-1725
Markl J, Kempter B, Linzen B, Bulholt MAC, Bruggen EFJ van (1981) Hemocyanins in spiders.
XVI. Subunit topography and a model of the quaternary structure of Eurypelma hemocyanin.
Hoppe-Seyler's Z Physiol Chern 362:1631-1641
*Marples BJ (1967) The spinnerets and epiandrous glands of spiders. J Linn Soc Zool 46 :209-223
Marples BJ (1968) The hypochilomorph spiders. Proc Linn Soc 179:11-31
Marples BJ (1969) Observations on decorated webs. Bull Br Arachnol Soc 1:13-18
Marples BJ (1983) Observations on the structure of the fore-gut of spiders. Bull Br Arachnol Soc
6:46-52
Marshall GAK (1902) The bionomics of South African insects. 4. Experiments on spiders in the
Karkloof. Trans Entomol Soc London 319-328
Marson JE (1947) Some observations on the ecological variation and development of the cruciate
zigzag camouflage device of Argiope pulchella (rhor.). Proc Zool Soc London 117 :219-227
Masters WM, Markl HS, Moffat AJM (1986) Transmission of vibration in a spider's web. In: Shear
WA (ed) Spiders: webs, behavior and evolution. Stanford Univ Press, Stanford CA, pp 49-69
Masuda H (1939) Biological notes on Pison iwatai Yasumatsu (Hymenoptera, Trypoxylonidae).
Mushi 12:114-146
Matthews RW, Matthews JR (1968) A note on Trypargilum arizonense in trap nests from Arizona,
with a review of prey preferences and cocoon structure in the genus (Hymenoptera:Sphecidae).
Psyche 75 :285 -293
Mauchamp B, Pennetier JL, Tarroux P, Lafont R (1977) Metabolisme de la diapause hivernale chez
Pieris brassicae. Aspects biochimiques et correlations endocrines. 4th Semin Physiol Insect,
Bordeaux
May ML (1985) Thermoregulation. In: Kergut GA, Gilbert L (eds) Comprehensive insect physio-
logy, biochemistry and pharmacology, vol 4. Pergamon Press, Oxford New York, pp 507-552
Maya M, Jaganathan S, Karuppaswamy SA (1981) Respiratory mechanism of the spider, Hippasa
pantherina (Family: Lycosidae). Proc Indian Acad Sci 90:539-543
Maynard-Smith J (1982) Evolution and the theory of games. Cambridge Univ Press, Cambridge
416 References
McCrone JD, Hatala JR (1967) Isolation and characterization of a lethal component from the
venom of Latrodectus mactans mactans. In: Russell FE, Saunders P (eds) Animal toxins. Perga-
mon Press, Oxford New York, pp 29-34
McCrone JD, Netzloff ML (1965) An immunological and electrophoretica1 comparison of the
venoms of the north american Latrodectus spiders. Toxicon 3:107 -110
McDaniel CN (1979) Haemolymph ecdysone concentrations in Hyalophora cecropia pupae, dauer
pupae and adults. J Insect PhysioI25:143-145
McDaniel RS, Oliver JH Jr (1978) Effects of two juvenile hormone analogs and i3-ecdysone on
nymphal development; spermatogenesis, and embryogenesis in Dermacentor variabilis (Say)
(Acari:Ixodidae). J Parasitol 64 :571-573
McIndoo NE (1911) The lyriform organs and tactile hairs of araneads. Proc Acad Nat Sci Phila 63:
375-418
McKeown KC (1936) Spider wonders of Australia. Angus & Robertson, Sidney
McLeod J (1884) Recherches sur la structure et la signification de l'appareil respiratoire des Arach-
nides. Arch BioI 5 :1-34
McQueen DJ (1978) Field studies of growth, reproduction, and mortality in the burrowing wolf
spider, GeolycoSil domifex (Hancock). Can J Zool 56:2037-2049
McQueen DJ (1979) Interactions between the pompilid wasp Anoplius relativus (Fox) and the
burrowing wolf spider Geolycosa domifex (Hancock). Can J Zool 57 :542-550
McQueen DJ (1980) Active respiration rates for the burrowing wolf spider Geolycosa domifex
(Hancock). Can J Zool 58:1066-1074
McQueen DJ (1983) Mortality patterns for a population of burrowing wolf spiders, GeolycoSil
domifex (Hancock), living in southern Ontario. Can J ZooI61:2758-2767
McQueen DJ, Mclay CL (1983) How does the intertidal spider Desis marina (Hector) remain under
water for such a long time? N Z J Zool 10 :383-392
*McQueen DJ, Jensen 1M, Dyer BS (1979) Resting and diel respiration rates for the burrowing wolf
spider GeolycoSil domifex (Hancock). Can J Zool 57 :1922-1933
McQueen OJ, Pannell LK, Mclay CL (1983) Respiration rates for the intertidal spider Desis marina
(Hector). N Z J Zoo 1 10:393--400
Meijer J (1976) A glandular secretion in the ocular area of certain erigonine spiders (Araneae, Liny-
phiidae). Bull Br Arachnol Soc 3:251-252
Melchers M (1963) Zur Biologie und zum Verhalten von Cupiennius salei (Keyserling), einer ameri-
kanischen Ctenide. Zoo I Jahrb Syst 91 :1-90
*Melchers M (1964) Zur Biologie der Vogelspinnen (Fam. Aviculariidae). Z Morphol Okol Tiere
53:517-536
Menge A (1866) Preussische Spinnen 1. Schr Naturforsch Ges Danzig 1:37
Meyer J (1977) The immigration of spiders (Araneida) into a new polder. Ecol EntomoI2:81-90
Michael AD (1896) The internal anatomy of Bdella. Trans Linn Soc 2 Ser 6:477 -528
Mikulska I (1961a) Rytm serea u swiezo wyleglych pajakow. Zesz Nauk UMK, Nauk Mat-przyr Bioi
6:21-33
Mikulska I (1961b) Changes in heart rate in spiders effected by increased temperature. Zool Pol
12:149-160
Mikulska I, Kokocinski W (1965) Le rythme du coeur dans l'abdomen isole de l'araignee. Bull Aead
Pol Sci CI2 13 :533-537
Miller LK (1969) Freezing tolerance in an adult insect. Science 166:105-106
Miller LK (1978) Physiological studies of arctic animals. Comp Biochem Physiol 59A:327 --334
*Millot J (1926) Contribution Ii l'histophysiologie des araneides. Bull Bioi Fr Beig SuppI8:1-238
Millot J (1930a) Le tissu reticule du cephalothorax des araneides et ses derives: nephrocytes et
cellules endocrines. Arch Anat Microsc 26:43-81
Millot J (1930b) Glandes venimeuses et glandes sericigenes chez les sicariides. Bull Soc Zool Fr
55 :150-175
Millot J (1935) Les glandes sericigenes des salticides. Bull Soc Zool Fr 60 :501-5 10
Millot J (1936) Le sens du gout chez les araignees. Bull Soc Zoo I Fr 61 :27
*Millot J (1938) Le developpement et la biologie larraire de ondcodides (= cyrtides), diptt\res para-
sites d'araignees. Bull Soc Zool Fr 63:162-181,183-197
References 417
Millot J (1949) Ordre des aramMes (Araneae). In: Grasse PP (ed) Traite de zoologie, vol VI. Mas-
son, Paris, pp 589-743
Millot J, Bourgin P (1942) Sur la biologie des Stegodyphus solitaire (Araneides Eresides). Bull BioI
Fr Belg 76 :298 -313
Millot J, Frontaine M (1937) La teneur en eau des araneides. Bull Soc Zool Fr 62:113-119
Minch EW (1978) Daily activity patterns in the tarantula Aphonopelma chalcodes Chamberlin. Bull
Br Arachnol Soc 4:231-237
Minch EW (1979) Annual activity patterns in the tarantula, Aphonopelma chalcodes Chamberlin.
Nov Arthropodae 1: 1-34
Mittelstaedt H (1985) Analytical cybernetics of spider navigation. In: Barth FG (ed) Neurobiology
of Arachnids. Springer, Berlin Heidelberg New York, pp 298-316
Miyashita K (1968a) Growth and development of Lycosa T·insignita Boes. et Str. (Araneae:Lyco-
sidae) under different feeding conditions. Appl Entomol Zool 3 :81-88
Miyashita K (l968b) Quantitative feeding biology of Lycosa T-insignita Boes. et Str. (Araneae:
Lycosidae). Bull Natl Inst Agric Sci Ser C No 22:329-344
Miyashita K (1969a) Seasonal changes of popUlation density and some characteristics of overwinter-
ing nymph of Lycosa T-insignita Boes. et Str. (Araneae:Lycosidae). Appl Entomol Zool4: 1-8
*Miyashita K (1969b) Effects of locomotory activity, temperature and hunger on the respiratory
rate of Lycosa T-insignita BOES. et STR. (Araneae:Lycosidae). Appl Entomol Zoo I 4:105-113
Miyashita T (1986) Growth, egg production, and popUlation density of the spider, Nephi/a clavata
in relation to food conditions in the field. Res Popul EcoI28:135-149
Moeur JE, Eriksen DH (1972) Metabolic responses to temperature of a desert spider, Lycosa (Par-
dosa) carolinensis (Lycosidae). Physiol Zool45 :290-30 1
Moles ML (1916) The growth and color patterns in spiders. J Entomol ZooI8:129-158
Moloney CL, Nicolson SW (1984) Water relations and haemolymph composition of two intertidal
spiders (order Araneae). J Eltp BioI EcoI83:275-284
*Mommsen TP (1977) Zusammensetzung und Funktion von eJttraintestinaler Verdauungsfliissigkeit
und Gift einer Spinne (Tegenaria atrica CL Koch, Agelenidae). Thesis, Univ Freiburg
Mommsen TP (1978a) Digestive enzymes of a spider (Tegenaria atrica Koch) - I. General remarks,
digestion of proteins. Comp Biochem PhysioI60A:365-370
Mommsen TP (1978b) Digestive enzymes of a spider (Tegenaria atrica Koch). II. Carbohydrases.
Comp Biochem PhysioI60A:371-375
Mommsen TP (1978c) Digestive enzymes of a spider (Tegenaria atrica Koch). III. Esterases, phos-
phatases, nucleases. Comp Biochem PhysioI60A:377 -3 82
Mommsen TP (1978d) Comparison of digestive a-amylases from two species of spiders (Tegenaria
atrica and Cupiennius salei). J Comp Physiol127 :355 -361
Mommsen TP (1980) Chitinase and p-N-acetylglucosaminidase from the digestive fluid of the spider,
Cupiennius salei. Biochim Biophys Acta 612:361-372
Monteith JL, Campbell GS (1980) Diffusion of water vapour through integuments-potential con-
fusion. J Therm BioI 5 :7-9
*Monterosso B (1928) Su la biologia degli scitodidi et la ghiandola glutinifera di essi. Arch Zool Ital
2:63-122
Montgomery TH (1903) Studies on the habits of spiders, particularly those of the mating period.
Proc Phil Acad Nat Sci 55 :80-90
Moore SJ (1976) Some spider organs seen by the scanning electron microscope, with special re-
ference to the book-lung. Bull Br Arachnol Soc 3 :177 -187
Morgan PN, Felton WW (1965) Utilization of mammalian cell cultures in spider venom studies.
Bacteriol Proc 65th Annu Meet, pp 120-127
Morgans D, Carroll PR (1977) The responses of isolated human temporal artery to the venom of
the Sydney funnel-web spider (Atrax robustus). TOlticon 15 :277
Morse D (1981) Prey capture by the crab spider Misumena vatia (Clerck) (Thomisidae) on three
common native flowers. Am Midi Nat 105 :358-367
Moss WW (1960) Description and mating behaviour of Allothrombium lerowci, new species (Aca-
rina:Trombidiidae), a predator of small arthropods in Quebec apple orchards. Can Entomol92:
898-905
418 References
Mothes U, Seitz KA (1980) Light and electron microscopy of the functional morphology of Tetra-
nychus urticae (Acari, Tetranychidae). I. Excretory systems. Zoo I Jahrb Anat 104:500-529
Mothes U, Seitz KA (1981) Fine structure and function of the prosomal glands of the two-spotted
spider mite Tetranychus urticae (Acari:Tetranychidae). Cell Tissue Res 221 :339-350
*Moulder BC, Reichle DE (1972) Significance of spider predation in the energy dynamics of forest
floor arthropod communities. Ecol Monogr 42:473-498
Moulder BC, Reichle DE, Auerbach SI (1970) Significance of spider predation in the energy dynamics
of forest floor arthropod communities. Oak Ridge Natl Lab Rep ORNL-4452 :170
Muic N, Stanic M, Meniga A (1956) Beitrag zur Kenntnis des Spinnengiftes von Latrodectus trede-
cimguttatus Rossi. Z Physiol Chern 305 :70 -74
Mullen GR (1969) Morphology and histology of the silk glands in Araneus sericatus CI. Trans Am
Microsc Soc 88 :232-240
Millier HG (1983) Ein Mermithide als Parasitoid von Coelotes inermis (L Koch 1855) (Arachnida:
Araneae:Agelenidae). Entomol Z 93 :358-360
Miiller HJ (1970) Formen der Dormanz bei Insekten. Nova Acta Leopoldina NS 191 :1-27
Muma MH, Jeffers WF (1945) Studies of the spider prey of several mud-<iauber wasps. Ann Ento-
mol Soc Am 38:245-255
Muma MH, Murna KE (1949) Studies on a population of prairie spiders. Ecology 30:485-503
Murakami Y (1983) Factors determining the prey size of the orb-web spider,Argiope amoena (1.
Koch) (Argiopidae). Oecologia (Berlin) 57 :72-77
Murphy JA, Platnick NI (1981) On Liphistius desultor Schiodte (Araneae, Liphistiidae). Bull Am
Mus Nat Hist 170:46-56
Mutvei H (1977) SEM studies on arthropod exoskeletons. 2. Horseshoe crab Limulus polyphemus
(1.) in comparison with extinct eurypterids and recent scorpions. Zoo I Scr 6 :203-213
Nakamura K (1972) The ingestion in wolf spiders. II. The expression of degree of hunger and
amount of ingestion in relation to spider's hunger. Res Popul Ecol 14 :82-96
Nakamura K (1974) A model of the functional response of a predator to prey density involving the
hunger effect. Oecologia (Berlin) 16 :265 -278
*Nakamura K (1977) A model of the functional response of a predator to varying prey densities;
based on the feeding ecology of wolf spiders. Bull Natl Inst Agric Sci Ser C No 31 :29-89
Nakamura K (1982) Prey capture tactics of spiders: an analysis based on a simulation model for
spider's growth. Res Popul Ecol 24:302-317
Nawabi S (1974) Histologische Untersuchungen an der Mitteldarmdriise von Stegodyphus pacijicus
(Pocock 1900) (Araneae, Eresidae). Thesis, Univ Bonn
Neck RW (1977) Reddish coloration in a green spider: evolutionary origin and subsequent adapta-
tion. J Zoo I (London) 184:267-269
*Needham AE (1974) The significance of zoo chromes. Zoophysiology and ecology, vol 3. Springer,
Needham AE (1978) Insect biochromes: their chemistry and role. In: Rockstein M (ed) Biochem-
istry of insects. Academic Press, London New York, pp 233 -305
Nemenz H (1954) tiber den Wasserhaushalt einiger Spinnen, mit besonderer Beriicksichtigung der
Transpiration. Osterr Zool Z 5:123-158
Nemenz H (1955) tiber den Bau der Cuticula und dessen Einfl~ auf die Wasserabgabe bei Spinnen.
Osterr Akad Wiss Math-Nat KI Abt 1164:65-76
Nentwig W (1980) The selective prey of linyphiid-like spiders and their space webs. Oecologia
(Berlin) 45 :236 -243
Nentwig W (1981) Insekten, Spinnennetze und Netzspinnen. Marburger Entomol Publl (5):1-139.
Thesis, Univ Marburg
*Nentwig W (1982a) Why do only certain insects escape from a spider's web? Oecologia (Berlin) 53:
412-417
Nentwig W (1982b) Beutetieranalysen an cribellaten Spinnen (Araneae:Filistatidae, Dictynidae,
Eresidae). Entomol Mitt Zool Mus Hamburg 7 :233-244
Nentwig W (1982c) Epigeic spiders: their potential prey and competitors: relationship between size
and frequency. Oecologia (Berlin) 55 :130-136
*Nentwig W (1983) The prey of web-building spiders compared with feeding experiments (Araneae:
Araneidae, Linyphiidae, Pholcidae, Agelenidae). Oecologica (Berlin) 56:132-139
References 419
Nentwig W (1985a) Feeding ecology of the tropical spitting spider Scytodes longipes (Araneae,
Scytodidae). Oecologia (Berlin) 65 :284-288
*Nentwig W (1985b) Prey analysis of four species of tropical orb weaving spiders (Araneae:Ara-
neidae) and a comparison with araneids of the temperate zone. Oecologia (Berlin) 66:580-594
Nentwig W (1985c) Social spiders catch larger prey: a study of Anelosimus eximius (Araneae:Theri-
diidae). Behav Ecol Sociobiol17 :79 -85
Nentwig W (1985d) Spiders eat crickets artificially poisoned with KCN and change the composi-
tion of their digestive fluid. Naturwissenschaften 72 :545 -546
*Nentwig W (1985e) Obligate kleptoparasitic behaviour of female flies at spider webs (Diptera:
Empidoidea:Microphoridae). Zool Anz 215 :348-354
Nentwig W (1985f) A mimicry complex between mutillid wasps (Hymenoptera:Mutillidae) and
spiders (Araneae). Stud Neotrop Fauna Environ 20:113-116
*Nentwig W (1986a) Non-webbuilding spiders: Prey specialists or generalists? Oecologia (Berlin) 69:
571-576
Nentwig W (1986b) Stabilimente in Radnetzen von Argiope-Arten (Araneae:Araneidae): Ein funk-
tionsloses Relikt? Verh DZG (in press)
Nentwig W, Christenson TE (1986) Natural history of the nonsolitary sheetweaving spider Anelos·
imus jucundus (Araneae:Theridiidae). Zoo I J Linn Soc 87:27 -35
Nentwig W, Heimer S (1983) Orb webs and single-line webs: An econo.mic consequence of space
web reduction in spiders. Z Zool Syst Evolutionsforsch 21 :26-37
*Nentwig W, Wissel C (1986) A comparison of prey lengths among spiders. Oecologia (Berlin) 68:
595-600
Neumann H (1977) Charakterisierung einiger Enzyme aus dem extraintestinalen Verdauungssaft
der Spinne Cupiennius salei Keyserling. Thesis, Univ Freiburg
Neville AC (1970) Cuticle ultrastructure in relation to the whole insect. Symp R Entomol Soc
London 5 :17 -39
*Neville AC (1975) Biology of the arthropod cuticle. Springer, Berlin Heidelberg New York
Neville AC, Luke BM (1969) A two-system model for chitin-protein complexes in insect cuticles.
Tissue Cell 1:689-707
Neville AC, Thomas MG, Zelazny B (1969) Pore canal shape related to the molecular architecture
of arthropod cuticle. Tissue CellI: 183 -200
Nicolai J (1973) Vogelleben. Belser, Stuttgart
Nicolson SW, Louw GN, Edney EB (1984) Use of a ventilated capsule and tritiated water to measure
evaporative water losses in a tenebrionid beetle. J Exp Bioi 108:477 -481
Niemann A (1957) Die Bedeutung einer Schneedecke flit die bedeckten Pflanzen, flit den Boden
und flit die schneenahe Luftschicht. Dtsch Gartenbauwirtsch 5 :168-171
Nitzsche R (1981) Beutefang und "Brautgeschenk" bei der Raubspinne Pisaura mirabilis (Cl.)
(Araneae:Pisauridae). Dipl Thesis, Univ Kaiserslautern
*Noirot Ch, Quennedey A (1974) Fine structure of insect epidermal glands. Annu Rev Entomol
19:61-80
Nq,rgaard E (1941) On the biology of Eresus niger Pet. (Aran.). Entomol Med 22:150-179
Nq,rgaard E (1951) On the ecology of two lycosid spiders (Pirata piraticus and Lycosa pullata)
from a Danish sphagnum bog. Oikos 3:1-21
Nq,rgaard E (1956) Environment and behaviour of Theridion saxatile. Oikos 7 :159-192
*Novak VJA (ed) (1975) Insect hormones. Chapman and Hall, London
Nyffeler M (1980) Juvenile Kreuzspinne als Nahrungsdieb im Netz einer adulten Kreuzspinne
Araneus quadratus. Mitt Entomol Ges Basel 31 :90
Nyffeler M (1982) Field studies on the ecological role of the spiders as insect predators in agroeco-
systems (abandonned grassland, meadows, and cereal fields). Thesis, Swiss Fed Inst Technol,
ZUrich
Nyffeler M, Benz G (1978) Die Beutespektren der Netzspinnen Argiope bruennichi (Scop.), Araneus
quadratus CI. und Agelena labyrinthica (Cl.) in {)dlandwiesen bei ZUrich. Rev Suisse Zool 85:
747-757
Nyffeler M, Benz G (1979) Jahreszeitliches und riiumliches Verteilungsmuster sowie Nahrungsoko-
logie der dominanten epigiiischen Spinnen von Winterweizenfeldern (Bodenfallenanalysen und
Freilandbeobachtungen). Mitt Schweiz Entomol Ges 52 :444-445
420 References
Nyffeler M, Benz G (1980) Kleptoparasitismus von juvenilen Kreuzspinnen und Skorpionsfliegen

in den Netzen adulter Spinnen. Rev Suisse Zoo187:907 -918
Nyffeler M, Benz G (1981) Einige Beobachtungen zur Nahrungsokologie der Wolfspinne Pardo sa
lugubris (Walck.) (Araneae, Lycosidae). Dtsch Entomol Z NS 28 :297 -300
Odell GV, Chan TK, Geren CR, Lee CK, Ward BC, Howell DE (1975) Chemical and biological
properties of tarantula and brown recluse venom toxins. Fed Proc 34 :504 -510
Oden NL (1981) The mating behavior of the spider Phidippus audax. Thesis, Michigan Univ, Ann
Arbor
Olive CW (1981a) Optimal phenology and body-size of orb-weaving spiders: foraging constraints.
Oecologia (Berlin) 49 :83 -87
Olive CW (1981 b) Co-adapted foraging traits in a guild of orb-weaving spiders. Oecologia (Berlin)
49:88-91
Oliveira P, Sazima I (1984) The adaptive bases of ant-mimicry in a neotropical aphantochilid spider
(Araneae:Aphantochilidae). Bioi J Linn Soc 22:145-·155
Open BD (1979) Revision of the genera and tropical American species of the spider family Ulobori-
dae. Bull Mus Comp ZooI148:443-549
Opell BD (1982) Cribellum, calamistrum and ventral comb ontogeny in Hyptiotes cavatus (Hentz)
(Araneae:Uloboridae). Bull Br Arachnol Soc 5 :338-343
Opel! BD (1984) A simple method for measuring dessication resistance of spider egg sacs. J Arachnol
12:245-247
Ori M (1975) Studies on the poisonous spider Chiracanthium japonicum as a pest of medical im-
portance.1. Envenomation by the spider. Jpn J San ZooI26:215-219
Otto C, Svensson BS (1982) Structure of communities of ground-living spiders along altitudinal
gradients. Holarct Ecol5 :35 --47
Oxford GS (1976) The color polymorphism in Enoplognatha ovatum (Clerck) (Araneae :Theridi-
idae) - temporal stability and spatial variability. Heredity 36 :369-381
Oxford GS (1983) Genetics of color and its regulation during development in the spider Enoplog-
natha ovata (Clerck) (Araneae :Theridiidae). Heredity 51 :621-634
Packard AS (1905) Change of color and protective coloration in a flower-spider (Misumena vatia
Thorell). J NY Entomol Soc 13 :85 -96
Padmanabhanaidu B (1962) Ionic composition of the blood of scorpion. Curr Sci 31 :21
Padmanabhanaidu B (1966) Ionic composition of the blood and blood volume of the scorpion
Heterometrusfulvipes. Comp Biochem PhysioI17:157-166
Palanichamy S (1985) Effect of temperature on food utilization, growth and egg production in the
spider Cyrtophora cicatrosa. J Therm Bioi 10:63-70
Palm en E, Soumalainen H (1945) Experimentel!e Untersuchungen tiber die Transpiration bei eini-
gen Arthropoden, insbesondere Kiifern. Ann Zool Soc Zoo I Bot Fenn Vanamo 11 :1-50
*Palmer JM (1985) The silk and silk production system of the funnel-web mygalomorph spider
Euagrus (Araneae, Dipluridae). J MorphoI186:195-207
Palmer JM, Coyle FA, Harrison FW (1982) Structure and cytochemistry of silk glands of the myga-
lomorph spider Antrodiaetus unicolor (Araneae, Antrodiaetidae). J MorphoI174:269-274
Palmgren P (1939) Okologische und physiologische Untersuchungen tiber die Spinne Dolomedes
fimbriatus (CL.). Acta Zool Fenn 24:1-42
Pansa MC, Migliori Natalizi G, Bettini S (1972) 5-Hydroxytryptamine content of Latrodectus
mactans tredecimguttatus venom from gland extracts. Toxicon 10:85-86
Parker GA (1984) The producer/scrounger model and its relevance to sexuality. In: Barnard CJ
(ed) Producers and scroungers. Chrom Helm, Beckenham, pp 127 -153
Parker JR, Roberts MT (1974) Internal and external parasites of the spider Pardosa nortensis
(Thorell) (Araneae :Lycosidae). Bull Br Arachnol Soc 3 :82-84
Parry DA (1951) Factors determining the temperature of terrestrial arthropods in sunlight. J Exp
Bioi 28 :445 -462
Parry DA (1954) On the drinking of soil capillary water by spiders. J Exp Bioi 31 :218-227
Parry G (1953) Osmotic and ionic regulation in the isopod crustacean Ligia oceanica. J Exp Bioi
30:567-574
Paul R (1986) Gas exchange and gas transport in the tarantula Eurypelma californicum - an over-
view. In: Linzen B (ed) Invertebrate oxygen carriers. Springer, Berlin Heidelberg New York,
pp 321-326
References 421
Paul R, Fincke T, Linzen B (in press) Respiration in the tarantula Eurypelma cali[ornicum: evidence
for diffusion lungs. J Comp Physiol B
Pauling L, Corey RB (1953) Two rippled-sheet configurations of polypeptide chains, and a note
about the pleated sheets. Proc Natl Acad Sci USA 39:253-256
Peakall DB (1964) Effects of cholinergic and anticholinergic drugs on the synthesis of silk fibroins
of spiders. CompBiochem PhysioI12:465-470
Peakall DB (1965) Regulation of the synthesis of silk fibroins of spiders at the glandular level. Comp
Biochem Physiol 15 :509 -515
Peakall DB (1966) Regulation of protein production in the silk glands of spiders. Comp Biochem
PhysioI19:253-258
·Peakall DB (1968) The silk glands. In: Witt PN, Reed CF, Peakall DB (eds) A spider's web: Pro-
blems in regulatory biology. Springer, Berlin Heidelberg New York, pp 5-28
Peakall DB (1969) Synthesis of silk, mechanism and location. Am Zool9 :71-79
·Peakall DB (1971) Conservation of web proteins in the spider, Araneus diadematus. J Exp Zool
176:257-264
Peakall DB, Witt PN (1976) The energy budget of an orb web-building spider. Comp Biochem
Physio154A:257 -264
Pearse AGE (1968) Histochemistry. Churchill, London
Peck WB, Whitcomb W (1968) Feeding spiders an artificial diet. Entomol News 79:233-236
Peckham GW, Peckham EG (1892) Ant-like spiders of the fam. Attidae. Occas Pap Nat Hist Soc
Wis 2:1-84
Pereira Lime FA, Schenberg S (1964) Caracteristicas e separa~o dos componentes do veneno de
Phoneutria [era. Cienc Cult (Sao Paulo) 16:187-192
Perret BA (1977) Proteolytic activity of tarantula venoms due to contamination with saliva. Toxi-
con 15 :505 -510
Perry SF (1978) Quantitative anatomy of the lungs of the red-eared turtle, Pseudemys scripta
elegans. Respir PhysioI35:245-262
Perry SF (1983) Reptilian lungs. Functional anatomy and evolution. Adv Anat Embryol Cell Bioi
79:1-81
·Peters HM (1955) Uber den Spinnapparat von Nephila madagascariensis (Radnetzspinnen, Fam.
Argiopidae). Z Naturforsch 103 :395 --404
Peters HM (1983) Struktur und Herstellung der Fangfiiden cribellater Spinnen (Arachnida:Araneae).
Verh Naturwiss Ver Hamburg NS 26:241-253
·Peters HM (1984) The spinning apparatus of Uloboridae in relation to the structure and construc-
tion of capture threads (Arachnida, Araneida). Zoomorphology 104:96-104
Peters HM, Kovoor J (1980) Un complement a l'appareil sericigene des Uloboridae (Araneae): Le
paracribellum et ses glandes. Zoomorphology 96:91-102
Petrunkevitch A (1922) The circulatory system and segmentation in Arachnida. J Morphol 36:
157 -189
·Petrunkevitch A (1933) An inquiry into the natural classification of spiders, based on a study of
their internal anatomy. Trans Conn Acad Arts Sci 31 :299-389
*Pianka ER (1966) Latitudinal gradients in species diversity, a review of concepts. Am Nat 100:33 -46
Pianka ER (1970) On r- and K-selection. Am Nat 104:592-597
Pichka VE (1977) On insects parasitizing in spiders, their cocoons and nests. Vestn Zool3 :92-93
Pickford GE (1942) Studies on the digestive enzymes of spiders. Trans Conn Acad Arts Sci 35:33-72
Piek TA, Simon-Thomas RT (1969) Paralysing venoms of solitary wasps. Comp Biochem Physiol
30:13-31
Piek TA, Buitenhuis A, Veldsema-Currie RD, Mantel P (1983) Smooth muscle contracting com-
pounds in venoms of sphecid wasps (Hymenoptera:Sphecidae). Comp Biochem Physiol C Comp
Pharmacol 75 :153-160
*Pielou EC (1975) Ecological diversity. Wiley, New York
*Pielou EC (1979) Biogeography. Wiley, New York
Pinkston KN, Frick JH (1973) Determinations of osmolarity on the fresh unfixed hemolymph of
four species of spiders. Am Entomol Soc Am 66:696 -6 97
Pizzi T (1975) Estudio histopatologico del aracnidismo necrotico por Loxosceles laera. Bol Chi!
Parasito130:34-39
422 References
Plateau F (1887) De l'absence de mouvements respiratoires perceptibles chez les arachnides. Arch
Bioi 7:331-348
Platnick NI (1971) The evolution of courtship behaviour in spiders. Bull Br Arachnol Soc 2:40-47
Platnick NI (1976a) Drifting spiders or continents? Syst ZooI25:101-109
Platnick NI (1976b) Concepts of dispersal in historical biogeography. Syst Zool 25 :294-295
Platnick NI (1977) Notes on the spiders from the Falkland Islands (Arachnida, Araneae). J Arach-
noI3:195-198
Platnick NI, Sedgwick WC (1984) A revision of the spider genus Liphistius (Araneae, Mesothelae).
Am Mus Nov 2781 :1-31
Platnick NI, Shadab MU (1978) A review of the spider genus Mysmenopsis. Am Mus Nov 2661:
1-22
*Plomley NJB (1947) Some notes on the biology of the Cyrtidae (Diptera) with speciaireference to
the genus Oncodes. Rec Queen Victoria Mus Lannceston 11:23-30
Pocock RI (1909) Mimicry in spiders. J Linn Soc 30:256-270
*Poinar GO Jr (1983) The natural history of Nematodes. Prentice Hall, Englewood Cliffs, 323 pp
Poinar GO Jr (1985) Mermithid (Nematoda) parasites of spiders and harvestmen. J Arachnol 13:
121-128
Poinar GO Jr, Benton CLB Jr (1986) Aranimermis aptispicula n. gen., n.sp. (Mermithidae:Nema-
toda), a parasite of spiders (Arachnida:Araneida). Syst ParasitoI8:33-38
Poinar GO Jr, Thomas GM (1985) Laboratory infection of spiders and harvestmen (Arachnida:
Araneae and Opiliones) with Neoaplectana and Heterorhabditis nematodes (Rhabditoidea). J
Arachnology 13:297-302
Poinar GO Jr, Welch HE (1981) Parasites of invertebrates in the terrestrial environment. In: Slusarski
W (ed) Rev Adv Parasitol. Pol Sci Publ, Warsaw, pp 947-954
Poinar GO Jr, Lane RS, Thomas GM (1976) Biology and redescription of Pheromermis pachysoma
a parasite of yellowjackets. Nematologia 22:360-370
Pointing PJ (1965) Some factors influencing the orientation of the spider Frontinella communis
(Hentz.) in its web (Araneae:Linyphiidae). Can EntomoI97:69-78
Polenec A (1962) Arahnidska fauna in asociacije pozirni. Bioi Vestn 10:71-83
Pollard SD (1984) Egg guarding by Clubiona cambridgei (Araneae, Clubionidae) against conspecific
predators. J Arachnolll:323-326
Pollard SD (in press) Prey capture in Dysdera crocata (Araneae:Dysderidae), a long fanged spider.
NZJ Zool
Pollard SD, Jackson RR (1982) The biology of Clubiona cambridgei (Araneae, Clubionidae): intra-
specific interactions. NZ J EcoI5:44-50
Pollard SD, Jackson RR (1984) Interpopulation variation in the life history of Clubiona cambridgei
(L. Koch) (Araneae: Clubionidae), a New Zealand hunting spider. NZ J Zoolll :413-416
*Porcheron P (1979) L'hormone de mue des arthropodes: dosage radioimmunologique, production,
divers aspects de son role physiologique. Thesis, Univ Pierre et Marie Curie, Paris
Porter S, Eastmond D (1982) Euryopis coki (Theridiidae), a spider that preys onPogonomyrmex
ants. J ArachnoI1O:275-277
Potzsch J (1966) Notizen zur Erniihrung und Lebensweise von Meta menardi Latr. (Araneae, Ara-
neidae). Abh Naturkd Mus Goerlitz 41 :1-23
Pound JM, Oliver JH (1979) Juvenile hormone: evidence its role in the reproduction of ticks.
Science 206 :355 -357
Powell JA (1967) Behavior of ground nesting wasps of the genus Nitelopterus, particularly N. cali-
fornicus (Hymenoptera:Sphecidae). J Kans Entomol Soc 40:331-346)
Praet H van, Kindt C (1979) Influence de la temperature sur Ie developpement embryonnaire
d'Erigone atra (BLACKWALL) et d'Oedothorax fuscus (BLACKWALL) (Araneida:Linyphi-
idae). BiolJaarb Dodonaea 47:107 -116
*Pratt GE, Brooks GT (ed) (1981) Juvenile hormone biochemistry. Action, agonism and antagonism.
Elsevier/North-Holland, Amsterdam New York
Precht H, Christophersen J, Hensel H, Larcher W (eds) (1973) Temperature and life. Springer, Berlin
Heidelberg New York
Prestwich KN (1977) The energetics of web-building in spiders. Comp Biochem Physiol57 A:321-
326
Prestwich KN (1982) The physiological and biochemical constraints on activity in spiders. Thesis,
Univ Florida
References 423
*Prestwich KN (1983a) Anaerobic metabolism in spiders. Physiol Zool56 :112-121

*Prestwich KN (1983b) The roles of aerobic and anaerobic metabolism in active spiders. Physiol
Zoo I56 :122-132
Price P (1977) General concepts on the evolutionary biology of parasites. Evolution 31 :405-420
Pritchard G, Scholefield P (1978) Observations of the food, feeding behaviour and associated sense
organs of Grylloblatta campodeiformis (Grylloblattodea). Can EntomolllO:205-212
Prosser CL (ed) (1973) Comparative animal physiology. 3rd edn. Saunders, Philadelphia London
Toronto
Puiz R (1985a) Diskontinuierliche Transpiration einer Vogeispinne, Eurypelma sp., bei einer kon-
stanten Umgebungstemperatur von 40 0 C (Araneae, Theraphosidae). Mitt Dtsch Ges Allg Angew
EntomoI4:358-361
Puiz R (1985b) Temperature-induced secretion in mygalomorph spiders. Verh Dtsch Zool Ges 78:
324 (in German)
Puiz R (1986a) Der Temperatursinn von Vogeispinnen. Thesis, Univ Kiel
Pulz R (1986b) Temperature-related behaviour and temperature perception in spiders: review of
present knowledge. In: Nachtigall W (Ser ed) Laudien H (ed) Thermal relations in animals and
man. Biona Rep 4. Akad Wiss Mainz. Fischer, Stuttgart, pp 41-48
Puiz R (in press) Patterns of evaporative water loss in tarantulas (Araneae, Theraphosidae): trans-
piration and secretion. Proc 9th Int Congr Arachnol, Panama 1983
Puiz R, Meeder T, Sommerschuh M (1984) Temperature perception in mygalomorph spiders:
responses to local thermal stimulation and to changing ambient temperatures. Verh Dtsch Zool
Ges 77 :320 (in German)
Puiz R, Schuiz R (1986) Phormictopus sp. (Araneae, Theraphosidae) in its natural habitat in Haiti.
In: Barrientos JA (ed) Actas X. Congreso Int Aracnologia, Jaca (Espana), vol I, p 282
Puntscher S (1979) Verteilung und Jahresrhythmik von Spinnen im zentralalpinen Hochgebirge
(Obergurgl, Otztaler Alpen). Thesis, Univ Innsbruck
Punzo F (1982) Hemolymph chemistry of lycosid spiders. Comp Biochem Physiol 7lB:703-707
*Punzo F (1983) Hemolymph chemistry of the spiders, Heteropoda venatoria (Sparassidae), Pisau-
rina mira (Pisauridae) and Amaurobius bennetti (Amaurobiidae). Comp Biochem l'hysiol A 75:
647-652
Punzo F, Jellies J (1983) Comparative waterrelations of araneid and lycosid spiderlings (Arachnida).
Comp Biochem Physiol A 74:981-985
Purcell WF (1909) Development and origin of the respiratory organs in Araneae. Q J Microsc Sci
54:1-110
Pyke GH (1984) Optimal foraging theory: a critical review. Annu Rev Ecol Syst 15 :523-575
Quinlan MC, Hadley NF (1982) A new system for concurrent measurement of respiration and
water loss in arthropods. J Exp ZooI222:255-263
Rabaud E (1923) Recherches sur la variation chromatique et l'homochromie des arthropodes ter-
restres. Bull Bioi Fr Belg 57: 1-59
Rakotovao LH, Rahandraha T (1973) Variations de la teneur en eau et du taux lipidique en fone-
tion de l'etat physiologique de Nephiltz matiagascariensis femelle adults. CR Acad Sci Ser D 277:
1517-1520
Ramousse R (1973) Body, web-building and feeding characteristics of males of the spider Araneus
diadematus (Araneae:Araneidae). Psyche 80:22-47
Ramousse R (1980a) Organisation temporelle du comportement constructeur chez huit especes
d 'Argiopidees. CR 5th Coli Arachnol, Barcelona, pp 203 -212
Ramousse R (1980b) Temporal patterns of web-building in Araneus diadematus Clerck. Proc 8th
Int Congr Arachnol, Vienna, pp 257 -260
Ramousse R, Davis F (1976) Web-building time in a spider: Preliminary application of ultrasonic
detection. Physiol Behav 17 :997-1000
Ramousse R, Le Guelte L (1979) Relations spatio-temporelles dans Ie comportement constructeur
chez l'Epeire diademe. Rev ArachnoI2:183-192
Ramousse R, Le Guelte L (1984) Strategies de construction de la toile chez deux especes d'araig-
nees (Araneus diadematus et Zygiella x-notata). Rev Arachnol 5 :255 - 265
Ramousse R, Wurdak E (1984) Biologie et comportement de la larve et de la premiere nymphe de
l'araignee Araneus suspicax (0. Pickard-<::ambridge). Rev Arachnol 5 :275 -286
424 References
Ramousse R, Le Guelte L, Le Berre M (1981) Organisation temporelle du comportement con-

structeur chez les Argiopidae. CR VIeme Coil Arachnol express franc (Modena-Pisa 1981). Atti
Soc Toscana Sci Nat Pisa Mem B SuppI88:159-172
Rathmayer W (1965) Neuromuscular transmission in a spider and the effect of calcium. Comp
Biochem PhysioI14:673-687
Ratte HT (1985) Temperature and insect development. In: Hoffmann KH (ed) Environmental
physiology and biochemistry of insects. Springer, Berlin Heidelberg New York, pp 33-66
Rau P (1928) The nesting habits of the wasp, Chalybion caeruleum. Ann Entomol Soc Am 21:
25-35
Rau P (1935a) The spider prey of the mud wasp Sceliphron caementarium (Araneae, Hymen:
Sphecidae). Entomol News 46 :267 -270
Rau P (1935b) The wasp, Chalybion cyaneum Fab., preys upon the black widow spider, Latro-
dectus mactans Fab. (Hymen., Araneae). Entomol News 46 :259-260
Rau P (1942) The sleep of Chalybion zimmermani. Can Entomo174:196
Rau P (1944) The prey and hunting habits of the wasp, Trypoxylon politum Say. Entomol News
55:9-10
*Raven RJ (1985) The spider infraorder Mygalomorphae (Araneae): cladistics and systematics.
Bull Am Mus Nat Hist 182:1-180
Ray-Lankester E (1905) The structure and classification of the Arachnida. Q J Microsc Sci V:XLVIII
Redmond JR (1968a) The respiratory function of hemocyanin. In: Ghiretti F (ed) Physiology and
biochemistry of haemocyanins. Academic Press, London New York, pp 5-21
Redmond JR (1968b) Transport of oxygen by the blood of the land crab, Geocarcinus lateralis.
Am Zool 8:471-479
Reed CF, Witt PN (1972) Growth rate and longevity in two species of orb-weaving spiders (Ara-
neae:Argiopid"ae). Bull Br Arachno1 Soc 2:111-112
Rekow MA, Civello DH, Geren CR (1983) Enzymatic and hemolytic properties of brown recluse
spider (Loxosceles reelusa) toxin and extracts of venom apparatus, cephalothorax and abdomen.
Toxicon 21 :441-444
Remane A, Storch V, Welsch U (1978) Kurzes Lehrbuch der Zoologie, 3rd edn. Fischer, Stuttgart
Remold H (1962) tiber die biologische Bedeutung der Duftdriisen bei den Landwilnzen (Geocorisae).
Z Vergl Physiol45 :636-694
Reukauf R (1931) Zur Biologie von Hyptiotes paradoxus. Z Morph tikol Tiere 21:691-701
Richter CJJ (1970) Aerial dispersal in relation to habitat in eight wolf spider species (Pardosa,
Araneae, Lycosidae). Oecologia 5 :200-214
Riddle WA (1981) Cold survival of Argiope aurantia spiderlings (Araneae, Araneidae). J Arachnol
9:343-345
Riddle WA (1985) Hemolymph osmoregUlation in several myriapods and arachnids. Comp Bio-
chern PhysiolA 80:313-323
Riddle WA, Markezich AL (1981) Thermal relations of respiration in the garden spider, Argiope
aurantia, during early development and overwintering. Comp Biochem Physiol A Comp Physiol
69:759-766
Riddle WA, Pugach S (1976) Cold hardiness in the scorpion,Paruroctonusaquilonalis. Cryobiology
12:248-253
*Riechert SE (1976) Web-site selection in the desert spider Agelenopsis aperta. Oikos 27 :311-315
Riechert SE (1982) Spider interaction strategies: communication vs coercion. In: Witt PN, Rovner
JS (eds) Spider communication. Princeton Univ Press, Princeton, pp 281-316
Riechert SE (1985) Decisions in multiple goal contexts: habitat selection of the spider, Agelenopsis
aperta (GERTSCH). Z Tierpsychol 70:53-69
Riechert SE, Cady AB (1983) Patterns of resource use and tests for competitive release in a spider
community. Ecology 64:899-913
Riechert SE, Gillespie RG (1986) Habitat choice and utilisation in web-building spiders. In: Shear
WA (ed) Spiders: webs, behavior and evolution. Stanford Univ Press, Stanford CA, pp 23-49
*Riechert SE, Luczak J (1982) Spider foraging: behavioural responses to prey. In: Witt PN, Rovner
JS (eds) Spider communication. Princeton Univ Press, Princeton, pp 353-386
*Riechert SE, Tracy CR (1975) Thermal balance and prey availability: bases for a model relating
web site characteristics to spider reproductive success. Ecology 56 :265 -284
Ring RA, Tesar D (1980) Cold-hardiness of the arctic beetie,Pyrho american us KIRBY Coeloptera,
Pythidae (Salpingidae). J Insect Physiol26 :763-774
References 425
Roberts NL (1936) Colour change in the leaf-curling spider (Araneus wagneri). Proc R Zoo 1 Soc
NSW 28-29
Robertson HG, Nicolson SW, Louw GN (1982) Osmoregulation and temperature effects on water
loss and oxygen consumption in two species of African scorpion. Comp Biochem Physiol A 71:
605-609
Robinson BC, Robinson MH (1974) The biology of some Argiope species from New Guinea: pre-
datory behaviour and stabilimentum construction (Araneae:Araneidae). Zool J Linn Soc 54:
145-159
Robinson BC, Robinson MH (1978) Developmental studies of Argiope argentata (Fabricius) and
Argiope aemula (Walckenaer). Symp Zool Soc London 42:31-40
Robinson GL, Paim U (1969) Regulation of external respiration by the book-lung spiracles of the
spiders, Araneus diadematus CLERCK and A. marmoreusCLERCK. Can J Zool 47 :355 -364
Robinson MH (1975) The evolution of predatory behavior in araneid spiders. In: Baerends G, Beer
C, Manning A (eds) Function and evolution in behavior. Claredon Press, Oxford
Robinson MH (1978) Symbioses between insects and spiders: an association between lepidopteran
larvae and the social spider Anelosimus eximius (Araneae:Theridiidae). Psyche 84:225-232
Robinson MH (1980) The ecology and behaviour of tropical spiders. Proc 8th Int Arachnol Congr,
Vienna, pp 13-32
*Robinson MH (1982a) The ecology and biogeography of spiders in Papua New Guinea. Monogr
Bioi 42:557 -581
*Robinson MH (1982b) Courtship and mating behavior in spiders. Annu Rev EntomoI27:1-20
Robinson MH (1983) Neotropical arachnology: Historic, ecological and evolutionary aspects.
IXth Lat Am Congr Zool-Inf Fin IX CI Peru, pp 89-100
Robinson MH, Mirick H (1971) The predatory behaviour of some araneid spiders and the origin of
immobilization wrapping. Psyche 76:487 -501
*Robinson MH, Olazarri J (1971) Units of behavior and complex sequences in the predatory behavior
of Argiope argentata (Fabricius): (Araneae:Araneidae). Smithson Contrib Zool 65: 1-36
*Robinson MH, Robinson BC (1970a) Prey caught by a sample population of the spider Argiope
argentata (Araneae:Araneidae) in Panama: a year's census data. Zool J Linn Soc 49:345-358
Robinson MH, Robinson BC (1970b) The stabilimentum of the orb web spider, Argiope argentata:
an improbable defence against predators. Can EntomoI102:641-655
Robinson MH, Robinson BC (1971) The predatory behavior of the ogre-faced spider Dinopis longi·
pes F. Cambridge (Araneae:Dinopidae). Am MidI Nat 85:85-96
Robinson MH, Robinson BC (1972) The structure, possible function and origin of the remarkable
ladder-web built by a New Guinea orb-web spider (Araneae:Araneidae). J Nat Hist 6:687-694
*Robinson MH, Robinson BC (1973a) Ecology and behavior of the giant wood spider Nephila macu·
lata (Fabricius) in New Guinea. Smithson Contrib ZooI149:1-76
Robinson MH, Robinson BC (1973b) The stabilimenta of Nephila clavipes and the origins of stabil-
imenturn-building in araneids. Psyche 80:277 -288
Robinson MH, Robinson BC (1974) Adaptive complexity: the thermoregulatory postures of the
golden-web spider Nephila clavipes at low altitudes. Am MidI Nat 92:386-396
Robinson MH, Robinson BC (1975) Evolution beyond the orb web: the web of the araneid spider
Pasilobus sp., its structure, operation and construction. Zool J Linn Soc 56:301-314
Robinson MH, Robinson BC (1976) The ecology and behavior of Nephila maculata: a supplement.
Smithson Contrib ZooI218:1-22
Robinson MH, Robinson BC (1977) Associations between flies and spiders: bibiocommensalism
and dipsoparasitism. Psyche 84:150-157
Robinson MH, Robinson BC (1978) Thermoregulation in orb-web spiders: new descriptions of
thermoregulatory postures and experiments on the effects of posture and coloration. Zool J Linn
Soc 64 :87 -102
Robinson MH, Valerio C (1977) Attacks on large or heavily defended prey by tropical salticid spi-
ders. Psyche 84:1-10
Robinson MH, Mirick H, Turner 0 (1969) The predatory behavior of some araneid spiders and the
origin of immobilization wrapping. Psyche 76 :487 --50 1
Robinson MH, Robinson BC, Graney W (1971) The predatory behaviour of the nocturnal orb web
spider Eriophora [uliginea (C.L. Koch) (Araneae :Araneidae). Rev Peru Entomol 14 :304-315
426 References
Rockstein M (1974) Physiology of insecta, vol V. Academic Press, London New York
Roesel AI (1761) Insectenbelustigung, part IV
Romer F, Gnatzy W (1981) Arachnid oenocytes: ecdysone synthesis in the legs of harvestmen
(Opilionidae). Cell Tissue Res 216 :449-453
Rosenberg 1, Seifert G (1977) The coxal glands of Geophilomorpha (Chilopoda): organs of osmo-
regulation. Cell Tissue Res 182:247-251
Ross K, Smith RL (1979) Aspects of the courtship behavior of the black widow spider, Latrodec-
tus hesperus (Araneae:Theridiidae), with evidence for the existence of a contact sex pheromone.
1 Arachnol 7:69-77
Roth V, Roth B (1986) On the size of spiders' eggs. Newsl Br arachnol Soc 45:5
Rovner IS (1966) Courtship in spiders without prior sperm induction. Science 152:543-544
Rovner IS (1967) Copulation and sperm induction by normal and palpless male linyphiid spiders.
Science 157 :835
Rovner IS (1976) Detritus stabilimenta on the webs of Cyclosa turbinata (Araneae, Araneidae). 1
Arachnol 4 :215 -216
Rovner JS, Knost SJ (1974) Post-immobilization wrapping of prey by lycosid spiders of the herba-
ceous stratum. Psyche 81 :398-415
Rubzov IA (1977) New species ofmermithids from spiders and earwigs. In: Fauna of Siberia. Nauka,
Novosibirsk, pp 16-22 (in Russian)
Rubzov IA (1978) Mermithids: Classification, importance and use. Nauka, Leningrad, 207 pp (in
Russian)
Rubzov IA (1979) New species of mermithids from arthropods in Azerbaidzhan. Izv Akad Nauk
Az SSR Ser Bioi Nauk 6:89-95 (in Russian)
Rubzov IA (1980) New species of mermithids from arthropods. In: Helminthes of insects. Nauka,
Moscow, pp 95 -102 (in Russian)
*Rudall KM, Kenchington W (1971) Arthropod silks: the problem of fibrous proteins in animal tissues.
Annu Rev Entomol16 :73-97
Riidiger W (1970) Gallenfarbstoffe bei wirbellosen Tieren. Naturwissenschaften 57:331-337
Rumney GR (1969) Climatology; and the worlds climates. Collier-Macmillan, Canada
Russell FE, Buess FW (1970) Gel electrophoresis: a tool in systematics. Studies with Latrodectus
mactans venom. Toxicon 8:81-84
Russell FE, Gertsch WJ (1982) Last word on araneism. Am ArachnoI25:7-10
Russell FE, Wainschel 1, Gertsch WJ (1973) Bites of spiders and others arthropods. In: Conn HF
(ed) Current therapy. Saunders, Philadelphia, London, Toronto, pp 868-870
Rypstra AL (1979) Foraging flocks of spiders. Behav Ecol Sociobiol5 :291-300
Rypstra AL (1981) The effect of kleptoparasitism on prey consumption and web relocation in a
Peruvian population of the spider Nephila clavipes. Oikos 37 :179-182
Rypstra AL (1983) The importance of food and space in limiting web-spider densities; a test using
field enclosures. Oecologia 59:312-316
Rypstra AL (1984) A relative measure of predation on web-spiders in temperate and tropical forests.
Oikos 43:129-132
Saaristo MI (1978) Spiders (Arachnida, Araneae) from Seychelles Islands, with notes on taxonomy.
Ann Zool Fenn 15 :99-126
Sabath LE (1969) Color change and life history observations in the spider Gea heptagon (Araneae:
Araneidae). Psyche 76:367-374
Salt RW (1950) Time as a factor in the freezing of undercooled insects. Can J Res Sect D 28:285-
291
*Salt RW (1961) Principles of insect cold-hardiness. Annu Rev Entomol6 :55 -74
Salt RW (1963) Delayed inoculative freezing of insects. Can Entomol 95: 1190-1202
Salt RW (1968) Location and quantitative aspects of ice nucleators in insects. Can 1 Zool46 :329-
333
*Sampayo RRL (1942) Latrodectus mactans y latrodectismo. EI Ateneo, Buenos Aires
Sannasi A, Subramonian T (1972) Hormonal rupture of larval diapause in tick Rhipicephalus san-
guineus (Latr.). Experientia 28:666-667
Savory TH (1930) Environmental differences of spiders of the genus Zilla. 1 EcoI18:384-385
Savory T (1977) Arachnida. 2nd edn. Academic Press, London New York
References 427
*Schaefer M (1974) Experimentelle Untersuchungen zur Bedeutung der interspezifischen Konkur-

renz bei 3 Wolfspinnen-Arten (Araneida:Lycosidae) einer Salzwiese. Zool Jahrb Syst 101:
213-235
*Schaefer M (1976a) Experimentelle Untersuchungen zum Jahreszyklus und zur Oberwinterung von
Spinnen (Araneida). Zoo I Jahrb Abt Syst Okol Geogr Tiere 103 :127 -289
Schaefer M (1976b) An analysis of diapause and resistance in the egg stage of Floronia bucculenta
(Araneida:Linyphiidae). Oecologia (Berlin) 25 :155-174
Schaefer M (1976c) Zur Steuerung der Jahresrhythmik bei Spinnen (Arachnida:Araneae). Entomol
Germ 3:125-129
Schaefer M (1976d) Wie ist das Eistadium der Deckennetzspinne Floronia bucculenta an Extrem-
faktoren des Winters angepl£t? Verh Dtsch Zoo I Ges p 226
*Schaefer M (1977a) Winter ecology of spiders (Araneida). Z Angew EntomoI83:113-134
Schaefer M (1977b) Zur Bedeutung des Wint~rs fUr die Populationsdynamik von vier Spinnenarten
(Araneida). Zool Anz 199:77 -88
Schaefer M (1977c) Untersuchungen Uber das Wachstum von zwei Spinnenarten (Araneida) im La-
bor und Freiland. Pedobiologia 17:189-200
Schartau W, Leidescher T (1983) Composition of the hemolymph of the tarantula Eurypelma cali-
fornicum. J Comp PhysioI152:73-77
Schatz H, S(jJmme L (1981) Cold-hardiness of some oribatid mites from the Alps. Cryo-Lett 2:
207-216
Schenberg S, Pereira Lima FA (1971) Phoneutria nigriventer venom - Pharmacology and biochem-
istry of its components. In: BUcher! W, Buckley EE (eds) Venomous animals and their venoms,
vol III. Academic Press, London New York, pp 279-297
*Schenberg S, Pereira Lima FA (1978) Venoms of Ctenidae. In: Bettini S (ed) Arthropod venoms.
Springer, Berlin Heidelberg New York, pp 217 -245
Schenone H, Semprevivo L, Schirmer E (1959) Consideraciones a prop6sito de dos casos de loxo-
scelismo cutaneo-visceral. Bol Chil Paras 14:17
*Schenone H, Suarez G (1978) Venoms of Scytodidae. Genus Loxosceles. In: Bettini S (ed) Arthro-
pod venoms. Springer, Berlin Heidelberg New York, pp 247-275
Schenone H, Letonja T, Knierim F (1975) Algunos datos sobre el aparate venenoso de Loxosceles
/aeta y toxicidad de su veneno sobre diversas especies animales. Bol Chi! Parasitol 30:37 -45
Schildknecht H, Kunzelmann P, Krauss D, Kuhn C (1972) Ober die Chemie der Spinnwebe. I. Na-
turwissenschaften 59:98-99
Schlinger EI (1952) The emergence, feeding habits, and host of Opsebius diligens Osten Sacken.
Pan-Pac EntomoI28:7-12
*Schlinger EI (1957) Generic revision of world Acroceridae. Thesis, Univ California, Davis
*Schlinger EI (1960) A revision of the genus Ogcodes Latreille, with particular reference to species
of the western hemisphere. Proc US Natl Mus 111(3429):227-336
*Schlinger EI (1969) In: Cole FR (eds) The flies of Western North America. Acroceridae. Univ Cali-
fornia Press, Berkeley, pp 219-224
Schlinger EI (1972) A new Brazilian panopine species Exetasis eickstedtae, reared from the thera-
phosid spider, Lasiodora klugi (Koch) with a description of its immature larval stages (Diptera,
Acroceridae). Pap Avulsos Zoo I 26:73-82
Schlinger EI (1981) Acroceridae. In: McAlpine JF et al. (eds) Manual of nearctic Diptera, chap. 43.
Agric Can, Monogr 27, Ottawa, pp 575 -584
Schlott M (1931) Biologische Studien an Agelena /abyrinthica Cl. Z Morphol Okol Tiere 24:1-77
Schlottke E (1936) tiber die Verdauungsfermente der Vogelspinnen. Sitzungsber Abh Naturforsch
Ges Rostock 3 Ser 6:89-105
Schneider P (1971) Ameisenjagende Spinnen (Zodariidae) an Cataglyphis-Nestern in Afghanistan.
Zool Anz 187 :199-201
Schoener TW (1971) Theory of feeding strategies. Annu Rev Ecol Syst 2 :369-404
Scholander PF, Flagg W, Walters V, Irving L (1953) Climatic adaptation in arctic and tropical poi-
kilotherms. Physiol Zool 26 :67-92
Scott H (1958) Tipulids with white tarsi hanging on spiders' webs and their dancing movements.
Entomol Mon Mag 94:247
428 References
Sedlak BJ, Gilbert LI (1979) Correlations between epidermal cell structure and endogenous hor-
mone titers during the fifth larval instar of the tobacco hornworm, Manduca sexta. Tissue Cell
11 :643-653
Seitz KA (1972) Elektronenmikroskopische Untersuchungen an den Guaninspeicherzel1en 'von
Araneus diadematus Clerck (Araneae, Araneidae). Z Morphol Tiere 72:385-397
*Seitz KA (1975) Licht- und elektronenmikroskopische Untersuchungen an den Malpighischen Ge-
f~en der Spinne Cupiennius salei Keys. (Ctenidae). Zool Jahrb Anat 94:413-440
Sekiguchi K (1952) On a new spinning gland found in geometric spiders and its function. Annot
Zool Jpn 25 :394-399
*Sekiguchi K (1955a) Differences in the spinning organs between male and female spiders. Sci Rep
Tokyo Kyoiku Daig Sect B 8:23-32
Sekiguchi K (1955b) The spinning organs in sub-adult geometric spiders and their changes accom-
panying the last moulting. Sci Rep Tokyo Kyoiku Daig Sect B 8:33-40
Seligy VL (1969) Biochemical aspects of pigment variation in the spider Enoplognatha ovata
(Clerck) (Araneae:Theridiidae). Can J ZooI47:1103-1105
Seligy VL (1971) Postembryonic development of the spider Enop/ognatha ovata (Clerck). Zoo 1 J
Linn Soc 50:21-31
Seligy VL (1972) Ommochrome pigments of spiders. Comp Biochem PhysioI42A:699-709
Sevacherian V, Lowrie DC (1972) Preferred temperature of two species of lycosid spiders, Pardosa
sierra andP. ramulosa. Ann Entomol Soc Am 65:111-114
Sewell MT (1955) The histology and histochemistry of the cuticle of a spider, Tegenaria domestica
(L.). Ann Entomol Soc Am 48:107 -118
*Seyfarth E-A (1980) Daily patterns of locomotor activity in a wand'ering spider. Physiol Entomol
5:199-206
*Seymour RS, Vinegar A (1973) Thermal relations, water loss and oxygen consumption of a North
American tarantula. Comp Biochem PhysioI44A:83-96
Sferra NJ (1986) The first record of Pterodontia flavipes Gray (Diptera:Acroceridae) larvae in the
mites Podothrombium (Acari:Trombidiidae) and Abrolophus (Acari:Erythraeidae). Entomol
News (in press)
Shafer GD (1949) The ways of the mud-dauber. Stanford Univ Press, Stanford
Shelly T (1983) Prey selection by the neotropical spider Alpaida tuonaba with notes on web-site
tenacity. Psyche 90:123-133
Sherman RG (1978) Insensitivity of the spider heart to solitary wasp venom. Comp Biochem
PhysioI61A:611-615
Sherman RG (1985) Neural control of the heartbeat and skeletal muscle in spiders and scorpions.
In: Barth FG (ed) Neurobiology of arachnids. Springer, Berlin Heidelberg New York, pp 319-336
Sherman RG, Pax RA (1968) The heartbeat of the spider Geolycosa missouriensis. Comp Biochem
Physiol 26 :529 -536
Sherman RG, Pax RA (1969) Electrical activity in single muscle cells of a spider heart. Comp Bio-
chern Physiol 28 :487 -489
Siebold CTE von (1843) Uber die Fadenwiirmer der Insekten. Entomol Z 4:78-84
Siebold CTE von (1848) Uber die Fadenwiirmer der Insekten. Entomol Z 9:290-300
Siebold CTE von (1854) Uber die Fadenwiirmer der Insekten. Entomol Z 15 :103-121
Simberloff D (1982) The status of competition theory in ecology. Ann Zool Fenn 19:241-253
Simpson GG (1964) Species density of North American recent mammals. Syst Zoo I13 :57 -73
Sinohara H, Tillinghast EK (1984) Carbohydrates associated with the orb web protein of Argiope
aurantia. Biochem Int 9:315-317
Sittertz-Bhatkar H (1980) Oral cleansing in spiders is gland mediated! Z Naturforsch 35 :669-673
Sivinski J, Stowe M (1980) A kleptoparasitic cecidomyiid and other flies associated with spiders.
Psyche 87 :337 -348
*Slama K, Romanuk M, Sorm F (1974) Insect hormones and bio-analogues. Springer, Berlin Heidel-
berg New York
Smith AP (1979) Life strategy and mortality factors of Sceliphron laetum (Smith) in Australia.
Aust J EcoI4:181-186
Smith CW, Micks DW (1968) A comparative study of the venom and other components of three
species of Loxosceles. Am J Trop Med Hyg 17:651-657
References 429
Smith DRR (1986) Population dynamics of Anelosimus eximius (Theridiidae). J Arachnol 14:
201-218
Smith RB, Mommsen TP (1984) Pollen feeding in an orb-weaving spider. Science 226:133-1333
Smith WJ (1977) The behavior of communicating. Harvard Univ Press, Cambridge Mass
Smith-Gill SJ (1983) Developmental plasticity: developmental conversion versus phenotypic mo-
dulation. Am Zool 23:47 -55
Smith-Trail DRR (1980) Predation by Argyrodes on solitary and communal spiders. Psyche 87:
349-355
St/>mme L (1965) Changes in sorbitol content and supercooling points in overwintering eggs of the
European red mite (Panonychus ulmi KOCH). Can J ZooI43:881-884
St/>mme L (1978) Notes on the cold-hardiness of prostigmate mites from Vestfjella, Dronning Maud
Land. Norw J EntomoI25:51-55
St/>mme L (1979) Overwintering ecology of alpine Collembola and oribatid mites from the Austrian
Alps. Ecol EntomoI4:175-180
St/>mme L (1981) Cold tolerance of alpine, Arctic and Antarctic Collembola and mites. Cryobiology
18:212-220
*St/>mme L (1982) Supercooling and winter survival in terrestrial arthropods. Comp Biochem Physiol
73A:519-543
St/>mme L, Conradi-Larsen EM (1977) Cold hardiness of collembolans and oribatid mites from wind-
swept mountain ridges. Oikos 29:118-126
St/>mme L, <I>stbye E (1969) Cold-hardiness in some winter active insects. Nor Entomol Tidsskr 16:
45-48
*Southwood TRE (1962) Migration of terrestrial arthropods in relation to habitat. Bioi Rev 37:
171-214
Southwood TRE (1978) Ecological methods, 2nd edn. Chapman and Hall, London
Spence I, Adams DJ, Gage PW (1977) Funnel-web spider venom produces spontaneous action po-
tentials in nerve. Life Sci 20:243-250
*Spindler KD, Keller R, O'Connor JD (1980) The role of ecdysteroids in the crustacean molting
cycle. In: Hoffmann JA (ed) Progress in ecdysone research. Elsevier/North-Holland, Amsterdam
New York, pp 409-430
Stamm-Menendez MD, Galarza Basanta AM (1961) Bioquimica de los omocromos. An Real Acad
Farm 2:115-140
Stanic M (1953) Beitrag zur Immunologie des Latrodektismus. Acta Trop 10:225-232
Stearns SC (1976) Life-history tactics: a review of the ideas. Q Rev Bioi 51 :3-47
Stehli FG, Douglas RG, Newell ND (1969) Generation and maintenance of gradients in taxonomic
diversity. Science 164:947-949
Steigen AL (1976) Energy metabolism in the wolf spider Pardosa palustris (1.) (Araneae, Lycosidae)
from Hardangervidda, southern Norway. Norw J EntomoI23:51-60
*Steinbrecht RA (1984) Arthropoda. Chemo-, hygro-, and thermoreceptors. In: Bereiter-Hahn J,
Matoitsy AG, Richards KS (eds) Biology of the integument. I. Invertebrates. Springer, Berlin
Heidelberg New York, pp 523-553
Steininger F (1934) Dber visuelle Anpassung bei Spinnen. Verh Dtsch Zool Ges 210-218
Stern H, Kullmann E (1975) Leben am seidenen Faden. Bertelsmann, Miinchen
Stewart DM, Martin AW (1970) Blood and fluid balance of the common tarantula Dugesiella
hentzi. Z Vergl Physiol 70:223-246
*Stewart DM, Martin AW (1974) Blood pressure in the tarantula Dugesiella hentzi. J Comp Physiol
88:141-172
*Stewart DM, Martin AW (1982) Moulting in the tarantula, Dugesiella hentzi. J Comp Physiol B
149:121-136
Stippich G (1981) Untersuchungen zur Okologie von Spinnen (Araneida) eines Kalkbuchenwaldes.
Dipl Thesis, Univ Gottingen
Stowe MK (1978) Observations of two nocturnal orb weavers that build specialized webs: Scolo-
derus cordatus and Wixia ectypa (Araneae:Araneidae). J ArachnoI6:141-146
Stradling DJ (1978) The growth and maturation of the "tarantula", Avicularia avicularia 1. Zool J
Linn Soc 62:291-303
Strand E (1914) Indoaustralische, papuanische und polynesische Spinnen des Senckenbergischen
Museums. Abh Senckenberg Ges 36:179-274
430 References
Strazny F, Perry SF (1984) Morphometric diffusing capacity and functional anatomy of the book
lungs in the spider Tegenaria spp. (Agelenidae). J MorphoI182:339-354
Strohmenger T, Nentwig W (1986) Adhesive and trapping properties of silk from different spider
species. Zool Anz (in press)
Striibing H (1958) Schneeinsekten. Die Neue Brehm-Biicherei. Ziemsen, Wittenberg, p 47
Suarez G, Biggemann U, Schenone H (1971) Estudios biochimicos del veneno de Loxosceles laeta
y de sus mecanismos de accidn. Bol Chil Parasit 26 :60
Sutcliffe DW (1963) The chemical composition of the hemolymph in insects and some other arthro-
pods in relation to their phylogeny. Comp Biochem Physiol 9:121-135
Suter RB (1981) Behavioural thermoregulation: solar orientation in Frontinella communis (Liny-
phiidae), a 6-mg spider. Behav Ecol SociobioI8:77-8
Sutherland SK (1972) The Sydney funnel-web spider (Atrax robustu s) . Fractination of the female
venom into five distinct components. Med J Aust 2:593-596
Sutherland SK (1973) Isolation, mode of action and properties of the major toxin (Atraxotoxin)
in the venom of the Sydney funnel-web spider (Atrax robustus). Proc Aust Soc Med Res 3:172
Sutherland SK (1980) Antivenom to the male Sydney funnel-web spider Atrax robustus. Preliminary
report. Med J Aust 2:437-441
Sutherland SK (1981) Management of spider bites in Australia. Aust Prescr 5 :83-88
Szlep R (1966) Evolution of the web spinning activities: the web spinning in Titanoeca albomacu-
lata Luc. (Araneae:Amaurobiidae). Isr J Zool15 :83-88
Tanaka K (1984) Rate of predation by a kleptoparasitic spider, Argyrodes /issifrons, upon a large
host spider,Agelena limbata. J ArachnoI12:363-367
Tanaka K, Ito Y (1982) Decrease in respiratory rate in a wolf spider,Pardo.l'll astrigera (1. Koch),
under starvation. Res Popul EcoI24:360-374
Tanaka K, Ito Y, Saito T (1985) Reduced respiratory quotient by starvation in a wolf spider, Par-
do.l'll astrigera (1. Koch). Comp Biochem PhysioI80A:415-418
Tauber MJ, Tauber CA (1976) Insect seasonality: diapause maintenance, termination, and post-
diapause development. Annu Rev EntomoI21:81-107
Taussig BL (1956) Arachnidism. Clin Med 3:971-974
Tenney SM, Remmers JE (1963) Comparative quantitative morphology of the mammalian lung:
diffusing area. Nature (London) 197 :54-56
Thaler E (1976) Nest und Nestbau von Winter- und Sommergoldhlihnchen (Regulus regulus und
R. ignicapillus). J Ornithol117 :121-144
Thaler K (1975) tiber einige Spinnen aus Spitzbergen (Arachnida, Aranei). Mitt Schweiz Entomol
Ges 48:437-442
Thaler K (1980) Die Spinnenfauna der Alpen: Ein zoogeographischer Versuch. Verh 8th Int
Arachnol Congr, Wien, pp 389-404
Thaler K, Steiner HM (1975) Winteraktive Spinnen auf einem Acker bei Grossenzersdorf (Nieder-
osterreich). Anz Schiidlingskd Pflanzenschutz Umweltschutz 48:184-187
*Thor S (1904) Recherches sur l'anatomie comparee des acariens prostigmates. Ann Sci Nat Zool
8th Ser 19:1-190
Thornhill R (1975) Scorpionflies as kleptoparasites of web-building spiders. Nature (London) 258:
709-711
Thornhill R, Alcock J (1983) The evolution of insect mating systems. Harvard Univ Press, Cam-
bridge Mass
Thorp RW, Woodson WD (1945) Black widow, America's most poisonous spider. Univ N Carol
Press, Chapel Hill
Tietjen WJ (1979) Tests for olfactory communication in four species of wolf spiders (Araneae,
Lycosidae). J Arachnol6 :197 -206
*Tietjen WJ, Rovner JS (1982) Chemical communication in lycosids and other spiders. In: Witt PN,
Rovner JS (eds) Spider communication. Princeton Univ Press, Princeton, pp 249-279
*Tillinghast EK (1984) The chemical fractionation of the orb web of Argiope spiders. Insect Bio-
chern 14:115-120
Tillinghast EK, Christenson T (1984) Observations on the chemical composition of the web of
Nephila clavipes (Araneae, Araneidae). ArachnoI12:69-74
References 431
Tillinghast EK, Kavanagh EJ (1977) The alkaline proteases of Argiope and their possible role in
web digestion. J Exp ZooI202:212-222
Tillinghast EK, Townley MA (1986) The independent regulation of protein synthesis in the major
ampullate glands of Araneus cavaticus Keyserling. J Insect Physiol 32:117-123
Tillinghast EK, Kavanagh EJ, Kolbjornsen PH (1981) Carbohydrates in the webs of Argiope spiders.
J MorphoI169:141-148
Tillinghast EK, Chase SF, Townley MA (1984) Water extraction by the major ampullate duct dur-
ing silk formation in the spider, Argiope aurantia Lucas. J Insect PhysioI30:591-596
Tischler W (1967) Zur Biologie und Okologie des OpilionidenMitopus morio F. BioI Zentralbl86:
473-484
Tischler W (1968) Getreidestoppeln als Winterlager fUr Kleintiere. ZoolJahrb Syst 95:523-541
Tobien H (1937) tiber Sinneshaare beiPterygotus (E"ettopterus) osiliensis Schmidt aus dem Ober-
sHur von Oesel. Palaeontol Z 19:254-266
Toft S (1976) Life-histories of spiders in a Danish beech wood. Nat JutlI9:5-40
Toft S (1978a) Phenology of some Danish beech-wood spiders. Nat Jutl 20:285 -304
Toft S (1978b) The life-history of Achaearanea lunata (CI.) in Denmark, with a note on Theridion
Jlarians Hahn (Araneae:Theridiidae). Bull Br Arachnol Soc 4:197-203
Toft S (1980) Humidity retaining function of the catching web of Tapinopa longidens (WIDER)
(Araneae :Linyphiidae). Entomol Medd 48:5-7
*Toft S (1983) Life cycles of Meta segmentata (Clerck, 1757) and Meta mengei (Blackwall, 1869) in
Western Europe (Arachnida:Araneae:Tetragnathidae). Verh Naturwiss Ver Hamb NS 26:265--
276
Tolbert WW (1975) Predator avoidance behaviours and web defensive structures in the orb weavers
Argiope aurantia and Argiope trifasciata (Araneae :Araneidae). Psyche 82 :29 -5 3
Tolbert WW (1979) Thermal stress of the orb weaving spider Argiope trifasciata (Araneae). Oikos
32:386-392
Toolson EC (1978) Diffusion of water through the arthropod cuticle: thermodynamic considera-
tion of the transition phenomenon. J Therm BioI 3 :69-73
Toolson EC (1980) Thermodynamic and kinetic aspects of water flux through the arthropod cutic-
le. J Therm Bioi 5 :1-6
Toolson EC, Hadley NF (1977) Cuticular permeability and epicuticular lipid composition in two
Arizona vejovid scorpions. Physiol Zoo150:323-330
Tretzel E (1954) Reife- und Fortpflanzungszeit bei Spinnen. Z Morphol Okol Tiere 42:634-691
Tretzel E (1955) Intragenerische Isolation und interspezifische Konkurrenz bei Spinnen. Z Morphol
Okol Tiere 44:43-162
Tretzel E (1961) Biologie, Okologie und Brutpflege von Coelotes terrestris (WIDER) (Araneae,
Agelenidae). Z Morphol ()kol Tiere 49:658-745
Troise E (1928) Preparacion de un suero immunizante contra el "Latrodectus mactans". Rev Soc
Argent Bioi 4:467 -4 75
Tsuneki K (1970) Gleanings on the bionomics of the east-Asiatic non~ocial wasps (Hymenoptera).
VI. Some species of Trypoxylonini. Etizenia 45 :1-20
Turnbull AL (1962) Quantitative studies of the food of Linyphia triangularis (Clerck) (Araneae,
Linyphiidae). Can EntomoI94:1233-1249
Turnbull AL (1964) The search for prey by a web-building spider Achaearanea tepidariorum (CL
Koch). Can EntomoI96:568-579
Turnbull AL (1965) Effects of prey abundance on the development of the spider Agelenopsis pot-
teri. Can Entomol 97 :141-147
TurnbullAL (1973) Ecology ofthe true spiders (Araneomorphae). Annu Rev Entomoll8:305 -348
Turner M (1979) Diet and feeding phenology of the green lynx spider,Peucetia Jliridans (Araneae:
Oxyopidae). J Arachnol 7 :149-154
Uchida H, Fujita K (1968) Mass occurrence and diurnal activity of Dicyrtomina rufescens (Collem-
bola:Dicyrtomidae) in winter. Sci Rep Hirosaki Univ 15 :36-48
Ueda Y (1974) A study on the function of the rectal bladder of spiders. Acta Arachnol 25 :47 -51
Uetz G, Johnson A, Schemske D (1978) Web placement, structure, and prey capture in orb-weav-
ing spiders. Bull Br Arachnol Soc 4: 141-148
432 References
Unzicker JD (1977) The Mimetidae of St. Helena Island (Arachnida, Araneae). Entomol News 88:
41-42
Uyemura T (1957) Colour change of two species of Japanese spiders. Acta Arachnol15 :1-10
*Vachon M (1938) Recherches anatomiques et biologiques sur la reproduction et Ie developpement
des pseudoscorpions. Ann Sci Nat Zoolll th Ser 1 :1-207
Vachon M (1949) Ordre des pseudoscorpions. In: Grasse pop (ed) Traite de zoologie, vol6. Masson,
Paris, pp 431-481
Vachon M (1957) Contribution a l'etude du cteveloppement postembryonnaire des araignees. Pre-
miere note. Generalites et nomenclature des stades. Bull Soc Zool Fr 82:337 -354
Vachon M (1966) Les conduits evacuateurs des glandes cheliceriennes chez les pseudoscorpions
(Arachn.). Senckenberg Bioi 47:29-33
Vachon M (1968) Piquiires et morsures d'arthropodes, araignees, scorpions, hymenopteres, abeilles
et gutlpes. Sandoz, Paris
Valerio CE (1974) Feeding on eggs by spiderlings of Achaearanea tepidariorum (Araneae, Theridi-
idae) and the significance of the quiescent instar in spiders. J Arachnol 2 :57 -63
Valerio CE (1980) Aral\as terafosidas de Costa Rica (Araneae:Theraphosidae). III Sphaerobothria,
Aphonopelma, Pterinopelma, Citharacanthus, Crypsidromus, y Stychoplastus. Rev Bioi Trop
28:271-296
Valerio CE (1984) Two alternative strategies for spider egg parasitoids. Rev Bioi Trop 32:123-128
Vannini M, Balzi M, Becciolini A, Carmignani I, Ugolini A (1985) Water exchange between mother
and larvae in scorpions. Experientia 41 :1620--1621
Vellard J (1934) Notes sur quelques parasites de mygales sud-americaines. Bull Soc Zool Fr 59:
293-295
*Vellard J (1936) Le venin des araignees, Masson, Paris
Vellard J (1954) Preparation de un serum contre Ie venin de "Loxosceles !aeta". CR Acad Sci 238:
2078
Versluys J, Demoll R (1922) Das Limulus-Problem. Ergebn Fortschr Zool5 :67 -376
Vicari G, Bettini S, Collotti C, Frontali N (1965) Action of Latrodectus mactans tredecimguttatus
venom and fractions on cells cultivated in vitro. Toxicon 3:101-106
Vincent LS (1985) The first record of a tachinid fly as an internal parasitoid of a spider (Diptera:
Tachinidae; Araneae:Antrodiaetidae). Pan-Pac EntomoI6:224-225
Vincent LS (1986) Pathogens and parasitoids of the fossorial mygalomorph spider Atypoides
riversi O.P.-Cambridge (Antrodiaetidae:Araneae) of various age and size classes. Proc 9th Int
Arachnol Congr, Panama (in press)
Vlijm L, Kessler-Geschiere AM (1967) The phenology and habitat of Pardo sa monticola, P. nigri-
ceps and P. pullata (Araneae, Lycosidae). J Anim EcoI36:31-56
Vlijm L, Kessler A, Richter CJJ (1963) The life history of Pardosa amentata (Cl.) (Araneae, Lyco-
sidae). Entomol Ber 23:75-80
Vollmer AT, McMahon JA (1974) Comparative water relations of five species of spiders from dif-
ferent habitats. Comp Biochem Physiol47 :75 3-765
Vollrath F (1976) Konkurrenzvermeidung bei tropischen kleptoparasitischen Haubennetzspinnen
der GattungArgyrodes. Entomol Germ 3:104-108
Vollrath F (1977) Zur Okologie und Biologie von kleptoparasitischenArgyrodes elevatus und syn-
oken Argyrodes-Arten. Thesis, Univ Freiburg
Vollrath F (1978) A close relationship between two spiders: Curimagua bayano synecious on a
Diplura species. Psyche 85 :347 -35 3
*Vollrath F (1979a) Behaviour of the kleptoparasitic spider Argyrodes elevatus. Anim Behav 27:
515-521
Vollrath F (1979b) Vibrations, their signal function for a spider kleptoparasite. Science 205:
1149-1151
*Vollrath F (1980) Male body size and fitness in the web-building spider Nephi!a c!avipes. Z Tier-
psychol 53 :61-78
Vollrath F (1981) Energetic considerations of a spider parasite - spider host system. Rev Arachnol
3:37-44
Vollrath F (1983) Relative and absolute growth in Nephila clavipes. Verh Naturwiss Ver Hamburg
26:277-289
References 433
Vollrath F (1984) Kleptobiotic interactions in invertebrates. In: Barnard CJ (ed) Producers and
scroungers. Chrom Helm, Beckenham, pp 61-94
Vollrath F (1986a) Eusociality and extraordinary sex ratios in the spider Anelosimus eximius (Ara-
neae:Theridiidae). Behav Ecol SociobioI18:283-287
Vollrath F (1986b) Foraging and growth: decisions in a sit-and-wait predator. Oecologia (in press)
Vollrath F (1986c) Environment, reproduction and the sex ratio of social spider Anelosimus eximius.
J ArachnoI14:267-281
Vollrath F (in prep) The economics of spider growth
Vollrath F, Houston A (1986) Previous experience and site tenacity in the orb web-building spider.
Nephila (Araneae, Araneidae). Oecologia 70 :305 -308
Vollrath F, Rohde-Arndt D (1983) Prey capture and feeding in the social spider Anelosimus eximius.
Z Tierpsychol61 :334 -340
Vuillaume M (1969) Les ommochromes: Les pigments des invertebres, Monogr 10. Masson, Paris,
pp 73-88
Wagner W (1888) La mue des araignees. Ann Sci Nat Zool Ser 7 6:281-393
Waitzbauer J (1983) Licht- und elektronenmikroskopische Untersuchungen der Spermiogenese von
Hermannia gibba (Oribatidae, Acari). Acarologia 24:95 -107
Walckenaer CA (1833) Une filaire parasite d'une Epeira diadema. Ann Soc Entomol Fr 2:74
Wanless FR (1980) A revision of the spider genera Asemonea and Pandisus (Araneae, Salticidae).
Bull Br Mus Nat Hist ZooI39:213-257
Wanless FR (1984a) A review of the spider subfamily Spartaeinae nom. n. (Araneae, Salticidae) with
descriptions of six new genera. Bull Br Mus Nat Hist ZooI46:135 --205
Wanless FR (1984b) A revision of the spider genus Cyrba (Araneae, Salticidae) with the description
of a new presumptive pheromone dispersing organ. Bull Br Mus Nat Hist Zool47 :445-481
Wanless FR (1985) A revision of the spider genera Holcolaetis and Sonoita (Araneae, Salticidae).
Bull Br Mus Nat Hist Zool48 :249-278
Ward GL (1972) Aggregations of Chalybion californicum (Saussure) near Centerville, Wayne County,
Indiana. Proc Indiana Acad Sci 81 :177-181
Ward P, Enders MM (1985) Conflict and cooperation in the group feeding of the social spider Stego·
dyphus mimosarum. Behaviour 94:167 -182
Wasowska S (1969) Particularites de l'appareil sericigene des males Clubiona phragmitis Koch. Zool
Pol 19 :505 -515
Watson JR (1922) The bite of Latrodectus mactans. Science (Corresp) 55 :539
Weese AO (1924) Animal ecology of an Illinois elm-maple forest. III Bioi Monogr 9 :1-93
Weigel G (1941) Farbung und Farbwechsel der Krabbenspinne Misumena vatia (L.). Z Vergl Physiol
29:195-248
Weis-Fogh T (1964) Diffusion in insect wing muscle, the most active tissue known. J Exp Bioi 41 :
229 256
Weygoldt P (1966a) Mating behavior and spermatophore morphology in the pseudoscorpion Dino·
cheirus tumidus Banks (Cheliferidae, Chernetidae). Bioi Bull 130:462-467
Weygoldt P (1966b) Vergleichende Untersuchungen zur Fortpflanzungsbiologie der Pseudoscor-
pione: Beobachtungen iiber das Verhalten, die Sameniibertragungsweisen und die Spermato-
phoren einiger einheimischer Arten. Z Morphol Okol Tiere 56 :39-92
Weygoldt P (1966c) Spermatophore web formation in a pseudoscorpion. Science 153: 1647 -1649
*Weygoldt P (1969a) The biology of pseudoscorpions. Harvard Univ Press, Cambridge Mass
Weygoldt P (1969b) Paarungsverhalten und Sameniibertragung bei dem Pseudoscorpion Withius
subruber Simon (Cheliferidae). Z TierpsychoI26:230-235
Wharton GW (1985) Water balance of insects. In: Kerkut GA, Gilbert L (eds) Comprehensive insect
physiology, biochemistry and pharmacology, vol4. Pergamon Press, Oxford New York, pp 565-
601
Wheeler AG, McCaffrey JP (1984) Ranzovius contubernalis: Seasonal history, habits, and descrip-
tion of fifth instar, with speculation on the origin of spider commensalism in the genus Ranzo·
vius (Hemiptera:Miridae). Proc Entomol Soc Wash 86:68-81
Whittow GC (1971) Evolution of thermoregulation. In: Whittow GC (ed) Comparative physiology
of thermoregulation, vol III. Academic Press, London New York, pp 202-258
434 References
Wiehle H (1927) Beitrage zur Kenntnis des Radnetzbaues der Epeiriden, Tetragnathiden und Ulo-
boriden. Z Morphol Okol Tiere 8:468-537
Wiehle H (1928) Beitriige zur Biologie der Araneen, insbesondere zur Kenntnis des Radnetzbaues.
Z Morphol Okol Tiere 11 :115--151
Wiehle H (1931) Neue Beitriige zur Kenntnis des Fanggewebes der Spinnen aus den Familien Argio-
pidae, Uloboridae und Theridiidae. Z Morphol Okol Tiere 22:349-400
Wiehle H (1937) Spinnentiere oder Arachnoidea VllI.26.Familie: Theridiidae oder Haubennetz-
spinnen (Kugelspinnen). In: Dahl M, Bischoff H (eds) Die Tierwelt Deutschlands, part 33.
Fischer, Jena, pp 119-222
Wiehle H (1949) Vom Fanggewebe einheimischer Spinnen. Die Neue Brehm Biicherei. Akad Ver-
lagsges, Leipzig; Ziemsen, Wittenberg/Lutherstadt
Wiehle H (1956) Spinnentiere oder Arachnoidea (Araneae). 28.Familie Linyphiidae - Baldachin-
spinnen. In: Dahl M, Bischoff H (eds) Die Tierwelt Deutschlands, part 44. Fischer, Jena, pp 1-
337
Wiener S (1956) The Australian red back spider (Latrodectus hasselti) , preparation of antiserum by
the use of venom adsorbed on aluminium phosphate. Med J Aust 1 :739-742
Wiener S (1957) The Sydney funnel-web spider (Atrax robustus). I. Collection of venom and its
toxicity in animals. Med J Aust 2:377-382
Wiesner W (ed) (1973a) In: Effects of temperature on ectothermic organisms. Springer, Berlin Hei-
delberg New York
Wieser W (ed) (1973b) Temperature relations of ectotherms: a speculative review. In: Effects of
temperature on ectothermic organisms. Springer, Berlin Heidelberg New York, pp 1-23
Wigglesworth VB (1947) The epicuticle in an insect Rhodnius prolixus (Hemiptera). Proc R Soc
London Ser B 134:163-181
Willem V (1917) Observation sur la circulation sanguine et la respiration pulmonaire chez les araig-
nees. Arch Neerl Phsiol Homme 1 :226-256
Willem V, Bastert C (1918) Essais d'inscription des pulsations cardiaques chez une araignee. Arch
Neerl Physiol Homme 2 :285 -289
Williams DS (1979) The feeding behaviour of New Zealand Dolomedes species (Araneae :Pisauridae).
N Z J Zool6 :95 -105
Williams G (1962) Seasonal and diurnal activity of harvestmen (phalangida) and spiders (Araneida)
in contrasted habitats. J Anim Ecol 31 :23-42
Willig A, Keller R (1973) Molting hormone content, cuticle growth and gastrolith growth in the
molt cycle of the crayfish Orconectes limosus. J Comp Physiol 86:377 -388
Willmer PG (1982) Microclimate and the environmental physiology of insects. Adv Insect Physiol
16:1-57
Wilson H (1943) Acute abdominal symptoms in arachnidism. Surgery 13 :924-928
Wilson RS (1962a) The structure of the dragline control valves in the garden spider. Q J Microsc
Sci 103:549-555
Wilson RS (1962b) The control of dragline spinning in the garden spider. Q J Microsc Sci 103:
557-571
*Wilson RS (1967) The heart-beat of the spider, Heteropoda venatoria. J Insect Physiol13:1309-
1326
Wingerden WKRE van (1977) Population dynamics of Erigone arctica (White) (Araneae, Linyphi-
idae). Thesis, Univ Amsterdam
Wingerden WKRE van (1980) Aeronautic dispersal of immatures of two linyphiid spider species.
Proc 8th Int Arachnol Congr, Vienna, pp 91-96
Winget RR, Herman WS (1976) Occurence of ecdysone in the blood of the chelicerate arthropod,
Limulus polyphemus. Experientia 32:1345-1346
Winget RR, Herman WS (1979) Influence of molt cycle and (J ecdysone on protein synthesis in the
chelicerate arthropod, Limulus polyphemus. Comp Biochem PhysioI62:119-122
*Wise DH (l975) Food limitation of the spider Linyphia marginata: Experimentaifield studies. Eco-
logy 56 :637 -646
Wise DH (l976) Variable rates of maturation of the spider, Neriene radiata (Linyphia marginata).
Am MidI Nat 96:66-75
References 435
Wise DH (1979) Effects of an experimental increase in prey abundance upon the reproductive rates
of two orb-weaving spider species (Araneae:Araneidae). Oecologia (Berlin) 41 :289-300
Wise DH (1982) Predation by a commensal spider, Argyrodes trigonum, upon its host: an experi-
mental study. J Arachnol10:111-116
Wise DH (1984a) Phenology and life history of the filmy dome spider (Araneae:Linyphiidae) in
two local Maryland popUlations. Psyche 91 :267-288
Wise DH (1984b) The role of competition in spider communities: insights from field experiments
with a model organism. In: Strong DR, Simberloff D, Abele LG, Thistle AB (eds) Ecological
issues: conceptual issues and the evidence. Princeton Univ Press, Princeton, pp 42-53
Wise DH (1986) Rearing studies with a spider exhibiting a variable phenology: no evidence of sub-
stantial genetic variation. Bull Br Arachnol Soc (in press)
Wise DH, Barata JL (1983) Prey of two syntopic spiders with different web structures. J Arachnol
11:271-282
With CJ (1906) The Danish expedition to Siam, 1899-1900. Chelonethi: an account of the Indian
false scorpions, together with studies on the anatomy and classification of the order. K Dan
Vidensk Selsk BioI Skr 7 II1:1-214
Witt PN (1963) Interrelationships between web-building behavior and amount of thread material in
the spider Araneus diadematus CI. Proc XVI Int Congr Zoo12:7
Witt PN (1975) The web as a means of communication. Biosci Commun 1 :7-23
*Witte H (1975) Funktionsanatomie der GenitaH)rgane und Fortpflanzungsverhalten bei den Miinn-
chen des Erythraeidae (Acari, Trombidiformes). Z Morphol Tiere 80: 13 7 -180
Witte H, Storch V (1974) Licht- und elektronenmikroskopische Untersuchungen an Hodensekreten
und Spermien der trombidiformen Milbe Abrolophus rubipes (Trouessard, 1888). Acarologia
15:441-450
Wolska H (1957) Preliminary investigations on the thermic preferendum of some insects and spiders
encountered in snow (in Polish). Folia Bioi 5 :195-208
Wood PA, Gabbutt PD (1979a) Silken chambers built by nymphal pseudoscorpions in laboratory
culture. Bull Br Arachnol Soc 4 :329-336
Wood PA, Gabbutt PD (1979b) Silken chambers built by adult pseudoscorpions in laboratory cul-
ture. Bull Br Aranchol Soc 4 :285 -293
Woodring JP (1970) Comparative morphology, homologies, and functions of the male system in
oribatid mites (Arachnida:Acari). J MorphoI132:425-452
Work RW (1976) The force-elongation behavior gfweb fibers and silks forcibly obtained from orb-
web spinning spiders. Text Res J 46:485-492
Work RW (1977a) Mechanisms of major ampullate silk fiber formation by orb-web-spinning spiders.
Trans Am Microsc Soc 96 :170-189
Work RW (1977b) Dimensions, birefringences and force-elongation behavior of major and minor
ampullate silk fibers from orb-web-spinning spiders - The effects of wetting on these properties.
Text Res J 47:650-662
Work RW (1981a) Web components associated with the major ampullate silk fibers of orb-web-
building spiders. Trans Am Microsc Soc 100:1-20
Work RW (1981b) A comparative study of the supercontraction of major ampullate silk fibers of
orb-web-building spiders (Araneae). J Arachnol 9 :299-308
Work RW (1984) Duality in major ampullate silk and pre cursive material from orb-web-building
spiders (Araneae). Trans Am MicroscSoc 103:113·-121
Work RW (1985) Viscoelastic behaviour and wet super contraction of major ampullate silk fibres
of certain orb-web-building spiders (Araneae). J Exp Bioi 118:379-404
Work RW, Morosoff N (1982) A physico-chemical study of the supercontraction of spider major
ampullate silk fibers. Text Res J 52:349-356
Workman C (1978) Individual energy budget of Trochosa te"icoia THORELL (Araneae:Lycosidae)
under constant and fluctuating temperature conditions. Symp Zoo ISoc London 42 :223-233
Workman C (1979) Life cycles, growth rates and reproductive effort in lycosids and other spiders.
Rep Kevo Subarct Res Stn 15 :48-55
Wright JE (1969) Hormonal termination of larval diapause in Dermacentor albipictus. Sciences
163 :390-391
436 References
*Yeargan KV (1975) Prey and periodicity of Pardosa ramulosa (McCook) in alfalfa. Environ Ento-
moI4:137 -141
Yokota SD (1979) Water, energy and nitrogen metabolism in the desert scorpion Paruroetonus
mesaensis. Thesis, Univ California, Riverside
Yokota SD (1984) Feeding and excretion in the scorpion Paruroctonus mesaensis: water and ma-
terial balance. J Exp Bioi 110 :253-265
Yoshikura M (1982) Notes on the mating plugs found in the female epigynum of a lynx spider,
Oxyopes sertatus (Araneae, Oxyopidae). Hepthathela 2:43-46
Yoshimoto CM (1964) Nesting activity and larval description of Trypoxylon (Trypoxylon) hieolor
Smith (Hymenoptera:Sphecidae) in Hawaii. Pac Insects 6 :517 -521
Young AM (1971) Foraging for insects by a tropical hummingbiid. Condor 73:36-45
Young SR (1980) Cold hardiness in a temperate cryptostigmatid mite. Oikos 35 :350-352
Young SR, Block W (1980) Experimental studies on the cold tolerance of Alaskozetes antaretieus.
J Insect Physiol26 :189-200
Zahl PA (1971) What's so special about spiders? National Geographic 140(2):190-219
Zemlin JC (1968) A study of the mechanical behavior of spider silks. Tech Rep 69-29-CM, AD684333,
US Army Natick Lab, Natick, MA 01760
Zimmermann W (1975) Biologische und rasterelektronenmikroskopische Feststellungen an Oecobi-
inae, Uroecobiinae und Urocteinae (Araneae:Oecobiidae) als Beitrag zum "Cribellaten-Ecribel-
laten-Problem ". Thesis, Univ Bonn
Subject Index
Ablemma 383 aeronautic 351,353 'Y-amino butyramide 203

absorption 42 aestivation 339 Ammoxenus 388
absorptive cells 232, 233, Agelastica 262 amorphous domains 206
237 Agelena 33,53,68,100,190, Amphinecta 387
Acanthoctenus 386 222,252,285,294,342, ampullate glands 127, 170,
Acari 9,223, see also mites 377, 378, 387 176,178,182,191,197,
accessory glands 165, 166 Agelenopsis 28,34,58,61, 205,208ff.
acetylcholine 147,150,209 65,102,300,321,326, amylase 233
Achaearanea 40 252,277 361,365,366,387 O!-amylase 235
335ff., 365, 384 aggregate glands 131, 183, anaerobic metabolism 88,
acidosis 158 184,188,192,203,205, 90ff., 94,104
aciniform glands 168, 176, 207,209 anal tubercle 170
178,180,182,190,205, aggregations 276,366 anaphylactoid 149
209 aggressive 151,251,261,275 Anapis 385
acinous glands 170, 172, - mimicry 138 Anapisona 385
184,186 Agroeca 214,388 anatomical diffusion factor
Acrocera 323,324 airborne 355 93
Acroceridae 306, 319ff. aktograph 373,374 Anelosimus 22, 222, 223,
actinomycin D 209 alanine 204,207 250,252,256,258,286,
Actinopus 382 albuminuria 153 384
action potentials 150 alcohols 14 annual 336ff.
active hunter 363 aldehydes 170 anticholinesterase 209
activity 34,46,89,90,94, Aliatypus 320, 381 antihistamines 159
101,131,261,265,271, alkaline protease 234, 237 antipredator 218,220
282,283,338,373,374, alkaloids 108,110 antiserum 158
378 Allomengea 70,76,334,335, Antistea 387
- time 65 342,385 antivenin 156, 158, 159
actual prey 250, 265 allomone 131 ant-mimicking 256
Aculepeira 130,385 Alopecosa 334, 340, 385 Antrodiaetus 170,320, 324,
adenine 239 Alpaida 250,385 381
adenocyte 123,126, 128, alphastat hypothesis 85 ants 250,251, 254ff., 257,
129, 131 Amaurobioides 82,387 261, see also Formicoidea
adhesion 212 Amaurobius 41,52,176, anus 44
adhesive 164, 197,201,203 178,184,197.222,253, Anyphaena 387
- spiral 207 321,368,373,375,386 aorta 96
- threads 187 amber 383 Aphantochilus 388
adrenalin 150 amidases 236 Aphidina 249,250,254,
adsorption 42 amines 203 261,271
advertising 17,218.219 amino acid composition 164, Aphonopelma 31,53,91,
Aelurillus 322, 388 204,205,207 102,106,107,320,373,
aerial dispersal 351, 354 - - sequences 84 373ff., 381
- plankton 352 - acids 81,182,237 aphrodisiac 129
438 Subject Index
apolysis 113 Australomimetus 216,222, Bolyphantes 53,70,75,76,

aposematic 219,262 385 269,385
Aptostiehus 320, 381 Austroehilus 382 bombadier beetle 262
aquatic habitats 212 autolytic 238 Bombyx 205,208
Araehnura 25.385 autonomous capture threads book-lung 36, 44ff., 63,
Araneae (as prey) 249,251, 194, 195 78ff., 86ff., 96, 324
253, 254, see also spiders Avicularia 229,376,381 Bothriocyrtum 320,381
araneism 143,151,152,156 avoidance 221,259 botulinus toxin 147
araneophagic 138 axial fibers 188ff., 194ff. Bouligand model 6
Araneus 20,22,25,70,73, - retraction 206 Brachycera 249,250,253,
74,76,111,115,169,180, axoneme 122 260
188,189,205,208,209, Aysha 387 Brachypelma 29,30, 35,36,
229,237,251,252,294, 47,104,105,381
300,314,316,317,320, bacterium 307 bradycardia 153
361,368,375,376,385 ballooning 54,255,349,351 breaking extension ratio 206
Araniella 27,70, 127,130, Baltoeurypterus 8 Brettus 137,388
385 barrier web 276 brood care 334,345
°
Aranimermis 302ff., 306 Ba"onopsis 321,387 Brown spiders 382
arborine 11 Baryehelus 381 Bug 251,260,261, see also
Arehaea 383 basal cells 232 Heteroptera
Aretosa 53,71,334,340, - lamina 169 bumble bees 257, 259
341,345,385 - membrane 234,238, burrow 58ff., 173,211,212,
Areys 216,385 244ff. 311,350,381
arginine 166, 168 Batesian mimicry 256 bursa copulatrix 130
Argiope 12,21,25,31,32, Bathyphantes 70,212,334, butterflies 251, 257, 260
45,50,53,63,64,73,102, 337,340,385
130,178,180,188,210, bees 251,256,257,259 caddishflies 259
213ff., 250ff., 257,260, beetles 254,256,261, see calamistrated strand 194,
280,335,338,344,354, also Coleoptera 196,197,199
355,365,375,385 benzoquinones 17 calamistrum 193, 194
Argyrodes 126, 128ff., 140, Bibionidae 261 calcium 81,156
254, 276ff., 384 bicarbonate buffering 85 Callilepis 261,388
Argyroneta 52,79,86,88, biennial 336ff. Calommata 381
90,94,212,300,301,387 bilins 17,18 Cambridgea 300,387
arhythmic 373,374 bilirubin 20 camouflage 17,212,214,
Ariadna 202,320, 325ff., biliverdin 19 218,219
383 biological clock 373, 379 capillaries 87, 91
Arrhenius-van't Hoff rules 30 biorhythms 371,373,376 capillary forces 125
arteries 81 birds 221,225 Caponia 382
artificial diet 258 birefringent 204,207,210 capture threads 162, 176,
aspartic acid 164 bite 152,156, 159,261 187,188,191
assimilation 292,293 black widow; see Latrodectus carbohydrases 236
association 275, 300 Blattodea 253 carbohydrates 76,131,207
astronomical navigation 378 blepharoconjunctivitis 152 carbonate 82
Atrax 144,145,150,151, blood flow 63 carbon dioxide 46ff.
381 body fluids 68,69 carboxyl groups 172,173,
atraxism 158 - size 103 178
atraxotoxin 150 - temperature 27ff., carboxylic esterases 236
°
atrium 86 59ff., 65, 67,75,150, carboxypeptidase A 235,236
atropine 159,209 152 cardiac acceleration 11
attack wrapping 215,251 - weight 100ff., 294, 361 - activity 105
attractant 131 Bohr effect 84,85 - frequencies 102
Atypoides 300,304,306, bola spider 191,201,202, - function 95
319,320,324,381 255 - ganglion 99, 108
Atypus 100,170,232,381 Bolostromus 350 - muscles 96
Subject Index 439
cardiac nerve 372 Cicadina 249,250,254 copper 82, 83

- systole 79 Cicurina 71, 387 copulation 60,122,128,
cardiostimulatoryeffect 108 circadian clock 378 131,222,269
carotenoids 18,21 - rhythm 372,376 copulatory bulb 122
Castillneira 387 circalunar activity 371 - ducts 130
catecholamines 147, 153, circannual rhythms 371 Corinna 38.7
158 circulatory system 86 coronatae 183
caves 68 cisternae 242,244 corticosteroids 156, 158
Cecidomyiidae 223,224,276 Cithaeron 388 Cory thalia 136, 137, 388
Celaenill 191,255,385 classification 380 Cosmophasis 322, 388
cell cycle 233 cloaca 230 cosmopolitan 352
cellular plug 238, 246 Clubiona 54,71,73,127, courtship 34,60,122,125,
cement 207 139,322, 334ff., 340, 387 134,135,139,140,222
- layer 4,12,14 clusters 351 coxa 239,240
central nervous system 99, clypeal glands 128,129 coxal fluid 51
344,372 Cobanus 322,326,388 - gland 44,48,50,51,128,
Centromerita 334,341,342, CO 2 production 88 163,168,239,240
385 CO 2 transport 85 crab spiders 388
Centromerus 70,75,334, coccids 257 cribellar capture threads
335,342,385 Coccinellidae 261 188,193,201
Ceratinella 46, 385 cockroaches 256, 257,260 - fibrils 197, 200, 201
Cesonill 300, 388 cocoon 67,76,212,214 - glands 176, 178, 184, 186
Chaetopelma 320,381 Coelotes 68,71,222,261, - silk 187,194,200,201
Chalybion 310ff. 257,268,300,387 wool 184
character displacement 282 coevolution 263,327 cribellate 194, 197, 199ff.,
chelicerae 229, 231 coexisting 274 262,382,384,386
chemical communication cohabitation 136 - silk 215,254,255
138 cold injury 68 - threads 187
- defence 262 resistance 69,72,74,75, cribellum 170, 174, 176,
- protection 251 339,355 184,187,193,197,200,
chemoreceptors 232, 262 tolerance 265 201
chewing mandibulae 261 cold-hardening 73 critical activity point 52
chill-coma 66,75 cold-hardiness 73 cryoprotectants 73, 266,
chilling 34lff. Coleoptera 223, 249ff., 272,339
chiracanthism 151,156 260, 271, see also beetles Cryphoeca 79
Chiracanthium 12 7, 151, Collembola 223,249,251, crypsis 17, 24
156,387 253, 254, 264, 267ff., cryptic 282,316
Chironomidae 261 340 crystalline domains 204,
chitin 4,93,236 colonization 349 206,210
- crystallites 6 colony 286 crystallisation 242
chitinase 233,235,236 color change 22, 23, 64 crystals 240,242, 244ff.
chitinisation 251, 261 coloration 16,29,219,282 Cteniza 320, 381
chitinolytic system 236 colour 64,214,284 Ctenus 386
chitobiase 236 colulus 170, 174 Culicidae 260
chloride 81 coma 158 Cupiennius 12ff., 44, 81, 82,
cholinergic agents 208 comb-footed spiders 384 84,85,90,216, 234ff.,
chromes 16 commensalism 222 241,262,354,355,374,
chromophore 20 competition 76,270,272, 386
Chrysomelidae 223,261,262 280,282,346,347 Curculionidae 223
Chrysopa 259 competitors 368 Curimagua 277,278,280,
Chrysso 23, 384 concealment 218,219 385
chylenteron 232 condensation 74 cutaneous form 157
chylusstomach 232 constant conditions 373,375 cuticle 3ff., 44ff., 359
chymotrypsine 236 contact pheromones 140 cuticular fibrils 6
Cicadellidae 271 convergences 263 - lipids 260
440 Subject Index
cuticulin layer 4,14 - pathway 91 ecribellate 184

cyanides 262 digestion 229,236,237,266 Ectatosticta 382
Cyatholipus 385 digestive enzymes 233ff., ectoparasitic 326
Cybaeus 387 272,286 ectothermic 75,264,265
cyclic oxygen consumption - fluid 145,229,233,235, egg 212,257,332,335,
89/90 262 336,368
cycling temperatures 31,33, - tract 230 - case 176, 182, 183,212,
35,64 Diguetia 176, 178,382 214,281
Cyc1ocosmia 381 Dinopis 25,255,372,384 - development 57
Cycloctenus 386 diplochronism 337ff. - parasites 214
Cyc10sa 70,176,178, 188, Diplura 277,278, 320, 381 - production 34
214,216,221,354,355, dipso-parasitic 276 - sac 54,57,58,280,335,
385 Diptera 223,249,254,265, 344,345,350
Cyllobellus 135,388 269ff., 319, 320 - stage 72,76
cymbium 126, 127 discrimination 216 ejaculation 121,122, 124
Cyrba 137, 138, 388 disguise 218,219 elastomers 207
Cyrtarachne 25,385 dispersal 348ff. embolus 125, 129, 130
Cyrtocarenum 172,173,381 display 134,136,137,140, embryogenesis 332,341
Cyrtophora 22ff., 34, 53, 58, 144 emisinid bugs 225
60,128,139,178,180,183, distribution range 348 emulsifying substances 236
222,277,284,385 disulfide groups 170 endemic 353
cysteine 165 diverticula 230,232,242 endocrine factor 132
division oflabour 286 - tissue 118
daddy long legs spiders 383 Dolomedes 26,41,43,102, endocuticle 5,114,125
damselflies 225,259,276, 216,333,376,386 endocuticular proteins 257
282 dome-shaped web 387 endocytosis 125,238,244
day-active 282 dormancy 303,333,338, endocytotic vesicles 242,245
decision 275,357,358,361, 340,342,343 endogenous 340,371,372,
364 dragline 128, 134, 136, 170, 374
defence 259 339,351,354 - heat production 28
defensive secretion 108 dragonflies 251, 259 - periodicities 373
dehydration 238 Drapetisca 25,70,76,334, endoparsitic 319,326
deprivation 287, 288 342,385 endoparasitoid 324
desiccation 43,45,46, 51ff., Drassodes 215,300,388 endoplasmic reticulum
67,210 Drassus 302 124ff., 163,233,242
Desis 82,88, 387 Dresserus 193,202,384 endosternite 231,239,240
desmosomes 232 drinking 43,50 endothermy 29,56
detoxification 261 dual respiratory system 93 end-plate potentials 147
development 32, 332, 336, Dugesiella 32,40,46,48,50, Enoplognatha 22, 333, 384
337,340,345 82,85,98,102,104,111, entapophyseal tracheae 79
Diaea 21,300,302,322,388 115,320,381 entapophyses 80, 86
diapause 32,76,113,117, Dunsopterus 8 entropy changes 206
265,324,332, 340ff., 362, dwarf spiders 385 environment 337,346,363,
378 Dyar's rule 333 364,369
diastole 96,98 Dysdera 79,122,139,140, enzymes 235
Dicrostichus 385 176,178,180,182,383 Epeira 385
Dictyna 40,41,47,53,178, Dysderina 383 epiandrous glands 124, 162,
200,201,253,387 dyspnea 158 168
dictyosomes 124, 126, 132, epicuticle 4,28,45,125
245 ecdysial gland 118 epidemic 151
diel activity 372,373,377 ecdysis 43,48,50,113, 116, epidemiology 151
diet 224,250,251, 254ff., 359ff.,366 epidermis 114
271, 359ff., 364 ecdysone 112ff. epigastric gland 123, 124
diffusion barrier 92ff. ecdysteroid 111ff., 344 Episinus 214,384
- capacity 91,93 ecophenotype 22 epistome 230
Subject Index 441
epithelial cells 125, 128 fat body 232 frost 67,68,74

- gland 131 - consumption 76 functional response 315
epithelium 230,239,240 fatalities 155,159 fungus spores 230,257
equilibrium 63 fatty acids 236 funnel weavers 387
Eresus 21,23,25,71,384 - acyl-amino acids 236 - webs 214,251,381,382,
Erigone 32, 33,43,385 Fecenia 386 386
Erigonidium 70,385 feeding rate 64,335,360,
Eriophora 250,354,355, 363,369 galactosamine 207
377,385 feigning death 261 Galeodes 10
Ero 214,216,222,223,256, femoral organs 141 Gallieniella 388
385 fibers 160,165,166,169, game theory 358,368
erythrocytes 150 170,172,173,192,198, gas exchange 47
escape hypothesis 349 204, 206ff., 237 Gasteracantha 21,62, 178,
esophageal valve 231,232 fibrils 5,187, 193ff., 198, 216,217,221,317,385
esophagus 230,231 200,202 Gea 23,385
esterase 149,235,237 fibroblasts 150 Gelanor 216,222,385
esterified bile acids 236 fibroin 209 genetic polymorphism 346
Euagrus 170,381 Filaria 301 - variation 361
Eugnatha 100,384 Filistata 31,35,40,43,79, genotypes 361
Eulonchus 322,324 88,91,94,100,102,107, Geolycosa 31,34,41,43,58,
eumelanins 21 170,178,180,184,186, 59,61,64,89,90,98,99,
Euophrys 71,388 197ff., 253, 290, 292,354, 104,107,108,300,321,
Euryattus 135,137,138, 382 372,385
388 filter system 230 giant crab spiders 386
eurychronism 337ff. fitness 286,357,358 gland 122
Euryopis 256 flagelliform glands 176,183, glandula aggregata 190
Eurypelma 45ff., 51, 8lff., 188,192,205 - flagelliformis 190
90,91,99,104,381 flies 251,254,259 glandulae coronatae 188
eurypterid 8 flooding 212 glomerin 108
evaporative cooling 35, 36, Floronia 23,24,31,32,70, glucose equivalents 233
48,63 76,334, 34lff., 385 glue 183, 184, 187ff., 207,
- heat loss 29 food availability 345, 347 215
Evarcha 388 - deprivation 292 gluey capture threads 187ff.,
evolution 141,168,174, - intake 34 193,201
215,221,222,263,286 - utilization 34 - silk 165,180
excreta 239ff., 244 - vacuoles 233 glutamic acid 164,207,236
excretion 47,48,238,239, foraging 281,282,315, glycerol 73
242,245,247 357ff. glycine 165,168,204,207
- organs 50 decisions 363, 364 glycocholate 236
excretory saccule 240 - style 101 glycogen 131,173,242,
Exetasis 325 - theory 362,364 244
Exlinea 321,386 foregut 230,231 glycoproteins 73,165,173,
exocuticle 5,114 Formicoidea 223,249, see 174,180,183,184,207
exocytotic 244 also ants glycoproteinaceous material
exogenous 373,374 fossile 383 132
exoskeleton 3,74,358 founder hypothesis 349 Gnaphosa 300,302,388
external digestion 272 free amino acids 349 gnathocoxae 229ff.
extraintestinal digestion 48 - fatty acids 14 gnathocoxal glands 36,44,
extra-<>ral 229 free-running rhythm 374 48,49,125,127,230
exuvia 43 freezing injury 68 Gnathonarium 41,385
exuviation 113 - point 73 Golgi apparatus 122,208
eye pigments 17 freezing-susceptible 69,266, - area 126
267,272 Gongylidium 70,334,341,
facies latrodectismi 152 freezing-tolerant 69 385
faeces 247, 248 Frontinella 63,385 gonopore 121, 124
442 Subject Index
Gordius 302 Heriaeus 21,388 hydrolase 237

gradient studies 57 Herphyllus 322,388 hydrostatic 81
gradients 51 Hersilio 123,125,178,180, - pressure 84
Gradungula 382 182,216,384 - skeletons 99
Grammostola 320,325,381 Heterodictyna 25, 387 - system 94
grasshoppers 250,251,256, Heterometrus 10 5-hydroxytryptamine 146
see also Orthoptera Heteromigas 382 hygroperception 52
growth 32, 357ff. heteronomous capture threads hygrothermal preferenda 54
- rate 294 194,196,202 Hymenoptera 249ff., 260,
guanine 16, 17,20,24,28, Heteropoda 96,98,102, 271,309
238ff. 372,386 hypermetamorphic larvae
guanocytes 20,24,239, 240, Heteroptera 253, 261, see 324
242 also bug hypertension 158, 159
gustatory cells 232 Heterorhabditis 307 Hypochilus 178, 182, 333,
gut 287,288 Hexathele 170,381 336, 382
Hexura 381 hypodermis 86,93,239,242
hackled silk 187 hibernacula 74 Hypomma 70,385
Hadrotarsus 384 hibernation 339,378 hypothermia 159
Hadrurus 1Off. - chamber 67 hypoxanthine 239
haemocyanin 8lff. Hickmania 382 Hyptiotes 25,184,186,
haemocytes 81 hiemal threshold 267 193,200,201,214,215,
haemolymph 63,69,73, hindgut 244 384
80ff., 84,91,92,237,239, Hippasa 90, 385
240,246,265,266,358 Hirudinea 223 ice 74
- osmolality 50 histamine 150,159,314 - crystals 68
- sinus 79 Histopona 71 - formation 69,74
haemolysis 149 Holarchaea 142,383 - nucleation 267,272
haemostasis 81 Holcolaethis 138 - nucleators 266
Hahnia 387 Holocnemus 222,383 Jdiops 381
Haplodrassus 388 Hololena 321,326,387 illness 155
Harpactea 212,383 Homalonychus 483 indole lactic acid 150
heart 63, 79ff. homogiomerin 108 indolic groups 166
- ganglion 38 Homoptera 250,251,253 infection 302, 306
- rate 30, 95ff., 108, 153, honey bees, see bees infracapitular gland 163
372 hormonal deficit 115 ingestion 233,289,290
heartbeat 372 hormone 112 inoculative freezing 74
heat death 52 - level 113 inorganic phosphates 73, 81
- flow 30 host 277, 280ff., 303, 307, inquilines 276,277,284,
- resistance 53 320,323,326,327 286
- shunting 64 host-parasitoid relationships insects 27,238,242,246,
- stupor 5lff. 326 342,343
- transfer 63,64 humidity 26,335,361 insertion 129
heating rate 64 - preferenda 51, 52 insulating 68
HeLa cells 150 hummingbird 225,276, integument 4,20,27,45
helial conformation 210 279,282 intermoult 333,334,341
helicoidal model 6ff. hunger 89, 287ff. - cycle 113
Heliophanus 322,388 huntsman spiders 386 - period 116
a-helix 210 Huttonio 383 intersegmental membranes
helpers 286 hyaline exocuticle 10,15 251
Hemiptera 249,250,256, hyaloplasm 125 interstitial cells 237ff.
271 hyaluronidase 146, 149, 150, - tissue 232,233,246,247,
hemo-, see haemo- 237 358
hepatopancreas 230 hydrocarbons 14 intestinal diverticula 239ff.,
Heptathela 125, l30, 170, hydrocortisone 159 245,246
174,175,222,381 hydrogen bonds 210 - juice 237
Subject Index 443
intima 86, 87 latrotoxin 146, 147 Lyssomanes 25,137,138,

intoxication 153 LD, 0 148ff. 388
ion exchange 210 leberidiocytes 358
ionic regulation 49 leg cleaning 37, 48 Machadonia 321,387
Ischnothele 258,277,354, Legendrena 388 Macrargus 53,54,70,333,
355, 381 Lepidoptera 223,251, 335, 385
Ischnothyreus 383 253ff., 259, 260, 271 macro fibres 7
isoenzymes 236 Leptoneta 125,130,383 Macrothele 381
Isopoda 223,253,254 Lepthyphantes 70, 126ff., magnesium 81, 85
isotherms 59 268,269,385 major ampullate duct 45,49
isozymes 235 lethality 155, 157 - - glands 49
Ixeuticus 199,387 Leucauge 23,180,188, Mallos 186,199,387
384 Malpighian tubules 50,232,
jet stream 352 life history 357,362 238ff., 244ff.
Julidae 262 life-cyle 332, 336ff., 345 Mamosa 387
Jumping spiders 388 ligaments 96 Marpissa 71,388
junction 204, 207 limiting factor 242 Mastophora 131,192,201,
juvenile hormones 118 Limulus 6,9 255,385
juvenoids 118 Linyphia 41,70,76,170, Matachia 321,386
178,189,252,294, mating plug 130
Kaira 130,131,385 333ff., 342, 361,385 maturation 334,337,338,
KeN 262 Liocranum 388 34Hf.,368
Kibramoa 320,382 lipase 149,235 mature 335
kinetic energy 206 Liphistius 96, 104, 170, 173, maturity 336, 369
- parameters 235,236 175,214,308,381 maxillary glands 125, 126,
klepto 274,278,280,284, lipid 11,14,42,45,76, 129,130
285 262,265,385 Maymena 385
kleptobiosis 275,284,286 - film 15 Mecicobothrium 381
kleptoparasites 225, 276, - layer 4 Mecistogaster 276
277,284,285 lipoids 146 Mecynogea 183,251,320,
kleptoparasitism 222 lipoproteins 146 354,355,364,369,385
Krogh's diffusion constant Lithyphantes 40, 384 Mecysmauchenius 383
91,93 liver 230, 232 Megadictyna 385
k -selected 281 locomotoric activity 48,373, Megaloprepus 276
kynurenine 18,21 374,376,378 melanins 18,21
long-<Jay 378 melanisation 5
labium 230,231 longevity 242,292,337,345 Melopoeus 172,173,381
labyrinth 240 Loxosceles 43,54,102, 144, melting point 69,73
lactate 91,104 149ff., 157,158,176,237, Menneus 384
- dehydrogenase 91 382 Mermis 301,302
lacunae 80, 96, 284 loxoscelism 157 Mermithidae 299ff.
lageniform 128 lung atrium 80 mesenteron 232, 233
lamellae 86, 87,92, 324 - sinus 80 mesocuticle 5
lamina unit 5,13,14 - surface 92 mesodeum 230
laminae 6 - volume 92 Meta 40,70,72,76, 127,
lamination 5 Lycosa 21,31,34,35,43, 222,252,333,362,384
lampshade-shaped webs 382 53,57,58,60, 62ff., 89, metabolic heat production
larval stage 332 91, 97ff., 102, 10Sff., 157, 29
Lasiodora 320,381 290,292,300,301,321, - overshoot 89
latitudinal diversity gradients 333,335,342,344,373, - rate 31, 87ff., 94, 100ff.,
354 385 234,266,292,293,343,
latrodectism 143ff., 15Hf. lycosidism 157 372
Latrodectus 13,14,22,25, lynx spiders 386 - water 42
127, 143ff., 150ff., 183, Lyovac 156 metabolism 90
184,307,316, 333ff., 384 lysis 237 Metacybra 102,388
444 Subject Index
Metaltella 386 mitochondria 90,91,232, Neopaltis 224

Metaphidippus 322 234,240,242,244 Neoscona 91,102,317,385
Metellina 334,337,342, Miturga 387 Nepentes 275
384 Modisimus 190,383 Nephila 36,40,53,62,63,
Metepeira 251,284,285, Moggridgea 320, 382 126, 128, 130, 169, 180,
364,385 molecular alignment 210 183, 188ff., 214, 217,221,
methaqualone 110 monophagous 255,257 250,252,276, 277, 279ff.,
Miagrammopes 178,196, morphine 156, 159 294,333,335,354,355,
197,200,202,215,384 mortality 315,345,346, 359ff., 364, 365, 368, 369,
Micrathena 60,216,217, 358,361,364 385
221,385 - rate 75 Nephilengys 183,385
microbial attack 67 moths 251 nephrocytes 239,240
microclimate 267,271 moulting 32,43,47,48,50, Neriene 284,336,337,346,
microclimatic conditions 64,90,93,111,115,136, 385
54,68 162,220,221,285,335, Nesticus 70,384
micro environments 58 340,342,358,359,367, neurogenic 108
microfibre 7 376 neurosecretory cells 344
microhabitat 28,266 - cycle 114, 116, 117 neurotoxin 153
Micromata 18,20,23,25, - hormones 112,118 niche 270,273
386 - platform 218,220,221 Nicodamus 385
micromatabilin 20 moults 333,334 Nigma 387
Micromygale 381 mouth cavity 230 night-active 282, 284
Microneta 70,385 mucilaginous 251 noctuid 286
Micropholcomma 385 mucin 174,180,182,186 nocturnal 374,375,377
Microphor 225 mucopolysaccharides 12,230 nonfunctional 221
Microstigmata 381
microtubules pattern 122
microvilli 125,126,128,
mucoprotein 182
mucosubstance 130, 172,
174,175,178,182,184
Nops 382
noradrenalin 15
nucleators 74
°
242, 244ff. mud-{joubers 108,311,313 Nuctenea 130,276,284,385
Micryphantes 300,385 mydriasis 152 nuptial bridge 222
midgut diverticula 242 Mynoglenes 129,385 - gift 222, see also wedding
midintestine 230, 231,234, myocardial cells 108 nursery-web spiders 386
238 myocardium 108 nycthemeral rhythms 372,
Migas 382 myopathic syndrome 154 375
migration 378 Myriapoda 138,253
milichiid 276,279 myrmecophagous 256 O 2 consumption 294, see
millipeds 251, 254 Mysmena 385 also oxygen
mimesis 24,216 Mysmenopsis 277,280,385 Ocnaea 323
mimetic 219,322 mytilliform organs 138 Odonata 251
Mimetus 216,222,256,385 Oecobius 180, 197,200,384
mimicry 25 Nabidae 271 Oedothorax 32,70,334,
miniature end-plate potentials N-acetyl galactosamine 207 341,385
147 (3-N-acetyl-glucosaminidase oenocytes 118
Ministigmata 381 233,235,236 Ogcodes 320ff.
miosis 152 Na+/K+ pumping mechanism ogre-faced spiders 384
Miridae 223,286 49 olfactory pheromones 13 7,
Miscophus 310,311 necrosis 157 140
Misumena 20ff., 25, 98, Nematocera 224, 249ff., oJigophagy 257
108,253,257,388 255,261,270 Olios 386
Misumenoides 102,388 nematode 299, 301, 308, om matins 17
Misumenops 102,275,300, 319 ommidins 17
388 Nemesia 125,172,373,381 ommins 17
mites 161,162,165,232, Neoaplectana 307 ommochromes 17, 21
266,342,343, see also Neocteniza 381 ontogenesis 219,220,376
Acari Neolana 387 Oonops 212,383
Subject Index 445
Opiliones 10,223,261,343 - wasps 249,250,254 physiostigmine 209

opisthosoma 62ff. parasitism 302,303,306 phytoecdysone 117
optimal foraging 363 parasitoids 285,319,323, pigments 19
orb weavers 385 324 pilferers 275
webs 191,203,204, Paratropis 381 pipe organ wasps 313
215ff., 220, 221,251,254, Pardosa 31,34,40,41,43, Pirata 40,52,53,71,253,
260,384,385 5Hf., 57,58, iI, 72, 88ff., 307,321,334,340,341,
Orsolobus 383 253,258,267, 287ff., 293, 344,385
Orthoptera 253,261,271, 300,302,306,321,323, pirate spiders 385
see also grasshoppers 337,373,377,378,385 pirates 275
osmolarity 82 paroxysms 152 Pisaura 76, IlIff., 122, 127,
osmolyte 81 Pasilobus 191,202,215,385 232,341,344,386
osmoregulation 49,50 Patu 385 Pison 310,311,313
osmotic gradient 238 PCO z 85 Pisonopsis 310,311,313
- pressure 82 pedicel 63, 96 pitcher plant 275
ostia 80,81,96 Pellenes 318,322,388 Pityohyphantes 22, 385
ovaries 118 Pentatomidae 261 planidium 323
overheating 35,57,64,65, peptidases 236 plankton 250,261
284 pericardial sinus 79,80 Plator 388
overwintering 67,70,338, periodogram 374 pleated sheet form 210
342 peritrophic membrane 230 Plectreurys 21,382
oviposition 340 petiole 80, 86, 240 plugs 244
oxygen 84 Peucetia 23,25,253,300, PO, 84,91
- consumption 29,31,32, 386 podocephalic glands 163
88,90,91,102,103,234, pH 82,84,85,91,145,235 - system 162
343,344 - optimum 235 Poecilopachys 191,202,385
oxygen-binding affinity 85 pharynx 230 poison gland 44,47,215,
oxygen-carrying capacity 84 phenotype 22 251
- pigment 82 phenoxazines 17 Polenecia 178,182,184,
oxygen~riving pressure 93 pheromone 110, 122, 126ff., 196,197,200,202,384
Oxyopes 122,321,378,386 131, 133ff., 215, 221, 255 pollen 230,257,276
- glands 129 polymorphism 22
pacemakers 99,372 Phiale 354, 355,388 poly-N-acetylglucosamine 4
Pachygnatha 70,215,334, Phidippus 88ff., 102, 107, polyols 73,266
340,341,384 136,253,291,300,302, polypeptides 150,204,208,
Paculla 383 322,335,373,388 210
palatibility 262 Philaeus 21,23,388 polysaccharide 170, 172ff.,
palp gland 130 Philodromus 43,71,73,322, 178,180,182,235
palpal chewing 125 386 Pompilidae 309
Palpimanus 383 Philoponella 218, 384 ponasterone A 117
pancreas 232 Phlegra 322, 388 population 281,285,306,
Panorpa 225,276 Pholcomma 46,384 315,339,346,349,352,
Panorpidae 225 Pholcus 71,123,170,178, 356,361,362,378
pantopods 232 183,190,251,252,257, pore canals 5, lOff., 45
paracribellar 196 307,334,340,383 Po"hotele 381
- fibrils 194 Phoneutria 143ff., 150, 151, Portia 137, 138,356,388
- glands 176, 184, 186 159,386 postimmobilization wrapping
paracribellum 194,197 Phonognatha 23, 385 215
paralysis 149 Phormictopus 29,35,36,49, posture 58,62,63
paralytic stage 147 212,213,216,381 potassium 81,237,238
paralyzed 314 photoperiod 340, 342ff., dihydrogen phosphate
Paraplectana 25,385 361 203
Pararchaea 383 photoperiodic regulation 378 - nitrate 203
parasites 214,299 Phrixotrichus 320,381 potential prey 249,250,256,
parasitic castration 306 physical gill 88, 212 257,271
446 Subject Index
preadaptation 251,256,275 pterins 18 rostrum 230,231

precipitation 74 Pterodontill 319,320,323 rough endoplasmic reticulum
- water 68 puffs 194,195,201 122,232
predation 222,285,345, pulmonary sinus 80 r-selected 281
346,364,367 purines 20 Rualena 321,326,387
predator length 259 puromycin 209 rubbers 206
predators 221,224,225, purse web 381 rudimentary 138, 239
258,274,319,363 pustiliform organs 138 Ryuthela 170,174,175,381
pregonoporal glands 124 Pycnothele 381
prekeratin 164, 165 pyriform glands 124, 176, sacculus 240,242
pre larval stage 332 178,180,184,205,209 sac spiders 387
prey 249 Salassina 216,217,221,385
- abundance 65 Q. 0 30,34,35,89,106,107 salicylic acid 262
- handling 65 quiescence 340,341,344 salivary glands 36,44, 153,
- length 259 quinazolinone alkaloids 108 162
- size 250 quinone tanning 5, 12 - organs 125
- specialisation 254, 256, Saloca 333,385
257 radial symmetry 215 Salticus 35,300,388
- spectrum 251,254,257 rain 201 Sassacus 322, 388
primer pheromones 136, Ranzovius 223,286 satiation 289,291
137,138 ray spiders 385 saturation point 287
proctodeum 230 re-absorption of water 50 scales 254,259,260
pro cuticle 4, 5 receptaculum seminis 121, scavenging 258
Prodidomus 388 123,125,130 SceUphron 31 Off.
Progradungula 382 recycle 250 Schizocosa 128,301,372,
programmed death 238 reduction 254 385
proline 164,168,182 regurgitated fluid 235 sclerotization 46
prosoma 62ff. regurgitation 237 Scoloderus 254,255,385
pro somatic gland 118 releaser pheromones 136, scorpion 10,232,234,238
prosthetic group 82 138 Scytodes 102,162,165,166,
protandrous 368 reproduction 33, 337ff. 170,176,183,184,251,
protease 149,207,236,237 reserves 359,360,366 252,354,355,382
protein 73, 76, 132,136, resilience 206, 207 secondary plant products
146, 175ff., 180, 183, 184, resistance data 52 262
203,207,234,235,237 resorption 244 secretion 36,47
- chains 206 - cells 240,242 secretory cells 232,233
- rubber 206 resorptive activities 240 - epithelium 169, 182
- synthesis 117, 208, 209 respiration rate 64,90,103, sector webs 215
proteolytic 149 292,293,344 sedation 108
- activities 145 respiratory cycles 372,373 sedatives 11 0
- enzymes 150,234,237 - lamellae 80 Segestria 71,122,125,170,
protuberances 129 - organs 46, 79 214,382
Psechrus 386 - quotient 76,294 Selenops 258, 354, 355, 386
Pseudanapis 385 - system 81 semasis 17,24,25
Pseudicus 300, 388 resynthesis 233 semi-autonomous capture
pseudoflagelliform glands retardation 341, 342 threads 197, 198
176, 178, 184, 186, 194, retreat 212,214,220,221, semiochemicals 133
197 250,316,326,377 senescence 242
pseudo scorpions 161, 162, rhabditids 299 Senoculus 386
164, 165 Rhagionidae 261 sensillae 8
Pseudostigmatidae 225 rhythm 37lff. Sericopelma 38,381
Psocoptera 223, 250 Ricinoides 10 serine 165, 168
Psychodidae 260 ricinuleid 10 serotherapy 158
Psyllina 250 RNA 209 serotonin 150
Pterinochilus 143 rostral gland 44, 47 serra ted 229
Subject Index 447
sex ratio 335 spermine 150 suboesophageal ganglion

sexual dimorphism 335 Sphaerobothria 373,381 240,314
- glands 130 Sphaerops 325,326 subsheet 250
sheet weeb 170,214,382, Sphecidae 309,310 subunits 83,84
385 sphingomyelinase 237 sucking stomach 230,231
- - spiders 385 Sphodros 381 suction pressure 43
shielding 218 spider bite 151,153 sugars 73
short -day 378 spiders (as prey) 256, see also sulci 129
Sicarius 832 Araneae sulfhydryl groups 166, 170,
sieve tracheae 79,81,87,88 spigot 169,174, 182ff., 188, 178,183
signal threads 214,381,383 193,194,200,201,210, sun shield 218
silk 49, 127, 128, 133, 135, spinnerets 44, 169, 170, super contraction 206
136,139,141, 160ff., 176, 173ff., 180, 182ff., 190, supercool 69,265,272
184,186,191, 203ff., 211, 208,209,215,216 supercooling point 69,70,
212,215,221,225,250, spiracles 46,47,49 72,74,266,339
276,351 spitting spiders 382 sur~val 294,335 ,364
- fiber 178, 182, 186 spool 169, 174, 184, 186 Sybota 196,384
- fibroin 204,208,237 spreading factor 146 symbionts 286
- glands 128, 161ff., stabilimentum 216ff. symbiosis 284
168ff., 173, 176,208,209, Staphylinidae 223 symbiotic 275,276,307,
211 starvation 76,234,238,266, 354
- ribbons 215 290ff. symphiles 277
- synthesis 208 static electric charges 202 Symphyta 261
Singa 70, 385 Steatoda 130,143,183, Syrnphytognatha 385
sit-and-wait predators 88, 321,384 synaptic vesicles 147
293,294,363 Stegodyphus 125,19733., synchronization 337,376
- strategy 89 201,233,253,384 syncytial 247, 248
Sitticus 301,388 Sternonyphantes 70,334, synthetic analogue 117
skin breathing 90 342,385 syrphid flies 257
smooth endoplasmic reticulum Stenochilus 383 systole 79,80,96
240,244,245 stenochronism 337ff.
snow 264, 266, 268 stercoral pocket 50, 239ff., Tachinidae 306,319
- cover 68,75 244,246,247 tachycardia 153,159
social 214,250,256,258, steroidogenous tissues 118 Tarna 384
259,285,377 sterols 14 tanning 5
sodium 81,238 stickiness 163, 201 tarantism 143, 157
solifugid 10 sticky 187,215,254,259 Tarentula 32,41,43,301,
somatolysis 25 - silk 261 302,321,385
Sonoita 138 - spiral 183, 203 taste-by-touch 262
Sossippus 301,385 - threads 221, 251 taurine 203
space web 138,251, 383ff. - webs 250 taurocholate 236
Speocera 383 stigma 79,86,90 taxonomic system 380
sperm 121ff., 131, 161, 165, sting 261 Tegenaria 12,40,76, 86ff.,
166,386 stink glands 261 90ff., 105, 107,127,176,
spermatheca 131 Stiphidius 387 214,229, 233ff., 252, 265,
spermathecal glands 129 stomodeum 230 321,335,340,359,361,
spermatogenesis 117 storage 238,242,244, 387
spermatophore 121ff., 126, 248 tegumentary cycle 113,114
131, 161, 162, 165, 166 - organ 233 Tekella 385
spermatozoa 122, 124 Storena 384 Telerna 79,81, 122ff., 130,
sperm competition 130 stress 248, 249 131, 162, 176, 178, 180,
- induction 121,125 stridulation 222 212,383
- priority 130 structural colors 21 temperature 332,334ff.,
- transmission 121,122 subnivean 265,267,268, 340,346,361, see also
- web 121,124,125,222 271 thermal
448 Subject Index
temperature adaptation 89 transpiration 44ff. viscosity 266

- perception 38 transport mechanisms 238 vitellogenesis 182, 340
- preferenda 51, 52 trapdoor 173,381
- regulation 56 treatment 156, 157 Wadotes 301
Tengella 387 triad 188,190 Walckenaera 129
termites 256,257 tributyrinase 235 wasp 106,225,256,259,
testes 121,122,131,165 Trichoptera 253 276,282,286
Tetrablemma 383 triglycerides 14 - venom 108
Tetragnatha 70,172, 188, Trite 136, 137,388 water content 40,42,74
214,252,301,334,341, Trochanteria 388 - gain 42
361,376,384 Trochosa 33,41,52ff.,57, - loss 14,26,28,36, 40ff.,
tetrapyrroles 18 67,68,75,321,385 44ff., 51ff., 64, 94
Teutana 69,70,72,127,384 tropical 68,336,354,355 - reabsorption 238
Textricella 385 tropopausic altitudes 352 - spider 387
Thaida 186,199,382 Trypoxylon 310ff. wax canal 11,13,14
Thanatus 71,73, 334,335, trypsine 236 - esters 14
341,386 tube 212,214,222 - layer 4
Thaumasia 354,355,386 - tracheae 78,79,81,86,90 web 358,362,363
Thelyphonus 10 tubule cells 238 - orientation 60
Theotima 383 tubuliform glands 176,178, - recycling 277
Theridion 53,54,57,70,72, 182,183 - reduction 191
222,252,257,301,321, turret spider 304, 306 - shaking 220
335,337,354,355,384 tyrosine 163,164,166,168, wedding gift 125, see also
Theridiosoma 385 172,178, 180ff., 186 nuptial
thermal adaptation 31, see - present 127
also temperature Uloborus 184,195, 200ff., Wendilgarda 385
- balance 63 219,384 winter 67,69,71,73
- - model 58,65 Ummidia 381 - activity 75,76, 264ff.,
- gradient 58 undercooling 265 338,339
- hysteresis 69 undulated fibers 197,198 - maturity 76
- preferenda 26,53,57,76 unsaturated fatty acids 73 winter-mature 269,339
- stress 31,58 uric acid 239,240,313 wolf spiders 385
thermal-hysteretic proteins Uroctea 128,176,178, wrapping 213,216,251,
266,339 180ff., 214, 215, 253, 257, 254,261,262
thermocouples 28,59 384 Wulfila 301,321,387
thermogenesis 29 Uroecobius 384
thermoreceptor 38 uterus externus 131 xanthommatin 17
thermoregulation 34, 57, 64, UV 20 Xenorhabdus 307
218 Xiphosura 9, 232
Thomisus 21,22,335,388 vacuoles 232 Xysticus 128,253,301,302,
Thysanoptera 250,251,253, valve mechanisms 245 306,322,388
261 vasa deferentia 121ff.,165
Tibellus 301, 386 vegetative center 153 yolk 230
tight junctions 244 venom 15,106, 142ff., 237,
Titanoeca 197, 386 314 Zearchaea 383
Tmarus 301,304,388 - glands 145,146,165, Zelotes 21,322,388
toxicity 150 237 Zodarion 376,384
toxin 146,147,157 ventilation 47 zoochromes 17
Toxopus 387 Verrucosa 301,385 Zora 301, 386
tracheae 46,47,78,90,92, vertical migrations 378 Zoropsis 124,193,232,386
94,96 vibrations 214, 221ff., 261, Zosis 178,184,195,200,
tracheal gland 163 282,316 218,384
- system 86 vibratory communication Zygaenidae 262
Trachelas 322,387 134 Zygiella 26,252,316,320,
transition temperatures 46 viscerocutaneous form 157 369,375,385
transmitter 147 viscid spiral 203, 204 zymogens 235

978 3 642 71552 5

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

978 3 642 71552 5

Uploaded by

Copyright:

Available Formats

Ecophysiology

With 133 Figures

Library of Congress Cataloging-in-Publication Data. Ecophysiology of spiders. Bibliography: p.

Recently another book on insect physiology was published. It was restricted

Regensburg, Autumn 1986 Wolfgang Nentwig

A Cuticle, Temperature and Respiration

I Chelicerate Cuticle Structure

B Glands, Silk and Webs

I Glandular Aspects of Sexual Biology

IV Comparative Structure and Histochemistry of Silk-Producing

I Structure and Function of the Digestive Tract

Nematode Parasites of Spiders

E Environment and Life History

I Life Cycles and Diapause

Appendix .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...... 380

References ............................... ..... 389

Subject Index .......................... ........ 437

Aitchison, Cassie W. 264 Macnab, Aynsley M. 133

It is almost impossible to exaggerate the significance of the cuticle in the life of an

1 Department of Zoology, The University, Manchester M13 9PL, England

2 Structure of the Arthropod Cuticle

2.1 Basic Subdivisions

2.4 Microstructural Features

2.5 Cuticular SensiUae

3 Structure and Ultrastructure of the Cuticles of the Chelicerate Groups

3.1 Eurypterid Cuticles

3.2 Xiphosuran Cuticles

3.3 Acarine Cuticles

3.4 Cuticles of Minor Chelicerate Groups

3.6 Spider Cuticle

tween successive layers; this is supported by evidence provided by densitometry and

Compared to the richness of integumental colors of numerous insects, coloration of

Whereas in other arthropods (especially in insects) compounds of most of the known

Ommochromes, as reported by Stamm-Menendez and Galarza Basanta (1961), Linzen

o o Fig. 6A,B. Structural formulas of

kynurenine and 3-hydroxykynurenine. All compounds were found to be granular in-

degradation products of the latter (Rudiger 1970). Compounds of the biliverdin IX

2.3 Purines: Guanine

2.4 Questionable or Unidentified Chromes

5.1 Morphological Color Change

in correspondence to the varying in tensity of environmental colors. Dark mature females,

5.2 Ontogenetic Color Change

Ontogenetic color change is developmental in Significance and the resulting coloration

5.3 Physiological Color Change

Based upon responses to environmental conditions, physiological color change is a rapid

cases color change results in a considerable darkening of the opisthosomal coloration

1 Zoologisches Institut der Universitiit Kiel, Zoophysiologie, Biologiezentrum, Olshausenstr. 40-60,

2 Background: Comparison of Spiders and Insects

GeneralAspects. Spiders,like all terrestrial arthropods, must keep fluctuations in body

3.1 The Body Temperature in Spiders

General Aspects. The Tb of an organism is determined by its metabolic heat produc-

Changes in Body Temperature. To summarize, Tb in spiders will always tend to ap-

3.2 Effects of Temperature

exhibits seasonal acclimatization as indicated by lower metabolic rates in June relative

3.2.2 Development, Growth and Reproduction

Fig. 11. Relation between the length of em-

3.2.3 Behavioural Activity

Stimulating Effects. Surface-dwelling lycosids, like Pardosa pullata, become most

Modifications. In Lycosa lenta and Filistata hibernalis, forced locomotion between

3.3 Evaporative Cooling

General Aspects. Evaporative cooling may not be expected to be significant in spiders,

3.4 Control of Temperature Responses

3.4.1 Temperature Perception