Phil. Trans. R. Soc. Lond.

B 311, 5-23 (1985) Printed GreatBritain in

[ 5 ]

Sedimentological, ecologicaland temporal patterns fossil of Lagerstaitten

BY A.
SEILACHER,

W.-E. REIF

AND F. WESTPHAL

Institut Museum Geologie Paldontologie, und fur und Sigwartstrasse 10, D-7400 Tiubingen, F.R.G.
[Plate 1]

Preservation of non-mineralized structures (including plants) and of articulated skeletons resultsfrom extraordinaryhydrographic,sedimentational and early diageneticconditions.The corresponding chiefcausativeeffects (stagnation,obrutionand bacterial sealing) define a conceptual continuum into which individual occurrences may be mapped. A more pragmatic, typological classification of conservation deposits, using a standard questionnaire, reveals ecological replacements,as well as trendsrelated to the evolution of the biosphere, throughgeological time. The theme of this symposium, 'extraordinary fossilbiotas: their evolutionary and ecological significance', recalls a project carried out some time ago in the Tubingen special research division 53 on palaeoecology. Our term 'fossil Lagerstatten' (alternative spelling: Lagerstaetten) is difficultto translate into English. Corresponding to economic Lagerstatten of minerals and ores, fossilLagerstatten were definedas rock bodies unusually rich in palaeontological information, either in a quantitative or qualitative sense. This means that the termembraces not only strata with an unusual preservation,but also less spectacular deposits such as shell beds, bone beds and crinoidal limestones.The concept also implies that there is no sharp boundary with 'normal' fossiliferous rocks. Rather, the preservationof any fossilis to be considered as an unusual accident that deserves attentionand questioning. The rationale behind such a broad approach is a sedimentologicalone. Fossil Lagerstatten are considered as end members of ordinary sedimentaryfacies,in which the unusual amount and quality ofthe palaeontological testbodies allows us to identify betterthe factorsresponsible forsuch a facies at all levels of its genesis: biotope conditions (palaeobiology), fate of the soft parts and organic skeletal material (necrolysis),sedimentarytransportand burial (biostratinomy) and fate of the mineralized skeletons (fossildiagenesis). On the other hand this approach is based on the contentionthat in spite of all the possible combinations and variations of biotic and abiotic factorsinvolved there is a finitenumber of situations that lead to the formationof fossilLagerstatten and that prospecting for them is a therefore realisticobjective. A first step in thisdirectionwas a geneticclassification(figure1), the broad outlines of which are still valid. In the meantime,our understandingofconcentration has deposits been greatlyimproved by the concepts of dynamic and event stratigraphywith far reaching applications in basin analysis (see Einsele & Seilacher (I982) and Bayer & Seilacher (I985) forexamples and references). With respectto conservation on deposits, which thissymposiumis focused,many new examples have been discovered and old ones analysed in more detail, including case historiespresented in thissymposium.Each of themhas,- course,its meritas a peephole in the screenofimperfect of

A. SEILACHER,

W.-E. REIF

AND

F. WESTPHAL

Ad4&~

^non-marine t

concentration deposits 0e>t .. t'<. ... condensationdepos. placer deposits concentration traps
conservation deposits

. ..j.-. .

~~~~~~~~~~marine-

V
%v<?0* ~~~ ~
a

g?

fv?<\

stagnationdeposits
~obrution deposits conservation traps

~ ~

~~~~

~~~

FIGURE

1. Synopsis of as Lagerstatten givenby Seilacher& Westphal(I97I). and classification fossil details and new, commonly

preservation, through which we can see unknown anatomical diagenetic boundary conditions in our classification.

problematic taxa. But each case is also important as an additional test for environmental and Instead of adding new examples, we shall here choose three well studied and representative examples from the Jurassic of southern Germany as a reference, and try to derive from them a questionnaire that can be applied to other conservation deposits. In doing so, we shall emphasize the invertebrate rather than the vertebrate, microfossil and plant records, because invertebrates cover a wider spectrum with regard to geological time, biomaterials and environments.

1. THE

ECHINODERM

LAGERSTATTE AN OBRUTION

OF GMUND DEPOSIT

(L. SINEMURIAN):

Being only a few centimetres thick and possibly no more than a few metres in lateral extent, this deposit can not compete with those of Holzmaden and Solnhofen. But its fossil content, the stratigraphic context and low level of genetic complexity as well as a careful analysis by Rosenkranz (I971) make it a very suitable prototype of what we call obrution deposits. situation (a) Stratigraphic Stratigraphically, the fossil-bearing lens of black shale is separated from contact with the

Keuper marls by a conglomeratic shell bed 0.5-0.6 cm in thicknessand underlyingterrestrial

PATTERNS

OF FOSSIL

LAGERSTATTEN

overlain by a similarconglomeraticlimestone.Palaeontological evidence, however,shows that this basal conglomerate of the Jurassic transgressionrepresentsin fact a rather complex environmentalhistory. The erosional nature of the Triassic-Jurassic contact is shown firstby the absence of the Rhaetic, which in this area is usually representedby coal-bearing swamp deposits which pass into marine sands and clays fartherto the west. Second, the conglomerate consistslargely of reworked calcareous concretions which are characteristic of the underlying Keuper marls ('Knollen-Mergel'). Third, the sole of the conglomeratic bed is covered by casts of vertical rhizocoralliidspreiteburrows,whose perfectly preservedscratchmarkssuggestthat the Keuper Such burrows marl, into which they were dug, was already well compacted and rather stiff. ichnofaciesof shallow marine firmgrounds. While other of the Glossifungites are characteristic are species of the ichnogenus Glossifungites oblique and in the size-class of the related the smaller ones at the base of the Jurassic can be compared to the vertical Rhizocorallium, and burrows of the modern amphipod Corophium may well be the works of related intertidal crustaceans. In any case these burrows show that the conglomerate is not simply the depositional phase of the event that made the erosion. Rather, the two processeswere separated by a quiescent firmgroundperiod lasting long enough to allow the establishment of the Glossifungites community.Moreover, deposition of the conglomerate bed at this place was not preceded by major erosion, which would have destroyedthe burrows. Nor does the conglomerate bed itselfrepresenta simple sedimentational process. Besides lithoclastsit contains many shells,predominantlyof byssate and cemented epifaunal bivalves (Plagiostoma, Lima, Entolium, Inoceramus, Liostrea).But poorly preserved specimens of Cardinia indicate that mud-burrowerswere also present originally, but were largely eliminated by diagenetic solution of theiraragonitic shells.Accordingly,we interpretthisbed as a composite record of muddy periods interruptedby major stormevents. These winnowed the mud away, was stopped by the growinglag ofshellsand pebbles, which later allowed but theirerosiveeffect the colonization by an epifaunal post-eventcommunityof epibyssate and cemented bivalves beforethe mud took over again. This interplay of sedimentation, shell production and storm reworking,comparable to situations known from modern littoral environments (Seilacher I985), was periodically interruptedby the deposition of black mud containing the echinoderms,which, unlike the previous and followingmud-layers,locally escaped subsequent stormerosion. The previous style of storm-winnowing, plus the lateral introductionof reworked Keuper in the thickerconglomeratic bed overlyingthe echinoderm horizon. But nodules, continued its top is modelled by large oscillation ripples,suggestingthat the eventual shiftto the muddy sedimentationofthe PsilonotusClays happened between thesand and the mud phase ofa single tempestite. and (b) Faunal spectrum preservation The well preserved echinoderm fauna (30 asteroid, 15 ophiuroid, 15 crinoid, 40 echinoid specimens; Rosenkranz (I 97I)) is embedded in thelower 1-2 cm oftheclay lens (figure2, 1-5). It represents ordinaryshellybottom communitywith respectto functionaladaptation, troan and age structure.Its only unusual featureis the well articulated preservationin phic diversity a muddy sedimentnot correspondingto the substrateon which such forms would normallylive. It is surprising, however, that groups other than echinodermsare not equally well represented.

8
'[

A. SEILACHER,
. pelagic
realm

W.-E. REIF

AND

F. WESTPHAL
washed-in |
-.

*,>

,V>

'S1>.
N

benthos

air lift

zCoo =
o
' 34

32

36

<9~~~~~3
".~~~3

i~~~~~~~~~~4 4 >F09t 0 ~
41

47(C)S

50

43~L

(Tb

46
i /447 201& K
,

0;

OV

412
4

ff

45'

realm pelagic

epiplanktic '~~~~ ~

1.benthic events

~~ ~27 ~
~~~~~~25
4,

bD N

~~~I i
4

14)

16

~~~~~~~~~~~~~~~~~~~~2

D26'

&Jinfaunal

;~~~~~~~~~~~~

~~~*4142

lli

fi
"0

~~~~~~~~~~~~~ ~~'

"~~~~?./ ~

ordinary ~~~benthos

4~~~~~~~~*

~~~~~~~~~~~~~~C
2~~~~~~~~~~~~~~~~~

(a)

PATTERNS

OF FOSSIL

LAGERSTATTEN

The only common non-echinoderm associate is a small oyster (Liostreairregularis) which encrustedpebbles as well as shellsofitsown kind. Such oysterclusters(figure2, 1) may comprise several generations,the changing orientationsof which attest to repeated overturnduring an extended exposure at the sediment surface. However, the generations could also represent different shellyphases separated by mud-coveredperiods. Nevertheless, Rosenkranz found that about 50 % ofthe oysters top ofthe basal shell bed were double-valved and probably became on victimsof the same event that killed the echinoderms. In contrast,epibyssalbivalves, being represented withinthe basal shell bed, are conspicuously lacking on its echinoderm-bearingtop. Rosenkranz's explanation is convincing: the shellybottom epifauna became buried alive by rapid mud sedimentation.Among the bivalves, this was fatal for the oysters,while pectinids and limids could escape by swimmingup. For the echinoderms,however, the fine sediment also had a smotheringeffectbecause it blocked their ambulacral systems.It is this 'Achilles heel' that accounts forthe high-leveltaxonomic selectivity theirpreservation. of (c) Other examples As Rosenkranz (I971) has shown, many, if not the majorityof articulated echinodermsin the fossilrecord are found in a correspondingsituation: at the tops of condensed tempestitic

FIGURE

2. Ecologicalspectraofrepresentative fossil in Jurassic Lagerstatten southern Germany. showsequencesofgenerations 1, Reclining clusters oysters of separatedbyoverturning burial?)events. (and articulated of diverse 2-5, The exclusive, preservation ecologically echinoderms that fell suggests they victims to a mud-smothering event. as 6-1l, Amongthe vertebrates well as the invertebrates Holzmaden fauna is dominatedby pelagic the Theirperfect film in organisms. preservation (associated aptychi, periostracal and zig-zagsiphuncle ammonites; inksackspreserved jet and other partsin coleoids8, 9) suggests in soft rapidsettling after deathand theabsence In ofbenthic scavengers. thebelemnites, bitesshowsoft the onlycarcassessunkby predator parts(including non-calcified lost proostracum, whichwereotherwise duringthenecroplanktonic 9), drifting stage (11). 12-21, Normally benthic groupsare mainlyrepresented forms by attachedto floating objectssuch as drift wood (12-24) and belemnite rostrum an drift specieswhosenon-calcified permitted extended necroplanktonic (15), or to live ammonites (16-21). For detailssee Seilacher(I982). are 22-23, Crustaceans represented of onlyby veryrarespecimens apparently benthic forms. The factthat corresponding speciesare also foundin depositsof the Solnhofen type (40-41; Osteno; Lebanon) suggests, thattheywerenot normalbenthos. however, 24-30, Trulybenthic organisms (including foraminifers ostracods) characterized smallsize and and are by occurrence particular on monotypic beddingplanes.This suggests thatthey colonizedthemudduring benthic in eventsand then became smothered an articulated fashion.Main representatives epibyssate are bivalve recliners echinoids is (24-26) and small, long-spined (27). Endobenthos represented tiered Chondrites horizons by bivalves(29, 30 from (28) and veryrareburrowing Riegraf(1977)). the 31-36, In Solnhofen pelagic guild is similarto Holzmaden,but containsadditionalelements such as crustacean larvae (34), jellyfish Saccocoma reconstruction), whichin placesis the (35) and themicrocrinoid (36, mostcommonfossil. is of absenceofdriftwood be a reason);butitincludes 37-39, Epiplankton lessdiverse (rarity ammonites, may bivalvesfrom floats assembled byssate shellsofhibolitid belemnites smallammonites and by necroplanktonic seaweeds(39) wereprobablywashedin from shore. (37). Overgrown the crustacean of 40-41, Amongthediverse fauna,thecommonoccurrence Mecochirus withtraceofdeath (40, of march)and ofEryon (41) is reminiscent Holzmaden. benthic do benthic but 42-49, In Solnhofen truly organisms notoccuras autochthonous horizons, as lateral import turbidity by currents. Some ofthemwerestillalive enoughto leave a short track(42-44), others dead, orientation contortion showingbelly-uplanding marks(5; figure current 5), (46) or postmortem possibly of indicative hypersaline bottom waters.Immobileand burrowing forms (excepta rareSolemya, werenot 43) imported. 50-51, Rare, but extremely diverseinsects(most commonly dragonflies), like the flying reptilesand Archaeopteryx, represent import air or surface by currents.

10

A. SEILACHER,

W.-E. REIF

AND

F. WESTPHAL

shell beds. It could be added that thissituationis mostlikelyto occur in the transgressive phases of larger or smaller cycles. In all cases, smothering finersedimentis the dominating process,while oxygen deficiency by in thepore water (gyttjacondition) was eitherabsent or helpfulonlyin thesense thatit inhibited infaunal scavengers or bulldozers that would have secondarily disarticulated the buried skeletons.

2. THE

BITUMINOUS

POSIDONIA

SHALES

OF

HOLZMADEN

(L.

TOARCIAN):

A STAGNATION

DEPOSIT

Since the famousfossilLagerstatteofHolzmaden and itsfossilcontenthave been adequately described (see Kauffman i98I; Riegraf I984), we can focus here on the points that are importantfora comparison with the other examples. (i) While the Gmund deposit is at the very base of theJurassic transgression, Toarcian the marks its peak. This situation favours the establishment of anoxic conditions in many epicontinental basins, whether this is due to changes in the hydrographic regime or to the mobilization of brines fromunderlyingsalt deposits (Jordan I974). (ii) In contrastto the Gmund echinoderm bed, here we are dealing with an enormous rock body that spreads over large parts of central Europe with a thicknessof tens of metres and a representing time span of the order of millionsof years. This difference importantto keep is in mind in view of the neverendingdiscussions as to whetherwe are dealing with a sapropel underneath a stratified,oxygen deficient water body or with a gyttja, in which oxygen was essentiallyrestricted the pore water. In such spatial and temporaldimensions, to deficiency both conditionsmust be expected at different timesand places, as theyare in modern stagnant basins (Savrda et al. I984). The question is simply,which situation dominated in space and time and was responsible for the unusual preservation (including softparts), with which we are concerned here. (iii) The ambiguitybetween the two alternative,or ratheralternating,models is expressed in contrasting sedimentologicalas well as ecological evidence. Sedimentologically,the presence lamination that can be correlatedover tensofkilometres(H. Roscher, personal of a millimetric communication) as well as landed ammonites that dropped their overgrowth before they eventually tiltedover without changing place or position (Seilacher I982) indicate veryquiet water. On theotherhand, the orientationand linear accumulation ofammoniteshellsand other currents up to 20 cm s-1. of at fossils many levels (Brenner 1976; Seilacher I 982) recorduniform (iv) In the ecological spectrum (figure2, 6-30) we observe the general absence of benthic organisms. Encrusting or reclining brachiopods, bivalves, serpulids and crinoids, otherwise standard elements of Liassic soft bottoms, are represented only as epiplanktonic float on ammonite shells (figure2, 16-20), belemnite shells with a non-calcifiedrostrum(figure2, 15) wood (figure2, 12-14, Seilacher I982), while nektonicforms(ammonites,coleoids, fish, or drift ichthyosaursand plesiosaurs) are the dominant fossils.On the other hand thereare individual horizons covered by monotypic small epibenthics, such as diademoid echinoids or byssate bivalves (Posidonia, figure2, 24-26), whose articulated state ofpreservationwould be in conflict with large-scale lateral import.More common are horizonswith abundant benthic microfauna (foraminifers, ostracodes; Riegraf 1985), while tieredbioturbationhorizons,dominated by the low-oxygentrace fossilChondrites (figure2, 28; Bromley & Ekdale 1984) are restricted a few to levels, which become more numerous towards the margins of the basin.

PATTERNS

OF FOSSIL

LAGERSTATTEN

11

This pictureis in accord with observationsin modern stagnant basins, where benthicfaunas decrease in body size and diversitywith decreasing oxygen levels and advance towards the centre of the basin during periods of reduced stagnation (Savrda etal. I984). In the shallower Posidonia shale basin, such benthic events were probably brought about by extreme storms, but they allowed only a few generationsto flourishbeforeabiotic conditions took over again. 3.
THE SOLNHOFEN AN

LITHOGRAPHIC

LIMESTONES DEPOSIT

(TITHONIAN):

OBRUTIONARY

STAGNATION

(a) Stratigraphic environmental and setting This fossil Lagerstatte, most famous for the preservation of feathered Archaeopteryx and medusae, was formedjust before the regressiveend of the Jurassic cycle. With a thicknessof up to 90 m it exceeds the south German Posidonia Shale, but both its lateral extent (basins a fewkilometres diameter in a buried sponge-reeftopography) as well as its time equivalent in (0.5 Ma) are considerably smaller. Also different the scarcityof fossils:had there been no is quarries, the unit would have possibly been mapped as 'non-fossiliferous'. Afterall the criteria for emergence (Limulustracks interpretedas bird tracks, desiccation cracks, rain drop impressions,etc.) have been discredited,the old lagoonal model is no longer relevant. Most authors now agree that we are dealing with permanentlysubmerged restricted basins, in which stormwave action was confinedto a few metresof depth and that there was some kind of stratification the water body excluding macrobenthos. This stratification in had probably the formof a halocline (Keupp 1977) and it is a secondary question, whetheror not the resultingstagnation was also expressed by an oxycline. In any case the bitumen content, even ifsecondarilylost by weathering,was originallylower than in the Posidonia Shales, owing to the higherrate ofcarbonate mud sedimentationand lower productivity the surfacewater. in (b) Ecologicalspectrum As in the Posidonia Shales, the fauna (figure2, 31-51) is dominated by pelagic organisms. But instead of ammonites, the most common fossilis the minute stemless crinoid Saccocoma (figures2, 36 and 3b), whose anatomy and preservation(see below) suggestsa medusoid-like mode of life. There is also a large varietyof fishand aquatic reptiles,but among these, landrelated formsare more common than the open-ocean ichthyosaurs. Epiplanktonic fauna is also present,but on a broader spectrumof floats.Driftwoodis hardly ever found, probably because in such small water bodies it would always be washed ashore. Since ammonites are rarer as a whole, the small percentage of overgrownshells amounts only to a few specimens. Some of them carry oysters,other lepadomorph barnacles (figure2, 38), which in Holzmaden occur only on wood (figure2, 13). Instead we find fouled belemnites, not in the formof an encrustationthat occurs when the rostralie at the bottom,but as a loose attachmentof byssal bivalves that preferthe conotheca ratherthan the rostrum(figure2, 37). Obviously, the now dominant hibolitid belemniteshad a longer necroplanktonicdrifting time, which made them more attractive for hitchhikers. The common association of several belemnites,or of belemnitesand ammonites, in such clusterssuggeststhat the bivalves could actively enlarge their float by byssal assemblage. Algal frondsare yet another kind of float, but in this case we deal with seaweeds that were torn offtheirrockylittoralhabitat by storms and washed into the basin. Benthic elements are by no means missing in the Solnhofen limestones, but unlike the

12

A. SEILACHER,

W.-E. REIF

AND

F. WESTPHAL

Holzmaden situation they never representautochthonous burrow or other benthic horizons (except perhaps of Foraminifera; Groiss I967). Instead we are dealing with inhabitantsof the shallow margins that became accidentally washed-in. This view is supported by several lines of evidence. Most spectacular are the death marches recorded by tracks behind carcasses in an environmentin which other tracks are never found. The length of these tracks relative to the animal is largest in the low-oxygen bivalve Solemya(figure 2, 43) and decreases (with (figure2, 40) and Eryon(figure2, 41). decreasing hardinessin Limulus(figure2, 42), Mecochirus Other crustaceans are never found with a track,but the marksleftby the dorsal side of Penaeus show that theseswimmers (figure2, 45 and figure5, plate 1), or by the dorsal finsofvarious fish, were already dead when they landed at the bottom (Mayr I967). A second indication is the selectivityof benthic import. With the exception of Solemya neitheris sessileepibenthos. (figure2, 43), no endobenthos (including burrows) is represented, Mobile epibenthic species are found,particularlyones that, on perturbation,would be able to swim up, such as Antedon (figure2, 47), Limulus(figure2, 42) and a varietyof crustaceans. It is also noteworthy that almost all importedlimulidsare juveniles not adults, whose much larger sizes can be reconstructed from tracks (not death marches) found in marginal areas of Solnhofen basins (Painten). This could be because juvenile limulids more readily swim up on more prone to be washed away than adults. disturbance and are therefore (c) Necrolytic features deformationsof drying A dorsal bend of the vertebral column, comparable to post mortem in SolnhofenArchaeopteryx, Pterodactylus modern carcasses (Schafer 1955) is a familiarfeature and Compsognathus, occurs also in fishes.In the lattercase it cannot be a desiccation feature, but because associated belly-up landing marks and tail finsdisrupted during the bending of the vertebralcolumn (figure7; Mayr I967) clearlyshow that the deformationtook place only after the carcass had sunk to the bottom. Thereforeit may be assumed that the bottom water was hypersaline to effect the necessarydehydration. deformations may be observed in articulated skeletonsofinvertebrates. Analogous necrolytic In a carcass of Penaeus (figures2 and 5) with a belly-up landing mark, the ventral flexureof the tail scraped offsediment afterthe body had tilted over. A possibly related phenomenon the feather-likeform of the arm tips is the contortion of Solnhofen crinoids. In Saccocoma, in is invariably obscured by intenseinward coiling (figure3 b). Antedon, contrast, (figure2, 36) coiled (figure3c) and sometimesdisrupted (figure2, has the proximal partsofthe arms intensely we 47), while the tipsare conspicuouslystraight.From such deformations may derive flexibility species (figure3d, e), but we may also patterns,and hence the mode of swimmingin different take them as environmentalsignals that seem to be lacking in Holzmaden-type deposits. (d) Biostratinomic features in azimuth The scarcityoffossils Solnhofenprecludes the statisticalmeasurementofpreferred orientations.Strong currentsare indicated, however, by tool and roll marks leftby ammonite shells (Seilacher I963) and other objects. Their association with incipient flute casts further position suggeststhat thesecurrentswere sediment-loaded. This is in accord with the preferred of the Solnhofenfossilsat the bases of beds and with the occurrence of grading in other Upper Jurassic occurrences (Nusplingen; Temmler I964). Turbidity currentsare also indicated by

PATTERNS

OF FOSSIL

LAGERSTATTEN

13

Cenomanian,Haqel (Lebanon)

(d)
1cm

Tithonian,Solnhofen

Kimmeridgian,Cerin FIGURE 3. Post mortem contortionof echinoderm carcasses in lithographic limestonesreveals flexibility patterns along

thearmsas wellas thedehydrating effect hypersaline of bottom waters.(a) Antedon U. Cerin. thiollieri, Jurassic, contortion thearmtipsin a current-oriented of carcass.(b) In Solnhofen Drag marks (dotted)suggest secondary Saccocoma contortion in thepinnulated is armtips(see figure 36) suggesting that (GPIT 1630/1) maximum 2, these weretheactivepartsin filter whilethearmbaseswereheldout as an umbrella.(c) Solnhofen swimming, in arm tips moved as oars by the moreflexible had stiff 'Antedon', contrast proximalparts. (d) This small comatulid a style contortion has of similar (c) (GPIT 1630/2). (e) In thelarger to armsections species, proximal are coiled likein (c), whilethedistalportion in deformed an arcuatefashion (GPIT 1630/3). Scale bars are I cm.

theradialcurrent directions fossil orientations aroundthemuchmorefossiliferous mappedfrom and betterexposedlithographic limestone basinsof the Lebanese Cretaceous(Huickel1970, is contrast theuniform to current of directions thePosidoniaShales figure which in marked 13),
(Bre-nne-r
TI976).

14

A. SEILACHER,

W.-E. REIF AND F. WESTPHAL

(e) Diagenetic features

As to be expected in such differentlithologies, certain biomaterials suffereddifferent in transformations bituminous shales and lithographiclimestones.In Solnhofen,forinstance, both vertebratecoprolites and the ink sacks of coleoid cephalopods are phosphatic, while in Holzmaden only coprolites are phosphatized, while the ink is pr'es-erved jet. In-both cases, as however, mineralizationhappened so early that these originallysoftmaterials did not become deformedby compaction. Aragonitesolutionseemsto have been a similarly earlyprocess,as shown by thepreservational historyof ammonites (Seilacher et al. 1976). Their compactional deformation,varying with different shell geometries,shows that the aragonite was dissolved not at the surfacebut within the sediment. On the other hand the plastic lateral deformationof Solnhofen ammonites in slumped layers ('Krumme Lagen') indicates that the shells had already been reduced to periostracal filmswithin a few metresbelow the sediment surface. The real problem is not the difference,but the identity of diagenetic signatures in the Holzmaden and Solnhofen cases in spite of their lithological differences.The flattened periostracal foil preservationof empty ammonite phragmocones that the two localities have in common contrastssharplywith 'normal' situations,in which such phragmocones are either or filledwith blocky lined with pyrite (non-bituminousLiassic black shales; Hudson (I982)) littlecompacted. This indicalcite (micriticlimestonesof the UpperJurassic) and are therefore were cates that the special conditionsresponsibleforthe exceptional preservationofotherfossils also on the pore water system. not restricted the open water, but had an effect to

(f) Prokaryotic scum: theneglectedfactor

viewed as theresultofphysicochemicalprocesses.Withinthe upper Diagenesis is traditionally of centimetres thesedimentcolumn, however,in which theobservedsoft-part permineralization and aragonite solution appears to have taken place, chemical conditionsare mainly controlled by microbiological activity. the microbiologyof ancient mud bottomsis largelya geochemical task. Still To reconstruct there are a few morphological clues. Scanning electron microscope studies by Keupp (I977) have demonstratedthat, apart fromcoccoliths (Hemleben 1977 b), coccoid cyanobacteria are

DESCRIPTION FIGURES

OF

PLATE

in 4-9. Cyanobacterial factor preservational can mats,probablya mostimportant processes, be inferred indirect evidence.In Solnhofen, their is and near-absence of from presence indicatednot onlyby lamination but features. erosion, also by surface fish tail (from FIGURE 4. Halo ofripped-up scumaroundswaying Barthel(1978), plate 61, figure 1). FIGURE 5. Preservation landingmarks of (rostral carina,eye tips,abdomen)besidesa fallen-over carcassofPenaeus and ofscrapemarkplus scumheap producedduring inflection tail end (GPIT 1630/4). of post mortem FIGURE 6. Ruffling aroundrollmarkofan ammonite shell(GPIT 1630/5) and drag marks(Eichstatt Museum). GPIT 1630/6) aroundfish with FIGURE 7. Cast ofreticulate ridgepattern(bottom surface, post mortem contortion ofvertebral column. FIGURE 8. Radial arrangement such ridgesaround an ammonite of shell,whose 'pedestal' elevationis a later feature. diagenetic (FromJanicke(I969), plate 6, figure 4.) FIGURE 9. Corresponding patternproduced by mechanicalcreasingof a depressurized blister.(Courtesyof Professor Otto, University Stuttgart.) Frei of

Phil. Trans. Soc.Lond. volume 1 R. 31 B,

7~~~~~~~~~~~~*,,K

et Seilacher, al., plate1

~~~~~~~~~~

*.~~~~~~~~~~~~~~~~~aek

f~~~~~~~~~~~~~~~1t

%~~~~~~~~~~ .~~~~~~~~~~~~~~

-'/~~~~~~~~~~~~~~~%

1,~ 147
-

~
9%20%7

~~~~~~~~~~
~~~~~~~~~~~~~~~~~~~~~~~~~a A

lb(b~

rv

FIUES49.Fr

ecrpio

eeopoie

4j~~~~~~~~~~(aigp

PATTERNS
_

OF FOSSIL

LAGERS TATTEN ~~~~~~~~~~~~(e)

15

.
(d) pelagic
(b) sessile,

~~~~~~~~~~~~~.

shalefacies)and Solnhofen bituminous FIGURE10. Stagnant basinsoftheHolzmadentype(mostly type(lithographic

short benthic benthos carbonates) bothlack autochthonous (exceptduring events)and are dominated drop by allowsalso thelateralimport, turbidity of fauna.The geometry Solnhofen-type basins,however, by currents, of oflittoral benthos, preferably vagileforms.

a major constituentof the Solnhofen muds. Their presence is also expressed macroscopically by a scum on the bed surface.This scum allowed tracksand roll marks to be preserved,caused the ruffling roll and drag marks (figure6; Mayr I967, plate 13, figures1-3) and was visibly of ripped offaround a swaying fishbody (figure4). Its presence is also expressed by reticulate ridges on the tops of Solnhofen beds (figure7), which have been variously explained as rain drop impressions(Mayr I967), syneresiscracks (Janicke I969) or load casting. Their radial arrangementaround an ammonite (figure8), however,suggeststhat we deal with the creasing of a filmsimilar to the tepee structuresof modern algal mats or the mechanical creasing of blistermembranes (figure9). From otherevidence Hemleben & Freels (I 977 a) have considered such prokaryoticscum as the chieffactorin the Cretaceous lithographiclimestonesof Hvar, Yugoslavia. Today, cyanobacterial filmsof this kind are largely restricted hypersalineenvironments. to in But, like cyanobacterial stromatolites, theycould have had a much wider distribution earlier would be (Keupp I977): times,particularlyin the Precambrian. Their effect (i) to protectsoft sediments against erosion; (ii) to favour the preservationof tracks and other markings; (iii) to serve as food source during benthic events; (iv) to protect carcasses against decay; (v) to act as a 'carbonate pump' into the sediment (Walker & Diehl, this symposium); (vi) to seal the particular microenvironment responsibleforthe absence of bioturbation (Krumbein I983) and forthe unusual preservationalhistoriesof ordinaryfossilssuch as ammonite shells. As yet, this eludes macroscopic investigation,but it may become more transparentthrough ultramicroscopic studies in ancient rocks and particularly through microbiological and geochemical analyses in adequate modern environments.

16

A. SEILACHER, 4.

W.-E. REIF
OTHER

AND

F. WESTPHAL

EXAMPLES

As epitomized by the faunal spectra and modes of preservation,the Holzmaden and the Solnhofen deposits have obviously much in common. Differences are minor and largely palaeogeographic frames (figure 11). On the one hand we have a explained by the different of fromthe margins,but withthepossibility storm-induced large stagnantbasin withlittleinflux establishmentof small-sized and monotypicbenthic faunas. water-mixingand the short-term only the shallow margins; In the much smaller basins of the Solnhofentype,stormscould affect currents, but the steeper slopes favoured the episodic introduction,by slumping and turbidity of fine-grainedsediment and transportable benthic organisms from the more oxygenated nearshore environments.In such an event, the imported bodies (some still alive) would reach soon to be covered by the mud settlingfromsuspension. Therefore,we would the bottom first, definethe Solnhofenlithographiclimestonesprimarilyas a stagnationdeposit,but one in which obrution was also an importantfactor. There are many counterpartsto these two typesin the Phanerozoic record,each with minor but significant pecularities.Thus the bituminousshales of the Lower Lias ofsouthernGermany (Olschiefer,Sinemurian) resemble Holzmaden except that, due to the thicknessof only a few decimetresand poor outcrops,theirfauna is much less spectacular; but theydo contain benthic horizons of small echinoids ('Cidaris' olifex)as well as tiered bioturbation horizons. Equally comparable are the Upper Triassic bituminous shales of southernSwitzerland, but the shales themselveslack periostracal impressionsof ammonite shells,which are only found in the hard dolomitic layers, where they were fixed by early diagenetic cementation (Rieber I973). A lessstagnantmodification theHolzmaden typeis foundin marginal areas ofthe Toarcian of basin, where benthic horizons are more common and the aragonitic ammonite shells are not dissolved. The same appears to be trueforparts ofthe Oxford and Kimmeridge clays in Britain, where the benthic element includes reclining and bone-encrusting oysters (Martill, this symposium; Aigner i 980). Nevertheless,the shales are bituminous and vertebrate skeletons have remained articulated. of The sponge reeftopographyand the regression the UpperJurassic has led to the formation of lithographic limestones in other parts of central Europe also. In Nusplingen (south of is Tubingen) fauna and preservationsare similar to Solnhofen except that Saccocoma lacking, some beds are clearly graded and the mud consistslargely of sponge spicules, indicating that sponge reefsaround the small basin were still growing (Temmler I964, I966). The faunal differenceof the French locality of Cerin may be largely due to its slightly older age (Kimmeridgian, Bernier et al. I983). In this case, however, the muds did at times emerge, as shown by horizons with dinosaur tracks. of representatives the Solnhofen type are in the Cretaceous The most similar extra-Jurassic of Lebanon. These lithographic limestones are associated and interbedded with huge slump subsiding pull-apart basins related to the nearby masses in what appear to be rhythmically Jordan rift system(Huckel I970; Hemleben I977 a). Fossils are very common and diversified, contortionsof fishand with washed-in benthos dominating over the pelagic rain. Post mortem echinoderm skeletons (figure3) suggest a halocline. In the mid-Triassic lithographicrocks of Alcover (Spain; Esteban Cerda et al. I977; Via Boada I977; Hemleben & Freels I977 b) the relation to an ancient reef topography in a regressing sea resembles Solnhofen. Minor differences preservationare due to the dolomitic nature of the rock, which, togetherwith in gypsum crystals,also suggestsincreased salinityin this case.

PATTERNS

OF FOSSIL

LAGERSTATTEN

17

Identification becomes-more difficult withincreasingtemporaldistancefrom thejurassic-type examples. The lithographic limestones of Monte Bolca (Eocene, N Italy; Sorbini I983), probably deposited in a volcanogenic topographyby graded turbiditesedimentation,can still be compared to Solnhofen,although many index groups of organisms have fallen victims to the Cretaceous extinction.But what about the 'lithographic' Green River Shales (Eocene) of North America? Being deposited in a basin too large to produce turbiditycurrents,are they rather a stagnation deposit comparable to the Posidonia shales, but in a salinityregime that favoured bacteria-mediated carbonate precipitation? Similarly enigmatic is the case of the thin-bedded chalks of the Niobrara formation.In addition to the famed vertebrateskeletons and the floatingcrinoid Uintacrinus shells that appear too large to have theycontain Inoceramus been epiplanktonic. However, the double-valved, closed preservationof these giant bivalves, equal encrustationon both valves and theirlying always on the same valve does not eitherfit a recliningmode oflife.Also recliningoysters, which are such a regular elementin otherchalks, are conspicuously absent. Could it be that the Niobrara chalks are the equivalent of a dark bituminousshale at a timewhen excessiveplanktoniccarbonate productiondiluted thesapropel into a white mud without the necessityof lateral carbonate import? On the other hand, the Lower Liassic deposits of Osteno (see Pinna, this symposium) are dark-coloured, finely laminated and bituminous. But they consist, like the lithographic limestonesof Nusplingen, largely of sponge spicules. Also theirfaunal spectrum relates them to Solnhofenratherthan to Holzmaden, with crustaceans,wormsand otherwashed-in benthos dominating over the pelagic guild and with benthic events (including bioturbation horizons) being completely absent. The preservation of the rare ammonites also resembles Solnhofen (flattened,non-pyritized, zig-zag siphuncle). What we can learn fromthe comparisonsso faris that lithology,colour and bitumen content should not be taken as the primary criteria in a genetic classificationof fossilLagerstatten, because they may change with the hydrographic situation and with shiftsin biological carbonate mud production. Nor should the salinityfactorbe rated too highly. The Jurassic deposits of the Karatau (Kazakhstan), forexample, are a valid counterpartof the Solnhofensituationalthough theywere deposited as dolomitic muds in a tectonic,probably hypersalinelake, in which the 'washed-in benthos' is mainlyreplaced by river-imported insects. Also deposited in a tectonic lake are the Eocene shales of Messel, Germany (Franzen, this washed-in land and flying symposium). Their ecological spectrum (fishes, vertebrates, insects) resemblesKaratau, but thissedimentis bituminousand non-calcareous withoutindicationsof increased salinity.Neverthelessthe complete absence of autochthonous benthos suggeststhat the stagnant condition was permanent and not interruptedby either storms (small basin geometry) or floods (thermocline). if Environmental identifications become still more difficult we deal with Palaeozoic or even Precambrian examples because modes of life are increasinglydifficult assess. There is no to problem in tracing back the obrution deposits of the Gmund type as long as we have echinodermsto go by. The Holzmaden type is also well representedby bituminousshales such as the Ohio shales ofthe Upper Devonian and Lower Carboniferousofthe easternUnited States (Barrow & Ettenson I980). But we do have problems to identify the Solnhofen type. The Silurian Mississinewa shales of Indiana, forinstance,were deposited in inter-reef basins, in which a turbiditicmode of sedimentationis indicated by grading and position of the fossils at the bases of the siltydolomite beds (Erdtmann & Prezbindowski I974~). But the fauna is more diverse than in Solnhofen and includes such immobile benthic forms as sponges,
2

Vol. 3

II.

18

A. SEILACHER,

W.-E. REIF

AND

F. WESTPHAL

brachiopods and a large variety of dendroid graptolites. Also the beds are commonly bioturbated from the top, indicating a rather diversifiedautochth-onousbenthos. But the dominant dendroid graptolite genus is Dictyonema, generally taken to have been pelagic. Also there are many kinds of cephalopods (little compacted, but not pyritized) and perfectly, three-dimensionally preservedcrinoid crowns,whose arms and pinnules are integratedinto an umbrella that would fita pelagic mode oflifewell. Thus we are dealing with a simil-ar situation to Solnhofencombined with a benthic elementunfamiliarin Mesozoic lithographiclimestones. There appears to be some similarity,however, to argillaceous dolomites of similar age in Wisconsin (Mikulic et al., this symposium). Also unfamiliar,from a Mesozoic point of view, are the Hunsruck Shales of the Lower Devonian of Germany (Kuhn i96I; Seilacher & Hemleben I966), because they combine soft-part preservation(non-mineralizedskeletonsofarthropods; articulation) withpyritization and a diversebenthicfauna including a varietyof burrows.Siltymicroturbidites a common are element that also allows the preservationof various arthropod tracks. Current action is also indicated by sedimentary structures and fossil orientations, while wave ripplesare conspicuously absent. These featuresand the predominance of echinoderms suggest that obrution was the major factor; but oxygen deficiencymust also have been involved, ifonly in the formofgyttja, to allow soft-partpreservation. It is also noteworthythat the echinoderms lack post mortem contortions. Still strangerformsof conservationdeposits are representedin the Cambrian. Among these we note the combination of a trilobite shell hash suggestive of storm winnowing with a bituminous lithologyand phosphatized arthropod 'soft parts' in the Upper Cambrian Alum Shales (Muller; thissymposium) and, of course, the famous Burgess Shale (Collins etal. I983; and the absence Conway Morris,thissymposium). Here the problematic nature of many fossils of more familiar groups such as echinoderms and molluscs make it difficult interpretthe to ecological spectrum.As a whole, the benthicelementappears to dominate in the BurgessShale, but in a washed-in rather than a benthic-eventfashion. However, there are some assumed pelagics such as Marrella, the most common faunal element. This looks like the Solnhofen spectrum,though a non-carbonate version. The discoveryof new localities in addition to the classical one suggests,however, that we deal not simply with a unique physical setting,but that a kind of time signature is also involved. The most challenging problem in this respect is the classification of the Ediacara-type conservation deposits of Vendian age (Fedonkin, this symposium): a task that is largely independent of the taxonomic problems involved (see Seilacher (I984) and this symposium). There is no question of dealing with the impressionsof soft-bodiedorganismsin a sandy facies that contrastswith the fine-grainedsedimentsdiscussed so far. It is true that we do know sandy obrution deposits also in the Phanerozoic. But they refer to vertebrateskeletonsin continental regimes,to exceptional turbiditeevents (fishdeposit of Sendenhorst; Siegfried I954) or to sand-smotheredechinoderms (Seilacher I968) and not to soft-bodiedorganisms. The Vendian fossiliferous sandstones,in contrast,appear to be largely storm-generated(Goldring & Curnow I967), that is, depositionally similar to many shallow marine sandstones in the Phanerozoic, in which such soft-bodyimpressionsnever occur. To invoke forthis unusual preservationa unique physiographicconstellationis no morejustified, given that Ediacara-type fossilshave been found in more than 20 localities all over the world (Glaessner 1984, figure1.8) and also in sandstonesthat are claimed to be deep-water turbidites

PATTERNS

OF FOSSIL

LAGERSTATTEN

19

(Anderson & Conway Morris I 982). Clearly thispreservationmustbe due to an 'anactualistic' factor (in a Phanerozoic view) of a more global nature. It has been claimed thatthisfactorwas a lack ofbioturbation.But burrowsmade by worm-like organismsand possibly coelenterateshave been found in many Ediacaran fossillocalities and with surprisingly advanced sediment feeding strategies. It should also be remembered that microbial degradation would destroya shallowly buried soft-bodiedcarcass. A more relevant clue may come from sedimentologicaldata (J. Gehling,unpublished,and personal observations cyanobacterial scum that protected in the field). They suggestthe presence ofa ratherresistant these sand surfacesagainst stormerosion and amalgamation and allowed flexiblesand-shards to be ripped off.This would go along with the general impression that cyanobacterial mats (with or without stromatoliticstructures)were much more widespread in Precambrian than featureshould have in later environments;but in exactly what way this non-uniformitarian influencedthe preservationof soft-bodiedorganismsstill remains to be explained. 5.

CONCEPTUAL

FRAME

AND

CLASSIFICATION

Our shortreviewofconservationdeposits is admittedlyveryincomplete. For instance it does not cover a group that we previouslycalled 'conservation traps' (Seilacher & Westphal I 97 I), cases as mammoths in permafrostcrevices, amber which is a holding bag for such different in our presentdiscussionare only the earlydiagenetic concretions insectsor bogs. Of significance in the formof coal balls (Scott & Rex, this symposium), 'orsten' (Muller, this symposium), Mazon Creek nodules (Baird etal., thissymposium) or the Lower Cretaceous fishconcretions fromBrazil (Muller, this symposium). Certainly,such concretionsare importantin that they facilitate fossil hunting and that they provide, through the right preparation techniques, But in a genetic sense, theyare only a subset of stagnation (and relativelyuncompacted fossils. obrution?) deposits. In the Posidonia Shales, for instance, beautifully preserved reptiles, fish and coleoid cephalopods come fromsuch concretions; but most of the nodules are barren and most of the fossilsare found in layers in which no such concretionsoccur.
stagnation (hydrographicregime) Holzmaden

Solnhofen
S

Bundenbach Gmund obrution (sedimentational regime) Ediacara

Hvar

bacterial sealing (early diagenetic regime)

FIGURE 11. Among the many factorsinvolved in the formationof conservation deposits, stagnation, obrution and

cyanobacterial sealing are considered the most dominant. They define a conceptual continuum, into which particular examples may be ' mapped'.
2-2

20

A. SEILACHER,
TABLE

W.-E. REIF AND F. WESTPHAL

QUESTIONNAIRE FOR CONSERVATION DEPOSITS

1. TENTATIVE

Tubingen questionnaire

Fossillagerstatten Conservation deposits

locality: age: (6) ecologicalspectrum burrows (episodic,continuous): tracks(withor without bodies): endobenthos: hemisessile vagile epibenthos (episodic,continuous): sessile vagile able to swim pelagics: nekton floaters epiplankton wood, cephalopod (on live or dead) shells, terrestrial organisms: tracks skeletons land plants(twigs, leaves,trunks) criteria (7) necrolytic death marches: landingmarks(live,dead): soft parts(impressions, films): organiccuticles: articulation (vertebrates, echinoderms, arthropods, bivalves, aptychi): (8) stratinomic criteria lifepositions: rollmarks(ofwhat?): convexup or down: current orientation (azimuth): wave orientation (azimuth): (9) diagenetic criteria aragonite preservation (what?): earlyaragonite solution (composite casts): pyritic steinkerns: concretionary cementation (nucleus, pressure shadow,buckle,pedestal): compactional deformation incoalation: phosphatization what?): (of replacement (shells, bones):
General conclusions: flyers

(1) basinsituation size (km): 10- 10? 101 102 103 setting: oceanic epicontinental terrestrial origin: tectonic volcanic astroblemic subrosional recifal sedimentary glacial frame: geographic limestones clastics crystalline (2) stratigraphy thickness(m): 0.01 0.1 1 10 100 in duration absolutetime: vertical context: transgressive peak transgression regressive fining-up cycles coarsening-up cycles lateralsequence: (3) sedimentology lithology: biograins (coccoliths, forams, radiolarians, spicules, etc.): structures: sedimentary lamination (varves, algal, etc.) slumphorizons gradedhorizons current ripples wave ripples emersion marks mud cracks, (tidal channels, etc.) (4) geochemistry evaporitic precipitates (aragonite, calcite,dolomite, gypsum, halite): concretions pyrite (globular, discoid): (particles, kerogene, bitumen): (5) taxonomic spectrum dominated (priority): by echinoderms cephalopods vertebrates crustaceans trilobites others:
Corg

isotopic deviations:

or stagnation (thermal halocline): obrution: algal sealing:

PATTERNS

OF FOSSIL

LAGERSTATTEN

21

In spite of its incompleteness this review has shown, however, that, apart from the two principal factorsof stagnation and obrution and theirvarious combinations, a large number of palaeogeographic, biological, sedimentological,diagenetic and time factorsmay contribute to the formationof conservationdeposits. The resultis that each case has a 'personality' that defies rigorousclassification. Of course, thereis some virtue in typological groupings. Thus we may distinguishbetween marine, hypersaline, lacustrine and swamp deposits, or between vertebrate, echinoderm, arthropod and plant Lagerstatten. Or group into Solnhofen, Holzmaden, Bundenbach and Burgesstypesofdeposits. Grouping according to age (Conway Morris,thissymposium;Walker & Diehl, thissymposium) may also be revealing. But while being usefulto emphasize patterns, such classifications can never be binding, because unlike in taxonomy there will be no agreement as to which criteriashould have primacy. Neverthelesswe should not be content with descriptiveregistrations.Instead of expecting the classificationto provide a standard set of pigeon holes, we should better consider it as a conceptual frameworkfor heuristic purposes. In figure 11 we propose to use the critical hydrographic, sedimentationaland early diagenetic exponentsto definea triangularspace into which each particular case can be mapped. In addition, we presenta tentativequestionnaire (table 1) in order to standardize this mapping job, in which by necessitya large number of specialists must be involved. What lies in front us, is more than fossil of hunting,the unravellingoftaxonomic relationships and facies analysis. Expanded over the whole fossilrecord, a comparative analysis of fossil Lagerstatten could become a genuine contributionof palaeontologists to the integratedview of Earth history.
REFERENCES

T. Aigner, I980 Biofabrics stratinomy theLowerKimmeridge and of Clay (U. Jurassic, Dorset,England). N. Jb. Geol.Paldont. Abh.159, 324-338. Anderson, M. & ConwayMorris, I982 A review, M. withdescriptions four of S. unusualforms, thesoft-bodied of fauna ofthe Conceptionand St John'sGroups (late Precambrian), Avalon Peninsula, Newfoundland. Third N. Am.Paleont. Proc.1, 1-8. Conv., F. Barrow,L. S. & Ettensohn, R. I980 A bibliography paleontology paleoecologythe and ofthe of Devonian-Missisippian black-shale in sequence North America. pages. Morgantown,West Virginia: U.S. DepartmentEnergy, 86 Information Center. K. Barthel, W. 1978 Solnhofen. BlickindieErdgeschichte. pages. Thun: Ott-Verlag. Ein 393 Bayer,U. & Seilacher, I985 Sedimentary evolutionary A. and cycles.In Friedmann, M. (ed.) Lecture in G. notes earth vol. sciences, 1, 465 pages. Berlin, New York,Tokyo: Springer. Heidelberg, Bernier, Barale,G. & Bourseau, P., J.-P. I983 Le chantier fouilles de paleoecologiques Kimmeridgien de superieur Exc. 1, 1-19. de Cerin (Jura,France). Int. Paleoecol. Guide Congr. Lyon, Brenner, K. 1976 Ammoniten-Gehause Anzeigervon Palao-Stromungen. Jb. Geol.Paldont., als Abh. 151, N. 101-118. Bromley, G. & Ekdale,A. A. I984 Chondrites: R. a tracefossil indicator anoxia in sediments. of Science, Wash. 224, - 872-873. Collins,D., Briggs, & ConwayMorris, I983 New Burgess D. S. Shale fossil sitesrevealMiddle Cambrianfaunal complex.Science, Wash.222, 163-167. Einsele,G. & Seilacher,A. (ed.) I982 Cyclic event and 536 stratification. pages. Berlin,Heidelberg,New York: Erdtmann, B.-E. & Prezbindowski, 1974 Niagaran (Middle Silurian) interreef D. fossil burial environments in Indiana. N. Jb. Geol.Paldont., Abh.144, 342-372. Esteban Cerda, M., Calzada, S. & Via Boada, L. 1977 Ambiente depositionaldelos yacimentos fosiliferos del Muschelkalk de superior Alcover-Mont-Ral. Cuadernos Iberica 189-200. Geol. 4, Freels,D. 1975 Plattenkalk-Becken Pietraroia(Prov. Benevento, bei S-Italien) als Voraussetzung einerFossillagerstattenbildung. Jb. Geol.Paldont., N. Abh.148, 320-352.
Springer.

22

A. SEILACHER,

W.-E. REIF AND F. WESTPHAL

Glaessner, F. I 984 Thedawn animal A biohistorical 244 pages. Cambridge:University M. of life. study. Press. R. Goldring, & Curnow,C. N. I967 The stratigraphy faciesof the late Precambrian Ediacara, South and at Australia. geol.Soc.,Aust.14, 195-.214. J. Groiss, T. I967 Mikropalaontologische J. der im Schichten Gebiet um Eichstiitt Untersuchungen Solnhofener (SiidlicheFrankenalb).Erlanger Abh.66, 75-92. geol. im Hemleben,C. 1977a Rote Tiden und die oberkretazischen Plattenkalke Libanon. N. Jb. Geol.Pal., Mh., pp. 239-255. und in Hemleben, I977 b Autochthone allochthone C. Sedimentanteile den Solnhofener Plattenkalken. Jb.Geol. N. Pal., Mh., pp. 257-271. und Hemleben,C. & Freels,D. 1977a Algen-laminierte gradierte Plattenkalke der Oberkreide in Dalmatiens N. Abh.154, 61-93. (Jugoslawien). Jb. Geol.Paldont., Hemleben,C. & Freels,D. 1977 b Fossilfuhrende dolomitisierte Plattenkalke dem 'Muschelkalksuperior'bei aus Montral(Prov. Tarragona,Spanien). N. Jb. Geol.Paldont., Abh.154, 186-212. in Hudson,J. D. I982 Pyrite ammonite-bearing shalesfrom Jurassic England and Germany. the of Sedimentology 29, 639-667. von des Huckel,U. 1970 Die Fischschiefer Haqel and Hjoula in der Oberkreide Libanon. N. Jb. Geol.Paldont., Abh.135, 113-149. Janicke, I969 Untersuchungen den Biotopder Solnhofener V. uber Plattenkalke. Mitt.bayer. Staatssamml. Palaont. hist. Geol.9, 117-181. als im Jordan,R. 1975 Salz- und Erdol/Erdgas-Austritt Fazies bestimmende Faktoren MesozoikumNordwestDeutschlands. Geol.Jb.,Reihe Heft13, 64 pages. A, E. Kauffman, G. I 98 I Ecologicalreappraisal theGermanPosidonienschiefer of basin (Toarcian) and thestagnant model.In Communities past (ed. J. Gray,A. J. Boucot & W. B. N. Berry);pp. 311-381. Stroudsburg ofthe Pa: Dowden. und Keupp, H. 1977 Ultrafazies Geneseder Solnhofener Plattenkalke Abh. (ObererMalm, sudliche Frankenalb). Ges. naturhistor. Nurnberg 128 pages. 37, Krumbein, E. I983 Microbial W. 330 chemistry. pages. Oxford:Blackwell. Schiefer. Kuhn, 0. I96I Die Tierwelt Bundenbacher der Neue Brehm-Bucherei 48 pages. Wittenberg. 274, der der Mayr,F. X. I 967 Palaobiologieund Stratinomie Plattenkalke Altmuhlalb. Erlanger Abh. 40 pages. geol. 67, Rieber,H. I 973 Ergebnisse in palaontologisch-stratigraphischer Untersuchungen derGrenzbitumenzone (Mittlere Trias) des Monte San Giorgio(Kanton Tessin,Schweiz).Ecl. geol.Helvetiae 667-685. 66, W. 1977 Goniomya in Riegraf, rhombifera (Goldfuss) thePosidoniaShales (Lias epsilon).N. Jb.Geol.Palaont. Mh., pp. 446-448. W. Riegraf, I 984 DerPosidonienschiefer. Fauna Faziesdessudwestdeutschen Biostratigraphie, und Untertoarciums.pages. 195 Enke. Stuttgart: W. Riegraf, I 985 Mikrofauna, und ToarciumSiiddeutschlands Vergleiche und Biostratigraphie Fazies im unteren mitbenachbarten Gebieten.Tubinger Mitt.3, 232 pages. Mikropal. Rosenkranz, I97I Zur Sedimentologie D. und Okologievon Echinodermen-Lagerstatten. Jb. Geol.Paldont., N. Abh.138, 221-258. Savrda, C. E., Bottjer, J. & Gorsline, S. I984 Development a comprehensive D. D. of oxygen-deficient marine biofacies model: evidencefrom Santa Monica, San Pedro,and Santa Barbara Basins,California Continental Borderland. Bull. Am.Ass.Petrol. Geol. 68, 1179-1192. Seilacher,A. I963 Umlagerungund Rolltransport von Cephalopoden-Gehausen. Jb. Geol.Paldont., N. Mh., pp. 593-619. A. of Seilacher, I 968 Originand diagenesis theOriskany Sandstone(LowerDevonian,Appalachians)as reflected in itsshellfossils. Recent developments incarbonate sedimentology in Central Europe, 175-185. Berlin, pp. Heidelberg, New York: Springer. A. Seilacher, I982 Posidoniashales (Toarcian, S. Germany) stagnant basin modelrevalidated. Paleontology, In essential historical of geology, Internat. meeting Venice 1981 (ed. E. Montanaro-Gallitelli), 25-55. Modena. pp. A. and Seilacher, I984 Late Precambrian earlyCambrianMetazoa: preservational real extinction? Patterns or In in of change earth evolution, Dahlem-Konferenzen 1984 (ed. H. D. Holland & A. F. Trendall), pp. 159-168. New York,Tokyo: Springer. Berlin, Heidelberg, A. Seilacher, I985 TheJerammodel: eventcondensation a modern in intertidal environment. Sedimentary In and evolutionary LectureNotesin EarthSciences(ed. G. M. Friedmann), cycles, vol. 1, 336-342. Berlin, Heidelberg, New York,Tokyo: Springer. Seilacher, Andalib,F., Dietl, G. & Gocht,H. I976 Preservational A., of history compressed Jurassic ammonites from Southern N. Germany. Jb. Geol.Paldont., Abh.152, 307-356. A. und Bildungstiefe Hunsruckschiefer Seilacher, & Hemleben,C. I966 Spurenfauna der Notizbl. (Unterdevon). hess. 94, Landesamt Bodenforsch 40-53. A. F. In Seilacher, & Westphal, I97i Fossil-Lagerstatten. Sedimentology ofCentral of parts Europe, Guidebook8. Int. Sediment. Congr.,pp. 327-335. Heidelberg.

PATTERNS

OF FOSSIL

LAGERSTATTEN

23

P. Ober-Senons. A 106, 1-36. Siegfried, 1954 Die Fisch.Fauna des westfailischen Palaeontographica Sorbini, I983 The fossil depositofBolca (Verona), Italy.Int.Congr. L. fish Paleoecol., Lyon, Guide Exc. llA, 23-33. Temmler, I964 Uber die Schiefer- Plattenkalke Weissen H. und des Jurader Schwabischen Alb (Wurttemberg). Arb.Geol.-Paldont. derTH, N.F. 43, 107 pages. Inst. Z. Temmler, I966 Uber die Nusplinger H. Fazies des WeissenJura Schwabischen (Wiirttemberg). deutsch. der Alb geol.Ges.116, 891-907. fosiliferos de Via Boada, L., Villalta,J.F. & EstebanCerda,M. I977 Paleontologia paleoecologia los yacimentos y Geol. del Muschelkalk entreAlcovery Mont-Ral. Cuadernos Iberica 247-256. superior 4,

Discussion R. RIDING (Department Geology, of University College, EXL, U.K.). I wish to query use CardifCF1 of the termobrution. Reference to stagnation and smothering(obrution) as alternativemodes of formation conservationLagerstattenis slightly of confusingbecause stagnation indicates the nature of the environmentwhereas smotheringis a type of asphyxiation. It would be clearer in thiscase to refer rapid burial ratherthan to smothering. to Stagnation and rapid burial could then be regarded as different processes, either of which can result in the asphyxiation of organisms. Asphyxiation occurs above the substratein the case of stagnation and below it in the case of smothering. This usage then clearly distinguishes environmental conditions (stagnation,rapid burial) fromthe actual mode of death which is, in both cases, asphyxiation. E. N. K. CLARKsoN-(Grant Institute Geology, of University Edinburgh, of WestMains Road,Edinburgh EH9 3JW, U.K.). The Lower Lias obrution deposit at Gmund is, as Dr Seilacher has noted, dominated by echinoderms and he had suggested that in such a case the active elements in the benthos mighthave escaped, whereas the echinodermscould not. Is it possible to distinguishsuch an obrution deposit fromone in which the original fauna was dominated by echinoderms and little else? I ask this with particular referenceto some horizons rich in intact echinodermsin the Scottish Silurian. H. B. WHITTINGTON, F.R.S. (Department Earth Sciences, of Sedgwick Museum,University Camof bridge).The Burgess Shale fauna includes the hard parts of characteristicMiddle Cambrian animals - trilobites,sponges, brachiopods, molluscs, hyolithidsand echinoderms- as well as a remarkable soft-bodiedfauna. I have argued in detail (WhittingtonI 97I) that Marrellawas a benthic animal, the thousands of specimens having been buried in varied orientationsin the deposit resultingfroma turbiditycurrent.The highlyfossiliferous layers in the Phyllopod bed of the BurgessShale appear to originatefromsuch a mode of transport and burial (Whittington I980). Comparison with the Solnhofen deposit would thus involve consideration of a similar mechanism forits formation. References
H. Whittington, B. I97i Redescription Marrellasplendens, of (Trilobitoidea)fromthe BurgessShale, Middle Cambrian,British Columbia. Geol.Surv. CanadaBull. 209, 19-20. H. Whittington, B. I980 The significance thefaunaoftheBurgess of Shale, Middle Cambrian,British Columbia. Proc.Geol.Ass.91, 129-132.

A. SEILACHER.'Obrution' means rapid burial, whether this process only preserved carcasses or also killed the organisms.But only in the latter case can we expect a selective preservation. In the meantime, the analysis of clusteredtrilobitesthat appear to have been selectivelykilled

24

A. SEILACHER,

W.-E. REIF

AND F. WESTPHAL

and then buried by the muddy clouds of stormeventshas added another beautifulexample of absence or under-representation non-trilobites of such obrution deposits. Whether the striking or non-echinodermsin these cases be considered an original featureor the outcome of better resistivity such accidents is a matter of ecological taste. to Reference
in trilobite assemblages theMiddleDevonianHamiltonGroup.Lethaia S. C. Speyer, E. & Brett, E. 1985 Clustered 18, 85-103.