BIOTROPICA 37(3): 416–420 2005 10.1111/j.1744-7429.2005.00055.

Foliage Density of Branches and Distribution of Plant-Dwelling Spiders1

Andréa Lúcia Teixeira de Souza2

Programa de Pós–Graduação em Ecologia, Departamento de Zoologia, IB, Universidade Estadual de Campinas, CP 6109,
13.081.900- Campinas, SP, Brasil

and

Rogério Parentoni Martins

Laboratório de Ecologia e Comportamento de Insetos, Departamento de Biologia Geral, ICB, Universidade Federal de Minas Gerais, CP 486,
30.161-970- Belo Horizonte, MG, Brasil

ABSTRACT
We compared the abundance of foliage-living spiders among seven widespread plant species comprising a gradient of levels of structural complexity in a tropical
savannah-like region in southeastern Brazil. Spider abundance among plant species was positively related to the foliage density of their branches. A field experiment
using artificial branches was carried out to isolate foliage density effects on spider abundance, thus controlling both biomass and texture effects. Artificial branches were
attached to branches of three plant species with similar foliage density, Baccharis dracunculifolia, Diplusodon virgatus, and Microlicia fasciculata. Two treatments were
set up: artificial branches with higher foliage density attracted more spiders than those with lower foliage density. The guild structure of hunting spiders was compared
among vegetative branches of three plant species with different levels of foliage density: B. dracunculifolia, D. virgatus, and Bidens gardneri. Stalker, and ambusher
spiders were more abundant on branches of B. dracunculifolia, which had the highest foliage density. Foliage-runners constituted the dominant guild on D. virgatus
and B. gardneri, which have lower foliage density branches. Our results suggest that branch architecture is the most important factor determining the abundance of
plant-dwelling spiders in the study area independently of branch biomass, leaf surface area or texture.

RESUMO
A abundância de aranhas foi comparada entre sete espécies de plantas, que formam um gradiente de nı́veis de complexidade estrutural, descrito através do número de
folhas por unidade de volume do ramo, numa região de cerrado no sudeste do Brasil. A abundância de aranhas entre as espécies de plantas foi positivamente relacionada
com a densidade de folhas de seus ramos. Um experimento de campo, usando ramos artificiais, foi conduzido para isolar o efeito da densidade de folhas dos efeitos de
biomassa. Ramos artificiais foram amarrados a ramos de três especies de plantas com densidade foliar semelhante, Baccharis dracunculifolia, Diplusodon virgatus and
Microlicia fasciculata. Dois tratamentos foram usados, ramos artificiais com alta densidade de folhas atraı́ram mais aranhas do que os com baixa densidade de folhas. A
distribuição de guildas de aranhas errantes foi comparada entre ramos de três espécies de plantas com diferentes nı́veis de densidade de folhas, B. dracunculifolia, D.
virgatus e Bidens gardneri. Aranhas saltadoras (Oxyopidae e Salticidae) e de tocaia (Thomisidae) foram mais abundantes nos ramos de B. dracunculifolia, cujos ramos
possuı́am uma maior densidade de folhas. Aranhas caçadoras noturnas foi a guilda dominante em D. virgatus e B. gardneri. Nossos resultados sugerem que a arquitetura
de ramos pode ser o fator mais importante que determina a abundância de aranhas que habitam plantas, independentemente da biomassa, área superficial de folhas e
textura dos ramos.

Key words: Brazil; habitat architecture; patch-choice; spider distribution.

PLANT STRUCTURAL COMPLEXITY, THE ARCHITECTURAL ARRANGE-
MENT OF BIOMASS IN SPACE, has been recognized as one of the main
factors that determine the diversity and abundance of plant-dwelling
spiders (Gunnarsson 1990, Scheidler 1990, Hurd & Fagan 1992,
Halaj et al. 1998). Although causal mechanisms remain obscure,
some variables have been proposed to explain this pattern, such
as prey abundance (Allen et al. 1975, Nentwig 1993), availabil-
ity of refuges against predators (Askenmo et al. 1977, Gunnarsson
1996), and milder microclimate conditions (Henschell & Lubin
1997). A positive influence between plant architecture and spi-
der abundance was demonstrated through indices of structural
complexity such as foliage density in big sage Artemisia tridentata
(Hatley & MacMahon 1980), density of needles in spruce Picea
1Received 7 June 2002; revision accepted 26 January 2005.
2
Corresponding author. Current address: Depto. de Biologia, Setor Ecologia –
CCBS, Universidade Federal de Mato Grosso do Sul – CP 549, CEP 79070-900,
Campo Grande, MS, Brazil; e-mail: altsouza@nin.ufms.br
abies (Gunnarsson 1990), leaf surface area in different species of
Eucalyptus (Evans 1997), and needle density and branching com-
plexity in Douglas-fir branches Pseudotsuga menziesii (Halaj et al.
2000). However, these studies were restricted to comparisons within
a particular plant species or a few plant species of the same genus in
temperate systems. In addition, the effect of foliage density on spider
abundance is usually confounded with the effect of biomass of the
branches.
Spiders form a very diverse group, and several species may
actively choose favorable microhabitats (Turnbull 1973, Morse &
Fritz 1982, Morse 1990, Nentwig 1993). Their requirements for
structural characteristics of microhabitats can change among guilds,
since they differ in relation to hunting strategies, refuge and repro-
ductive behavior, and morphological and physiological features such
as visual acuity, tolerance to sunlight and shade, moisture, and ther-
mal conditions (Uetz et al. 1978, Uetz 1991, Foelix 1996). Thus,
we expected that spider guild distribution could be associated with

416
 Distribution of Plant-Dwelling Spiders 417

characteristics of the microhabitat such as leaf size and foliage den-

sity (Hatley & MacMahon 1980, Robinson 1981, Scheidler 1990,
Mason 1992).
In this study, we examined the spider distributions among plant
species of different genera and families in the Brazilian Cerrado. We
tested the structural complexity hypothesis that spider abundance
is positively related to foliage density of branches and leaf surface
area, and examined whether guilds of spiders vary among plants
with different levels of structural complexity.

METHODS

This study was carried out in an old-field site at the Estação

Ecológica do Panga, located 40 km south of Uberlândia district, SE
Brazil (19◦ 09 S, 48◦ 23 W). The ecological station includes plant
formations such as savannah-like (Cerrado), old fields, and forest
(Schiavini & Araújo 1989).
A component of plant structural complexity can be described
as a result of plant branching (see Kuppers 1989). Here, as a measure
of this component we used a foliage density index (FDI), defined
as: FDI = number of leaves on the branch/branch volume. We esti-
mated the ratio between the number of leaves and branch volume to
prevent the possible effect of branch volume on spider abundance
(see Scheidler 1990). Seven plant species that were abundant in the
study area and with different FDIs were sampled to test the effect of
structural complexity on spider abundance: Baccharis dracunculifolia
DC.(Asteraceae), Bidens gardneri Baker (Asteraceae), Chromolaena
laevigata (Lam.) King & H.Rob. (Asteraceae), Diplusodon virga-
tus Pohl (Lythraceae), Microlicia euphorbioides Mart. (Melastomat-
aceae), Microlicia fasciculata Mart.ex Naud. (Melastomataceae), and
Vochysia tucanorum Mart. (Vochysiaceae). Initially, a sample of ten
branches (one per individual plant) was collected from each plant
species to determine FDI. From each branch, we recorded the num-
ber of leaves and estimated the branch volume from measures of the
three dimensions, height, width, and depth, using the formula for
ellipsoid volume as proposed by Hacker and Steneck (1990). From
each branch, we also estimated leaf surface area from ten leaves with
a digital scanner. As FDI was highly positively correlated with mean
leaf surface area (r = 0.995, P < 0.001), we used only mean FDI
to test the effect of plant architecture on spider abundances. In the
analyses, we considered only the mean values of these variables for
each plant species.
SAMPLING OF SPIDERS.—Branches of the seven plant species, B. dra-
cunculifolia (N = 30 individual plants), B. gardneri (N = 21), C.
laevigata (N = 20), D. virgatus (N = 21), M. euphorbioides (N =
25), M. fasciculata (N = 15), and V. tucanorum (N = 14), were sam-
pled between December 1997 and February 1998, between 1700 h
and 1900 h to prevent possible effects of spider activity time in
the results. All sampled plants were 1–2 m high, and one branch
from each individual plant was collected. Each branch was carefully
enclosed in a plastic bag and cut at about 40 cm from its tip. The
plastic bags were large enough to avoid contact with the branches,
thus preventing the spiders from escaping (see Gunnarsson
FIGURE 1. Schematic design of artificial branches used as treatments.
Branches with high (A) and low (B) foliage density were attached to completely
cleared natural branches.
1983, Halaj et al. 1998). No spider was observed escaping from
the branches. The plastic bags were sealed in the field, and filled
with CO 2 at the laboratory before examination. The spiders were
separated and fixed in 70 percent ethanol.
The relationship between spider abundance and FDI was tested
with a complete randomized experimental design with two factors,
plant species and foliage density. Three plant species were used:
B. dracunculifolia, M. fasciculata, and D. virgatus. The foliage den-
sity of the branches was manipulated by replacing natural branches
with artificial ones. The artificial branches consisted of wire and
plastic branch models with 24 leaf stems, and leaf models made of
cloth to prevent the effects of chemical components and texture of
the natural branches (Fig. 1). In each individual plant, a branch
was cleared of leaves and leaf stems to be used as support of the
branch model. A branch model was tied to the support branch
with a thin wire. Two treatment levels, high and low foliage density
were used. Each branch model had 24 leaf stems (Fig. 1a and b).
In the high foliage density treatment, each leaf stem had four model
leaflets, totaling 96 model leaflets (Fig. 1a). In the low foliage density
treatment, two model leaflets were attached to 12 leaf stems total-
ing 24 model leaflets, whereas the remaining stems had no leaves
(Fig. 1b). In the high foliage density branches, each leaf model area
was 1.5 cm2 , whereas in the low foliage density branches each model
leaf area was 6.0 cm2 . Thus, total leaf surface area and biomass were
kept constant in both treatments. Thirty-two artificial branches of
each treatment were attached to different plant species: 24 plants
were used for B. dracunculifolia and M. fasciculata (12 in each
treatment), and 16 in D. virgatus (eight in each treatment). After
1 week, the artificial branches were recovered also between 1700 h
418 Souza and Martins
and 1900 h, as made for natural branches. The branches were col-
lected with the same procedure used to sample natural branches, to
allow data comparisons.
GROUP STRUCTURE.—The abundance of each group was compared
by sampling natural branches of three plant species with different
levels of foliage density. The FDI was estimated from ten individuals
of each plant species as described above. Baccharis dracunculifolia
(FDI = 28.82 ± 1.50) (mean ± 1 SD, N = 10) had higher foliage
density, whereas D. virgatus (FDI = 10.34 ± 1.28) and B. gard-
neri (FDI = 1.37 ± 2.03) presented intermediate and low foliage
density values, respectively. The three species had significantly dif-
ferent values of mean foliage density (F 2,27 = 98.49, P < 0.001;
all differences significant at P < 0.05 according to Tukey’s multi-
ple comparisons test). To evaluate the influence of plant structural
complexity and spider guild structure, 90 branches from each of the
three plant species were sampled, one branch from each individual
plant.
The spiders are usually classified into two large groups, wan-
dering and web-building spiders, due to differences in behavior
and use of silk to capture prey (Turnbull 1973, Uetz et al. 1999).
In this study, the web-builders Araneidae, Theridiidae, and Dic-
tynidae made up only 11.4 percent of the individuals and were
pooled into a single group. The guilds of spiders followed Uetz
et al. (1999), and those used in the frequency analyses were foliage-
runners (Anyphaenidae and Clubionidae), stalkers (Oxyopidae and
Salticidae), and ambushers (Thomisidae). The remaining wander-
ing spiders represented only 5.2 percent of the sample and were not
analyzed.
DATA ANALYSIS.—The relationship between spider abundance and
foliage density (FDI) was analyzed with linear regression. Each repli-
cate consisted of the mean number of spiders per branch in each
plant species. The experiment using artificial branches with two
structural complexity levels was evaluated with a one-way ANOVA.
Residuals of each analysis were graphically checked for nonnormal-
ity, and data were log10 transformed to obtain a normal distribution
and equal variances. G-tests were used to test differences on spider
group abundance (Sokal & Rohlf 1995).
RESULTS
The mean number of spiders per branch was linearly related to
the structural complexity index (Y = −1.723 + 0.334X , P =
0.014, R 2 = 0.73; Fig. 2). The most abundant spider families in
the sampled plant species were Anyphaenidae (23.1%), Oxyopidae
(20.0%), and Salticidae (18.5%). The other families recorded were
Clubionidae, Thomisidae, and Pisauridae comprising 9.2 percent
of the total; the remaining 29.2 percent were web-building spiders
(Araneidae, Theriididae, Dictynidae, and Scytodidae).
The mean number of colonizing spiders on branch models
was similar among the three plant species (F 2,58 = 0.089, P =
0.873). However, more spiders colonized the high foliage density
FIGURE 2. Relationship between the mean number of spiders per branch
and foliage density index (FDI), in logarithmic scale. Line is the least-squares
fit, bars indicate the standard deviation.
branches (0.625 ± 0.117, mean ± SE) relative to the low density
ones (0.219 ± 0.074), a significant difference (F 1,58 = 2.503,
P = 0.007). The different colonization trends between the two
treatments followed the same pattern among plant species, indicated
by the nonsignificant plant species × treatment interaction (F 2,58 =
0.005, P = 0.992). The only arthropods that occurred on the
artificial branches were spiders, in both treatment levels.
A total of 211 spiders was recorded on branches of B. dra-
cunculifolia, D. virgatus, and B. gardneri. Patterns of spider guild
abundance differed among the three plant species (G = 22.48, df =
6, P = 0.001). The abundance of web-building spiders did not differ
among the three plant species (G = 1.75, df = 2, P = 0.417). When
each wandering spider guild was tested separately, foliage-runners
was the most abundant guild comprising about 52.0 percent of all
spiders on both D. virgatus and B. gardneri, and its frequency also
did not differ between the plant species (G = 1.79, df = 2, P =
0.409). On the other hand, the abundance of stalker and ambusher
spiders were significantly different from that expected by chance
(G = 23.09, df = 2, P < 0.001; G = 8.39, df = 2, P = 0.015,
respectively; Fig. 3).
DISCUSSION
The abundance of plant-dwelling spiders was positively correlated
with the branch foliage density on which they occurred. This trend
seems to be independent of plant taxon since most of the variance
on spider abundance (73%) was explained by only this component
of branch structural complexity, even comparing plant species from
different genera and families. In addition, the experiment using
vegetative branch models strongly suggests that branch architecture
have significant effects on spider abundance even though branch
biomass, texture, and leaf surface area available for colonization to
spiders were kept constant. The effect of branch structural com-
plexity on the abundance of spiders and other arthropods has been
exhaustively demonstrated in the literature (Hatley & MacMahon

FIGURE 3. Relative abundance of spider groups in different plant species:
BAD: Baccharis dracunculifolia; DIV: Diplusodum virgatus; BIG: Bidens gardneri.
1980; Gunnarsson 1990, 1992; Scheidler 1990; Halaj et al. 2000).
However, in all of these studies the effects of biomass and/or surface
area available for colonization were confounded with the architec-
ture of branches. Romero and Vasconcellos-Neto (2005) carried out
an experiment comparing three levels of leaf reduction. Their re-
sults showed that leaf cutting negatively influenced the abundance
of jumping spider Psecas chapoda. Experiments involving the ma-
nipulation of branches through leaf or branchlet removal and tying
branches together to alter foliage density also altered the biomass
and surface area available for colonization by arthropods, both to
spiders and their prey, a problem already raised by some authors (e.g.,
Halaj et al. 1998, 2000). The experimental design used in this study
enabled us to maintain both biomass and surface area constant. Fur-
thermore, the treatments consisted of artificial material, rendering
these patches less attractive to insects, the potential spider prey. In
fact, no other arthropods were found on treatment branches. The
positive relationship between the number of spiders and branch
foliage density is usually explained by the fact that patches with
higher structural complexity attract a larger number of prey, and
because they provide more hiding places and milder microclimates
to spiders (Askenmo et al. 1977, Hurd & Fagan 1992, Nentwig
1993, Gunnarsson 1996, Evans 1997, Henschell & Lubin 1997).
At least for some spider species, however, refuge availability and/or
mild microclimate conditions may be more important than prey
availability (Robinson 1981). We found a strong positive relation-
ship between spider abundance and foliage density independently
of plant taxon, an interesting result since different plant species vary
in their nutritional content, potentially influencing the abundance
of herbivorous insects. In our field experiment, no arthropods, ex-
cept spiders, were attracted to artificial branches made of plastic,
wire, and cloth, and more spiders were found in the higher foliage
density treatment. In several systems, especially in the tropics, prey
availability may not be a limiting factor to spiders (Wise 1993).
Experiments manipulating the availability of prey to spiders could
help clarify this question.
Distribution of Plant-Dwelling Spiders 419
Spider guild structure differed among branches of B. dracun-
culifolia, D. virgatus, and B. gardneri. When the distribution of each
group was analyzed separately, the distribution of stalker and am-
busher spiders on the three plant species was significantly different
from that expected by chance. In addition, foliage-runner was the
dominant guild on D. virgatus and B. gardneri, species with low
foliage density (Fig. 3). On the plant species with highest foliage
density, B. dracunculifolia, we observed no dominant spider guild.
Scheidler (1990) and later Halaj et al. (1998, 2000) suggested that
differences in the spider fauna on different plant species indicate
the existence of spider associations for specific characteristics of the
habitat. The narrower diet of some spider guilds can constrain their
distribution (see Nyffeler et al. 1994), whereas predation can be
the most important factor determining the distribution of other
guilds (see Provencher & Vickery 1988, Ehmann & MacMahon
1996). Spiders are frequently preyed upon by birds (Askenmo et. al.
1977, Gunnarsson 1983) and arthropods such as mantids (Moran
& Hurd 1994), wasps, and other spiders (Wise 1993, Nyffeler
& Sterling 1994, Wise & Chen 1999). From all hunting spiders
recorded on vegetative branches of the three plant species studied,
only foliage-runners, mostly Anyphaenidae, are regarded as noctur-
nal, spending the diurnal periods sheltered inside tubular retreats
built of silk threads in the foliage; the remaining guilds have diur-
nal activity (see Foelix 1996, Uetz et al. 1999). Our results agree
with the hypothesis that night-active groups such as clubionids and
anyphaenids should differ in their distribution and requirements
relative to diurnal-active spiders such as oxyocids, salticids, and
thomisids. Diurnal spider groups are probably more vulnerable to
predation by visually oriented hunters (see Wise 1993, Gunnarsson
1996). Thus, it is possible that anyphaenids and clubionids can
escape from visually oriented predators such as Salticidae, Oxyopi-
dae, or birds, mantids, and wasps on plants with low foliage density.
However, further experiments are necessary to evaluate the effect
of refuge availability on the distribution of diurnal families such
as Thomisidae, Oxyopidae, and Salticidae as opposed to nocturnal
ones such as Anyphaenidae, Miturgidae, and Clubionidae.
ACKNOWLEDGMENTS
We are grateful for useful comments of M. O. Tanaka, T. Lewin-
sohn, and J. Halaj. We thank A.A.A Barbosa for plant identification
and Universidade Federal de Uberlândia for logistical support. This
study and the research scholarship of the senior author were funded
by CNPq.
LITERATURE CITED
ALLEN, J. D., H. J. ALEXANDER, AND R. GREENBERG. 1975. Foliage arthropod
communities of crop and fallow fields. Oecologia 22: 49–56.
ASKENMO, C., A. VON BRÖMSSEN, J. EKMAN, AND C. JANSSON. 1977. Impact of
some wintering birds on spider abundance in spruce. Oikos 28: 90–94.
EHMANN, W. J., AND J. A. MACMAHON. 1996. Initial tests for priority effects
among spiders that co-occur on sagebrush shrubs. J. Arachnol. 24: 173–
185.
420 Souza and Martins
EVANS, T. A. 1997. Distribution of social crab spiders in eucalypt forests. Aust.
J. Ecol. 22: 107–111.
FOELIX, R. F. 1996. Biology of spiders, 2nd ed, 330 p. Oxford University Press,
Oxford, England.
GUNNARSSON, B. 1983. Winter mortality of spruce-living spiders: Effects of
spider interactions and bird predation. Oikos 40: 226–233.
———. 1990. Vegetation structure and the abundance and size distribution of
spruce-living spiders. J. Anim. Ecol. 59: 743–752.
———. 1992. Fractal dimension of plant and body size distribution in spiders.
Funct. Ecol. 6: 636–641.
———. 1996. Bird predation and vegetation structure affecting spruce-living
arthropods in a temperate forest. J. Anim. Ecol. 65: 389–397.
HACKER, S. D., AND R. S. STENECK. 1990. Habitat architecture and the abun-
dance and body-size-dependent habitat selection of a phytal amphipod.
Ecology 71: 2269–2285.
HALAJ, J., D. W. ROSS, AND A. R. MOLDENKE. 1998. Habitat structure and prey
availability as predictors of the abundance and community organization
of spiders in western Oregon forest canopies. J. Arachnol. 26: 203–
20.
———, ———, AND ———. 2000. Importance of habitat structure to the
arthropod food-web in Douglas-fir canopies. Oikos 90: 139–52.
HATLEY, C. L., AND J. A. MACMAHON. 1980. Spider community organiza-
tion: Seasonal variation and the role of vegetation architecture. Environ.
Entomol. 9: 632–639.
HENSCHELL, J. R., AND Y. D. LUBIN. 1997. A test of habitat selection at two
spatial scales in a sit-and-wait predator: A web spider in the Namib
Desert dunes. J. Anim. Ecol. 66: 401–413.
HURD, L. E.,, AND W. F. FAGAN. 1992. Cursorial spiders and succession: Age or
habitat structure? Oecologia 92: 215–221.
KUPPERS, M. 1989. Ecological significance of above-ground architectural pat-
terns in woody plants: A question of cost-benefit relationships. T. Ecol.
Evol. 4: 375–379.
MASON, R. R. 1992. Populations of arboreal spiders (Araneae) on Douglas-firs
and true firs in the interior Pacific Northwest. Environ. Entomol. 21:
75–80.
MORSE, D. H. 1990. Leaf choices of nest-building crab spiders (Misumena vatia).
Behav. Ecol. Sociobiol. 27: 265–267.
———, AND R. S. FRITZ. 1982. Experimental and observational studies of patch
choice at different scales by the crab spider Misumena vatia. Ecology 63:
172–82.
MORAN, M. D., AND L. E. HURD. 1994. Short-term responses to elevated
predator densities: Noncompetitive intraguild interactions and behavior.
Oecologia 98: 269–273.
NENTWIG, W. 1993. Spiders of Panama: Biogeography, investigation, phenology,
check list, key and bibliography of a tropical spider fauna, 274 p. Sandhill
Crane Press, Gainesville, USA.
NYFFELER, M., AND W. L. STERLING. 1994. Comparison of the feeding niche of
polyphagous insectivores (Araneae) in a Texas cotton plantation: Esti-
mates of niche breath and overlap. Environ. Entomol. 23: 1294–1303.
———, W. L. STERLING, AND D. A. DEAN. 1994. Insectivorous activities of
spiders in United States field crops. J. Appl. Entomol. 118: 113–128.
PROVENCHER, L., AND W. VICKERY. 1988. Territoriality, vegetation complexity,
and biological control: The case for spiders. Am. Nat. 132: 257–266.
ROBINSON, J. V. 1981. The effect of architectural variation in habitat on a spider
community: An experimental field study. Ecology 62: 73–80.
ROMERO, G. Q., AND J. VASCONCELLOS-NETO. 2005. The effects of plant struc-
ture on the spatial and microspatial distribution of a bromeliad-living
jumping spider (Salticidae). J. Anim. Ecol. 74: 12–21.
SCHEIDLER, M. 1990. Influence of habitat structure and vegetation architecture
on spiders. Zool. Anz. 225: 333–340.
SCHIAVINI, I., AND G. M. ARAÚJO. 1989. Considerações sobre a vegetação da
Reserva Ecológica do Panga (Uberlândia). Soc. Nat. Uberlândia 1: 61–
66.
SOKAL, R. R., AND F. J. ROHLF. 1995. Biometry. 3rd edition, 887 p. Freeman,
New York, USA.
TURNBULL, A. L. 1973. The ecology of true spiders (Araneomorphae). Annu.
Rev. Entomol. 18: 305–348.
UETZ, G. W. 1991. Habitat structure and spider foraging. In S. S. Bell, E. D.
McCoy, and H. R. Mushinsky (Eds.). Habitat structure: The physical ar-
rangement of objects in space, pp. 325–348. Chapman & Hall, London,
England.
———, J. HALAJ, AND A. B. CADY. 1999. Guild structure of spiders in major
crops. J. Arachnol. 27: 270–280.
———, A. D. JOHNSON, AND D. W. SCHEMSKE. 1978. Web placement, web
structure, and prey capture in orb-weaving spiders. Bull. Br. Arachnol.
Soc. 4: 141–148.
WISE, D. H. 1993. Spiders in ecological webs, 328 p. Cambridge University
Press, Cambridge, England.
———, AND B. CHEN. 1999. Impact of intraguild predators on survival of a
forest-floor wolf spider. Oecologia 121: 129–137.