Copyright ᭪ 2000, The Paleontological Society

Taphonomy and paleobiology

Anna K. Behrensmeyer, Susan M. Kidwell, and Robert A. Gastaldo

Abstract.—Taphonomy plays diverse roles in paleobiology. These include assessing sample quality relevant to ecologic, biogeographic, and evolutionary questions, diagnosing the roles of various taphonomic agents, processes and circumstances in generating the sedimentary and fossil records, and reconstructing the dynamics of organic recycling over time as a part of Earth history. Major advances over the past 15 years have occurred in understanding (1) the controls on preservation, especially the ecology and biogeochemistry of soft-tissue preservation, and the dominance of bi- ological versus physical agents in the destruction of remains from all major taxonomic groups (plants, invertebrates, vertebrates); (2) scales of spatial and temporal resolution, particularly the relatively minor role of out-of-habitat transport contrasted with the major effects of time-averaging; (3) quantitative compositional fidelity; that is, the degree to which different types of assemblages reflect the species composition and abundance of source faunas and floras; and (4) large-scale var- iations through time in preservational regimes (megabiases), caused by the evolution of new bod- yplans and behavioral capabilities, and by broad-scale changes in climate, tectonics, and geochem- istry of Earth surface systems. Paleobiological questions regarding major trends in biodiversity, major extinctions and recoveries, timing of cladogenesis and rates of evolution, and the role of environmental forcing in evolution all entail issues appropriate for taphonomic analysis, and a wide range of strategies are being developed to minimize the impact of sample incompleteness and bias. These include taphonomically robust metrics of paleontologic patterns, gap analysis, equal- izing samples via rarefaction, inferences about preservation probability, isotaphonomic compari- sons, taphonomic control taxa, and modeling of artificial fossil assemblages based on modern an- alogues. All of this work is yielding a more quantitative assessment of both the positive and neg- ative aspects of paleobiological samples. Comparisons and syntheses of patterns across major groups and over a wider range of temporal and spatial scales present a challenging and exciting agenda for taphonomy in the coming decades.

Anna K. Behrensmeyer. Department of Paleobiology, National Museum of Natural History, Smithsonian Institution, MRC 121, Washington, D.C. 20560. E-mail: [email protected] Susan M. Kidwell. Department of Geophysical Sciences, University of Chicago, 5734 South Ellis Avenue, Chicago, Illinois 60637. E-mail: [email protected] Robert A. Gastaldo. Department of Geology, Colby College, Waterville, Maine 04901-4799. E-mail: [email protected]

Accepted: 30 June 2000

What is Taphonomy? characterize more generally as ‘‘the study of The fossil record is rich in biological and processes of preservation and how they affect ecological information, but the quality of this information in the fossil record’’ (Behrens- information is uneven and incomplete. The meyer and Kidwell 1985). Since the 1950s, the same might be said for many types of neo- analysis of postmortem bias in paleobiologic biological information, but in such cases, sam- data has been one of the prime motivations of pling biases are imposed by scientists and are the field, but taphonomy has always been a explicable as part of a research design. With multi-tasking science (e.g., see historical re- fossils, natural processes have done the sam- views in Behrensmeyer and Kidwell 1985; Ca- pling and created the biases before research de´e 1991), and this remains true today. States begins. Taphonomy seeks to understand these of preservation of biotic remains are not only processes so that data from the fossil record (1) indicators of how faithfully biological his- can be evaluated correctly and applied to pa- tory has been recorded (issues of paleobiolog- leobiological and paleoecological questions. ic data fidelity and resolution), but are also (2) Efremov (1940: p. 85) first defined taphon- testaments to environmental conditions (the omy as ‘‘the study of the transition (in all its aegis of ‘‘taphofacies’’), and (3) evidence of details) of animal remains from the biosphere important aspects of biological evolution into the lithosphere,’’ naming a field that we (skeletal and biochemical novelties, live/dead

᭧ 2000 The Paleontological Society. All rights reserved. 0094-8373/00/2604-0005/$1.00 104 ANNA K. BEHRENSMEYER ET AL. interactions and feedbacks), because organ- nipulative experiments, analyses of synoptic isms not only produce potential fossils but data sets, probabilistic models) and scientific also are highly effective recyclers of plant and disciplines (tools and expertise from biogeo- animal material. Strictly speaking, the logical chemistry, geomicrobiology, isotope geo- limits of taphonomy are defined by its focus chemistry, geochronology, ecology, biome- on processes and patterns of fossil preserva- chanics, archeozoology, anthropology, sedi- tion1, but in practice, taphonomy serves a mentology, sequence stratigraphy; see recent broader role in stimulating research on all reviews and syntheses in Wilson 1988a; Al- types of biases affecting paleontological infor- lison and Briggs 1991a; Donovan 1991; Gif- mation, including those introduced by collect- ford-Gonzalez 1991; Lyman 1994; Brett 1995; ing, publication, and curation methods on the Briggs 1995; Haglund and Sorg 1997; Claas- one hand, and stratigraphic incompleteness sen 1998; Martin 1999). on the other (see also Lyman 1994; Donovan Taphonomy still is strongly oriented to- and Paul 1998; Holland this volume). ward modern analogues as a means of iden- Taphonomy today is focused first and fore- tifying and quantifying processes, but in- most on a geobiological understanding of the creasingly exploits the stratigraphic record earth, grounded on the postmortem process- for hypothesis testing. Reliance on the fossil es that recycle biological materials and affect record to ‘‘bear its own witness’’ is an abso- our ability—positively and negatively—to lute necessity for some facies and taxa, but reconstruct past environments and biotas. constitutes a powerful independent method The classic flowchart of taphonomic transfor- even for environments and groups that are mations (Fig. 1) is now underpinned by a well represented in the Recent world. Re- much fuller and quantitative understanding gardless of subject, however, most taphono- of interim states and pathways of fossiliza- mists remain determinedly empirical in ap- tion, owing to an explosion of interest in the proach, dedicated to assembling baseline in- field since the early 1980s. Some of the most formation on taphonomic patterns and pro- notable advances have been in (1) microbial, cesses. Such work usually targets individual biogeochemical, and larger-scale controls on fossil assemblages or modern analogues for the preservation of different tissue types; (2) particular groups of organisms (protists to processes that concentrate biological re- vertebrates) and types of environments (gla- mains; (3) the spatio-temporal resolution and cial to abyssal plain). This fact-gathering fo- ecological fidelity of species assemblages; cus is typical of a relatively new field of and (4) the outlines of ‘‘megabiases’’ (large- study, but a theoretical component also is be- scale patterns in the quality of the fossil rec- ginning to develop, with proposals for gen- ord that affect paleobiologic analysis at pro- eral models for organic preservation (e.g., Ly- vincial to global levels and at timescales usu- man 1994; Kowalewski 1997). There have ally exceeding ten million years). These ad- been a number of forays into the realm of vances are highlighted in this review because taphonomic theory by paleobiologists seek- of their impact on paleobiologic analysis and ing to distinguish sampling biases from bio- their promise as research themes in the com- logical patterns. These include attempts to ac- ing decades. Such advances reflect an in- count for preservational biases using as- creasingly ecumenical approach in terms of sumptions of random preservation and ‘‘hol- scientific methods (field measurements, ma- low curve’’ models for original taxonomic abundance as well as models that test the ef- fects of incomplete fossilization, stratigraphic 1 It often falls to Taphonomy to answer the most basic of incompleteness, nonrandom distributions of paleontological questions, ‘‘What is a fossil?’’ Material definitions concerning degree of mineralization and cri- facies and hiatuses, and blurring of genera- teria based on age considerations are problematic for tions by time-averaging on our ability to eval- Holocene to sub-Recent organic remains. Hence, we pre- uate phylogenies, rates of evolution, and tem- fer a more flexible definition: ‘‘A fossil is any nonliving, biologically generated trace or material that paleontol- po and mode of speciation (Marshall 1990, ogists study as part of the record of past life.’’ 1994; Gilinsky and Bennington 1994; Foote TAPHONOMY AND PALEOBIOLOGY 105

FIGURE 1. The main pathways for organic remains from death to paleobiological inference. Each path is af- fected by taphonomic processes and circumstances that filter the information as it passes to the next stage. Taphonomy is the study of how biological, chemical, and physical processes operating between each stage preserve or destroy organic remains and affect information in the fossil record (Behrensmeyer and Kidwell 1985).

1996; Foote and Raup 1996; Roopnarine 1999; some promising directions for the future. This Wagner 2000a; Simo˜es et al. 2000a; and see review is organized by scales of processes in papers in this volume by Wagner, Holland, order to underscore two key points. One is the Alroy et al., and Jablonski). Arguments about wide array of different qualities of the fossil whether the lack of a fossil record is evidence record that paleobiologic analysis depends for original absence (e.g., Vrba 1995; Foote et upon and that taphonomic analysis is relevant al. 1999; Valentine et al. 1999; Novacek et al. to; these qualities range from the preservation 2000) also draw upon underlying assump- of DNA molecules to the analytic comparabil- tions about how taphonomy works at a more ity of samples from disparate regions and general level. geologic periods (Table 1). A second is the Here we review highlights of taphonomic multidisciplinary nature of taphonomic anal- research from the past 15 years, since the tenth ysis at all scales, illustrated by the variety of anniversary issue of Paleobiology, and suggest new techniques and lines of evidence that are 106 ANNA K. BEHRENSMEYER ET AL.

TABLE 1. Research in taphonomy has demonstrated many sources of potential bias affecting qualities of the fossil record relevant to various paleobiological questions. (Adapted from Kidwell and Brenchley 1996.)

Aspects of quality in the fossil record Sources of bias Biochemical fidelity Shifting of original compositions (e.g., isotopic and molecular) by diagenesis and metamorphism Anatomical fidelity Destruction or incomplete mineralization of soft tissues; disar- ticulation, fragmentation, recrystallization, and physical de- formation Spatial fidelity Transport out of life position, rearrangement within life habi- tat, transport out of life habitat or biogeographic province (e.g., necroplanktonic organisms, pollen) Temporal resolution Mixing of noncontemporaneous remains within single sedi- mentary units via physical or biological processes (tapho- nomic time-averaging) Compositional fidelity Selective destruction/preservation of species, morphs, dis- carded body parts; bias from introduction of exotics and noncontemporaneous remains Completeness of time series Episodicity in deposition; taphonomic or diagenetic oblitera- tion of fossils in surviving lithofacies (producing gaps and condensation of the record); poor preservation of some envi- ronments (deposits thin, localized, or readily eroded) Consistency in preservation over Major shifts in intrinsic and extrinsic properties of organisms, Geologic time including morphology and behavior in relation to other or- ganisms—or shifts in the global environment, which can cause secular or long-term cyclic changes in preservation (megabiases) now being brought to bear on both established bles in Gastaldo 1988; Kidwell and Bosence and new issues in paleobiology.2 1991; Speyer and Brett 1991; Behrensmeyer and Hook 1992; Martin 1999). Highlights and Research Outlooks Detailed studies show that taphonomic sys- Much of the progress in taphonomy has oc- tems are more complex than originally sup- curred via an environment-by-environment posed, but many of these complications are search for patterns and processes. Although shared across major environments and taxo- many environments have not been explored nomic groups, which is good news for data fully, it is clear from available actualistic and comparability and the potential for unifying stratigraphic studies that depositional context theory. Contrary to the impressions given by is extremely important in controlling the qual- basic paleoecology texts, some taphonomic ity and nature of fossil preservation. Environ- features previously thought to be diagnostic mental setting determines such important fac- of a particular taphonomic process or circum- tors as the likelihood of immediate burial, ex- stance are now recognized as having a differ- humation and reworking, the biogeochemis- ent dominant cause or resulting from multiple try of the early diagenetic environment, and processes (the concept of ‘‘equifinality’’ [Ly- the nature of the local community that gen- man 1994]). A good example is the disarticu- erates or is capable of recycling tissues (i.e., is lation and fragmentation of animal hardparts: mortality typically attritional or catastrophic, a growing body of actualistic evidence indi- are biominerals undersaturated or in sur- cates that, in both continental and marine set- plus?). From such considerations the general tings, such damage is overwhelmingly taphonomic attributes of most major fossil- biogenic (from predation, scavenging, etc.) preserving facies now can be sketched as a framework for more detailed testing (e.g., ta- rather than an index of physical energy (e.g., Haynes 1991; Jodry and Stanford 1992; Beh- rensmeyer 1993; Cade´e 1994; Cate and Evans 2 Taphonomy also provides guidelines concerning how humans can become fossilized. See Mirsky 1998 and 1994; Lyman 1994; Oliver and Graham 1994; Haglund and Sorg 1996 for user-friendly reviews. Best and Kidwell 2000a; for Cambrian exam- TAPHONOMY AND PALEOBIOLOGY 107 ple see Pratt 1998). Moreover, in the absence steady rates of postmortem modification, of recycling metazoans, damage is dependent which are linked to chaotic aspects of the ex- upon the state of decay of connective tissues trinsic environment and over long periods of rather than the distance of hydraulic transport time could appear to be linear. (e.g., Allison 1986; Kidwell and Baumiller 1990; Greenstein 1991; Ferguson 1995). Simi- Controls on the Preservation of Biological larly, rounding of hardparts is more likely to Remains result from repeated reworking within a high- Most individual organisms never become fos- energy environment than from abrasion dur- sils, but taphonomic research has discovered ing long-distance transport, as demonstrated much about the circumstances that capture rich by comparing indigenous shells from beaches samples of past life. These samples may be quite versus exotic shells in turbidites, or bones that different from those of living systems because have been trampled or chewed versus those of postmortem processes, but there is plenty to transported in rivers (Behrensmeyer 1982, work with, whether the tissues of interest (1) are 1990; Potts 1988; Davies et al. 1989; Andrews composed exclusively of volatile organics (e.g., 1990; Meldahl and Flessa 1990; Kidwell and nucleic acids, amino acids, simple sugars, Bosence 1991; Spicer 1991; Gastaldo 1994; Ly- starch; see Briggs this volume), (2) include re- man 1994; Llona et al. 1999; Nebelsick 1999). fractory organics (lignin, collagen, cellulose, chi- Paleoanthropologists and archeologists have tin, glycolipids, resins, sporopollenin [see learned that many taphonomic agents, includ- Briggs 1993]), or (3) are mineralized during life ing humans, can cause similar patterns of (major biominerals are aragonite, calcite, apatite, bone modification, skeletal-part representa- various forms of silica). tion, and faunal composition; these patterns Preservation depends on an array of pro- are heavily influenced by which bones and cesses and conditions operating at different taxa are the most durable and identifiable in scales (Fig. 1). These are the face of destructive processes (Grayson 1989; Gifford-Gonzalez 1991; Lyman 1994). 1. the supply side of the equation:rateofinput, A second complicating realization, derived total volume, and composition (durability) of primarily from experiments on marine ma- biological remains delivered to the environ- croinvertebrates, is that many taphonomic pro- ment; cesses are inconstant in rate over time. Car- 2. the nature of the pre-burial environment: se- casses of regular echinoids, for example, frac- lectivity and intensity of modification by local ture like live echinoids until microbial decay is physical, chemical and biological agents at the sufficiently advanced for connective ligaments sediment-air or sediment-water interface. to be weaker than the calcite plates, a period of Modification may be destructive (as in the case ‘‘ambiguous’’ behavior that lasts a few hours in of bioerosion, scavenging, and dissolution) or tropical temperatures but days or weeks in cold stabilizing (as in the case of bioencrustation water; once this decay threshold is passed, the and den/burrow formation; disintegration of the test proceeds at a much 3. the rate (immediacy) and permanence of buri- faster rate than in pre-threshold specimens al, which determines how long tissues are ex- (Kidwell and Baumiller 1990). As a second ex- posed to processes operating on the sediment ample, the postmortem ‘‘disappearance,’’ surface as opposed to those within the sedi- probably by dissolution, of aragonitic shells mentary column; and from early postlarval mollusks in Texas la- 4. diagenetic conditions within the upper part of goons is very rapid initially but slows logarith- the sedimentary column (highly dynamic mixed mically, so that loss is best described as a taph- zone), where organic remains and sediments onomic half-life (Cummins et al. 1986a,b). In are still subject to bioturbation, meteoric ef- contrast, the episodic movement of plant debris fects, microbial processes, and possible phys- downstream, the alternating burial/exposure ical reworking. Postburial modification may of shells on seafloors and the reworking of stabilize (e.g., mineral coatings, infillings, re- bones in channels all provide examples of non- placements) or reduce biochemical fidelity 108 ANNA K. BEHRENSMEYER ET AL.

FIGURE 2. Schematic portrayal of information changes in two different types of fossil assemblages compared with a hypothetical living community at one point in time. Each cell is a species characterized by two variables, ecological abundance during life and taphonomic durability of the remains; the living community consists of equal numbers (50:50) of species that are permanent residents (black cells) and transients (gray cells, e.g., highly mobile forms or those on seasonal or longer-term population cycles). White cells in the fossil assemblages indicate species that are not preserved. The single-event assemblage (census) will capture most of the resident and some of the transient species (in this example, 80% and 37% of the cells (species) in the life assemblage, respectively) and is not strongly affected by taphonomic durability. In contrast, the attritional, time-averaged assemblage will be biased toward higher durability and more abundant species, capturing 44% of the resident and 52% of the transient species. A similar graphic model could be applied to all types of organisms or organic parts, with varying results for census versus time-averaged assemblages depending on the range of durability, type of community, environment, and length of time-averaging. and anatomical detail (hydrolysis, continued interiors and in continental rift margins or au- maceration, dissolution, recrystallization). locogens (failed branches of continental rifts) that have escaped tectonic recycling. Exam- Beyond these near-surface factors, which ples include Archaean and Proterozoic earli- embed the remains in a consolidated sedi- est life deposits (Grotzinger 1994; Walter et al. mentary matrix, the long-term survival of fos- 1995) and Devonian through Carboniferous silized material is determined by land plants and animals (Kidston and Lang 5. the fate of the larger sedimentary body. The 1920; Rolfe et al. 1994). key factors here are strongly linked to tectonic setting, which determines rates of sediment Fossil assemblages commonly are parsed aggradation and compaction, depth of even- according to the way remains initially accu- tual burial (and thus nature of later diagenesis mulated in the depositional system, i.e., in and metamorphism), and structural defor- terms of major types of supply-side input (Fig. mation. The longest-surviving fossil-bearing 2). Attritional (time-averaged) assemblages sequences occur in stable cratonic margins or reflect the release of discarded organic prod- TAPHONOMY AND PALEOBIOLOGY 109 ucts (e.g., pollen, leaves) and input from nor- phatization proceeds rapidly enough to pre- mal mortality over periods of years to millen- serve undegraded volatile muscle and visceral nia. Single-event (census) assemblages reflect tissues in three dimensions (including embry- unusual events such as sudden anoxia, severe os [Xiao and Knoll 1999]), whereas calcite and storms, pathogen outbreaks, droughts, and pyrite are sufficient to preserve structures volcanic eruptions that kill large numbers of composed of more slowly decaying chitin, lig- individuals at one time (minutes to months). nin, and collagen (Allison 1988; Allison and In the former case, it may take considerable Briggs 1991b; Underwood and Bottrell 1994). time and slow net sediment accumulation to Instances of successful in vitro precipitation of amass a dense concentration of organic re- minerals provide further insights into the dy- mains in a single bed in the absence of some namics of fossilization of soft tissues: ‘‘phos- other concentrating process; in the latter, this phatization’’ may consist of (1) fine-grained may happen literally overnight. 0.3-␮ apatite that precipitates in the tissues Soft-Part Preservation. The taphonomy and themselves (subcell features preserved), (2) diagenetic biogeochemistry of metazoan soft 1-␮ apatite that replaces invasive bacteria and tissues and biomolecules have been the sub- creates a fully 3-D pseudomorph of cells or tis- ject of intense field and laboratory study in the sues, and (3) comparable replacement but of last 15 years (Allison and Briggs 1991b; Hen- noninvasive bacterial coats, which replicate wood 1992a; Briggs 1993; Briggs and Kear only the outlines of cells or tissues (Wilby and 1994a,b; Allison and Pye 1994; Westall et al. Briggs 1997; and see Franzen 1985; Martill 1995; Bartley 1996; Briggs et al. 1997, 1998; 1990; Xiao and Knoll 1999; also Taylor 1990 Bartels et al. 1998; Davis and Briggs 1998; and Evans and Todd 1997 for replication of Duncan et al. 1998; Orr et al. 1998; Briggs this soft tissues by biological overgrowth). volume). Delicate molecules like DNA are ex- Laboratory and field studies on animal soft- tremely difficult to preserve, as would not sur- tissue preservation and konservat-lagersta¨tten prise anyone who has struggled to extract have greatly elaborated and deepened our un- good material from living organisms, and the derstanding of the multiple advantages of an- oldest confidently identified DNA is less than aerobic decomposition. Anaerobic decompo- 100 k.y. old (Bada et al. 1999; Wayne et al. sition is in fact slower (Kristensen et al. 1995) 1999). Although cinematically fabled, amber is and far less effective (efficient) in decompos- not favorable for DNA preservation. Resins ing refractory material than aerobic decom- are not airtight, and so generally only the position, thus prolonging the window for most refractory portions of insects are fossil- preservation. If it is linked to low oxygen in ized (Stankiewicz et al. 1998; but see preser- overlying water, this excludes predators and vation of volatile structures in amber via de- scavengers and keeps them from destroying hydration [Henwood 1992a,b]). tissues before these can be encased by micro- Laboratory degradation of metazoan car- bial mats (fostering local anoxic conditions) or casses under oxic and anoxic conditions dem- become buried in sediment below the redox onstrates the relative reactivities of tissue discontinuity level (Seilacher 1984; Seilacher et types, means of retarding decay, and absolute al. 1985; Wilby et al. 1996; Palaios 1999; and rates of decay. Such data are used not only to see Janzen 1977 on microbial strategies (1) rank tissue reactivities, but (2) rank rates against metazoan scavenging). Even though of mineral precipitation in fossil specimens, anaerobic decomposition proceeds almost as (3) establish criteria for recognizing oxic and fast as aerobic decomposition on volatile ma- anoxic subenvironments (diagenetic minerals terial, only anaerobic microbial processes lib- precipitate in distinctive Eh-pH fields created erate the appropriate cations and levels of al- by different anaerobic microbial communi- kalinity to precipitate early diagenetic phos- ties), and (4) provide absolute time limits on phate, calcite, pyrite, siderite, and other min- the contemporaneity of co-occurring fossils erals (Allison and Briggs 1991b). In terms of (Allison 1988; and see McGree 1984, and for soft-tissue preservation, a little decay of or- plants Ferguson 1995). Apparently, only phos- ganic matter is thus good (Allison 1988; Chaf- 110 ANNA K. BEHRENSMEYER ET AL. etz and Buczynski 1992) because, by depleting lular to tissue-grade structures with high fi- oxygen, the local chemical environment is delity (e.g., Knoll 1985; Scott 1990). Anoxia is a driven to anaerobic conditions that favor min- highly efficient agent of konservat-lagersta¨tte eral precipitation in and around the organics, formation in aquatic systems when it affects which is essential to their long-term preser- overlying waters (accomplishing points a–c vation. above). This may result from elevated temper- A general model for superb soft-part pres- ature or organic matter overload (e.g., a phy- ervation thus has the following requisites: (a) toplankton bloom), and is the cause rather than a carcass (microbe, catkin, worm, wombat) in the effect of metazoan mortality (e.g., Stachow- good condition at time of death (death with- itsch 1984). Although catastrophic death is im- out significant morbidity or other damage to portant (for point a), it turns out that mass body parts); (b) postmortem isolation of the death is not a prerequisite for superb preser- carcass from scavengers and physical disrup- vation of multiple individuals in one sedimen- tion; (c) decomposition retarded until miner- tary layer, as evidenced by the wide spacing of alization is accomplished; and (d) avoidance metazoan specimens within classic konservat- of later reworking (advantages of entombment lagersta¨tten (Seilacher et al. 1985). Even in over- within tree stump, incised valley fill, karst de- all aerobic environments, single carcasses can pressions, structural graben, or aulocogen be highly effective in depleting oxygen from [e.g., Lyell and Dawson 1853; Dawson 1882; the immediate environment within sediment Archer et al. 1991; Cunningham et al. 1993]). or under microbial mats, creating their own lo- Points a–c can be accomplished via cata- cally anaerobic conditions favorable to diage- strophic burial (obrution), for example from netic mineralization (e.g., Scha¨fer 1972; Spicer ash falls and sediment avalanches on land and 1980; Martill 1985; Baird et al. 1986; Allison et from various sedimentary processes in water al. 1991b). (Baird et al. 1986; Demko and Gastaldo 1992; Taphonomic Feedback. Both soft and hard bi- Wing et al. 1993; Crowley et al. 1994; Rolfe et ological remains can develop positive feed- al. 1994; Yang and Yang 1994; Downing and back systems that significantly enhance their Park 1998; Brett et al. 1999; Feldmann et al. own chances for preservation, especially 1999; Hughes and Cooper 1999; Labandeira where remains are densely concentrated. For and Smith 1999). example, skeletal hardparts can alter water However, contrary to stereotypes, enclosed flow dynamics, promote trapping and bind- water bodies having acidic, hypersaline, or an- ing of sediment, and increase the erosion re- oxic conditions are highly effective environ- sistance of seafloors (Kidwell and Jablonski ments for preservation without unusual sedi- 1983; Seilacher 1985; Behrensmeyer 1990). ment burial events, and in fact this is a more Also, unusually high inputs of carcasses can common means of konservat-lagersta¨tte for- overwhelm the capacity of normal recycling mation (e.g., Seilacher et al. 1985; Whittington processes (e.g., scavengers faced with a surfeit and Conway Morris 1985; Martill 1988; Barthel of carcasses; oxygen depletion by dead organ- et al. 1990; Briggs and Crowther 1990; Brett ic matter in aquatic systems). Concentrations and Seilacher 1991; Schaal and Ziegler 1992; of remains also can create favorable diagenetic Bartley 1996; Bartels et al. 1998). Heat and conditions, ‘‘self-buffering’’ local porewaters chemical transformation of volatiles to more re- to reduce overall hardpart dissolution or pro- fractory forms (e.g., charcoal, kerogen, graphite moting replacement of associated remains [Butterfield 1990; Lupia 1995; Vaughan and (e.g., Kotler et al. 1992; Schubert et al. 1997). Nichols 1995]), before or after burial, is one Concentrations can have negative effects as means to lengthen the window of opportunity well. For example, shell-rasping grazers be- for mineralization (point c above). Alternative come a major destructive force only where paths are acidity, which is antimicrobial, and dead shells reach a critical abundance in tidal pickling (subaqueous dehydration via salt); channels (Cutler 1989), and drought concen- both can retard decomposition sufficiently for trations of animals around water holes focuses slowly polymerizing silica to replace subcel- death and skeletal input but also may increase TAPHONOMY AND PALEOBIOLOGY 111 physical destruction and exhumation via construction and ecology in determining dif- trampling and digging (Haynes 1985, 1988, ferent hardpart fates in the same environment; 1991). ‘‘comparative taphonomy’’ of Brett and Baird Preservation of Mineralized and other Refractory 1986, with implications for differential repre- Tissues. Most aspects of this subject—the sentation of taxa sensu Johnson 1960). For ex- supply side, pre-burial effects, rates of burial, ample, in modern shallow marine settings, early diagenesis, and permanent incorpora- mollusk shell fragmentation commonly varies tion into the stratigraphic record—have re- independently of water energy or bears direct ceived heightened attention in the last 15 evidence of being the product of predators years, and we refer the reader to several ex- and scavengers rather than physical environ- cellent review volumes for details (Donovan ment itself (Feige and Fu¨ rsich 1991; Cade´e 1991; Allison and Briggs 1991a; Lyman 1994; 1994; Cate and Evans 1994; Best and Kidwell Martin 1999; Martin et al. 1999a). 2000a), and branching colony form among A major focus for work on the postmortem scleractinian corals significantly increases sedimentology and biology of hardparts (i.e., postmortem disintegration relative to massive biostratinomy) has been paleoenvironmental and encrusting forms (Greenstein and Moffat analysis (taphofacies analysis, sensu Speyer 1996; Pandolfi and Greenstein 1997a; and for and Brett 1986; Parsons and Brett 1991), and bryozoans see Smith and Nelson 1994). these efforts are proving to have direct as well Most actualistic studies of this type previ- as indirect value to paleobiology. For benthic ously focused on variation among taxa within systems, for example, analysis of styles of fos- a single major group, or variation among en- sil preservation and concentration reveal bed- vironments for a single taxon, but now include by-bed and facies-level differences in key eco- benthic foraminifera (Martin et al. 1999b), gas- logical factors such as frequency of storm re- tropods (Walker 1989, 1995; Taylor 1994; working and oxygen levels that would be un- Walker and Voight 1994), bivalves (Parsons detectable from inorganic matrix alone (e.g., 1989; Meldahl and Flessa 1990; Parsons and Norris 1986; Parsons et al. 1988; Meyer et al. Brett 1991; Cutler and Flessa 1995; Best and 1989; Brett et al. 1993; Ausich and Sevastopulo Kidwell 2000b), echinoids (Greenstein 1993; 1994). Determinations of the extent to which Nebelsick 1995), crinoids (Meyer and Meyer different marine and continental environ- 1986; Llewellyn and Messing 1993; Silva de ments bear distinctive ‘‘taphonomic signa- Echols 1993; Baumiller et al. 1995), brachio- tures’’ also provide a means of recognizing ex- pods (Daley 1993; Kowalewski 1996a), and otic, out-of-habitat material in fossil assem- various shell-encrusters (Bishop 1989; Walters blages (e.g., Davies et al. 1989; Miller et al. and Wethey 1991; Lescinsky 1993, 1995; Mc- 1992), or material reworked from older depos- Kinney 1996). In continental settings, inten- its that are ecologically or evolutionarily irrel- sive work on rates of litter decomposition evant to the host deposits (e.g., Argast et al. (Boulton and Boon 1991; Ferguson 1995) and 1987; Plummer and Kinyua 1994; Trueman on sources and signatures of macrofloral ma- 1999). terial in deltas and other organic-rich coastal Insights from hardpart condition derived environments (Gastaldo et al. 1987) provides from analysis of death and fossil assemblages a valuable basis for comparison with the increasingly are complemented by actualistic stratigraphic record. Lab and field investiga- experiments on rates and controls on modifi- tions have also targeted arthropods (e.g., Wil- cation. This work provides insights into recy- son 1988b; Henwood 1992a; Martinez-Delclos cling processes themselves, for example, the and Martinell 1993; Labandeira and Smith huge importance of organisms as agents of 1999; Wilf and Labandeira 1999; Smith 2000; skeletal transport and modification in modern Labandeira et al. in press), fish, birds, and oth- systems (described above; thus affecting the er lower vertebrates (Elder and Smith 1984; limits to back-extrapolation over geologic Smith et al. 1988; Wilson 1988c; Oliver and time) and of factors intrinsic to the hardpart Graham 1994; Blob 1997; Stewart et al. 1999; producers themselves (i.e., the roles of body Llona et al. 1999), and mammals including hu- 112 ANNA K. BEHRENSMEYER ET AL. mans (Frison and Todd 1986; Haynes 1988, assemblages overall), how such clocks vary 1991; Fiorillo 1989; Andrews 1990; Blumen- among groups, and how they behave with schine 1991; Behrensmeyer 1993; Kerbis Peter- elapsed time-since-death (do rates of deterio- hans et al. 1993; Sept 1994; Tappen 1994a; ration decrease, increase, or remain steady for Haglund and Sorg 1997; Cruz-Uribe and a specimen held under ‘‘constant’’ postmor- Klein 1998; Cutler et al. 1999). Zooarcheolo- tem conditions)? gists and paleoanthropologists have contrib- Comparisons of death assemblages with lo- uted important actualistic research linking cal live communities are one powerful means damage patterns to taphonomic processes in of assessing out-of-habitat transport and time- their efforts to distinguish human from non- averaging that has been applied to many ma- human bone modification and assemblage rine and continental groups (reviewed by Kid- formation. Over the past 15 years, zooarcheol- well and Flessa 1995; and next section). Direct ogists have made important advances in char- dating of mollusk shells in death assemblages acterizing bone modification patterns for spe- is increasingly used to explore time-averaging cific taphonomic agents and developing more and taphonomic clocks in marine systems accurate methods for analyzing skeletal-part (e.g., Powell and Davies 1990; Flessa et al. ratios (e.g., inclusion of limb shaft fragments, 1993; Flessa and Kowalewski 1994; Martin et which were formerly omitted from such anal- al. 1996; Meldahl et al. 1997a; Kowalewski et yses, has a significant impact on archeological al. 1998), and the results (1) settled disputes inferences [Bartram and Marean 1999]). This on scales of time-averaging (commonly thou- work is featured in some major volumes (Bon- sands of years even for intertidal and shallow nichsen and Sorg 1989; Solomon et al. 1990; subtidal assemblages, and tens of thousands Hudson 1993; Lyman 1994; Oliver et al. 1994) on the open shelf, contrary to rapid rates of as well as individual field and laboratory individual shell destruction that can be mea- studies of hyenas (Blumenschine 1986, 1988, sured experimentally); (2) established the 1991; Marean 1992), lions (Dominguez-Rod- highly probabilistic and unsteady rather than rigo 1999), predatory birds and small mam- monotonic accrual of damage with elapsed mals (Andrews 1990; Cruz-Uribe and Klein time-since-death (owing to erratic exposure to 1998; Stewart et al. 1999), and other pre- and taphonomic agents); and (3) established the postdepositional processes (Lyman 1985, probabilistic nature of down-core stratigraph- 1994; Noe-Nygaard 1987; Marean et al. 1991, ic ordering in shell ages (linked to relative 1992; Tappen 1994b). rates of sediment aggradation and physical Actualistic studies of hardpart modification and bioadvection). A very promising direction are also determining the security—and pit- of new research involves comparisons of ma- falls—of ‘‘traditional’’ paleontologic inferenc- jor co-occurring taxa, such as mollusks versus es about spatial resolution and time-averaging benthic foraminifera (Martin et al. 1996; An- of skeletal assemblages (and see next section). derson et al. 1997) and lingulid brachiopods Among the questions amenable to experimen- (Kowalewski 1996a,b), where bioclasts have tation and measurement are, How far are fos- disparate postmortem durabilities and thus silizable materials transported outside the high potential for ‘‘disharmonious’’ scales of original life habitat? What proportion of ma- time-averaging. terial is moved (how great is the dilution fac- For the paleobiologist collecting in the field, tor for indigenous material in the ultimate one of the most obvious taphonomic aspects host deposit)? Over what periods can biolog- of the record is the concentration of fossils in ical materials survive in various environ- select beds or horizons and the nonrandom ments, how different are those periods, and to quality of fossil preservation. Work on this what extent can these periods of potential topic continues to be primarily stratigraphic time-averaging be interpreted from fossil con- rather than actualistic, and such studies con- dition? These questions have generated re- sider (1) how concentrations are distributed search on possible ‘‘taphonomic clocks’’ of with respect to gradients in biological input, damage accrual (for individual specimens or sediment reworking, and net sediment accu- TAPHONOMY AND PALEOBIOLOGY 113 mulation; (2) whether such concentrations can over multiple years (Walker 1988; Callender et be utilized for basin analysis (stratigraphic al. 1994; Best and Kidwell 1996; Walker et al. applications, including error-bars in biostra- 1998; Parsons et al. 1999; Kennish and Lutz tigraphy [see Holland this volume]); and (3) 1999) and in situ assessment of porewater geo- whether diverse concentration types have im- chemistry (Goldstein et al. 1997; Walker et al. plications for the qualities of paleontologic in- 1997; Best et al. 1999). Rather than deducing formation (e.g., positive versus negative ef- early diagenetic conditions or taphonomic fects of hiatuses in sedimentation, likely scales consequences, these can be measured directly, of time-averaging, and selective preservation). and the new multiyear rate information pro- Much of this work is framed in a sequence- vides more explicit links to radiocarbon-cali- stratigraphic context and encompasses a brated studies of skeletal deterioration. range of continental (Behrensmeyer 1987, Finally, the microscopic modification and 1988; Dodson 1987; Eberth 1990; Behrensmey- breakdown of mineralized microstructures er and Hook 1992; Gastaldo et al. 1993a; Rog- before and during shallow burial—that is, ers 1993; Smith 1993; Badgley and Behrens- ‘‘weathering’’ and early diagenesis—still re- meyer 1995; Smith 1995; Wilf et al. 1998; Rog- ceive relatively little attention, notwithstand- ers and Kidwell 2000) and marine settings ing their huge importance in recycling biolog- (Kidwell 1991, 1993; Doyle and Macdonald ical materials. These relatively ordinary and 1993; Fu¨ rsich and Oschmann 1993; Ausich and pervasive processes are a counterpart to the Sevastopulo 1994; Brett 1995; Rivas et al. 1997; extraordinary processes that preserve soft tis- Kondo et al. 1998; Ferna´ndez-Lo´pez 2000; also sues, and deserve the same highly focused various papers in Kidwell and Behrensmeyer level of geologic, geochemical, and geomicro- 1993). This and other research in the strati- biological analysis. Although the signatures of graphic record is yielding new evidence for such processes may be less obvious than other particular modes of accumulation, including kinds of damage, and require SEM to fully predation (Wilson 1988c; Ferna´ndez-Jalvo et identify (e.g., Cutler and Flessa 1995), skeletal al. 1998; Andrews 1990), trapping (Richmond materials in all environments are subject to at- and Morris 1996), fluvial reworking (Schmude tack from some combination of the following: and Weege 1996; Smith and Kitching 1996), physical oxidation, hydrolysis, and UV light and drought-related or other types of mass (especially in continental settings); microbor- death (Sander 1989; Rogers 1990; Fiorillo 1991; ing (by algae, fungi, larvae, etc. everywhere); see also Eberth et al. 1999). Actualistic studies microbial maceration (of microstructural or- based on core samples in modern environ- ganic matrix, in both aerobic and anaerobic ments (e.g., Gastaldo and Huc 1992), as well conditions); and dissolution (of mineral phase as studies that track hardpart reworking un- within hardparts; including back-precipita- der different energy and net-sedimentation tion and recrystallization of minerals, which conditions, would be valuable additions to may reset isotopic ratios [Budd and Hiatt continuing stratigraphic efforts. 1993]). Limited actualistic work to date indi- In terms of future directions, marine studies cates strong environmental differences in have continued to focus on midlatitude set- rates and specific pathways, but except in cas- tings, but attention to fully tropical settings is es of rapid permineralization (e.g., Downing increasing. This includes both reefs and as- and Park 1998), hardparts generally become sociated pure carbonate sediments (Miller less resistant to destruction during reworking 1988; Parsons 1989; Miller et al. 1992; Dent and time-averaging. For example, compared 1995; Perry 1996, 1999; Stoner and Ray 1996; with bones in dry, highly seasonal, savannah Zuschin and Hohenegger 1998) as well as sil- settings (Behrensmeyer 1978a; Lyman and Fox iciclastic and mixed composition seafloors 1989), those in rainforests appear to weather (Best and Kidwell 2000a), which rival carbon- more slowly but are soft and spongy from the ate sediments volumetrically on modern trop- activities of bioeroders such as fungi (Kerbis ical shelves. In addition, many field surveys Peterhans et al. 1993; Tappen 1994b; and see now include time-lapse experimental arrays Cade´e 1999 for supratidal example). Degra- 114 ANNA K. BEHRENSMEYER ET AL. dation in temperate and arctic settings gen- size fraction; for review and synthesis see Kid- erally is slow (Noe-Nygaard 1987; Andrews well and Bosence 1991), and the Texas study and Cook 1989; Sutcliffe 1990), indicating that supports this by illuminating how biological bones on the surface have a longer opportu- information can be captured in shelly death nity for burial in cold environments. Bone assemblages, even under conditions that may weathering stages based on actualistic studies seem unfavorable on the basis of short-term (e.g., Behrensmeyer 1978a) have been applied loss rates. to the fossil record (e.g., Potts 1986; Fiorillo 1988; Cook 1995) with some success, although Spatial and Temporal Resolution distinguishing primary weathering damage Postmortem import and export of remains from similar features (e.g., cracking) acquired to an accumulation site, and the mixing of after burial or during diagenesis can be prob- multiple generations of organisms and/or lematic. communities during time-averaging, deter- In marine settings, there is growing evi- mine the spatial and temporal resolving pow- dence that microbial processes are at least as er of a fossil assemblage. Along with the dif- important as physico-chemical ones in the ferential destruction of species that occurs ‘‘dissolution’’ of molluscan shell both on the during these processes, space- and time-av- seafloor and during shallow burial, preferen- eraging of organic input also affect composi- tially attacking organic-rich microstructures tional fidelity of a fossil assemblage (Fig. 2). In and proceeding at similar rates in both anaer- this paper, ‘‘fidelity’’ refers to how closely obic and aerobic settings (Poulicek et al. 1988; (faithfully, accurately, truthfully) the fossil Cutler and Flessa 1990; Glover and Kidwell record captures original biological informa- 1993; Clark 1999; for brachiopods see Emig tion, be it spatial patterning or the presence/ 1990; Daley 1993; Daley and Boyd 1996). Body absence and relative abundances of species; size clearly has a strong effect on the preser- and ‘‘resolution’’ refers to the acuity or sharp- vation of macrobenthic shells (as also among ness of that record, i.e., the finest temporal or continental bone assemblages [e.g., Behrens- spatial bin into which the fossil remains can meyer and Dechant Boaz 1980]), and there is confidently be assigned. growing evidence that rate of shell disintegra- Although much more work is required for a tion declines over time during early diagene- full taxonomic and environmental picture, a sis (Cummins et al. 1986b; Glover and Kidwell taphonomic highlight of the past 15 years has 1993). Thus the dynamics of ‘‘loss budgets’’ been the tremendous advance in quantifying may be complex. For example, in an innovative the magnitudes and selectivities of postmor- and highly influential set of field experiments tem transport and time-averaging, both in in Texas lagoons, Cummins et al. (1986b) doc- modern systems and the stratigraphic record, umented taphonomic ‘‘half-lives’’ as short as using a diverse array of scientific methods for 60 days for mm-scale postlarval shells, sug- different groups (Figs. 3, 4). Some key hypoth- gesting very high rates of carbonate shell re- eses of paleontologic reconstruction, for ex- cycling (and see Staff et al. 1985, 1986). How- ample order-of-magnitude estimates of time- ever, they subsequently calculated that nearly averaging, down-core stratigraphic mixing of all shell carbonate produced in those sedi- cohorts, and how damage accrues over ments must be preserved to obtain the ob- elapsed time, have been tested directly via ra- served shell content in the long-term record; diometric and other dating of modern death that is, virtually all of the larger shells that assemblages (particularly the series of papers constitute the bulk of the skeletal biomass pro- on molluscan assemblages of the Gulf of Cal- duced by the live community must survive ifornia; citations in preceding section). Both (Powell et al. 1992). Other actualistic and time-averaging and its relationship to ‘‘spatial stratigraphic evidence shows that the mollus- averaging’’ also have been explored produc- can fossil record is time-averaged but relative- tively via probabilistic modeling (various au- ly high-fidelity (various live/dead studies and thors in Kidwell and Behrensmeyer 1993). paleontologic analyses based on the Ͼϳ2-mm The expanding baseline of information on TAPHONOMY AND PALEOBIOLOGY 115

FIGURE 3. Estimated limits on time-averaging of selected types of continental plant tissues and vertebrate and marine invertebrate assemblages. The different categories (tissues versus deposits) reflect the fact that paleobota- nists regard tissue type as playing the most important role in time-averaging for plant remains, while paleozool- ogists regard depositional environment or process as more important. Modified from Kidwell and Behrensmeyer 1993.

modern and ancient systems also is fostering life position (the highest possible spatial fi- conceptual models, such as the reciprocal na- delity and resolution), biological remains can ture between the durability of remains and be transported out of their original life habi- their likely temporal and spatial acuity (Ko- tats, thereby becoming allochthonous and po- walewski 1997), and how the attributes of tentially problematic from the standpoint of temporal, spatial, and compositional fidelity paleocommunity reconstruction. Allochthon- vary independently. For example, a mass-buri- ous or ‘‘exotic’’ wind-dispersed spores and al event from which mobile species and adults pollen can account for high proportions of escaped can produce an assemblage with high taxa in some samples, especially in areas with time- and space-resolution but low composi- little local vegetation (e.g., middle of large tional fidelity (e.g., Fig. 5), whereas if the live lakes, offshore marine environments, and ice community is transported en masse during [Farley 1987; Calleja et al. 1993; Traverse the fatal event (e.g., avalanching, turbidity 1994]). In contrast, animal-pollinated pollen, currents), the temporal resolution and ecolog- leaves, and other macroscopic phytodebris are ical fidelity of the assemblage may be high but relatively heavy and their records tend to have spatial fidelity very low. Alternatively, a time- quite high spatial fidelity, although deposi- averaged assemblage in which hundreds or tional context must be considered. For exam- thousands of years of input are mixed (rela- ple, on temperate and tropical forest floors, ac- tively low temporal resolution) may nonethe- tualistic tests indicate that litter sampled at less contain virtually all preservable species any one point is derived largely from the sur- that lived in the area over that time, and per- rounding 1000–3000 m2 of vegetation (Burn- haps even in fairly accurate proportions (thus ham et al. 1992; and see Gastaldo et al. 1987; facies-level spatial resolution, and high eco- Burnham 1989, 1993, 1994; Meldahl et al. 1995; logical fidelity of a durable subset of the orig- and for pollen see Jackson 1994). Such easily inal community) (Fig. 5). degraded material must be buried quickly to Spatial Fidelity. Although the presence of a be preserved, but careful sampling of pre- taxon in a fossil assemblage suggests that it served spatial associations of taxa can capture occupied that site, particularly if ‘‘rooted’’ in extremely high-resolution macrofloral records 116 ANNA K. BEHRENSMEYER ET AL.

do et al. 1987; Jackson 1989; Traverse 1990; Burnham 1990; Thomasson 1991; Webb 1993). The skeletal hardparts of vertebrates and benthic invertebrates almost always are pre- served out of life position, but actualistic stud- ies indicate that out-of-life-habitat transport generally affects relatively few individuals in a given fossil assemblage (see reviews by Rich 1989; Allen et al. 1990; Behrensmeyer 1991; Kidwell and Bosence 1991; Behrensmeyer and Hook 1992; Kidwell and Flessa 1995; also spe- cific studies by Behrensmeyer and Dechant Boaz 1980; Miller 1988; Miller et al. 1992; Ne- belsick 1992; Greenstein 1993; Stoner and Ray 1996; Anderson et al. 1997; Flessa 1998; Cutler et al. 1999). Again, depositional context is cru- cial in determining spatial fidelity, and in pro- viding warning flags for highly biased assem- blages (Fig. 4). For example, in settings dom- inated by gravity-driven or surge transport of normal sediments, bioclasts may be entirely exotic in origin (e.g., in washover fans, tidal channels and their deltas, turbidites, base-of- slope settings). Organisms can be important transporters of biological remains, but there is great variation in the magnitude of this trans- FIGURE 4. Spatial and temporal representation in fossil port: some predators leave debris at the kill assemblages for different major groups of organisms, in continental and benthic marine depositional settings. A, site; others concentrate it in a den or midden Continental settings: dotted lines show areas of the within the prey’s life habitat (e.g., hyenas, time/space plot occupied by vertebrate remains, dashed crabs, fish, most owls), although in some cases lines plant remains; estimate for pollen excludes trees because certain morphotypes can be transported hun- prey remains end up outside of their life hab- dreds of miles by water (e.g., Farley 1987) or thousands itat (wolf dens, diving seabirds). Finally, the of miles by wind (e.g., Calleja et al. 1993) prior to set- sprinkling onto seafloors of rocky intertidal tling from the water or air column, respectively. B, Ben- thic marine settings include shelly macroinvertebrates shells rafted by seaweed (Bosence 1979), ver- and exclude nektonic and planktonic contributions to tebrate debris from necroplanktic carcasses the fossil assemblage, because spatial resolution of these (‘‘bloat and float’’ [Scha¨fer 1972]), down- components can depend upon current drift. stream transport of bones (Behrensmeyer 1982; Dechant Boaz 1994; Aslan and Behrens- sufficient for detailed reconstructions of di- meyer 1996), and wind-transport of remains versity and community interrelationships on land (Oliver and Graham 1994) can be (Wing and DiMichele 1995; Gastaldo et al. highly effective modes of transport, but can be 1993b, 1996, 1998; Davies-Vollum and Wing taphonomically subtle in terms of recognition 1998). Moss polsters and small-diameter and impact on the composition of fossil as- ponds collect pollen (both wind- and animal- semblage. pollinated types) from smaller areas of source Natural history observations contribute to vegetation than do large ponds (Jackson our conception of the possible, and net effects 1994). Fluvial channels and river deltas typi- of transport on composition have been inves- cally include—but usually are not dominated tigated by lab and field experiments (Behrens- by—plant remains from upstream parts of meyer 1982; Frison and Todd 1986; Lask 1993; their drainage basin (e.g., Scheihing and Pfef- Prager et al. 1996; Blob 1997; and others pre- ferkorn 1984; and see Collinson 1983; Gastal- viously cited; see also many live/dead com- TAPHONOMY AND PALEOBIOLOGY 117

FIGURE 5. Schematic diagram in which each axis represents the summed results of preservational processes af- fecting time resolution, spatial fidelity, and compositional fidelity relative to the living community, showing that virtually all combinations of these aspects of record quality are possible. Note that we are restricting Compositional Fidelity to the accuracy by which the fossils represent the species present, species abundances, and population structure of the original community. In addition to the major sources of taphonomic bias represented in this dia- gram, some organic parts, such as pollen, wood, and bovid teeth, have inherent limitations with respect to taxo- nomic assignment, and these restrict the attainable level of fidelity even when preservation is excellent. SIV ϭ shelly invertebrates, P ϭ plants, V ϭ vertebrates, TA ϭ terrestrial arthropods (mostly insects). parisons discussed in next section). It is assemblage, i.e., the period of time represent- difficult to retrodict or to reconstruct the ef- ed by the biological components of any fossil fects of extinct organisms as agents, but over- assemblage, varies over many orders of mag- all, out-of-habitat postmortem transport does nitude (from virtually zero to millions of not appear to be an overwhelming taphonom- years; Fig. 3) and depends upon many factors ic problem in ordinary depositional settings. (see papers in Kidwell and Behrensmeyer Biological remains do not become homoge- 1993; Kidwell 1998). These include (1) tissue nized in composition across broad environ- types, (2) the habitat and specifically the fre- mental gradients either in modern or in an- quency of burial events and exhumation cient settings (see review by Kidwell and Fles- events, and (3) the depth of bioturbation with- sa 1995). in the sedimentary column relative to net sed- Time-Averaging. Because population turn- iment accumulation rates (Fu¨ rsich and Aber- over rates of individual taxa are less (often han 1990; Kidwell and Bosence 1991; Behrens- much less) than net rates of sediment accu- meyer and Hook 1992; Cutler 1993; Martin mulation, the biological remains of successive, 1993; Kowalewski 1997). Robust hardparts noncontemporaneous populations of organ- can survive multiple reworking events, even isms may be admixed within a single bed, a within slowly accumulating sedimentary rec- concept first articulated by Walker and Bam- ords (stratigraphic condensation), and can bach (1971). Multiple lines of evidence indi- also survive exhumation and incorporation cate that the degree of time-averaging within any into quite different younger sedimentary de- 118 ANNA K. BEHRENSMEYER ET AL. posits (stratigraphic leakage) (e.g., Cade´e pollen). Moreover, mass mortalities of animals 1984; Henderson and Frey 1986). Highly dis- rarely capture complete communities because parate ages of co-occurring fossils generally agents of death often are taxon or age-class are linked to settings of erosion (such as mod- specific (Greenstein 1989), and organism size ern coastlines of Pleistocene or Tertiary strata and mobility also are important factors [Wehmiller et al. 1995]), or prolonged low net (Krantz 1974). In aquatic systems, mass mor- sedimentation (e.g., modern sediment-starved talities tend to capture unusual communities continental shelves, where shells from 20,000 or communities in unusual states (e.g., anoxia years ago to present co-occur in thin sedimen- is more common in small shallow bodies of tary veneers from most recent marine trans- water than in large open ones; mass death of gression [Kidwell and Bosence 1991; Flessa single-species aggregations may occur follow- and Kowalewski 1994; Flessa 1998]). Such ex- ing spawning events [Brett and Seilacher amples involve tissues that are particularly 1991]). durable at death (mollusk shells, vertebrate Because of the diverse ecological and taph- teeth, pollen and spores) or have been made onomic scenarios that are possible, time-av- more durable by diagenesis during temporary eraging can have a number of effects on the burial (much vertebrate material associated species diversity and composition of fossil as- with marine lags; steinkerns or concretions of semblages. However, for organisms that pro- delicate shells or refractory skeletons [Fu¨ rsich duce durable materials, the usual effect is to 1978; Kidwell 1991]). inflate diversity compared with what an ecol- Does time-averaging significantly impact ogist would measure (‘‘census’’) at any single our understanding of paleobiological sys- moment (Fu¨ rsich and Aberhan 1990). For ex- tems? The answer to this question depends on ample, among 81 different data sets compar- the scale of time-averaging and the quality of ing live marine mollusks and dead shells from information required to answer the paleobio- the same sediments, all contain two to ten logic question(s) at hand (Paul 1998). Time-av- times more dead species than species cen- eraged blurring of critical paleobiological sused alive in the same habitat, even when the events, such as the demise of the dinosaurs numbers of live individuals outnumber dead (Rigby et al. 1987; Argast et al. 1987), can have individuals (Kidwell in press; same phenom- an obvious effect on evolutionary reconstruc- enon for vertebrates, see Behrensmeyer 1993). tions. But there are many more subtle conse- When additional live-censuses are taken and quences of time-averaging. Catastrophic buri- their species pooled, the known live fauna be- al events such as volcanic ash falls (e.g., Voor- gins to more closely resemble the richness of hies 1992) can capture instantaneous samples the death assemblage. This demonstrates that, of landscapes and organisms caught in the in contrast to any single, instantaneous census ‘‘wrong place at the wrong time.’’ From a taken by an ecologist or captured by a cata- preservational perspective, the trapped inver- strophic mortality, time-averaged death as- tebrates, vertebrates, and macroflora may pro- semblages are fundamentally different types vide highly ‘‘correct’’ spatial and proportional of samples of communities, summing biolog- representations of the community at that spot ical input over longer periods (Peterson 1977; and instant in time. However, these cata- other examples in Kidwell and Bosence 1991). strophically trapped organisms may be ad- Other studies of marine and continental bio- mixed with (or at least deposited within very mineralizing groups indicate that the proba- close stratigraphic proximity to) seeds, paly- bility of incorporation into a time-averaged nological components, and bones already pre- death assemblage declines with tissue dura- sent in the soil. Depending upon the type of bility—from ϳ95% for shelled mollusks to soil and its maturity (Retallack 1990), these ϳ75% for echinoids and land mammals and pre-event remains may represent populations ϳ50% for marine decapods (Kidwell and Fles- that existed in the area prior to the catastroph- sa 1995). These general relationships contrast ic event, and/or populations that never lived with time-averaged assemblages of low-du- at the locality (e.g., some wind-transported rability organisms, in which species richness TAPHONOMY AND PALEOBIOLOGY 119 may be significantly undersampled relative to gone very little time-averaging. In leaf assem- the actual number of species that occur in the blages it is even possible to infer greater time- living community (e.g., leaf assemblages averaging than is actually the case. For [Wing and DiMichele 1995]). example, plant debris resting on a volcanic ash In very specific depositional settings, it is may represent canopy leaves shed as an im- possible to use stratigraphic evidence or bio- mediate response to ash loading (Burnham logical inference to constrain the absolute time and Spicer 1986), but could be mistaken for lit- over which biological remains have accumu- ter from a longer-term recolonization of the lated. The best circumstances are where there ash-fall deposit. is high temporal resolution based on radio- Paleobiologists have hoped to find a signa- metric dates (e.g., Potts et al. 1999), or a nat- ture of degrees of time-averaging in the state ural cyclicity within the biotic system (e.g., of fossil preservation, but so far this has seasonal deciduousness [Gastaldo et al. 1996]) proved elusive. Although old shells are more or in sediment deposition (e.g., lacustrine consistently in poor condition than young varves [Bell et al. 1987; Wilson 1993; Wilson shells (Powell and Davies 1990; Flessa et al. and Barton 1996; Briggs et al. 1998]). Variabil- 1993; Meldahl et al. 1997a,b) and shell ages ity in bone weathering stages has been used tend to increase down-core (Kershaw et al. as an indicator of time-averaging (e.g., Potts 1988; Cutler and Flessa 1990; Flessa et al. 1986), and geochemical signals acquired early 1993), neither the taphonomic grade (e.g., de- in diagenesis show promise as a way of cali- gree of abrasion or encrustation) nor the pre- brating relative degrees of time-averaging in cise relative stratigraphic positions of skeletal vertebrate accumulations (Trueman 1999). remains in the sedimentary column are infal- Generally, however, paleontologists estimate lible criteria for reconstructing the ages of in- the absolute and relative durations of time-av- dividual elements within molluscan assem- eraging by a process of elimination (see pa- blages. Individual shells within the same in- pers in Kidwell and Behrensmeyer 1993; Kid- tertidal assemblage can vary in 14C ages by well 1998). Assemblages with a high propor- more than 1000 years, quantifying the scale of tion of life-positioned and/or articulated time-averaging within a ‘‘bed,’’ and the age specimens, and especially those incorporating range increases to ϳ20,000 years for assem- nonmineralized tissues with known rates of blages from offshore subtidal areas (Flessa decay, can be categorized as snapshot-type and Kowalewski 1994; Kowalewski et al. census assemblages with minimal time-aver- 1998). This is a consequence of overall robust- aging (but see discussion above) (Fig. 2), ness of molluscan shells compared with other whereas the opposite extreme of highly con- shelly macroinvertebrates (Kidwell and Beh- densed or lag material may be recognized by rensmeyer 1993; Kidwell and Flessa 1995; Ko- the highly disparate diagenetic styles or bio- walewski 1996b) and can result in an ‘‘over- stratigraphic ages of co-occurring material complete’’ record when net sedimentation and (usually) close association with a signifi- rates are low—i.e., time represented by fossils cant stratigraphic discontinuity surface (Fu¨r- is greater than that represented by sediment sich 1978). Interpretation of material of inter- (Kowalewski 1996b; condensed assemblages mediate-scale time-averaging, which accounts of Fu¨ rsich 1978; Kidwell and Bosence 1991). for the vast majority of land vertebrate and This contrasts with the relatively low du- shelly invertebrate assemblages, depends on rability of weathered bone material, which is depositional context (Kidwell and Bosence less likely than fresh bone to survive to be- 1991; Behrensmeyer and Hook 1992) and, less come fossilized in continental environments confidently, state of fossil preservation (see be- (Behrensmeyer 1978a). Most transported low). The less durable the material, the shorter and/or attritional fossil bone assemblages the window for time-averaging and accrual of consist of durable, unweathered elements such progressive damage. In fact, most assemblag- as teeth, jaws, and fragmentary limb parts, es of non-woody plant material and nonmi- and the average state of fragmentation or dis- neralized invertebrates have probably under- integration is a poor index of the duration of 120 ANNA K. BEHRENSMEYER ET AL. surface exposure or the degree of time-aver- time-averaging capture the long-term spatial aging. However, high variability in weather- variability of populations in an area? In other ing state, fragmentation, or abrasion in a sin- words, does time-averaging equal spatial-av- gle assemblage can indicate a complex taph- eraging? Given an environment characterized onomic history, which should, on average, cor- by time-averaged death assemblages, would relate with greater time-averaging. The one expect to find within a single-point sam- relationship between bone damage variability ple a record of almost all the preservable taxa and time interval of accumulation needs test- that ever occupied the environment (McKin- ing via comparative studies in both modern ney 1991). Time-space equivalence would de- environments and the stratigraphic record. pend on two conditions. One is that, over the For example, weathering or abrasion features period of time-averaging, the physico-chemi- could be examined in concert with new chem- cal properties of the sample site must vary suf- ical approaches to time-averaging in bone de- ficiently to permit colonization by the entire posits, which suggest that variability in rare- range of organisms in the community. This earth elements is correlated with mixed spa- condition will generally be met only for sites tial bone sources, hence greater time-averag- that are large relative to the size of the organ- ing (Trueman and Benton 1997; Trueman ism, for example hundreds of square meters 1999). The mixed preservational quality of a for sessile invertebrates or plants. A second single type of shell also is taken as the best cri- condition is that, following burial, all com- terion for time-averaging within marine as- ponents of the spatially variable faunas or flo- semblages (Johnson 1960; Fu¨ rsich 1978; Kid- ras must have an equal chance for preserva- well and Aigner 1985; papers in Kidwell and tion. This condition could be met for taxa with Behrensmeyer 1993). similar kinds of organic or biomineralized re- Such extrinsic and intrinsic time-averaging mains. It is not possible at present to provide factors, along with analytical time-averaging an answer to the time-space equivalency ques- (i.e., postcollection pooling of specimens from tion, although the possibility is tantalizing for different sites or stratigraphic intervals [Beh- paleoecologists with good vertical but poor rensmeyer and Hook 1992]), reduce the re- lateral exposures. Multiple spatial samples of solving power of fossil assemblages for many time-equivalent fossil assemblages are needed paleobiological questions, especially those to test this hypothesis; evaluating diversity concerning species interactions, community over the sample area thus provides a way to composition, and fine-scale patterns of evo- assess potential spatial completeness. For ex- lution, compared with what is possible in ample, Bennington and Rutherford (1999) studies of modern biotas or fossil records used multiple, small samples across the ex- dominated by census assemblages (i.e., mac- posure and then calculated cluster confidence rofloral and nonmineralizing animals). Given intervals to estimate completeness. the thousands of years of time-averaging that are apparently common within modern mol- Compositional Fidelity of Fossil luscan assemblages, for example, Kowalewski Assemblages (1996b) has concluded that many paleobiolog- In the last 15 years, taphonomists have ap- ical questions below a millennial timescale plied a variety of research approaches to eval- cannot be addressed (and see limits on recon- uating the compositional fidelity of fossil as- structing environmental change by Roy et al. semblages, i.e., the quantitative faithfulness of 1996; Behrensmeyer 1982; Olszewski 1999; the record of morphs, age classes, species rich- and see Martin et al. 1999b). Anderson et al. ness, species abundances, trophic structure, (1998) believe that it is possible to isolate etc. to the original biological signal (e.g., Fig. shorter-term preservational and community 2). Research has included (1) extrapolations trends in time, but the hardpart assemblage from laboratory and field measurements of must be the product of episodic rather than rates of destruction of tissue types in modern continuous time-averaging. systems; (2) deductive analyses of fossil as- Relationship between Space and Time. Does semblages, in which the preservational quality TAPHONOMY AND PALEOBIOLOGY 121 of individual specimens and sedimentary con- tological information—is between organisms text are used to infer likely postmortem mod- having mineralized or highly refractory tis- ification of taxonomic composition (informed sues and those lacking such materials (‘‘soft- by point 1); and (3) actualistic live/dead stud- bodied taxa’’) (Fig. 2). Soft-bodied taxa have ies, in which the composition of a death as- negligible preservation potential under ordi- semblage (shells, bones, leaf litter, pollen) is nary environmental conditions, such as oxy- compared with the local living community. genated seafloors and lake beds, and land sur- These empirical approaches are complement- faces characterized by moist and/or warm ed by probabilistic models and computer sim- conditions (see earlier section), and the de- ulations aimed at testing both taphonomic struction of these organisms can represent a and ecological (supply-side) controls on the substantial loss in biological information. In nature of the record (Cutler and Flessa 1990; marine sands and muds, for example, such Miller and Cummins 1990, 1993; Behrensmey- taxa constitute 30–100% of species (Schopf er and Chapman 1993; Cutler 1993; and see 1978; Staff et al. 1986; Kidwell and Bosence Roopnarine et al. 1999 and Roopnarine 1999 1991; Masse´ 1999; and for hardground exam- for simulations of taphonomic effects on spe- ple, see Rasmussen and Brett 1985), and in the ciation patterns). macroflora a large proportion of the non- Most live/dead tests of fidelity have fo- woody (herbaceous) species can be missing cused on single taxonomic groups in a limited from litter samples (Scheihing 1980; Burnham suite of environments—as in ecological stud- 1989; Burnham et al. 1992). Thus, unless based ies, there are logistical limits to the scope of on konservat-lagersta¨tten with census-level an investigation. Methodological differences time resolution, most ‘‘reconstructions’’ of can make comparisons difficult across taxo- food webs and energy flows by paleoecolo- nomic and environmental divides (e.g., single gists differ fundamentally from those of living versus multiple pooled censuses of the live communities, and are useful only for compar- community; visual survey versus sieving of ison with similarly preserved (isotaphonom- upper sedimentary column for dead hard- ic) assemblages (Scott 1978; Behrensmeyer parts; methods of estimating individuals from and Hook 1992) or simulations based on living collections of disarticulated and discarded communities (e.g., Behrensmeyer and Chap- body parts). However, we are beginning to de- man 1993; Miller and Cummins 1993). velop a clear sense of how the construction One clear pattern from existing studies is and life habits of organisms and their post- that there is tremendous variance in fidelity mortem environment combine to determine even among ‘‘preservable’’ groups, linked to death assemblage fidelity for several systems the durability of their hardparts (for review (e.g., various papers in Martin et al. 1999a). see Kidwell and Flessa 1995). In the marine There also is an increasing number of com- realm, this is a function of hardpart construc- parative taphonomic studies across taxonomic tion: mollusks and nonagglutinating benthic and environmental boundaries (e.g., Jackson forams appear to have approximately equal and Whitehead 1991; Martin et al. 1996; An- durabilities and high fidelities, in terms of derson et al. 1997). Virtually all of these live/ species representation, and are followed in de- dead studies have been concerned with the creasing order by scleractinian corals, echi- species compositions of assemblages, rather noids, decapods, and agglutinating foraminif- than with trophic group, age-class and morph era (and for freshwater mollusks see Briggs et composition, or population size (species abun- al. 1990; Warren 1991; Cummins 1994). There dance) (but see Cummins 1986a; Palmqvist are few actualistic data for the postmortem 1991, 1993; Behrensmeyer 1993), and with nu- durability of brachiopods (but see Daley 1993; merical rather than biomass metrics (but see Kowalewski 1996a) and bryozoans (but see Behrensmeyer and Dechant Boaz 1980; Staff et Smith and Nelson 1994; Hageman et al. 2000), al. 1985). and no live/dead comparisons or direct age- One of the clearest contrasts in fossil pres- dating for these phyla to our knowledge. ervation—and thus in the fidelity of paleon- Hence information for these and other groups 122 ANNA K. BEHRENSMEYER ET AL. remains largely based on inferences from the bone survival compared with low latitudes, fossil record (e.g., lithology-specific diagenetic and within each of these zones, dry land sur- selection against small specimens of trilobites faces can be more favorable than moist ones [Chatterton and Speyer 1997]). (Behrensmeyer 1978a; Noe-Nygaard 1987; Research on reef corals is expanding from Kerbis Peterhans et al. 1993; Sept 1994; Tappen analysis of damage styles (Scoffin 1992; Pan- 1994a,b; Elias et al. 1995; Stewart et al. 1999). dolfi and Greenstein 1997a) to evaluations of Thus, taxonomic fidelity should be greater in ecological fidelity (Greenstein and Pandolfi cool and/or dry climates where bones have 1997; Pandolfi and Greenstein 1997b; Green- higher preservation potential. For the conti- stein et al. 1998). Results so far are mixed: nental macroplant record, Burnham (1989, deepwater settings yield high live/dead tax- 1993) demonstrated differences in systematic onomic agreements like those for mollusks, representation of taxa within modern forest- whereas in shallow water, environmental zo- floor leaf litters and channel deposits, and also nation is preserved but taxonomic congruence found that different depositional settings is low, with strong underrepresentation of within the same regime provide dissimilar fi- massive growth forms and overrepresentation delities. Within the subtropical fluvial regime, of (rapidly growing) branching forms among for example, channel and channel margin (for- the dead. Reef systems present quite different ebank) accumulations of leaves represent 13– conditions for live/dead analysis than soft 47% and 38–48% of the riparian vegetation, re- sediments: (1) dead specimens are commonly spectively. Counterintuitively, autochthonous overgrown and thus more difficult to detect levee and floodplain settings adjacent to these than live (see also this problem for reef-en- primarily allochthonous assemblages may crusting and -boring bivalves, where dead provide a fidelity record with as little as 29% richness is lower than live richness, contrary to of local vegetation represented (range from unlithified seafloors [Zuschin et al. 2000]); (2) 29% to 58% depending upon sample site). Be- resolution of corallite skeletal anatomy is es- cause arborescent plants along the river mar- sential for species-level identification of coral gin act as a barrier to the lateral movement of death assemblages, and thus systems with canopy materials, there is very little mixing greater time-averaging (and thus potential for among microhabitats. Different climatic re- taphonomic modification) or higher propor- gimes are characterized by different levels of tions of fragile forms (e.g., Indo-Pacific versus fidelity, and using comparative work in sub- Caribbean) will yield lower taxonomic fideli- tropical, tropical, and temperate climates, ties; (3) among colonial organisms, the per- Burnham developed ways to extrapolate and centage of an ‘‘individual’’ that is alive or calculate credible values of standing taxonom- dead must be estimated rather than simply ic richness by applying the appropriate cli- scored live/dead, and decisions must be made matic factor (Burnham 1993; see also Gastaldo about how to count fragments on the seafloor, and Staub 1999). if at all; and (4) similarly, decisions must be Tests of the relative fidelity of macrofloral made about whether dead material sieved and pollen records underscore the limitations from sedimentary pockets should be counted of any single type of paleoecological record and how best to integrate this with live/dead and the benefits of a comparative approach data for in situ corals based on stretched-line (Gastaldo and Ferguson 1998). For example, scuba transects. Gastaldo et al. (1998) incorporated megafloral, In the continental realm, fidelity among carpological (fruits and seeds), palynological, land mammals is strongly affected by body- and biogeochemical data to evaluate a late Ol- size distributions within habitats, agent of ac- igocene abandoned fluvial channel. They cumulation, and climate. Natural-history ob- demonstrated that leaf fossils recorded decid- servations and the few existing live/dead uous riparian plants; fruits and seeds not only studies of bones on open land surfaces sug- confirmed the presence of riparian elements gest that cool temperatures associated with but increased alpha diversity nearly threefold high latitudes and altitudes promote longer because these body parts represented under- TAPHONOMY AND PALEOBIOLOGY 123 story and herbaceous ground-cover plants ysis. All of these results underscore the poten- that were not preserved as wood or leaves; tial value of death assemblages for environ- palynological and palynofacies debris con- mental impact and other studies bearing on firmed the presence of some, but not all, ri- conservation biology (Powell et al. 1989; Da- parian taxa and added evidence for other local vies 1993; and see review by Kidwell and Fles- (algae) and regional components; and the bio- sa 1995). geochemical data reflected variations in me- Several more fundamental difficulties still gafloral contribution to the channel. For other challenge our application of these insights to examples see Jackson and Whitehead 1991 assessing fidelity in the older metazoan fossil and Ferguson 1995. record. One is the problem of evolutionary In addition to the need for quantitative es- ecology: not only have the durabilities of an- timates of fidelity in more environments and imal hardparts changed over time (e.g., with taxonomic groups, existing actualistic data changes in mineralogy, microstructure, body sets could be examined for taphonomic ‘‘rules size, skeletal robustness), but organisms that of thumb’’ applicable to the fossil record. For interact with skeletal hardparts also have example, do fidelity levels improve as data are evolved. Such organisms are some of the most pooled from increasingly large geographic ar- important agents of postmortem destruction eas (i.e., within sample, within facies/habitat, in modern systems (e.g., shell and bone-crush- within basin, within province) (Kidwell and ing predators and scavengers, various bioer- Bosence 1991; Cutler 1991; Wing and Di- oders, sediment-irrigating bioturbators, and, Michele 1995; Olszewski and West 1997; Had- of course, fungi and other microbes, whose ly 1999)? Temporally nested studies and sim- roles and histories as biological recyclers may ulations could reveal how stable death-assem- be important but still are difficult to assess blage composition is during the first few hun- [Robinson 1990, 1991]). How reasonable is it— dreds of thousands to one million years of and how far back into the past is it reason- burial (i.e., live versus dead in modern envi- able—to extrapolate present-day estimates of ronments, uplifted Holocene strata, and/or death-assemblage fidelity into the paleoeco- Pleistocene fossils). The degree of fidelity logical past (i.e., taphonomic uniformitarian- might be expected to decline because of the ism)? The Cenozoic and Cretaceous may be cumulative wear and tear of diagenesis, lim- within the reach of modern estimates of mol- ited outcrop areas for sampling different fa- luscan and scleractinian fidelity, but what do cies, changes in biogeographic range and we do with older records? To move beyond ar- community structure, and, eventually, extinc- guments based on taphonomic uniformitari- tion. However, data so far indicate that the anism, for both Paleozoic and post-Paleozoic agreement between live and dead floras/fau- material, it is essential to determine whether nas can remain rather high over periods of a death assemblages that appear to be highly bi- million years or more (e.g., Wolff 1975; Da- ased in composition have distinctive damage muth 1982; Valentine 1989; Greenstein and patterns—i.e., to link taphofacies studies of Moffat 1996), so much so that it permits rec- damage (e.g., dead specimens of Species X are ognition of the uniqueness of recent environ- in poor condition, but those of Species Y are mental degradation (Greenstein et al. 1998). in good condition) to information on live/ For example, Valentine (1989) reports that dead agreement. Greenstein (1999) has begun Pleistocene marine faunas in California in- such work on reef corals, and this should be clude 77% of the living mollusk species from incorporated into other live/dead investiga- the Californian province, with most ‘‘live- tions. only’’ species being numerically rare and re- Abundance of species in modern ecosys- stricted to deeper-water habitats that are not tems is a key variable for characterizing di- well represented in onshore Pleistocene out- versity and various measures of dominance, crops. Data from other marine and continental and reconstructing such information from the groups could be similarly tested for sensitiv- fossil record is important for investigating the ity to geographic and temporal scale of anal- history of biodiversity. Taphonomic processes 124 ANNA K. BEHRENSMEYER ET AL. have the potential to alter abundances in tures of morphologic variability in fossil ma- many significant ways—for example, via dif- terial, and distinguishing between taphonom- ferential destruction during time-averaging ic, sample-size, and biological controls on this and because of different population turnover variability, bears on issues of numbers of spe- rates of local species—even if the import of ex- cies and their stability in taxonomy. Thus, otic species is minor. Many live/dead inves- these considerations are critical to biostratig- tigations have generated adequate data to test raphy, evolutionary analysis, and estimates of agreement in rank order and relative abun- species richness (Hughes and Labandeira dance, but the numbers of studies are still too 1995). Increasing numbers of morphometric few for most groups to provide a credible ba- studies are based on taphonomically astute sis for using abundance data in paleobiologi- sampling, for example restricting samples to cal reconstructions. For marine mollusks, single bedding planes or horizons of constant meta-analysis indicates high live/dead agree- and (hopefully) known time-resolution, and ments (Kidwell 1999; and for freshwater mol- keeping close track of lithologic context to lusks, see Briggs et al. 1990; Warren 1991; control for or assess ecopheny; but the ques- Cummins 1994). Quaternary lake deposits tion of bias in morphologic representation at also provide a firm basis for assessing reli- present is still largely a qualitative assess- ability in palynological abundance data. For ment. the wind-pollinated part of plant communi- ties, pollen assemblages are faithful recorders Megabiases of plant relative abundances in the source ‘‘Megabias’’ refers to bias in relatively large- area, especially when the forest is relatively scale paleobiologic patterns, such as changes homogeneous, but animal-pollinated plants in diversity and community structure over are almost always grossly underrepresented tens of millions of years, and variation in the because of low pollen yield per tree and be- quality of the record between mass and back- cause most of this large and heavy pollen falls ground extinction times or among different very near the source tree (see reviews by Jack- climate states, biogeographic provinces, and son 1994 and by Jackson and Overpeck this tectonic settings (Behrensmeyer and Kidwell volume). In contrast, vertebrate paleontolo- 1985; and see treatments by Kowalewski and gists regard both relative abundances and Flessa 1996; Martin 1999). Baseline informa- rank-ordering of species as suspect (i.e., guilty tion accumulated since then has stimulated of bias unless proven otherwise) (e.g., Badgley new thinking on two reciprocal fronts: (1) 1982, 1986; Barry et al. 1991; but see Behrens- broad-scale changes in climate, plate tectonics, meyer and Dechant Boaz 1980). ocean-atmosphere chemistry, and biological Finally, morphological fidelity of fossil pop- evolution as likely drivers of secular change in ulations is also a taphonomic concern, but has taphonomic processes (Fig. 6); and (2) the received relatively little work. Preserved mor- probable impact of such changes on the qual- phologic variance might be affected in several ity of paleontologic evidence used to recon- ways, for example from time-averaging of struct and parameterize geological and bio- multiple generations (broadening variance) logical phenomena. and from differential destruction of fragile or Given that Earth history can be divided into small morphs (skewing variance or limiting periods with different atmospheric and sur- recognition of true polymorphism) (Kidwell face conditions, and given that the history of and Aigner 1985). Bell et al. (1987, 1989) pro- life also presents intervals with distinct bod- vide a powerful empirical example where, be- yplans and life habits, it seems plausible that cause of the occurrence of both mass-mortal- the geologic record would be characterized by ity (census) and time-averaged assemblages in a series of discernable ‘‘taphonomic domains’’ the same stratigraphic unit, the effects of time- (Fig. 6). Reflecting secular changes in the na- averaging on morphological variance and ture of life and environments on Earth at a character association could be evaluated en- global scale, these domains would constitute tirely with fossil evidence. Capturing true pic- the broadest-scale biases in the quality of pa- TAPHONOMY AND PALEOBIOLOGY 125 leobiological information. Superimposed on Gastaldo 1992; Allison and Briggs 1993a; Oost them would be province-scale and/or shorter- and de Boer 1994). term secular and cyclical variation in tapho- Although the potential impact of such large- nomic processes, and intervals of unique taph- scale processes has been recognized for some onomic conditions, e.g., those associated with time (Efremov 1940), they remain largely un- regional or global mass extinctions and/or explored aside from some aspects of the ma- major perturbations in Earth’s environmental rine record. Various case studies illustrate the condition. We refer to all such broad-scale continuing debate over the relative importance taphonomic patterns and trends—affecting of taphonomic versus biologic signals, and of paleontologic analysis at provincial to global intrinsic versus extrinsic taphonomic effects. levels, over timescales of Ͼ10 m.y., or among For example, in the fossil record of unique major taxonomic groups—as megabiases. events, such as the profound biotic changes at Important taphonomic shifts may result the Precambrian/Cambrian boundary, did from the evolution of organic form and be- multicellular life really ‘‘explode’’ about 550 havior that makes organisms intrinsically m.y. ago, or are we simply seeing the opening more or less likely to fossilize, and from of a new taphonomic window? This particular changes in extrinsic biotic and abiotic controls change appears to represent a linked shift in on preservation (Fig. 6). Examples of intrinsic metazoan evolution and organic recycling. changes include evolution in the composition Given the extensive field work on late Precam- and structure of mineralized skeletons, body brian deposits over the past two decades, it is size, mobility/life habit including burrowing very unlikely that shelly organisms existed in behavior and pollination, deployment of life any abundance prior to the end of the Precam- forms into new environments, and (for plants) brian. The evolution of biomineralization in- evolution of deciduous versus perennial deed represents a major taphonomic event in the intrinsic preservation potential of multicel- growth habits (for references, see discussion lular organisms. On the other hand, debate below). Examples of extrinsic biotic changes continues over the role of taphonomic process- include the increasing depth of bioturbation es in the concomitant disappearance of the through the Phanerozoic (Thayer 1983; Retal- globally distributed soft-bodied Ediacaran fau- lack 1990; Droser and Bottjer 1993; Buatois et na (Fedonkin 1994; McIlroy and Logan 1999). al. 1998; McIlroy and Logan 1999), the evolu- Researchers have hypothesized that, given the tion of more effective shell and bone crush- overall aerobic environments of deposition, ers/ingestors (Vermeij 1977, 1987; Behrens- some Ediacaran organisms had tougher body meyer and Hook 1992) and other biodegraders construction or ways of life that enhanced their such as fungi (Robinson 1990, 1991), and the intrinsic preservational characteristics, and that shift to detritivore and herbivore dominance the geological disappearance is a signal of ac- on land (DiMichele and Hook 1992; Laban- tual biological extinction (e.g., Seilacher 1984, deira 1998) (Fig. 6). Extrinsic physical changes 1994), whereas others have inferred a sharp de- include fluctuations in the temperature and cline at this time in the extrinsic environmental geochemistry of Earth’s atmosphere and conditions that permitted the development and oceans (Berner and Canfield 1989; Maliva et early cementation of microbial mats (Fedonkin al. 1989; Berner 1991; Martin 1995; Malinky 1994). Capable of preserving soft-bodied or- and Heckel 1998; Stanley and Hardie 1998), ganisms as ‘‘death masks,’’ this taphonomic and tectonic and climatic effects on the origi- mode existed in late Precambrian seas as long nal extent and preservation of particular en- as effective grazers and bioturbators (‘‘grave vironments through geological time (e.g., the robbers’’) were absent, but disappeared when ‘‘wetlands bias’’ in the global plant record, such organisms invaded this environmental variation in total sedimentary rock volume, zone (Gehling 1999). A comparable taphonom- the proportion of tropical continents, and lag- ic mode persists in Recent hypersaline tidal ersta¨tte-preserving lithographic limestone ba- flats and lagoons where most metazoans are sins [Tardy et al. 1989; Sepkoski et al. 1991; excluded, but in normal marine environments 126 ANNA K. BEHRENSMEYER ET AL.

FIGURE 6. Intrinsic and extrinsic changes with the potential for major effects on taphonomic processes and organic preservation over geologic time. This chart provides a preliminary framework for examining hypotheses concerning changing ‘‘taphonomic domains’’ through the fossil record for the marine and continental realms as well as possible TAPHONOMY AND PALEOBIOLOGY 127 it became extinct as a means of preserving soft- 1994; Kah and Knoll 1996; Schubert et al. bodied multicellular organisms at the begin- 1997). ning of the Cambrian (for related studies, see The histories of sedimentary basins on Sepkoski et al. 1991; Knoll et al. 1993; Knoll and timescales of 106–107 years could impart sig- Sergeev 1995; Kah and Knoll 1996). nificant trends in the quality of fossil records, Through the Phanerozoic, other proposed within the broader domains described above. megabiases in the marine realm relate to both Continental depositional systems could exhib- intrinsic and extrinsic biotic and abiotic fac- it several distinct phases of organic preserva- tors. At the intrinsic end of the spectrum, tion. For example, as a foreland basin changed styles of echinoid preservation and thus qual- from underfilled to overfilled, physical and ities of data appear to have diversified as echi- chemical conditions should favor plant pres- noid constructional morphology diversified ervation in early phases (high water table, low (Greenstein 1992); the frequency of lingulid oxidation) and vertebrate preservation in later brachiopod preservation has declined, possi- phases (paleosols with CaCO3, concentration bly because of decreasing biomineralization of vertebrate remains through fluvial rework- during life (Kowalewski and Flessa 1996); and ing) (e.g., as suggested in Demko et al. 1998; increases in the physical scale, taphonomic for analogous tectonic and sequence-strati- complexity and probable time-averaging of graphic variation in the quality of marine fos- shell beds is linked to shelly macrobenthos ac- sil records, see Kidwell 1991, 1993; Brett 1995). quiring biomechanically tougher hardparts Climate change also should impose major and expanding into more energetic environ- shifts in the quality of the fossil record, with ments (Kidwell and Brenchley 1994, 1996; Li wet cool conditions favoring plant preserva- and Droser 1997; Simo˜es et al. 2000b; and see tion, drier warmer times bone phosphate, and Ausich 1997 for intrinsic factors in encrinites). fluctuating CCD levels governing the preser- In addition, aragonitic and calcitic biominer- vation of deep-sea microfauna (Martin 1999; als may confer different degrees of resistance and for possible storm-bed effects, see Brandt to predation as well as to postmortem destruc- and Elias 1989). This potentially affects the tion (Stanley and Hardie 1998). In contrast, fossil record at a wide range of timescales, primarily extrinsic factors have been invoked from regular fluctuations in preservation to explain the changing frequency of meta- caused by orbital cycling, to longer-term shifts zoan konservat-lagersta¨tten over the Phaner- in latitudinal gradients that modify the extent ozoic (bioturbation, basin type, clay mineral- of habitable and preservable biotic space, to ogy) (Aronson 1992; Allison and Briggs the drift of continents across major global cli- 1993b; Butterfield 1995; Oost and de Boer mate belts. Moreover, processes and circum- 1994), and both intrinsic and extrinsic factors stances favorable to preservation of one major appear to have played a role in changing pat- group often are less favorable for others, re- terns of marine mineralization (both replace- sulting in potential temporal disjunctions be- ment of hardparts and early cementation of tween the marine and continental record, and sediments) over Precambrian and Phanero- the plant and land vertebrate record. zoic time (ocean saturation states, ecology and Taphonomic processes and circumstances biomineralogy of target taxa, abundance of associated with mass extinction events could other organisms as elemental sources) (Walker constitute a recurring set of phenomena with and Diehl 1985; Knoll et al. 1993; Grotzinger a different set of biases relative to the intervals

← links between these two realms. The points of inception of potentially important changes in intrinsic and extrinsic variables are indicated on the chart (see text for references). Shading variations on the vertical bars indicate possible taphonomic domains for shelly invertebrates based on depth of bioturbation, for plants based on the development of relatively refractory tissues and evolution within biodegraders such as fungi, and for vertebrates based on body size and changes in bone-processing capabilities of predators and scavengers. Revised dates for period boundaries from D. Erwin (personal communication 2000). 128 ANNA K. BEHRENSMEYER ET AL. between such events. For instance, after a mass sarily being willfully blase´, but may reason extinction event, a depleted array of consum- (geologically, statistically) that taphonomic ers could lead to reduced biological recycling, bias is relatively small and thus a second-or- allowing better preservation during the peri- der effect, or conclude that bias is random rel- od of biotic recovery (Hunter 1994; Williams ative to the variable under study and thus un- 1994; Cutler and Behrensmeyer 1996; D’Hondt likely to create artificial patterns. In other sit- et al. 1998). It is also possible that climatic or uations, the strategy is to normalize the data chemical stress prior to a mass extinction or adjust the metric to compensate for proba- could affect taphonomic processes, either di- ble or known biases (e.g., rarefaction, compar- rectly with increasing frequency of mass isons of data trends with unevenness in sam- deaths or indirectly through the organisms, pling intensity, which may be a function of with poorer preservation indicating increased rock availability as well as taphonomic pro- competition for scarce organic or biomineral cesses), use of taphonomic control taxa, Laz- resources. The record of mass death events arus and other gap analyses (see discussion also can be affected by long-term changes in below; also Sepkoski and Koch 1996; Foote taphonomic processes that control destruction this volume; Holland this volume). A variant and permanent burial. In many marine and of this strategy is analytic time- and space-av- continental settings where bioturbators or eraging (Behrensmeyer and Hook 1992), physical reworking can mix sediments over 10 whereby paleontologists group (bin) data to 100ϩ cm, instantaneous inputs from local more coarsely than nature and thereby reduce mass death may have no net stratigraphic sig- noise introduced by variation at lower scales; nature because the debris is rapidly homoge- this is axiomatic in many macroevolutionary nized with background attritional input (as in and global analyses. Another strategy, once Greenstein 1989; and see Badgley 1982; Mel- the quality of the record has been evaluated, dahl 1990; Behrensmeyer and Chapman 1993; is simply to sidestep biased or incomplete in- Cutler and Behrensmeyer 1996). Through the formation by rephrasing the hypothesis or Phanerozoic, the stratigraphic frequency of shifting the emphasis of the study to suit the documented mass mortalities thus may be quality of the data (see discussion in Paul partly controlled by evolutionary changes in 1992). bioturbation depth and intensity. The construction and analysis of paleobio- logical data sets usually entails many tapho- Strategies for Addressing Taphonomic nomic assumptions, and several points must Biases be kept in mind in designing strategies. One One of the goals of taphonomy is to estab- is the precise meaning of the terms ‘‘bias’’ and lish patterns, and preferably quantitative ‘‘incompleteness.’’ Bias by definition is neither magnitudes and causes, of bias in the fossil re- uniform nor random (occurring unpredict- cord. Paleontologists are developing new ably), but instead is a skewing of information ways to evaluate shortcomings both in the re- in some systematic way. This is a different cord itself and in our sampling of it, and many concern from the incompleteness of data, which, researchers now approach these issues rigor- as used in paleontology, refers to the extent of ously and proactively. Generally speaking, knowledge—that is, how fully the available approaches to biases, both real or hypothe- pool of information has been sampled and sized, have ranged from assuming or reason- thus how stable and detailed our picture of ing that the record itself—or our knowledge of that system is thought to be (Paul 1992, 1998). it—is too poor for biological analysis, to as- Incomplete information can provide a fair and suming or reasoning that the record can be true (unbiased) sampling of reality, and con- taken at face value. These usually reflect dif- fidence intervals can be calculated for it (e.g., ferent starting points for analysis—guilty of Sadler 1981; Strauss and Sadler 1989; Marshall overwhelming bias until proven otherwise, or 1990, 1994; McKinney 1991; see also various innocent until proven biased. Paleontologists papers in Gilinsky and Signor 1991). Collec- who take this latter approach are not neces- tion curves and other growth-of-information TAPHONOMY AND PALEOBIOLOGY 129 curves are longstanding examples (see Paul maxima, ratios (e.g., predator–prey) rather 1992). But if gaps in information preferentially than absolute numbers of taxa, rates and pat- fall, for whatever reason, within particular bi- terns of change rather than specific trajecto- ota, segments of time, or environments within ries, evenness rather than total species rich- the scope of the analysis, then incompleteness ness (e.g., Foote this volume; Alroy et al. this can be transformed into the more serious volume). Factoring out bias, on the other hand, problem of bias. requires analytic dissection of the data at A second point is that bias and incomplete- hand. For example, from an innocent-until- ness can both be either natural or analytical in proven-guilty stance, does evolutionary rate, origin. For example, gaps can result from hi- geographic range, or numbers of species per atuses and barren intervals in the stratigraph- genus actually covary with preservation cate- ic record, but gaps also are generated analyt- gories, such that using samples of diverse ically by coarse sampling schemes and by var- preservational quality in a single data set iation in monographic effort. Natural tapho- would yield a misleading paleobiologic inter- nomic bias includes the tendency for pretation (e.g., Jablonski 1988; Foote and Raup small-bodied individuals or species in a 1996)? From a guilty-until-proven-innocent group to be underrepresented, and for trans- stance, do observed spikes of taxonomic first gressive records to be thin and/or faunally and last appearances in stratigraphic sequenc- condensed relative to regressive deposits. Po- es rise above levels expected at those horizons tential sources of analytic bias include relying because of stratigraphic truncation and con- heavily on North American and European rec- densation (e.g., Holland 1995, 1996; Holland ords in the construction of global data sets, and and Patzkowsky 1999), or because of collector- interpreting at face value species richness data induced variation in sample size? from samples with disparate scales of time- By virtue of their size and scope, data sets averaging or positions on growth-of-knowl- on broad-scale paleobiological patterns entail edge curves. a particular set of taphonomic issues. For ex- A third point to be clear on, whether devis- ample, the marine metazoan diversity pattern ing a strategy or evaluating one used by oth- (Sepkoski 1978, 1993), the continental plant ers, is the reality that is being targeted—the record (Knoll et al. 1984; Kendricks and Crane actual history of life, the fossil record of life (a 1997), and major branching points in verte- taphonomically filtered subset of informa- brate evolution (Maxwell and Benton 1990; tion), the known fossil record (an analytic sub- Benton 1998) are interpreted to reflect biolog- set influenced by geopolitics), or a data set ical history at a global, stage-sized bin level, based on some analytic subset of the known and this may be correct at that scale of anal- fossil record (published data, unpublished ysis, although as yet there has been no com- data, and/or new fieldwork). Data that are prehensive analysis of biases. Some major po- complete at one scale (e.g., compendia from tential sources of bias, such as those pertain- the published fossil record) may be neither ing to variation in outcrop area and volume, complete nor unbiased at another more inclu- the ‘‘pull of the Recent’’ (intensive sampling sive scale (e.g., the actual fossil record). Con- of extant fauna extends the stratigraphic rang- versely, data that are incomplete and biased at es of poorly sampled fossils), monographic ef- a fine scale (e.g., major gaps in the local record fort per geologic period, and effect of hyper- because of facies controls on the original pres- rich lagersta¨tten (involving taxa that occur in ence or postmortem preservation of species) a single interval), have been addressed, at might yield adequate information at a broader least coarsely (e.g., for the marine record, see scale of analysis (e.g., presence of the higher Sepkoski et al. 1981). Moreover, the basic pat- taxon in the region anytime within a coarser terns (e.g., trends in numbers of families over time interval). time) have remained stable in spite of expand- Finally, some metrics will be less sensitive ing knowledge of the fossil record (Sepkoski to incomplete data than others—for instance, 1993; Benton 1998). Such ‘‘growth of knowl- medians rather than absolute minima and edge’’ curves (Paul 1992, 1998) support con- 130 ANNA K. BEHRENSMEYER ET AL.

fidence in the adequacy of information on the rived from an equivalent suite of natural sam- known fossil record for trends at this scale, and pling conditions. One strategy is to limit anal- this type of analysis can be applied at many ysis to samples from a specific habitat or suite levels in the taxonomic, temporal, and spatial of habitats or facies (e.g., Bambach 1977); this hierarchies (see examples in Donovan and assures greater taphonomic equivalence, al- Paul 1998; and for morphospace occupation, though it also limits the universality of the re- see Foote 1997). sults. The longer the interval of time, the more Research is now focusing on biodiversity important it is to sample within a single taph- patterns and biases at finer geographic scales, onomic domain: if natural taphonomic re- with the effects of environmental (facies) and gimes have shifted through time, and thereby biogeographic variation on numbers of re- altered the proportional preservation of bod- corded taxa and range limits (e.g., Raymond yplans, growth stages, or habitats (megabias), and Metz 1995; Vrba 1995; Wing and Di- then a major assumption behind biological in- Michele 1995; Alroy 1996; Behrensmeyer et al. terpretation of rarefied data is potentially vi- 1997), including the effects of nonrandom pat- olated, notwithstanding the within-habitat terns in facies and hiatuses through strati- design. For example, unconsolidated lower graphic sequences (Holland 1995). To what ex- shoreface facies from the Cenozoic record, tent, for example, does the acknowledged which include faunally condensed shell grav- weighting of published fossil data toward els, are likely to yield higher species richness- North America and Europe influence the per- es for both taphonomic and analytic reasons ceived global pattern? Determining whether than predominantly lithified units from the subregions have distinctly different biodiver- Mesozoic (Kidwell and Jablonski 1983; Kid- sity patterns will indicate the extent to which well and Brenchley 1996), and, owing to bio- uneven sampling across the globe might bias turbation, neither of these records preserves our perception to date of global trends. When as high a proportion of discrete storm-bed diversity through time within particular geo- concentrations as the Paleozoic (Brandt 1986; logical intervals is examined closely, different Sepkoski et al. 1991). Thus, for a variety of rea- patterns do emerge for different regions, and sons, taphonomic biases might be expected to ongoing research is testing the relative roles of inflate the raw alpha diversity of benthic com- intrinsic biologic factors, biological response munities in each erathem to a greater degree to regional environmental conditions, and than the one before it, making a biological in- taphonomic issues linked to regional environ- terpretation of diversity increase ambiguous ments (Miller and Foote 1996; Miller 1997; unless sampling is standardized. Cross-time Waisfeld et al. 1999). Databases that track the analysis of comparable samples, even if piece- environmental context and taphonomic char- meal—e.g., lithified Mesozoic with lithified acter of paleontologic occurrences—that is, Cenozoic, Paleozoic storm-beds with Meso- that are not done in a lithologic vacuum—are zoic storm-beds—is one obvious next step. an important next step in evaluating both nat- Another approach would be to characterize ural (taphonomic) and anthropogenic (analyt- taphonomic regimes and study their effects on ic) sampling bias with respect to global bio- apparent taxonomic diversity using sampling diversity patterns. standardization within each regime and in In many kinds of studies, but especially in combined samples. At basinal scales, nonran- evolutionary ecology, rarefaction (Sanders dom geographic and temporal variations in 1968; Raup 1975; Miller and Foote 1996) has completeness and taphonomy can in fact ac- been widely used as a means to standardize count for a large part of apparent faunal turn- samples (e.g., to compare species richness in over patterns through time (Brett 1995; Hol- samples of disparate size). To infer that differ- land 1995; Behrensmeyer et al. 1997), and thus ences in rarefied diversity over time or space scaled-up versions of similar biases might af- have biological explanations, however, one fect regional to global patterns. Benton (1998), must assume that the samples are isotaphon- for example, suggests that some marked shifts omic, i.e., that information for each bin is de- in Phanerozoic vertebrate diversity through TAPHONOMY AND PALEOBIOLOGY 131 time are not evidence of evolutionary events paleosol settings. Many macroplant assem- but are an artifactual ‘‘lagersta¨tten effect’’ of blages are from specific types of wetlands en- pooling data from scattered horizons of su- vironments (e.g., wet floodplain, proximal perb preservation with information from de- channel, abandoned channel, and channel), posits of more ordinary fossil preservation making them somewhat isotaphonomic by de- (this has been rejected as insignificant in the fault (ditto the records of soft-bodied ani- periodicity of marine metazoan extinctions mals), but water chemistry, rates of aggrada- [Sepkoski 1990; and see Foote this volume and tion, and climate can vary in ways that affect Alroy et al. this volume]). that preservation (Gastaldo 1994; Demko et al. At both large and small scales of investi- 1998; Gastaldo and Staub 1999). ‘‘Next-gen- gation, establishing taphonomic equivalence eration’’ analysis of macroevolutionary trends (isotaphonomy) has become an important re- requires such critical appraisal, both as a search goal. The aim is to achieve meaningful means of estimating confidence limits and as biological comparisons across space and time, a means of testing environmental forcing fac- and in particular to reduce dependency on tors. Although challenging, this is a logical modern analogues. Isotaphonomy is particu- next step toward integrating taphonomic ad- larly critical for establishing the credibility of vances into mainstream paleobiology. species abundance trends and other diversity An additional challenge in paleobiologic measures derived from the fossil record. Cri- analysis, especially of broad-scale patterns, is teria for defining isotaphonomic assemblages the significance of missing taxa, and the use should include as many lines of evidence as of ‘‘taphonomic control taxa’’ has been pro- possible without prohibitively limiting sam- posed as one way to determine when absences ple size: geological setting, general climatic are meaningful (Bottjer and Jablonski 1988; Ja- regime, lateral and vertical scale of the fossil- blonski et al. 1997). These are biologically iferous unit, paleogeochemistry, body-part abundant taxa with hardparts that are com- representation, and other indicators of taph- parably or less robust than those of the target onomic processes including time-averaging. taxon, and preferably relatively close taxo- This approach is relatively new and several nomically. The reasoning is that if the control strategies are possible. Benton (1998; also taxon is present, then the target should have Briggs and Clarkson 1990) suggests compar- also been preserved if it co-occurred in that ing similar types of lagersta¨tten at different unit; moreover, the author who reports the times as snapshots of ‘‘true’’ diversity, al- control taxon would have been likely to report though such deposits commonly record un- the target taxon if present. Thus, cyclostomes usual conditions of mortality and environ- may serve as control taxa for tracking the en- ment that limit their fidelity with respect to vironmental and evolutionary expansion of true or average diversity (see previous sec- cheilostome bryozoans, isocrinids for milleri- tions). Moreover, and as a general caveat, there crinid crinoids, and other small infaunal ve- are huge pitfalls (commonly unacknowl- neroids for tellinid bivalves. Likewise, teeth of edged) to extrapolating regional or global sig- Hipparion have been used as a control for the nals from single locales. An alternative start- appearance datum of the similarly sized horse ing point is to establish broader equivalence in Equus in Africa (Behrensmeyer 1978b), and depositional environments or taphofacies; turtles as a control for crocodiles in an anal- hence, vertebrate faunas from channel fill ver- ysis of climatic effects in the Cretaceous/Ter- sus levee versus floodplain paleosol settings tiary extinction (Markwick 1998). might be compared through time, and all are Foote et al. (1999) used similar taphonomic combined for a representation of the diversity reasoning to challenge molecular evidence for of the fluvial system as a whole. Clyde and an Early Cretaceous origin of modern Euthe- Gingerich (1998) take such an approach in ex- rian mammal orders, arguing instead that amining mammalian community response to these groups did not arise much before their environmental change in the late Paleocene earliest known geologic record in the latest using isotaphonomic samples from floodplain Cretaceous or Paleocene. The frequent lack of 132 ANNA K. BEHRENSMEYER ET AL. ancestral taxa in the fossil record often is at- Aronson 1992). These concerns would apply tributed to evolution in stratigraphically un- to other taxa that lack transitional forms in the derrepresented peripheral or poorly pre- fossil record; credible evidence that a taxon is served habitats (e.g., for ‘‘uplands’’ habitats absent for evolutionary rather than taphonom- for Cretaceous angiosperms and vertebrates, ic reasons requires a good understanding of see Olson 1966 and Retallack and Dilcher both the completeness of sampling and the 1986; and for Mesozoic plant taxa now rec- possible taphonomic bias in the fossil-bearing ognized in the Upper Carboniferous and deposits (and see discussion on bias relevant Permian, see DiMichele et al. in press). How- to bio-events in Sepkoski and Koch 1996). ever, Foote et al. (1999) take an important step In contrast to inverse-type models, where beyond this by quantifying preservation rates one works back from paleontologic patterns for other Cretaceous mammals and arguing (‘‘what bias could generate this pattern?’’), that, unless the taphonomy of the earliest eu- quantitative forward modeling of taphonomic therians differed radically from other mam- processes and effects from the input perspec- mal taxa in the Mesozoic (i.e., body sizes, hab- tive is in its infancy. Existing studies illustrate itats/facies, life habits) and in such a way that the great potential of this strategy for address- severely reduced the quality of their record, ing biases and limits to the resolution of the then the probability of missing modern euthe- fossil record. This modeling can be conducted rian orders throughout the Cretaceous is very at many scales and provides an important low (barring, they note, a ‘‘Garden of Eden’’ means of bridging the gap between actualistic in which all these orders both originated and data and stratigraphic patterns at the assem- remained in an undersampled region for tens blage level. Behrensmeyer and Chapman of millions of years). (1993), for example, take this approach in us- This approach touches upon a problem that ing computer simulations to create artificial needs resolution if we are to calibrate tapho- time-averaged vertebrate assemblages, based nomic effects on appearance events—namely, on a modern African assemblage and known the effect of a taxon’s live abundance, spatial rates of bone input. They show that hundreds distribution, and temporal range on its prob- to thousands of years of time-averaging is ability of preservation. Rare, localized, and needed to capture all of the major taxa in po- geologically short-lived species, such as might tential fossil localities (i.e., time-averaging is initiate a major new lineage, could be partic- good, and even necessary, for producing an ularly vulnerable to taphonomic bias (see ear- accurate portrayal of species presence and lier discussion of Valentine 1989), and the re- rank order) (also see Miller and Cummins cord of continental taxa might suffer dispro- 1990, 1993 on marine taxa). Building upon portionately for both taphonomic (patchy empirical evidence that modern molluscan preservation) and biologic (e.g., greater im- death assemblages are dominated by the portance of endemism in origination?) rea- shells of recent cohorts (Meldahl et al. 1997a), sons. For example, detailed study of Cenozoic Olszewski (1999) models time-averaging and vertebrate assemblages indicates that distinct predicts the sample size necessary to ensure faunal communities can exist in adjacent ba- retrieval of specimens from each component sins or portions of basins for long periods of time segment in the source assemblage, i.e., a time (106–107 yr) (Behrensmeyer 1978b; Bown complete sample of the entire span of time-av- and Beard 1990). Whether this segregation is eraging. There have also been immense taphonomic or ecologic in origin, it indicates strides in conceptualizing and testing the ef- how a taphonomic control approach might fects of incomplete preservation and strati- impart misleading results on vertebrate pres- graphic gaps on evolutionary patterns in the ence/absence at this temporal and spatial broadest sense, including tempo and mode of scale, and raises the issue of how such facies speciation, and in using estimates of phylog- control might ‘‘scale up’’ to global patterns eny to infer the quality of the fossil record. For over tens to hundreds of millions of years and recent entries to this large literature, see Car- at higher taxonomic levels (DiMichele and roll 1997, Roopnarine et al. 1999, and Wagner TAPHONOMY AND PALEOBIOLOGY 133

2000a; also chapters by Alroy et al., Foote, Hol- including the wide differences in susceptibility land, and Wagner in this volume. to time-averaging of major taxonomic groups. Is it possible to develop a general index of rel- Focal Areas for the Future ative ‘‘preservation potential,’’ including likely Taphonomy’s continuing challenge is to degrees of time-averaging, for different body- evaluate the prolific but problematic fossil rec- plans, life strategies, and ecological settings? ord for systematic patterns in preservation Can we establish a basis for recognizing ‘‘ab- that may constitute bias in information quali- normal preservation,’’ indicating profound ty, and to develop accurate ways to measure shifts in taphonomic regimes, such as post-ex- and use such patterns in paleobiologic analy- tinction differences in shelly faunas? It should sis. Over the last few decades, paleontologists be possible to take assemblage-level processes have become more sanguine about the quality and biases, and develop hypotheses about how of paleontologic data. There are two aspects to these operate at a larger scale as a basis for de- this: (1) an appreciation that all data, paleon- fining ‘‘normal’’ circumstances of preservation tologic or otherwise, are incomplete, and that for individual taxa and for different types of the critical question is whether they are ade- communities. quate to address the question at hand (Paul To advance these aims, and to summarize 1998); and (2) a realization that taphonomic our preceding highlights of the discipline, we comparability or noncomparability of samples recommend the following key focal areas for across time and space must be taken into con- future taphonomic research relevant to paleo- sideration in deriving biological patterns from biology: paleontological data. This means that, ideally, samples used to examine temporal and spatial 1. Field and lab experiments on the budget trends should be from comparable deposition- of input and permanent burial and on the al contexts and preservational states, even if rates, agents, pathways, and conditions of re- absolute scales of time-averaging or spatial fi- cycling of biological materials, especially the delity cannot be specified. Alternatively, when relatively subtle geochemical and geomicro- the point is to compare or combine biological biological aspects of ‘‘weathering’’ on and just factors such as diversity across environments, below the depositional interface. regions, and geological domains, an opposite 2. Quantification of time-averaging for a approach is required that takes into account broader array of taxonomic groups and de- and compensates for clearly different qualities positional settings, including the relative con- of data. Today this is largely done by tapho- tributions of successive cohorts of material. In nomic uniformitarianism, i.e., extrapolating part, such work can test and amplify the hy- modern-day rates and error estimates back in pothesis that time-averaged assemblages are time. A challenge for next-generation research dominated by the most recent cohorts, as sug- is to assess the very real limits imposed by gested by recent empirical work on marine secular changes in fossilization through the mollusks. Recognizing scales of time-averag- geological record. ing via damage levels and other tangible clues There are particular focal areas for taphon- in surviving fossil material is a key aspect of omy that are likely to generate important con- this research. tributions in the next several decades. Com- 3. Actualistic estimates of the compositional parison across plants, invertebrates and verte- fidelity (for species richness, abundances, age brates is a promising growth area for taphon- groups, etc.) of assemblages for a broader ar- omy’s broader contributions to understanding ray of taxonomic groups and depositional set- geobiological processes and to developing a tings, including explicit attention to whether theoretical basis for the field. Enough now is fidelity can be inferred from observed levels of known about each major group to suggest damage (i.e., taphofacies information) and to some common denominators, such as the ef- how fidelity varies as a function of geographic fects of bioturbators and bioeroders, as well as scale of investigation and the geologic aging some contrasts in approaches and problems, of an assemblage. 134 ANNA K. BEHRENSMEYER ET AL.

4. Long-term stratigraphic trends in the qual- logic phenomena such as evolutionary rates ities (Table 1) of the fossil record, including and diversity through time in marine versus use of the Pleistocene or Neogene as a reflec- continental organisms. tion of the Recent in order to investigate taph- Recent and future contributions of taphon- onomic modification associated with longer omy are relevant to an array of paleobiologic is- periods of time/space averaging, lithification, sues including the following (Jablonski 1999): and other aspects of the ‘‘permanent’’ fossil record. 1. Paleocommunity structure and composition, 5. A major initiative in probabilistic and other and how this changes through time in response to quantitative modeling as a means of testing ex- environmental perturbations, especially climate isting hypotheses and formulating new hy- shifts. Establishing paleocommunity structure potheses to test in the stratigraphic record and depends heavily on studies in recent ecosys- Recent systems. tems, but neontologic and paleontologic views of communities differ in their focus on what After several decades of intensive research, controls species distribution and behavior taphonomists now visualize the fossil record (live versus dead), their selective treatment of of taxa with mineralized or highly refractory particular taxonomic groups, and especially tissues as dominated by time-averaged assem- their degrees of temporal sampling. More ex- blages, with widely spaced horizons and in- change between neo- and paleoecologists, tervals bearing higher resolution records of with an emphasis on collecting new types of taxa and paleocommunities. This contrasts field data and modeling fossil assemblages us- with groups lacking readily preserved tissues: ing actualistic data, could generate new in- macroplant and soft-bodied animal records sights and a stronger foundation for recon- clearly are subject to much less time-averag- structing paleocommunities. As a basis for ing per assemblage than is true for pollen, ma- this, we need more quantitative information rine shelly faunas, or land vertebrates, and on the potential spatial fidelity, temporal mix- preservation is limited to a narrower range of ing, and compositional fidelity (percent living environmental conditions. The result is a se- and preservable species) for a wider array of ries of geographically and temporally narrow environments and taxonomic groups. By em- windows of high anatomical and temporal pirically linking levels of bias (inferred quali- resolution, relatively widely separated in ties) to damage profiles (observed states of space and time, producing a historical record preservation of species) and depositional with many gaps. Hence the fundamental modes, we also can develop criteria for isota- trade-off now recognized in taphonomy: the phonomic equivalence across space and time, better the preservation of individual organ- both between Recent and ancient biological isms and the finer the temporal resolution of systems and through comparative work with- individual samples, the less likely these are to in the stratigraphic record. be repeated at close and regular intervals 2. The history of biodiversity dynamics at dif- through geological time. The tendency for the ferent scales, from individual assemblages to global taphonomically most robust groups to exhibit tallies of diversity (numbers of species, genera, fam- the greatest time-averaging (and thus spatial ilies, etc.). There is a clear need for better un- averaging) (reciprocal model of Kowalewski derstanding of taphonomic effects on Phan- 1997) is a key corollary of this pattern. Ta- erozoic (and Precambrian) diversity patterns. phonomy’s agenda for the future revolves Large-scale shifts in taphonomic regime (links around better understanding the genesis and between organic preservation and the chemi- fidelity of these different types of records cal and physical states of the earth as well as (time-averaged and time-specific), how their faunal/floral evolution) or between recurring attributes are affected by local to global-scale taphonomic states (e.g., due to climate) as sug- tectonic, climatic, and biotic conditions, and gested in Figure 6, may be contributing con- how these taphonomic differences affect our founding patterns to the diversity curves. assessment and understanding of paleobio- These megabiases will not necessarily be elim- TAPHONOMY AND PALEOBIOLOGY 135 inated by approaches such as rarefaction, and tory of the biotic system to secular and cyclic geo- modeling what happens to diversity measures chemical and geophysical changes in Earth and its over stratigraphic shifts in taphofacies could atmosphere. Taphonomy’s continuing role is to help to clarify the effects of taphonomy on rar- characterize sampling biases that affect mac- efaction ‘‘universes.’’ Controlling taphonomic roevolutionary reconstructions, but it contrib- biases using isotaphonomic approaches may utes an additional perspective on macroevo- permit us to develop robust Phanerozoic di- lution through its focus on the recycling of or- versity patterns for particular environments, ganic and inorganic materials. Such processes and this should be complemented by greater have undoubtedly responded to and also af- efforts to develop nonactualistic means of es- fected environmental changes on Earth. Thus, timating and compensating for bias when is- improved information on geologic intervals otaphonomy is impossible. And the explosion and settings where physical, chemical, and bi- of understanding of the chemical and physical ological recycling has been particularly effec- aspects of fossil preservation (soft and hard tive or particularly ineffective in breaking parts) provides a framework for assessing the down organic materials is essential. Investi- stratigraphic patterning of census assemblag- gation of the macrotaphonomic history of the es (including konservat-lagersta¨tten) and biotic system will involve integrating different judging what we are missing in other parts of scales of evidence for plants, invertebrates, the fossil record. and vertebrates, and developing hypotheses 3. Rates of evolutionary events (originations, ra- about how taphonomic patterns through time diations, extinctions, and rebounds after extinc- relate to Earth’s physical and chemical history. tions), including major periods of faunal and floral change at the end of the Permian, the K/T bound- Beyond the paleobiologic issues discussed ary, and the Pleistocene. Establishing rates de- in this review, taphonomy has much to con- pends on accurate biostratigraphic records of tribute to the fields of ecology, biogeochemis- taxonomic presence, absolute dating of these try, sedimentary geology and stratigraphy, records, and comparisons that are matched for paleoanthropology, and conservation biology. the durations over which change was mea- We look forward to even greater cross-disci- sured. An important goal for taphonomy is to plinary collaborations of information and sci- develop more rigorous measures of ‘‘preser- entific methods to apply to taphonomic data vation potential’’ for different types of organ- and questions. Facts—the next generation of isms and provide alternative tests of biostrati- data acquisition and analysis of individual graphic range limits for comparison with taxa and assemblages—will continue to be those based on abundance patterns and gap fundamentally important to all aspects of ta- analysis. There should be a search for sub-Re- phonomy in the coming decades. But there cent and Plio-Pleistocene analogues in which also should be much more attention to syn- known extinctions or appearances are record- thesis, at the local, regional, and global levels. ed in stratigraphic sequences with high tem- Next-generation research can target data ac- poral resolution to provide comparisons for quisition that feeds into the search for larger the more distant geological record. New un- patterns and provides tests for interim hy- derstanding of processes of preservation and potheses about global-scale changes in taph- destruction at the molecular to sequence onomic regimes and megabiases affecting the stratigraphic scales could feed into simula- largest-scale paleobiological interpretations of tions of real versus apparent records of taxo- the history of life. nomic ranges. Likewise, increased apprecia- Acknowledgments tion of the limits to resolution provided by space- and time-averaging can help to provide We thank the editors of this special issue for reality checks and quantification of error-bars their encouragement and suggestions during in correlating environmental change with ma- the writing of this article, and A. I. Miller, R. jor biotic events in Earth history. R. Rogers, and D. Jablonski for helpful re- 4. Correspondence of the macroevolutionary his- views. 136 ANNA K. BEHRENSMEYER ET AL.

Literature Cited future challenges. Philosophical Transactions of the Royal So- ciety of London B 354:77–87. Allen, J. R. L., R. A. Spicer, and A. K. Behrensmeyer. 1990. Trans- Badgley, C. E. 1982. How much time is represented in the pre- port—hydrodynamics. Pp. 227–235 in Briggs and Crowther sent? The development of time-averaged modern assemblag- 1990. es as models for the fossil record. In B. Mamet and M. J. Co- Allison, P. A. 1986. Soft-bodied animals in the fossil record: the peland, eds. Proceedings of the Third North American Pale- role of decay in fragmentation during transport. Geology 14: ontological Convention 1:23–28. 979–981. ———. 1986. Counting individuals in mammalian fossil assem- ———. 1988. Konservat-Lagersta¨tten: cause and classification. blages from fluvial environments. Palaios 1:328–355. Paleobiology 14:331–344. Badgley, C. E., and A. K. Behrensmeyer. 1995. Preservational, Allison, P. A., and D. E. G. Briggs, eds. 1991a. Taphonomy, re- paleoecological and evolutionary patterns in the Paleogene of leasing the data locked in the fossil record. Plenum, New York. Wyoming–Montana and the Neogene of Pakistan. Palaeo- ———. 1991b. The taphonomy of soft-bodied animals. Pp. 120– geography, Palaeoclimatology, Palaeoecology 115:319–340. 140 in Donovan 1991. Baird, G. C., S. D. Sroka, C. W. Shabica, and G. J. Kuecher. 1986. ———. 1993a. Paleolatitudinal sampling bias, Phanerozoic spe- Taphonomy of Middle Pennsylvanian Mazon Creek area fos- cies diversity, and the end-Permian extinction. Geology 21: sil localities, northeast Illinois: significance of exceptional fos- 65–68. sil preservation in syngenetic concretions. Palaios 1:271–285. ———. 1993b. Exceptional fossil record: distribution of soft-tis- Bambach, R. K. 1977. Species richness in marine benthic habitats sue preservation through the Phanerozoic. Geology 21:527– through the Phanerozoic. Paleobiology 3:152–167. 530. Barry, J. C., M. E. Morgan, A. J. Winkler, L. J. Flynn, E. H. Lind- Allison, P. A., and K. Pye. 1994. Early diagenetic mineralization say, L. L. Jacobs, and D. Pilbeam. 1991. Faunal interchange and and fossil preservation in modern carbonate concretions. Pa- Miocene terrestrial vertebrates of southern Asia. Paleobiology laios 9:561–575. 17:231–245. Allison, P. A., C. R. Smith, H. Kukert, J. W. Deming, and B. Ben- Bartels, C., D. E. G. Briggs, and G. Brassel. 1998. The fossils of nett. 1991. Deep-water taphonomy of vertebrate carcasses: a the Hunsru¨ ck Slate: marine life in the Devonian. Cambridge whale skeleton in the bathyal Santa Catalina Basin. Paleobi- University Press, Cambridge. ology 17:78–89. Barthel, K. W., N. H. M. Winburne, and S. Conway Morris. 1990. Alroy, J. 1996. Constant extinction, constrained diversification, Solnhofen: a study in Mesozoic paleontology. Cambridge and uncoordinated stasis in North American mammals. Pa- University Press, Cambridge. laeogeography, Palaeoclimatology, Palaeoecology 127:285– Bartley, J. K. 1996. Actualistic taphonomy of cyanobacteria: im- 311. plications for the Precambrian fossil record. Palaios 11:571– Alroy, J., P. L. Koch, and J. C. Zachos. 2000. Global climate 586. change and North American mammalian evolution. In D. H. Bartram, L. E., Jr., and C. W. Marean. 1999. Explaining the ‘‘Kla- Paleobiology Erwin and S. L. Wing, eds. Deep time: ’s perspec- sies pattern’’: Kua ethnoarchaeology, the Die Kelders Middle tive. Paleobiology 26(Suppl. to No. 4):259–288. Stone Age archaeofauna, long bone fragmentation and car- Anderson, L. C., B. K. Sen Gupta, R. A. McBride, and M. R. nivore ravaging. Journal of Archaeological Science 26:9–29. Byrnes. 1997. Reduced seasonality of Holocene climate and Baumiller, T. K., G. Llewellyn, C. G. Messing, and W. E. Ausich. pervasive mixing of Holocene marine section: northeastern 1995. Taphonomy of isocrinoid stalks: influence of decay and Gulf of Mexico shelf. Geology 25:127–130. autonomy. Palaios 10:87–95. Anderson, L. C., R. A. McBride, M. J. Taylor, and M. R. Byrnes. Behrensmeyer, A. K. 1978a. Taphonomic and ecological infor- 1998. Late Holocene record of community replacement pre- mation from bone weathering. Paleobiology 4:150–162. served in time-averaged molluscan assemblages, Louisiana ———. 1978b. Correlation in Plio-Pleistocene sequences of the chenier plain. Palaios 13:488–499. northern Lake Turkana Basin: a summary of evidence and is- Andrews, P. 1990. Owls, caves, and fossils. University of Chi- sues. Pp. 421–440 in W. W. Bishop, ed. Geological background cago Press, Chicago. to fossil man. Scottish Academic Press, Edinburgh. Andrews, P., and J. Cook. 1989. Natural modifications to bones ———. 1982. Time resolution in fluvial vertebrate assemblages. in a temperate setting. Man 20:675–691. Paleobiology 8:211–227. Archer, M., S. J. Hand, and H. Godthelp. 1991. Riversleigh: the ———. 1987. Miocene fluvial facies and vertebrate taphonomy story of animals in ancient rainforests of inland Australia. in northern Pakistan. In F. G. Ethridge, R. M. Flores, and M. Reed Books, Balgowlah, New South Wales. D. Harvey, eds. Recent developments in fluvial sedimentolo- Argast, S., J. O. Farlow, R. M. Gabet, and D. L. Brinkman. 1987. Transport-induced abrasion of fossil reptilian teeth: impli- gy. SEPM Special Publication 39:169–176. cations for the existence of Tertiary dinosaurs in the Hell ———. 1988. Vertebrate preservation in fluvial channels. Pa- Creek Formation, Montana. Geology 15:927–930. laeogeography, Palaeoclimatology, Palaeoecology 63:183– Aronson, R. B. 1992. Decline of the Burgess Shale fauna: eco- 199. logic or taphonomic restriction? Lethaia 25:225–229. ———. 1990. Transport/hydrodynamics of bones. Pp. 232–235 Aslan, A., and A. K. Behrensmeyer. 1996. Taphonomy and time in Briggs and Crowther 1990. resolution of bone assemblages in a contemporary fluvial sys- ———. 1991. Terrestrial vertebrate accumulation. Pp. 291–335 in tem: the East Fork River, Wyoming. Palaios 11:411–421. Allison and Briggs 1991. Ausich, W. I. 1997. Regional encrinites: a vanished lithofacies. ———. 1993. The bones of Amboseli: bone assemblages and Pp. 509–519 in C. E. Brett and G. C. Baird, eds. Paleontological ecological change in a modern African ecosystem. National events: stratigraphic, ecological and evolutionary implica- Geographic Research 9:402–421. tions. Columbia University Press, New York. Behrensmeyer, A. K., and R. E. Chapman. 1993. Models and Ausich, W. I., and G. D. Sevastopulo. 1994. Taphonomy of Lower simulations of time-averaging in terrestrial vertebrate accu- Carboniferous crinoids from the Hook Head Formation, Ire- mulations. Pp. 125–149 in Kidwell and Behrensmeyer 1993. land. Lethaia 27:245–256. Behrensmeyer, A. K., and D. E. Dechant Boaz. 1980. The Recent Bada, J. L., X. S. Wang, and H. Hamilton. 1999. Preservation of bones of Amboseli Park, Kenya, in relation to East African pa- key biomolecules in the fossil record: current knowledge and leoecology. Pp. 72–92 in A. K. Behrensmeyer and A. P. Hill, TAPHONOMY AND PALEOBIOLOGY 137

eds. Fossils in the making. University of Chicago Press, Chi- Bosence, D. W. J. 1979. Live and dead faunas from coralline algal cago. gravels, Co. Galway. Palaeontology 22:449–478. Behrensmeyer, A. K., and R. W.Hook. 1992. Paleoenvironmental Bottjer, D. J., and D. Jablonski. 1988. Paleoenvironmental pat- contexts and taphonomic modes. Pp. 15–136 in A. K. Beh- terns in the evolution of post-Paleozoic benthic marine inver- rensmeyer, J. D. Damuth, W. A. DiMichele, R. Potts, H.-D. tebrates. Palaios 3:540–560. Sues, and S. L. Wing, eds. Terrestrial ecosystems through Boulton, A. J., and P. I. Boon. 1991. A review of methodology time. University of Chicago Press, Chicago. used to measure leaf litter decomposition in lotic environ- Behrensmeyer, A. K, and S. M. Kidwell. 1985. Taphonomy’s con- ments: time to turn over an old leaf? Australian Journal of Ma- tributions to paleobiology. Paleobiology 11:105–119. rine and Freshwater Research 42:1–43. Behrensmeyer, A. K., N. E. Todd, R. Potts, and G. E. McBrinn. Bown, T. M., and K. C. Beard. 1990. Systematic lateral variation 1997. Late Pliocene faunal turnover in the Turkana Basin, in the distribution of fossil mammals in alluvial paleosols, Kenya and Ethiopia. Science 278:1589–1594. lower Eocene Willwood Formation, Wyoming. Geological So- Bell, M. A., M. S. Sadagursky, and J. V. Baumgartner. 1987. Util- ciety of America Special Paper 243:135–151. ity of lacustrine deposits for study of variation within fossil Brandt, D. S. 1986. Preservation of event beds through time. Pa- samples. Palaios 2:455–466. laios 1:92–96. Bell, M. A., C. E. Wells, and J. A. Marshall. 1989. Mass-mortality Brandt, D. S., and R. J. Elias. 1989. Temporal variation in tem- layers of fossil stickleback fish: catastrophic kills of polymor- pestite thickness may be a geologic record of atmospheric phic schools. Evolution 43:607–619. CO2. Geology 17:951–952. Bengtson, S. 1994. Early life on Earth (Nobel Symposium No. Brett, C. E. 1995. Sequence stratigraphy, biostratigraphy, and ta- 84). Columbia University Press, New York. phonomy in shallow marine environments. Palaios 10:597– Bennington, J. B., and S. D. Rutherford. 1999. Precision and re- 616. liability in paleocommunity comparisons based on cluster- Brett, C. E., and G. C. Baird. 1986. Comparative taphonomy: a confidence intervals: how to get more statistical bang for your key to paleoenvironmental interpretation based on fossil sampling buck. Palaios 14:506–515. preservation. Palaios 1:207–227. Benton, M. J. 1998. The quality of the fossil record of the ver- Brett, C. E., and A. Seilacher. 1991. Fossil Lagersta¨tten: a taph- tebrates. Pp. 269–300 in S. K. Donovan and C. R. C. Paul, eds. onomic consequence of event sedimentation. Pp. 283–297 in G. The adequacy of the fossil record. Wiley, New York. Einsele, W. Ricken, and A. Seilacher, eds. Cycles and events in stratigraphy. Springer, Berlin. Berner, R. A. 1991. A model for atmospheric CO2 over Phaner- ozoic time. American Journal of Science 291:339–376. Brett, C. E., A. J. Boucot, and B. Jones. 1993. Absolute depths of Berner, R. A., and D. E. Canfield. 1989. A new model for atmo- Silurian benthic assemblages. Lethaia 26:25–40. spheric oxygen over Phanerozoic time. American Journal of Brett, C. E., T. E. Whiteley, P. A. Allison, and E. L. Yochelson. Science 289:333–361. 1999. The Walcott-Rust quarry: Middle Ordovician trilobite Best, M. M. R., and S. M. Kidwell. 1996. Bivalve shell taphonomy Konservat-Lagersta¨tten. Journal of Paleontology 73:288–305. Briggs, D. E. G. 1993. Fossil biomolecules. Bulletin of the Bio- in tropical siliciclastic environments: preliminary experimen- chemical Society 15:8–12. tal results. In J. E. Repetski, ed. Sixth North American pale- ———. 1995. Experimental taphonomy. Palaios 10:539–550. ontological convention, Abstracts of papers. Paleontological Briggs, D. E. G., and E. N. K. Clarkson. 1990. The late Palaeozoic Society Special Publication 8:34. radiation of malacostracan crustaceans. In P. D. Taylor and G. ———. 2000a. Bivalve taphonomy in tropical mixed siliciclastic- P. Larwood, eds. Major evolutionary radiations. Systematics carbonate settings. I. Environmental variation in shell con- Association Special Volume 42:165–186. Clarendon, Oxford. dition. Paleobiology 26:80–102. Briggs, D. E. G., and P. R. Crowther, eds. 1990. Palaeobiology, a ———. 2000b. Bivalve taphonomy in tropical mixed siliciclastic- synthesis. Blackwell Science, Oxford. carbonate settings. II. Effect of bivalve life habits and shell Briggs, D. E. G., and A. J. Kear. 1994a. Decay of Branchiostoma: types. Paleobiology 26:103–115. Implications for soft-tissue preservation in conodonts and Best, M. M. R., S. M. Kidwell, T. C. W.Ku, and L. M. Walter. 1999. other primitive chordates. Lethaia 26:275–287. The role of microbial iron reduction in the preservation of ———. 1994b. Decay and mineralization of shrimps. Palaios 9: skeletal carbonate: bivalve taphonomy and porewater geo- 431–456. chemistry in tropic siliciclastics vs. carbonates. Geological So- Briggs, D. J., D. D. Gilbertson, and A. L. Harris. 1990. Molluscan ciety of America Abstracts with Programs 31:419–420. taphonomy in a braided river environment and its implica- Bishop, J. D. D. 1989. Colony form and the exploitation of spatial tions for studies of Quaternary cold-stage river deposits. Jour- refuges by encrusting bryozoa. Biological Reviews 64:197– nal of Biogeography 17:623–637. 218. Briggs, D. E. G., P. R. Wilby, B. P. Pe´rez-Moreno, J. L. Sanz, and Blob, R. W. 1997. Relative hydrodynamic dispersal potentials of M. Fregenal-Martı´nez. 1997. The mineralization of dinosaur soft-shelled turtle elements: implications for interpreting soft tissue in the Lower Cretaceous of Las Hoyas, Spain. Jour- skeletal sorting in assemblages of non-mammalian terrestrial nal of the Geological Society, London 154:587–588. vertebrates. Palaios 12:151–164. Briggs, D. E. G., B. A. Stankiewicz, D. Meischner, A. Bierstedt, Blumenschine, R. 1986. Carcass consumption sequences and the and R. P. Evershed. 1998. Taphonomy of arthropod cuticles archaeological distinction of scavenging and hunting. Journal from Pliocene late sediments, Willershausen, Germany. Pa- of Human Evolution 15:639–659. laios 13:386–394. ———. 1988. An experimental model of the timing of hominid Briggs, D. E. G., R. P. Evershed, and M. J. Lockheart. 2000. The and carnivore influence on archaeological bone assemblages. molecular paleontology of continental fossils. In D. H. Erwin Journal of Archaeological Science 15:483–502. and S. L. Wing, eds. Deep time: Paleobiology’s perspective. Pa- ———. 1991. Hominid carnivory and foraging strategies and leobiology 26(Suppl. to No. 4):169–193. the socio-economic function of early archaeological sites. Buatois, L. A., A. G. Ma´ngano, J. F.Genise, and T. N. Taylor.1998. Philosophical Transactions of the Royal Society of London B The ichnologic record of the continental invertebrate invasion: 334:211–221. evolutionary trends in environmental expansion, ecospace Bonnichsen, R., and M. H. Sorg. 1989. Bone modification. Insti- utilization, and behavioral complexity. Palaios 13:217–240. tute for Quaternary Studies, University of Maine, Orono. Budd, D. A., and E. E. Hiatt. 1993. Mineralogical stabilization of 138 ANNA K. BEHRENSMEYER ET AL.

high-magnesium calcite: geochemical evidence for intracrys- Cook, E. 1995. Taphonomy of two non-marine Lower Creta- tal recrystallization within Holocene porcellaneous forami- ceous bone accumulations from southeastern England. Pa- nifera. Journal of Sedimentary Petrology 63:261–274. laeogeography, Palaeoclimatology, Palaeoecology 116:263– Burnham, R. J. 1989. Relationships between standing vegetation 270. and leaf litter in a paratropical forest: implications for paleo- Crowley, S. S., D. A. Dufek, R. W. Stanton, and T. A. Ryer. 1994. botany. Review of Palaeobotany and Palynology 58:5–32. The effects of volcanic ash disturbances on a peat-forming en- ———. 1990. Paleobotanical implications of drifted seeds and vironment: environmental disruption and taphonomic con- fruits from modern mangrove litter, Twin Cays, Belize. Pa- sequences. Palaios 9:158–174. laios 5:364–370. Cruz-Uribe, K., and R. G. Klein. 1998. Hyrax and hare bones ———. 1993. Reconstructing richness in the plant fossil record. from modern South African eagle roosts and the detection of Palaios 8:376–384. eagle involvement in fossil bone assemblages. Journal of Ar- ———. 1994. Patterns in tropical leaf litter and implications for chaeological Science 25:135–147. angiosperm paleobotany. Review of Palaeobotany and Paly- Cummins, H. 1994. Taphonomic processes in modern fresh- nology 81:99–113. water molluscan death assemblages: implications of the fresh- Burnham, R. J., and R. A. Spicer. 1986. Fossil litter preserved by water fossil record. Palaeogeography, Palaeoclimatology, Pa- volcanic activity at El Chico´n, Mexico: a potentially accurate laeoecology 108:55–73. record of the pre-eruption vegetation. Palaios 1:158–161. Cummins, H., E. N. Powell, R. J. Stanton, and G. M. Staff. 1986a. Burnham, R. J., S. L. Wing, and G. G. Parker. 1992. The reflection The size frequency distribution in palaeoecology: effects of of deciduous forest communities in leaf litter: implications for taphonomic processes during formation of molluscan death autochthonous litter assemblages from the fossil record: Pa- assemblages in Texas bays. Palaeontology 29:495–518. leobiology 18:30–49. ———. 1986b. The rate of taphonomic loss in modern benthic Butterfield, N. J. 1990. Organic preservation of non-mineralizing habitats: how much of the potentially preservable community organisms and the taphonomy of the Burgess Shale. Paleo- is preserved? Palaeogeography, Palaeoclimatology, Palaeoe- biology 16:272–286. cology 52:291–320. ———. 1995. Secular distribution of Burgess Shale-type pres- Cunningham, C. R., H. R. Feldman, E. K. Franseen, R. A. Gas- ervation. Lethaia 28:1–13. taldo, G. Mapes, C. G. Maples, and H.-P. Schultze. 1993. The Cade´e, G. C. 1984. Macrobenthos and macrobenthic remains on Upper Carboniferous Hamilton fossil Lagersta¨tte in Kansas: the Oyster Ground, North Sea. Netherlands Journal of Sea Re- a valley fill, tidally influenced deposit. Lethaia 26:225–236. search 18:160–178. Cutler, A. H. 1989. Shells survive—loss, persistence and accu- ———. 1991. The history of taphonomy. Pp. 3–21 in Donovan mulation of hardparts in shallow marine sediments. Geolog- 1991. ical Society of America Abstracts with Programs 21:A71. ———. 1994. Eider, shelduck, and other predators, the main ———. 1991. Nested faunas and extinction in fragmented hab- producers of shell fragments in the Wadden Sea—paleoeco- itats. Conservation Biology 5:496–505. logical implications. Palaeontology 37:181–202. ———. 1993. Mathematical models of temporal mixing in the ———. 1999. Bioerosion of shells by terrestrial gastropods. Le- fossil record. Pp. 169–187 in Kidwell and Behrensmeyer 1993. thaia 32:253–260. Cutler, A. H., and A. K. Behrensmeyer. 1996. Models of verte- Calleja, M., M. Rossignol-Strick, and D. Duzer. 1993. Atmo- brate mass mortality events at the K/T Boundary. In G. Ryder, spheric pollen content off West Africa. Review of Palaeobo- D. Fastovsky, and S. Gartner, eds. The Cretaceous-Tertiary tany and Palynology 79:335–368. Event and other catastrophes in earth history. Geological So- Callender, W. R., E. N. Powell, and G. M. Staff. 1994. Taphonom- ciety of America Special Paper 307:375–380. ic rates of molluscan shells placed in authochthonous assem- Cutler A. H., and K. W. Flessa. 1990. Fossils out of sequence: blages on the Louisiana contental slope. Palaios 9:60–73. computer simulations and strategies for dealing with strati- Carroll, R. L. 1997. Patterns and processes of vertebrate evolu- graphic disorder. Palaios 5:227–235. tion. Cambridge University Press, Cambridge. ———. 1995. Bioerosion, dissolution and precipitation as taph- Cate, A. S., and I. Evans. 1994. Taphonomic significance of the onomic agents at high and low latitudes. Senckenbergiana biomechanical fragmentation of live molluscan shell material Maritima 25:115–121. by a bottom-feeding fish (Pogonias cromis) in Texas coastal Cutler, A. H., A. K. Behrensmeyer, and R. E. Chapman. 1999. bays. Palaios 9:254–274. Environmental information in a recent bone assemblage: roles Chafetz, H., and C. Buczynski. 1992. Bacterially induced lithi- of taphonomic processes and ecological change. Palaeogeog- fication of microbial mats. Palaios 7:277–293. raphy, Palaeoclimatology, Palaeoecology 149:359–372. Chatterton, B. D. E., and S. E. Speyer. 1997. Ontogeny. Pp. 173– 247 in H. B. Whittington et al. Arthropoda 1, Trilobita, re- Daley, G. M. 1993. Passive deterioration of shelly material: a vised. Part O of R. C. Moore and C. Teichert, eds. Treatise on study of the Recent eastern Pacific articulate brachiopod Ter- invertebrate paleontology. Geological Society of America and ebratalia transversa Sowerby. Palaios 8:226–232. University of Kansas, Boulder, Colo. Daley, R. L., and D. W. Boyd. 1996. The role of skeletal micro- Claassen, C. 1998. Shells. Cambridge Manuals in Archaeology. structure during selective silicification of brachiopods. Jour- Cambridge University Press, Cambridge. nal of Sedimentary Research 66:155–162. Clark, G. R., II. 1999. Organic matrix taphonomy in some mol- Damuth, J. 1982. Analysis of the preservation of community luscan shell microstructures. Palaeogeography, Palaeoclima- structure in assemblages of fossil mammals. Paleobiology 8: tology, Palaeoecology 149:305–312. 434–446. Clyde, W. C., and P. D. Gingerich. 1998. Mammalian community Davies, D. J. 1993. Taphonomic analysis as a tool for long-term response to the latest Paleocene thermal maximum: an iso- community baseline delineation: taphoanalysis in an environ- taphonomic study in the northern Bighorn Basin, Wyoming. mental impact statement (EIS) for proposed human seafloor Geology 26:1011–1014. disturbances, Alabama continental shelf. Geological Society Collinson, M. E. 1983. Accumulations of fruits and seeds in of America Abstracts with Programs 25:A459. three small sedimentary environments in southern England Davies, D. J., E. N. Powell, and R. J. Stanton. 1989. Taphonomic and their palaeoecological implications. Annals of Botany 52: signature as a function of environmental process: shells and 583–592. shell beds in a hurricane-influenced inlet on the Texas coast. TAPHONOMY AND PALEOBIOLOGY 139

Palaeogeography, Palaeoclimatology, Palaeoecology 72:317– tiary of Riversleigh, Queensland, Australia. Palaeontology 41: 356. 835–851. Davies-Vollum, K. S., and S. L. Wing. 1998. Sedimentological, Eberth, D. A. 1990. Stratigraphy and sedimentology of verte- taphonomic, and climatic aspects of Eocene swamp deposits brate microfossil localities in uppermost Judith River For- (Willwood Formation, Bighorn Basin, Wyoming). Palaios 13: mation (Campanian) of Dinosaur Provincial Park, south-cen- 28–40. tral Alberta, Canada. Palaeogeography, Palaeoclimatology, Davis, P. G., and D. E. G. Briggs. 1998. The impact of decay and Palaeoecology 78:1–36. disarticulation on the preservation of fossil birds. Palaios 13: Eberth, D., R. Rogers and T. Fiorillo, convenors. 1999. Bonebeds: 3–13. genesis, analysis, and paleoecological significance (program Dawson, J. W. 1882. On the results of recent explorations of erect for a symposium). Journal of Vertebrate Paleontology trees containing animal remains in the coal formation of Nova 19(Suppl. to No. 3):7A. Scotia. Philosophical Transactions of the Royal Society of Efremov, J. A. 1940. Taphonomy: new branch of paleontology. London B 173:621–659. Pan American Geologist 74:81–93. Dechant Boaz, D. 1994. Taphonomy and the fluvial environ- Elder, R. L., and G. R. Smith. 1984. Fish taphonomy and paleo- ment. Pp. 377–413 in R. S. Corruccini and R. L. Ciochon, eds. ecology. Geobios Me´moire Spe´cial 8:287–291. Integrative paths to the past: paleoanthropological advances Elias, S. A., T. R. Van Devender, and R. De Baca. 1995. Insect in honor of F. Clark Howell. Prentice-Hall, Englewood Cliffs, fossil evidence of late glacial and Holocene environments in N.J. the Bolson de Mapimi, Chihuahuan Desert, Mexico: compar- Demko, T. M., and R. A. Gastaldo. 1992. Paludal environments isons with the paleobotanical record. Palaios 10:454–464. of the Lower Mary Lee coal zone, Pottsville Formation, Ala- Emig, C. C. 1990. Examples of post-mortality alteration in Re- bama: stacked clastic swamps and peat mires. International cent brachiopod shells and (paleo)ecological consequences. Journal of Coal Geology 20:23–47. Marine Biology 104:233–238. Demko, T. M., R. F. Dubiel, and J. T. Parrish. 1998. Plant taphon- Evans, S., and J. A. Todd. 1997. Late Jurassic soft-bodied wood omy in incised valleys: implications for interpreting paleocli- epibionts preserved by bioimmuration. Lethaia 30:185–189. mate from fossil plants. Geology 26:1119–1122. Farley, M. B. 1987. Palynomorphs from surface water of the east- Dent, S. R. 1995. A taphofacies model of the Recent South Flor- ern and central Caribbean Sea. Micropaleontology 33:254– ida continental shelf: a new perspective for a classic, exposed 262. carbonate environment. Ph.D. dissertation. University of Cin- Fedonkin, M. A. 1994. Vendian body fossils and trace fossils. Pp. cinnati, Cincinnati, Ohio. 370–388 in Bengtson 1984. D’Hondt, S., P. Donaghay, J. C. Zachos, D. Luttenberg, and M. Feige,A.,andF.T.Fu¨ rsich. 1991. Taphonomy of the Recent mol- Lindinger. 1998. Organic carbon fluxes and ecological recov- luscs of Bahı´a la Choya (Gulf of California, Sonora, Mexico). ery from the Cretaceous-Tertiary mass extinction. Science 282: Zitteliana 18:89–113. 276–279. Feldmann, R. M., T. Villamil, and E. G. Kauffman. 1999. Deca- DiMichele, W. A., and R. B. Aronson. 1992. The Pennsylvanian- pod and stomatopod crustaceans from mass mortality Lag- Permian vegetational transition: a terrestrial analogue to the ersta¨tten: Turonian (Cretaceous) of Colombia. Journal of Pa- onshore-offshore hypothesis. Evolution 46:807–824. leontology 73:91–101. DiMichele, W. A., and R. W. Hook. 1992. Paleozoic terrestrial Ferguson, D. K. 1995. Plant part processing and community re- ecosystems. Pp. 205–325 in A. K. Behrensmeyer, J. D. Damuth, construction. Eclogae Geologicae Helvetiae 88:627–641. W. A. DiMichele, R. Potts, H.-D. Sues, and S. L. Wing, eds. Ferna´ndez-Jalvo, Y., C. Denys, P. Andrews, T. Williams, Y. Dau- Terrestrial ecosystems through time. University of Chicago phin and L. Humphreys. 1998. Taphonomy and paleoecology Press, Chicago. of Olduvai Bed-I (Pleistocene, Tanzania). Journal of Human DiMichele, W. A., S. H. Mamay, D. S. Chaney, R. W. Hook, and Evolution 34:137–172. W. J. Nelson. In press. An early Permian flora with Late Perm- Ferna´ndez-Lo´pez, S. 2000. Ammonite taphocycles in carbonate ian and Mesozoic affinities from north-central Texas. Journal epicontinental platforms. Fifth international symposium on of Paleontology. the Jurassic System (Vancouver, B.C.). GeoResearch Forum 6: Dodson, P. 1987. Microfaunal studies of dinosaur paleoecology, 293–300. Judith River Formation of southern Alberta (Canada). Palaeo- Fiorillo, A. R. 1988. Taphonomy of Hazard Homestead Quarry geography, Palaeoclimatology, Palaeoecology 10:21–74. (Ogalalla Group), Hitchcock County, Nebraska. Contribu- Dominguez-Rodrigo, M. 1999. Flesh availability and bone mod- tions to Geology University of Wyoming 26:57–97. ifications in carcasses consumed by lions: palaeoecological ———. 1989. An experimental study of trampling: implications relevance in hominid foraging patterns. Palaeogeography,Pa- for the fossil record. Pp. 61–72 in R. Bonnichsen and M. H. laeoclimatology, Palaeoecology 149:373–388. Sorg, eds. Bone modification. Institute for Quaternary Stud- Donovan, S. K., ed. 1991. The processes of fossilization. Colum- ies, University of Maine, Orono. bia University Press, New York. ———. 1991. Taphonomy and depositional setting of Careless Donovan, S. K., and C. R. C. Paul, eds. 1998. The adequacy of Creek Quarry (Judith River Formation), Wheatland County, the fossil record. Wiley, New York. Montana. Palaeogeography, Palaeoclimatology, Paleoecology Downing, K. F., and L. E. Park. 1998. Geochemistry and early 81:281–311. diagenesis of mammal-bearing concretions from the Sucker Flessa, K. W. 1998. Well-traveled cockles: shell transport during Creek Formation (Miocene) of southeastern Oregon. Palaios the Holocene transgression of the southern North Sea. Geol- 13:14–27. ogy 26:187–190. Doyle, P., and D. I. M. Macdonald. 1993. Belemnite battlefields. Flessa, K. W., and M. Kowalewski. 1994. Shell survival and time- Lethaia 26:65–80. averaging in nearshore and shelf environments: estimates Droser, M. L., and D. J. Bottjer. 1993. Trends and patterns of from the radiocarbon literature. Lethaia 27:153–165. Phanerozoic ichnofabrics. Annual Review of Earth and Plan- Flessa, K. W., A. H. Cutler, and K. H. Meldahl. 1993. Time and etary Sciences 21:205–225. taphonomy: quantitative estimates of time-averaging and Duncan, I. J., D. E. G. Briggs, and M. Archer. 1998. Three-di- stratigraphic disorder in a shallow marine habitat. Paleobi- mensionally mineralized insects and millipedes from the Ter- ology 19:266–286. 140 ANNA K. BEHRENSMEYER ET AL.

Foote, M. 1996. On the probability of ancestors in the fossil re- tic alluvial channel-fills. Review of Palaeobotany and Paly- cord. Paleobiology 22:141–151. nology 91:1–21. ———. 1997. Sampling, taxonomic description, and our evolv- Gastaldo, R. A., W. Riegel, W. Pu¨ ttmann, U. H. Linnemann, and ing knowledge of morphological diversity. Paleobiology 23: R. Zetter. 1998. A multidisciplinary approach to reconstruct 181–206. the Late Oligocene vegetation in central Europe. Review of ———. 2000. Origination and extinction components of taxo- Palaeobotany and Palynology 101:71–94. nomic diversity: general problems. In D. H. Erwin and S. L. Gehling, J. G. 1999. Microbial mats in terminal Proterozoic sil- Wing, eds. Deep time: Paleobiology’s perspective. Paleobiology iciclastics: Ediacaran death masks. Palaios 14:40–57. 26(Suppl. to No. 4):74–102. Gifford-Gonzalez, D. 1991. Bones are not enough: analogues, Foote, M., and D. M. Raup. 1996. Fossil preservation and the knowledge, and interpretive strategies in zooarchaeology. stratigraphic ranges of taxa. Paleobiology 22:121–140. Journal of Anthropological Archaeology 10:215–254. Foote, M., J. P. Hunter, C. M. Janis, and J. J. Sepkoski Jr. 1999. Gilinsky, N. L., and J. B. Bennington. 1994. Estimating numbers Evolutionary and preservational constraints on origins of bi- of whole individuals from collections of body parts: a taph- ologic groups: divergence times of Eutherian mammals. Sci- onomic limitation of the paleontological record. Paleobiology ence 283:1310–1314. 20:245–258. Franzen, J. L. 1985. Exceptional preservation of Eocene verte- Gilinsky, N. L., and P. W. Signor, eds. 1991. Analytical paleobi- brates in the lake deposits of Grube Messel (West Germany). ology. Short Courses in Paleontology No. 4. Paleontological Philosophical Transactions of the Royal Society of London B Society, Knoxville, Tenn. 311:181–186. Glover, C. P., and S. M. Kidwell. 1993. Influence of organic ma- Frison, G. C., and L. C. Todd. 1986. The Colby Mammoth Site: trix on the post-mortem destruction of molluscan shells. Jour- taphonomy and archaeology of a Clovis kill in northern Wy- nal of Geology 101:729–747. oming. University of New Mexico Press, Albuquerque. Goldstein, S. T., P. Van Cappellen, A. Roychoudhury, and C. Ko- Fu¨ rsich, F. T. 1978. The influence of faunal condensation and retsky. 1997. Preservation of salt-marsh foraminifera in ex- mixing on the preservation of fossil benthic communities. Le- perimental arrays deployed below the sediment-water inter- thaia 11:243–250. face, Sapelo Island, Georgia (USA). Geological Society of Fu¨ rsich, F. T., and M. Aberhan. 1990. Significance of time-aver- America Abstracts with Programs 30:A383. aging for paleocommunity analysis. Lethaia 23:143–152. Grayson, D. K. 1989. Bone transport, bone destruction, and re- Fu¨ rsich, F. T., and W. Oschmann. 1993. Shell beds as tools in ba- verse utility curves. Journal of Archaeological Science 16:643– sin analysis: the Jurassic of Kachchh, western India. Journal of 652. the Geological Society, London 150:169–185. Greenstein, B. J. 1989. Mass mortality of the West-Indian echi- Gastaldo, R. A. 1988. A conspectus of phytotaphonomy. In W. noid Diadema antillarum (Echinodermata: Echinoidea): a nat- A. DiMichele and S. L. Wing, eds. Methods and applications ural experiment in taphonomy. Palaios 4:487–492. of plant paleoecology: notes for a short course. Paleontolog- ———. 1991. An integrated study of echinoid taphonomy: pre- ical Society Special Publication 3:14–28. dictions for the fossil record of four echinoid families. Palaios ———. 1992. Taphonomic considerations for plant evolutionary 6:519–540. investigations. The Palaeobotanist 41:211–223. ———. 1992. Taphonomic bias and the evolutionary history of ———. 1994. The genesis and sedimentation of phytoclasts with the family Cidaridae (Echinodermata: Echinoidea). Paleobi- examples from coastal environments. Pp. 103–127 in A. Tra- ology 18:50–79. verse, ed. Sedimentation of organic particles. Cambridge Uni- ———. 1993. Is the fossil record of regular echinoids really so versity Press, Cambridge. poor a comparison of living and subfossil assemblages? Pa- Gastaldo, R. A., and D. K. Ferguson. 1998. Reconstructing Ter- laios 8:587–601. tiary plant communities: introductory remarks. Review of Pa- ———. 1999. Taphonomy of reef-building corals II: shallow and laeobotany and Palynology 101:3–6. deep reef environments of the tropical western Atlantic. Geo- Gastaldo, R. A., and A. Y. Huc. 1992. Sediment facies, deposi- logical Society of America Abstracts with Programs 31:A420. tional environments, and distribution of phytoclasts in the Greenstein, B. J., and H. A. Moffat 1996. Comparative taphon- Recent Mahakam River delta, Kalimantan, Indonesia. Palaios omy of modern and Pleistocene corals, San Salvador, Bahama. 7:574–591. Palaios 11:57–63. Gastaldo, R. A., and J. R. Staub. 1999. A mechanism to explain Greenstein, B. J., and J. M. Pandolfi. 1997. Preservation of com- the preservation of leaf litter lenses in coals derived from munity structure in modern reef coral life and death assem- raised mires. Palaeogeography, Palaeoclimatology, Palaeoe- blages of the Florida Keys: implications for the Quaternary cology 149:1–14. fossil record of coral reefs. Bulletin of Marine Science 61:431– Gastaldo, R. A., D. P. Douglass, and S. M. McCarroll. 1987. Or- 452. igin, characteristics, and provenance of plant macrodetritus Greenstein, B. J., L. A. Harris, and H. A. Curran. 1998. Com- in a Holocene crevasse splay, Mobile Delta, Alabama. Palaios parison of Recent coral life and death assemblages to Pleis- 2:229–240. tocene reef communities: implications for rapid faunal re- Gastaldo, R. A., T. M. Demko, and Y. Liu. 1993a. Application of placement of recent reefs. Carbonates and Evaporites 13:23– sequence and genetic stratigraphic concepts to Carboniferous 31. coal-bearing strata: an example from the Black Warrior Basin, Grotzinger, J. P. 1994. Trends in Precambrian carbonate sedi- USA. Geologische Rundschau 82:212–226. ments and their implication for understanding evolution. Pp. Gastaldo, R. A., G. P. Allen, and A. Y. Huc. 1993b. Detrital peat 245–258 in Bengtson 1984. formation in the tropical Mahakam River delta, Kalimantan, Hadly, E. A. 1999. Fidelity of terrestrial vertebrate fossils to a eastern Borneo: formation, plant composition, and geochem- modern ecosystem. Palaeogeography, Palaeoclimatology, Pa- istry. In J. C. Cobb and C. B. Cecil, eds. Modern and ancient laeoecology 149:389–410. coal-forming environments. Geological Society of America Hageman, S. J., N. P. James, and Y. Bone. 2000. Cool-water car- Special Paper 286:107–118. bonate production from epizoic bryozoans on ephemeral sub- Gastaldo, R. A., H. Walther, J. Rabold, and D. Ferguson. 1996. strates. Palaios 15:33–48. Criteria to distinguish parautochthonous leaves in Cenophy- Haglund, W. D., and M. H. Sorg, eds. 1997. Forensic taphonomy, TAPHONOMY AND PALEOBIOLOGY 141

the post-mortem fate of human remains. CRC Press, New Jackson, S. T., and D. R. Whitehead. 1991. Holocene vegetation York . patterns in the Adirondack Mountains. Ecology 72:641–653. Haynes, G. 1985. On watering holes, mineral licks, death, and Janzen, D. H. 1977. Why fruits rot, seeds mold, and meat spoils. predation. Pp. 53–71 in D. Meltzer and J. I. Mead, eds. Envi- American Naturalist 111:691–713. ronments and extinctions in late glacial North America. Cen- Jodry, M. A., and D. J. Stanford. 1992. Stewart’s Cattle Guard ter for the Study of Early Man, University of Maine, Orono. Site: an analysis of bison remains in a Folsom kill-butchery ———. 1988. Mass deaths and serial predation: comparative campsite. Pp. 101–168 in D. J. Stanford and J. S. Day, eds. Ice taphonomic studies of modern large-mammal deathsites. Age hunters of the Rockies. Denver Museum of Natural His- Journal of Archaeological Science 15:219–235. tory and University Press of Colorado, Denver. 378 pp. ———. 1991. Mammoths, mastodonts and elephants. Cam- Johnson, R. G. 1960. Models and methods for analysis of the bridge University Press, Cambridge. mode of formation of fossil assemblages. Geological Society Henderson, S. W., and R. W. Frey. 1986. Taphonomic redistri- of America Bulletin 71:105–1086. bution of mollusk shells in a tidal inlet channel, Sapelo Island, Kah, L. C., and A. H. Knoll. 1996. Microbenthic distribution of Georgia. Palaios 1:3–16. Proterozoic tidal flats: environmental and taphonomic con- Henwood, A. A. 1992a. Exceptional preservation of dipteran siderations. Geology 24:79–82. flight muscle and the taphonomy of insects in amber. Palaios Kendricks, P., and P. R. Crane. 1997. The origin and early diver- 7:203–212. sification of land plants: a cladistic study. Smithsonian Insti- ———. 1992b. Soft-part preservation of beetles in Tertiary am- tution Press, Washington, D.C. ber from the Dominican Republic. Palaeontology 35:901–912. Kennish, M. J., and R. A. Lutz. 1999. Calcium carbonate disso- Holland, S. M. 1995. The stratigraphic distribution of fossils. Pa- lution rates in deep-sea bivalve shells on the East pacific Rise leobiology 21:92–109. at 21ЊN: results of an 8-year in-situ experiment. Palaeogeog- ———. 1996. Recognizing artifactually generated coordinated raphy, Palaeoclimatology, Palaeoecology 154:293–299. stasis: implications of numerical models and strategies for Kerbis Peterhans, J. C., R. W. Wrangham, M. L. Carter, and M. field tests. Palaeogeography, Palaeoclimatology, Palaeoecol- D. Hauser. 1993. A contribution to tropical rain forest taphon- ogy 127:147–156. omy: retrieval and documentation of chimpanzee remains ———. 2000. The quality of the fossil record: a sequence strati- from Kibale Forest, Uganda. Journal of Human Evolution 25: graphic perspective. In D. H. Erwin and S. L. Wing, eds. Deep 485–514. time: Paleobiology’s perspective. Paleobiology 26(Suppl. to No. Kershaw, P. J., D. J. Swift, and D. C. Denoon. 1988. Evidence of 4):148–168. recent sedimentation in the eastern Irish Sea. Marine Geology Holland, S. M., and M. E. Patzkowsky. 1999. Models for simu- 85:1–14. lating the fossil record. Geology 27:491–494. Kidston, R., and W. H. Lang. 1920. Old Red Sandstone plants Hudson, J. 1993. From bones to behavior. Occasional Paper No. showing structure, from the Rhynie Chert bed, Aberdeenshi- 21. Center for Archaeological Investigations, Southern Illinois re, Part 2. Transactions of the Royal Society of Edinburgh 52: 603–627. University, Carbondale. Kidwell, S. M. 1991. The stratigraphy of shell concentrations. Pp. Hughes, N. C., and D. L. Cooper. 1999. Paleobiologic and taph- 211–290 in Allison and Briggs 1991. onomic aspects of the ‘‘granulosa’’ trilobite cluster, Kope For- ———. 1993. Taphonomic expressions of sedimentary hiatus: mation (Upper Ordovician, Cincinnati region). Journal of Pa- field observations on bioclastic concentrations and sequence leontology 73:306–319. anatomy in low, moderate and high subsidence settings. Geo- Hughes, N. C., and C. C. Labandeira. 1995. The stability of spe- logische Rundschau 82:189–202. cies in taxonomy. Paleobiology 21:401–403. ———. 1998. Time-averaging in the marine fossil record: over- Hunter, J. 1994. Lack of a high body count at the K-T boundary. view of strategies and uncertainties. Geobios 30:977–995. Journal of Paleontology 68:1158. ———. 1999. High fidelity of species relative abundances in ma- Jablonski, D. J. 1988. Estimates of species durations. Science 240: rine molluscan death assemblages. Geological Society of 969. America Abstracts with Programs 31:A419. ———. 1999. The future of the fossil record. Science 284:2114– ———. In press. Ecological fidelity of molluscan death assem- 2116. blages. In S. A. Woodin, J. Y. Aller, and R. C. Aller, eds. Or- ———. 2000. Micro- and macroevolution: scale and hierarchy in ganism-sediment interactions. Belle Baruch Institute Volume. evolutionary biology and paleontology. In D. H. Erwin and S. University of South Carolina Press, Columbia. L. Wing, eds. Deep time: Paleobiology’s perspective. Paleobi- Kidwell, S. M., and T. Aigner. 1985. Sedimentary dynamics of ology 26(Suppl. to No. 4):15–52. complex shell beds: implications for ecologic and evolution- Jablonski, D. J., S. Lidgard, and P. D. Taylor. 1997. Comparative ary patterns. Pp. 382–395 in U. Bayer and A. Seilacher, eds. ecology of bryozoan radiations: origin of novelties in cyclo- Sedimentary and evolutionary cycles. Springer, Berlin. stomes and cheilostomes. Palaios 12:505–523. Kidwell, S. M., and T. Baumiller. 1990. Experimental disintegra- Jackson, S. T. 1989. Postglacial vegetational change along an ele- tion of regular echinoids: roles of temperature, oxygen and vational gradient in the Adirondack Mountains (New York): a decay thresholds. Paleobiology 16:247–271. study of plant macrofossils. New York State Museum Bulletin Kidwell, S. M., and A. K. Behrensmeyer, eds. 1993. Taphonomic 465. approaches to time resolution in fossil assemblages. Short ———. 1994. Pollen and spores in Quaternary lake sediments Courses in Paleontology No. 6. Paleontological Society, Knox- as sensors of vegetation composition: theoretical models and ville, Tenn. empirical evidence. Pp. 253–286 in A. Traverse, ed. Sedimen- Kidwell, S. M., and D. W. J. Bosence. 1991. Taphonomy and time- tation of organic particles. Cambridge University Press, Cam- averaging of marine shelly faunas. Pp. 115–209 in Allison and bridge. Briggs 1991. Jackson, S. T., and J. T. Overpeck. 2000. Responses of plant pop- Kidwell, S. M., and P. J. Brenchley. 1994. Patterns in bioclastic ulations and communities to environmental changes of the accumulation through the Phanerozoic: changes in input or in late Quaternary. In D. H. Erwin and S. L. Wing, eds. Deep destruction? Geology 22:1139–1143. time: Paleobiology’s perspective. Paleobiology 26(Suppl. to No. ———. 1996. Evolution of the fossil record: thickness trends in 4):194–220. marine skeletal accumulations and their implications. Pp. 142 ANNA K. BEHRENSMEYER ET AL.

290–336 in D. Jablonski, D. H. Erwin, and J. H. Lipps, eds. Evo- bionts on the scallops Chlamys hastata (Sowerby 1843) and lutionary paleobiology. University of Chicago Press, Chicago. Chlamys rubida (Hinds 1845). Palaios 8:267–277. Kidwell, S. M., and K. W. Flessa. 1995. The quality of the fossil ———. 1995. The life orientation of concavo-convex brachio- record: populations, species, and communities. Annual Re- pods: overturning the paradigm. Paleobiology 21:520–551. view of Ecology and Systematics 26:269–299. Li, X., and M. L. Droser. 1997. Nature and distribution of Cam- Kidwell, S. M., and D. Jablonski. 1983. Taphonomic feedback: brian shell concentrations: evidence from the Basin and ecological consequences of shell accumulation. Pp. 195–248 in Range province of western United States (California, Nevada M. J. S. Tevesz and P. L. McCall, eds. Biotic interactions in re- and Utah). Palaios 12:11–1126. cent and fossil benthic communities. Plenum, New York. Llewellyn, G., and C. G. Messing. 1993. Compositional and Knoll, A. H. 1985. Exceptional preservation of photosynthetic taphonomic variations in modern crinoid-rich sediments organisms in silicified carbonates and silicified peats. Philo- from the deep-water margin of a carbonate bank. Palaios 8: sophical Transactions of the Royal Society of London B 311: 554–573. 111–122. Llona, A., C. Pinto, and P. Andrews. 1999. Amphibian taphon- Knoll, A. H., and V. N. Sergeev. 1995. Taphonomic and evolu- omy and its application to the fossil record of Dolina (middle tionary changes across the Mesoproterozoic-Neoproterozoic Pleistocene, Atapuerca, Spain). Palaeogeography, Palaeocli- transition. Neues Jahrbuch fu¨ r Geologie und Pala¨ontologie, matology, Palaeoecology 149:411–430. Abhandlungen 195:289–302. Lupia, R. 1995. Paleobotanical data from fossil charcoal: an ac- Knoll, A. H., K. Niklas, P. G. Gensel, and B. Tiffney. 1984. Char- tualistic study of seed plant reproductive structures. Palaios acter diversification and patterns of evolution in early vas- 10:465–477. cular plants. Paleobiology 10:34–47. Lyell, C., and J. W. Dawson. 1853. On the remains of a reptile Knoll, A. H., I. J. Fairchild, and K. Swett. 1993. Calcified mi- (Dendrerpeton acadianum, Wyman and Owen), and of a land crobes in Neoproterozoic carbonates: implications for our un- shell discovered in the interior of an erect fossil tree in the derstanding of the Proterozoic/Cambrian transition. Palaios coal measures of Nova Scotia. Quarterly Journal of the Geo- 8:512–525. logical Society of London IX:58–63. Kondo, Y., S. T. Abbott, A. Katamura, P. J. J. Kamp, T. R. E. Naish, Lyman, R. L. 1985. Bone frequencies: differential transport, in T. Kamataki, and G. S. Saul. 1998. The relationship between situ destruction, and the MGUI. Journal of Archaeological Sci- shellbed type and sequence architecture: examples from Ja- ence 12:221–236. pan and New Zealand. Sedimentary Geology 122:109–127. ———. 1994. Vertebrate taphonomy. Cambridge Manuals in Ar- Kotler, E., R. E. Martin, and W. D. Liddell. 1992. Experimental chaeology, Cambridge University Press, Cambridge. analysis of abrasion and dissolution resistance of modern Lyman, R. L., and G. L. Fox. 1989. A critical evaluation of bone reef-dwelling Foraminifera: implications for the preservation weathering as an indication of bone assemblage formation. of biogenic carbonate. Palaios 7:244–276. Journal of Archaeological Science 16:293–317. Kowalewski, M. 1996a. Taphonomy of a living fossil: the lin- Malinky, J. M., and P. H. Heckel. 1998. Paleoecology and ta- Glottidia palmeri gulide brachiopod Dall from Baja California, phonomy of faunal assemblages in gray ‘‘core’’ (offshore) Mexico. Palaios 11:244–265. shales in Midcontinent Pennsylvanian cyclothems. Palaios 13: ———. 1996b. Time-averaging, overcompleteness, and the geo- 311–334. logical record. Journal of Geology 104:317–326. Maliva, R. G., A. H. Knoll, and R. Siever. 1989. Secular change ———. 1997. The reciprocal taphonomic model. Lethaia 30:86– in chert distribution: a reflection of evolving biological par- 88. ticipation in the silica cycle. Palaios 4:519–532. Kowalewski, M., and K. W. Flessa. 1996. Improving with age: Marean, C. W. 1991. Measuring the post-depositional destruc- the fossil record of lingulide brachiopods and the nature of tion of bone in archaeological assemblages. Journal of Ar- taphonomic megabiases. Geology 24:977–980. chaeological Science 18:677–694. Kowalewski, M., G. A. Goodfriend, and K. W. Flessa. 1998. ———. 1992. Captive hyaena bone choice and destruction, the High-resolution estimates of temporal mixing within shell Schlepp effect and Olduvai archaeofaunas. Journal of Archae- beds: the evils and virtues of time-averaging. Paleobiology 24: ological Science 19:101–121. 287–304. Markwick, P. J. 1998. Crocodilian diversity in space and time: Kranz, P. M. 1974. The anastrophic burial of bivalves and its pa- the role of climate in paleoecology and its implications for un- laeoecological significance. Journal of Geology 82:237–265. derstanding K/T extinctions. Paleobiology 24:470–497. Kristensen, E., S. I. Ahmed, and A. H. Devol. 1995. Aerobic and Marshall, C. R. 1990. Confidence intervals on stratigraphic rang- anaerobic decomposition of organic matter in marine sedi- es. Paleobiology 16:1–10. ment: which is fastest? Limnology and Oceanography 40: 1430–1437. ———. 1994. Confidence intervals on stratigraphic ranges: par- Labandeira, C. C. 1998. Early history of arthropod and vascular tial relaxation of the assumption of randomly distributed fos- plant associations. Annual Review of Earth Planetary Scienc- sil horizons. Paleobiology 20:459–460. es 28:153–193. Martill, D. M. 1985. The preservation of marine vertebrates in Labandeira, C. C., and D. M. Smith. 1999. Forging a future for the Lower Oxford Clay (Jurassic) of central England. Philo- fossil insects: thoughts on the First International Congress of sophical Transactions of the Royal Society of London B 311: Paleoentomology. Paleobiology 25:154–157. 155–165. Labandeira, C. C., K. R. Johnson, and P. Lang. In press. Insect ———. 1988. Preservation of fish in the Cretaceous Santana For- herbivory across the Cretaceous/Tertiary boundary: major mation of Brazil. Palaeontology 31:1–18. extinction and minimum rebound. In J. H. Hartman, K. R. ———. 1990. Macromolecular resolution of fossilized muscle Johnson, and D. J. Nichols, eds. The Hell Creek Formation and tissue from an elopomorph fish. Nature 346:171–172. the Cretaceous-Tertiary boundary in the northern Great Martin, R. E. 1993. Time and taphonomy: actualistic evidence Plains—an integrated Continental record at the end of the for time-averaging of benthic foraminiferal assemblages. Pp. Cretaceous. Geological Society of America Special Paper. 34–56 in Kidwell and Behrensmeyer 1993. Lask, P. B. 1993. The hydrodynamic behavior of sclerites from ———. 1995. Cyclic and secular variation in microfossil biom- the trilobite Flexicalymene meeki. Palaios 8:219–225. ineralization: clues to the biogeochemical evolution of Phan- Lescinsky, H. L. 1993. Taphonomy and paleoecology of epi- erozoic oceans. Global and Planetary Change 11:1–23. TAPHONOMY AND PALEOBIOLOGY 143

———. 1999. Taphonomy, a process approach. Cambridge Uni- Miller, A. I., and H. Cummins. 1990. A numerical model for the versity Press, Cambridge. formation of fossil assemblages: estimating the amount of Martin, R. E., J. F. Wehmiller, M. S. Harris, and W. D. Liddell. post-mortem transport along environmental gradients. Pa- 1996. Comparative taphonomy of bivalves and foraminifera laios 5:303–316. from Holocene tidal flat sediments, Bahı´a la Choya, Sonora, ———. 1993. Using numerical models to evaluate the conse- Mexico (Northern Gulf of California): taphonomic grades and quences to time-averaging in marine fossil assemblages. Pp. temporal resolution. Paleobiology 22:80–90. 150–168 in Kidwell and Behrensmeyer 1993. Martin, R. E., R. T. Patterson, S. T. Goldstein, and A. Kumar, eds. Miller, A. I., and M. Foote. 1996. Calibrating the Ordovician ra- 1999a. Taphonomy as a tool in paleoenvironmental recon- diation of marine life: implications for Phanerozoic diversity struction and environmental assessment. Palaeogeography, trends. Paleobiology 22:304–309. Palaeoclimatology, Palaeoecology 149(special issue). Miller, A. I., G. Llewellyn, K. M. Parsons, H. Cummins, M. R. Martin, R. E., S. P. Hippensteel, D. Nikitina, and J. E. Pizzuto. Boardman, B. J. Greenstein, and D. K. Jacobs. 1992. Effect of 1999b. Artificial time-averaging and the recovery of ecologi- Hurricane Hugo on molluscan skeletal distributions, Salt Riv- cal signals preserved in the subfossil record: linking the tem- er Bay, St. Croix, U. S. Virgin Islands. Geology 20:23–26. poral scales of ecology and paleoecology. Geological Society Mirsky, S. 1998. I shall return. Earth 7:48–53. of America Abstracts with Programs: A356. Morales, M., ed. 1996. The continental Jurassic. Museum of Martinez-Delclos, X., and J. Martinell. 1993. Insect taphonomy Northern Arizona Bulletin 60. experiments: their application to the Cretaceous outcrops of Nebelsick, J. H. 1992. Echinoid distribution by fragment iden- lithographic limestones from Spain. Kaupia (Darmsta¨dter tification in the northern Bay of Safaga, Red Sea, Egypt. Pa- Beitra¨ge zur Naturgeschichte) 2:133–144. laios 7:316–328. Masse´, H. L. 1999. Les carbonates associe´s a la macrofaune des ———. 1995. Comparative taphonomy of clypeasteroids. Eclo- sables fins littoraux en Me´diterrane´e nord-occidentale. gae Geologica Helvetiae 88:685–693. Oceanologica Acta 22:413–420. ———. 1999. Taphonomy of Clypeaster fragments: preservation Maxwell, W. D., and M. J. Benton. 1990. Historical tests of the and taphofacies. Lethaia 32:241–252. absolute completeness of the fossil record of tetrapods. Pa- Noe-Nygaard, N. 1987. Taphonomy in archaeology, with special leobiology 16:322–335. emphasis on man as a biasing factor. Journal of Danish Ar- McGree, H. 1984. On food and cooking: the science and lore of chaeology 6:7–62. the kitchen. Scribner, New York. Norris, R. D. 1986. Taphonomic gradients in shelf fossil assem- McIlroy, D., and G. A. Logan. 1999. The impact of bioturbation blages: Pliocene Purisima Formation, California. Palaios 1: on infaunal ecology and evolution during the Proterozoic- 256–270. Cambrian transition. Palaios 14:58–72. Oliver, J. S., and R. W. Graham. 1994. A catastrophic kill of ice- McKinney, F. K. 1996. Encrusting organisms on co-occurring trapped coots: time-averaged versus scavenger-specific dis- disarticulated valves of two marine bivalves: comparison of articulation patterns. Paleobiology 20:229–244. living assemblages and skeletal residues. Paleobiology 222: Oliver, J. S., N. E. Sikes, and K. M. Stewart, eds. 1994. Early hom- 534–567. inid behavioural ecology. Academic Press, London. McKinney, M. L. 1991. Completeness of the fossil record: an Olson, E. C. 1966. Community evolution and the origin of mam- overview. Pp. 66–83 in Donovan 1991. mals. Ecology 47:291–302. Meldahl, K. H. 1990. Sampling, species abundance, and the Olszewski, T. D. 1999. Taking advantage of time-averaging. Pa- stratigraphic signature of mass extinction: a test using Holo- leobiology 25:226–238. cene tidal flat molluscs. Geology 18:890–893. Olszewski, T. D., and R. R. West. 1997. Influence of transporta- Meldahl, K. H., and K. W. Flessa. 1990. Taphonomic pathways tion and time-averaging in fossil assemblages from the Penn- and comparative biofacies and taphofacies in a Recent inter- sylvanian of Oklahoma. Lethaia 30:315–329. tidal/shallow shelf environment. Lethaia 23:43–60. Oost, A. P., and P. L. de Boer. 1994. Tectonic and climatic setting Meldahl, K. H., D. Scott, and K. Carney. 1995. Autochthonous of lithographic limestone basins. Geobios Me´moire Spe´cial leaf assemblages as records of deciduous forest communities: 16:321–330. an actualistic study. Lethaia 28:383–394. Orr, P. J., D. E. G. Briggs, and S. L. Kearns. 1998. Cambrian Bur- Meldahl, K. H., K. W. Flessa, and A. H. Cutler. 1997a. Time-av- gess Shale animals replicated in clay minerals. Science 281: eraging and postmortem skeletal survival in benthic fossil as- 1173–1175. semblages: quantitative comparisons among Holocene envi- Palaios. 1999. Unexplored microbial worlds (theme issue). Vol. ronments. Paleobiology 23:209–229. 141 Palaios. 1999. Unexplored microbial worlds (theme issue). Meldahl, K. H., O. G. Yajimovich, C. D. Empedocles, C. S. Gus- Vol. 14, No. 1. tafson, M. M. Hidalgo, and T. W. Reardon. 1997b. Holocene Palmqvist, P. 1991. Differences in the fossilization potential of sediments and molluscan faunas of Bahı´a Concepcı´on: a mod- bivalve and gastropod species related to their life sites and ern analog to Neogene rift basins of the Gulf of California. trophic resources. Lethaia 24:287–288. Geological Society of America Special Paper 318:39–56. ———. 1993. Trophic levels and the observational completeness Meyer, D. L., and K. B. Meyer. 1986. Biostratinomy of Recent cri- of the fossil record. Revista Espan˜ ola de Paleontologı´a 8:33– noids (Echinodermata) at Lizard Island, Great Barrier Reef, 36. Australia. Palaios 1:294–302. Pandolfi, J. M., and B. J. Greenstein. 1997a. Taphonomic alter- Meyer, D. L., W. I. Ausich, and R. E. Terry. 1989. Comparative ation of reef coral: effects of reef environment and coral taphonomy of echinoderms in carbonate facies: Fort Payne growth form. I. The Great Barrier Reef. Palaios 12:27–42. Formation (Lower Mississippian) of Kentucky and Tennessee. ———. 1997b. Preservation of community structure in death as- Palaios 4:533–552. semblages of deep-water Caribbean reef corals. Limnology Miller, A. I. 1988. Spatial resolution in subfossil molluscan re- and Oceanography 42:1505–1516. mains: implications for paleobiological analyses. Paleobiolo- Parsons, K. M. 1989. Taphonomy as an indicator of environ- gy 14:91–103. ment: Smuggler’s Cove, St. Croix, U.S.V.I. In D. K. Hubbard, ———. 1997. Dissecting global diversity patterns: examples ed. Terrestrial and marine ecology of St. Croix, U.S. Virgin Is- from the Ordovician radiation. Annual Review of Ecology lands. West Indies Laboratory Special Publication 8:135–143. and Systematics 28:85–104. Parsons, K. M., and C. E. Brett. 1991. Taphonomic processes and 144 ANNA K. BEHRENSMEYER ET AL.

biases in modern marine environments: an actualistic per- Retallack, G. J. 1990. Soils of the past: an introduction to paleo- spective on fossil assemblage preservation. Pp. 22–65 in Don- pedology. Unwin Hyman, London. ovan 1991. Retallack, G. J., and D. L. Dilcher. 1986. Reconstructions of se- Parsons, K. M., C. E. Brett, and K. B. Miller. 1988. Taphonomy lected seed ferns. Annals of the Missouri Botanical Garden 75: and depositional dynamics of Devonian shell-rich mudstones. 1010–1057. Palaeogeography, Palaeoclimatology, Palaeoecology 63:109– Rich, F. J. 1989. A review of the taphonomy of plant remains in 140. lacustrine sediments. Review of Palaeobotany and Palynolo- Parsons-Hubbard, K. M., W. R. Callender, E. N. Powell, C. E. gy 58:33–46. Brett, S. E. Walker, A. L. Raymond, and G. M. Staff. 1999. Rates Richmond, D. R., and T. H. Morris. 1996. The dinosaur death- of burial and disturbance of experimentally-deployed mol- trap of the Cleveland-Lloyd Dinosaur Quarry, Emery County, luscs: implications for preservation potential. Palaios 14:337– Utah. Pp. 533–546 in Morales 1996. 351. Rigby, J. K., Jr., K. R. Newman, J. Smit, S. Van Der Kaars, R. E. Paul, C. R. C. 1992. How complete does the fossil record have to Sloan, and J. K. Rigby. 1987. Dinosaurs from the Paleocene be? Revista Espan˜ola Paleontologı´a 7:127–133. part of the Hell Creek Formation, McCone County, Montana. ———. 1998. Adequacy, completeness and the fossil record. Pp. Palaios 2:296–302. 1–22 in S. K. Donovan and C. R. C. Paul, eds. The adequacy Rivas, P., J. Aguirre, and J. C. Braga. 1997. Entolium beds: hiatal of the fossil record. Wiley, New York. shell concentrations in starved pelagic settings (middle Liaas- Perry, C. T. 1996. The rapid response of reef sediments to chang- sic, SE Spain). Eclogae Geologica Helvetiae 90:293–301. es in community composition: implications for time averag- Robinson, J. M. 1990. Lignin, land plants, and fungi. Biological ing and sediment accumulation. Journal of Sedimentary Re- evolution affecting Phanerozoic oxygen balance. Geology 15: search 66:459–467. 607–610. ———. 1999. Reef framework preservation in four contrasting ———. 1991. Phanerozoic atmospheric reconstructions: a ter- modern reef environments, Discovery Bay, Jamaica. Journal of restrial perspective. Palaeogeography, Palaeoclimatology, Pa- Coastal Research 15:796–812. laeoecology 97:51–62. Peterson, C. H. 1977. The paleoecological significance of unde- Rogers, R. R. 1990. Taphonomy of three dinosaur bone beds in tected short-term temporal variability. Journal of Paleontol- the Upper Cretaceous Two Medicine Formation of north- ogy 51:976–981. western Montana: evidence for drought-related mortality. Pa- Plummer, T., and A. M. Kinyua. 1994. Provenancing of hominid laios 5:394–413. and mammalian fossils from Kanjera, Kenya, using EDXRF. ———. 1993. Systematic patterns of time-averaging in the ter- Journal of Archaeological Science 21:553–563. restrial vertebrate record: a Cretaceous case study. Pp. 228– Potts., R. 1986. Temporal span of bone accumulations at Olduvai 249 in S. M. Kidwell and A. K. Behrensmeyer, eds. Taphonom- Gorge and implications for early hominid foraging behavior. ic approaches to time resolution in fossil assemblages. Short Paleobiology 12:25–31. Courses in Paleontology 6 (Paleontological Society). ———. 1988. Early hominid activities at Olduvai. Aldyne de Rogers, R. R., and S. M. Kidwell. 2000. Associations of verte- Gruyter, New York. brate skeletal concentrations and discontinuity surfaces in Potts, R., A. K. Behrensmeyer, and P. Ditchfield. 1999. Paleo- continental and shallow marine records: a test in the Creta- landscape variation and early Pleistocene hominid activities: ceous of Montana. Journal of Geology 108:131–154. Members 1 and 7, Olorgesailie Formation. Journal of Human Rolfe, W. D. I., E. N. K. Clarkson, and A. L. Panchen, eds. 1994 Evolution 37:747–788. (1993). Volcanism and early terrestrial biotas. Transactions of Poulicek, M., G. Goffinet, C. Jeuniaux, A. Simon, and M. F. Voss- the Royal Society of Edinburgh (Earth Sciences) 84. Foucart. 1988. Early diagenesis of skeletal remains in marine Roopnarine, P. D. 1999. Breaking the enigma of stratophenetic sediments: a 10 years study. Actes Colloque Recherches series: a computational approach to the analysis of microevo- Oce´anographiques en Mer Me´diterrane´e, Universite´ Etat, lutionary mode. Geological Society of America Abstracts with Liege 107–124. Programs 31(7):A42. Powell, E. N., and D. J. Davies. 1990. When is an ‘‘old’’ shell re- Roopnarine, P. D., G. Byars, and P. Fitzgerald. 1999. Anagenetic ally old? Journal of Geology 98:823–844. evolution, stratophenetic patterns, and random walk models. Powell, E. N., G. Staff, D. J. Davies, and W. R. Callender. 1989. Paleobiology 25:41–57. Macrobenthic death assemblages in modern marine environ- Roy, K., J. W. Valentine, D. Jablonski, and S. M. Kidwell. 1996. ments: formation, interpretation and application. CRC Criti- Scales of climatic variability and time averaging in Pleisto- cal Reviews in Aquatic Science 1:555–589. cene biotas: implications for ecology and evolution. Trends in Powell, E. N., R. J. Stanton Jr., A. Logan, and M. A. Craig. 1992. Ecology and Evolution 11:458–463. Preservation of Mollusca in Copano Bay, Texas. The long-term Sadler, P. M. 1981. Sediment accumulation and the completeness record. Palaeogeography, Palaeoclimatology, Palaeoecology of stratigraphic sections. Journal of Geology 89:569–584. 95:209–228. Sander, P. M. 1989. Early Permian depositional environments Prager, E. J., J. B. Southard, and E. R. Vivoni-Gallart. 1996. Ex- and pond bonebeds in central Archer County, Texas. Palaeo- periments on the entrainment threshold of well-sorted and geography, Palaeoclimatology, Palaeoecology 69:1–21. poorly sorted carbonate sands. Sedimentology 43:33–40. Sanders, H. L. 1968. Marine benthic diversity: a comparative Pratt, B. R. 1998. Probable predation on Upper Cambrian trilo- study. American Naturalist 102:243–282. bites and its relevance for the extinction of soft-bodied Bur- Schaal, S., and W. Ziegler, eds. 1992. Messel: an insight into the gess Shale-type animals. Lethaia 31:73–88. history of life and of the Earth. Translated by M. Shaffer-Feh- Rasmussen, K. A., and C. E. Brett. 1985. Taphonomy of Holocene re. Clarendon, Oxford. cryptic biotas from St. Croix, Virgin Islands: information loss Scha¨fer, W. 1972. Ecology and paleoecology of marine environ- and preservation biases. Geology 13:551–553. ments. University of Chicago Press, Chicago. Raup, D. M. 1975. Taxonomic diversity estimation using rare- Scheihing, M. 1980. Reduction of wind velocity by the forest faction. Paleobiology 1:333–342. canopy and the rarity of non-arborescent plants in the Upper Raymond, A., and C. Metz. 1995. Laurussian land-plant diver- Carboniferous fossil record. Augumenta Palaeobotanica 6: sity during the Silurian and Devonian: mass extinction, sam- 133–138. pling bias, or both? Paleobiology 21:74–91. Scheihing, M. H., and H. W. Pfefferkorn. 1984. The taphonomy TAPHONOMY AND PALEOBIOLOGY 145

of land plants in the Orinoco Delta: a model for the incor- in the Permian sequences of the Parana´ Basin: implications for poration of plant parts in clastic sediments of late Carbonif- the Phanerozoic trend in bioclastic accumulations. Revista erous age of Euramerica. Review of Palaeobotany and Paly- Brasileira de Geocieˆncias 30:495–499. nology 41:205–240. Smith, A. M., and C. S. Nelson. 1994. Selectivity in sea-floor pro- Schmude, D. E., and C. J. Weege. 1996. Stratigraphic relation- cesses: taphonomy of bryozoans. Pp. 177–180 in P. J. Hay- ship, sedimentology, and taphonomy of Meilyn, a dinosaur ward, J. S. Ryland, and P. D. Taylor, eds. Biology and palaeo- quarry in the basal Morrison Formation of Wyoming. Pp. 547– biology of bryozoans. Proceedings of the ninth international 554 in Morales 1996. bryozoology conference, 1992. Olsen and Olsen, Fredensborg, Schopf, T. J. M. 1978. Fossilization potential of an intertidal fau- Denmark. na: Friday Harbor, Washington. Paleobiology 4:261–270. Smith, D. M. 2000. Beetle taphonomy in a recent ephemeral lake, Schubert, J. K., D. L. Kidder, and D. H. Erwin. 1997. Silica-re- southeastern Arizona. Palaios 15:152–160. placed fossils through the Phanerozoic. Geology 25:1031– Smith, G. R., R. F. Stearley, and C. E. Badgley. 1988. Taphonomic 1034. bias in fish diversity from Cenozoic floodplain environments. Scoffin, T. P. 1992. Taphonomy of coral reefs: a review. Coral Palaeogeography, Paleoclimatology, Palaeoecology 63:263– Reefs 11:57–77. 273. Scott, A. C. 1990. Anatomical preservation of fossil plants. Pp. Smith, R. M. H. 1993. Vertebrate taphonomy of Late Permian 263–266 in Briggs and Crowther 1990. floodplain deposits in the southwestern Karoo Basin of South Scott, R. W. 1978. Approaches to trophic analysis of paleocom- Africa. Palaios 8:45–67. munities. Lethaia 11:1–14. ———. 1995. Changing fluvial environments across the Perm- Seilacher, R. W. 1984. Late Precambrian and Early Cambrian ian-Triassic boundary in the Karoo Basin, South Africa and metazoa: preservational or real extinctions? Pp. 159–168 in H. possible causes of tetrapod extinctions. Palaeogeography, Pa- D. Holland and A. F. Trendall, eds. Patterns of change in earth laeoclimatology, Palaeoecology 117:81–104. evolution. Springer, Berlin. Smith, R. M. H., and J. Kitching. 1996. Sedimentology and ver- ———. 1985. The Jeram Model: event condensation in a modern tebrate taphonomy of the Tritylodon Acme Zone: a reworked intertidal environment. Pp. 336–341 in U. Bayer and A. Sei- palaeosol in the Lower Jurassic Elliot Formation, Karoo Su- lacher, eds. Sedimentary and evolutionary cycles. Springer, pergroup, South Africa. Pp. 531–532 in Morales 1996. Berlin. Solomon, S., I. Davidson, and D. Watson. 1990. Problem solving ———. 1994. Early multicellular life: late Proterozoic fossils and in taphonomy. Tempus: archaeology and material culture the Cambrian extinction. Pp. 389–400 in Bengtson 1984. studies in anthropology, Vol. 2. University of Queensland, St. Seilacher, A., W. E. Reif, and F. Westphal. 1985. Sedimentologi- Lucia, Queensland, Australia. cal, ecological, and temporal patterns of fossil Lagersta¨tten. Speyer, S. E., and C. E. Brett. 1986. Trilobite taphonomy and Philosophical Transactions of the Royal Society of London B Middle Devonian taphofacies. Palaios 1:312–327. 311:5–23. ———. 1991. Taphofacies controls: background and episodic Sepkoski, J. J., Jr. 1978. A kinetic model of Phanerozoic taxonom- processes in fossil assemblage preservation. Pp. 501–545 in ic diversity: I. Analysis of marine orders. Paleobiology 4:223– Allison and Briggs 1991. 251. Spicer, R. A. 1980. The importance of depositional sorting to the ———. 1990. The taxonomic structure of periodic extinction. In biostratigraphy of plant megafossils. Pp. 171–183 in D. L. V. Sharpton and P. Ward, eds. Global catastrophes in earth Dilcher and T. N. Taylor, eds. Biostratigraphy of fossil plants. history. Geological Society of America Special Paper 247:33– Dowdon, Hutchinson, and Ross, New York. 44. ———. 1991. Plant taphonomic processes. Pp. 71–113 in Allison ———. 1993. Ten years in the library: new data confirm pale- and Briggs 1991. ontological patterns. Paleobiology 19:43–51. Stachowitsch, M. 1984. Mass mortality in the Gulf of Trieste: the Sepkoski, J. J., Jr., and C. F. Koch. 1996. Evaluating paleontologic course of community destruction. Marine Ecology (Publica- data relating to bio-events. Pp. 21–34 in O. H. Walliser, ed. zioni della Stazione Zoologica di Napoli) 5:243–264. Global events and event stratigraphy in the Phanerozoic. Staff, G. M., E. N. Powell, R. J. Stanton Jr., and H. Cummins. Springer, Berlin. 1985. Biomass: is it a useful tool in paleocommunity recon- Sepkoski, J. J., Jr., R. K. Bambach, D. M. Raup, and J. W.Valentine. struction? Lethaia 18:209–232. 1981. Phanerozoic marine diversity and the fossil record. Na- Staff, G. M., R. J. Stanton Jr., E. N. Powell, and H. Cummins. ture 293:435–437. 1986. Time averaging, taphonomy and their impact on paleo- Sepkoski, J. J., Jr., R. K. Bambach, and M. L. Droser. 1991. Secular community reconstruction: death assemblages in Texas bays. changes in Phanerozoic event bedding and the biological Geological Society of America Bulletin 97:428–443. overprint. Pp. 298–312 in G. Einsele, W. Ricken, and A. Sei- Stankiewicz, B. A., H. N. Poinar, D. E. G. Briggs, R. P. Evershed, lacher, eds. Cycles and events in stratigraphy. Springer, Ber- and G. O. Poinar Jr. 1998. Chemical preservation of plants and lin. insects in natural resins. Proceedings of the Royal Society of Sept, J. M. 1994. Bone distribution in a semi-arid riverine habitat London B 265:641–647. in eastern Zaire: implications for the interpretation of faunal Stanley, S. E., and L. A. Hardie. 1998. Secular oscillations in the assemblages at early archaeological sties. Journal of Archae- carbonate mineralogy of reef-building and sediment-produc- ological Science 21:217–235. ing organism driven by tectonically forced shifts in seawater Silva de Echols, C. M. H. M. 1993. Diatom infestation of Recent chemistry. Palaeogeography, Palaeoclimatology, Palaeoecol- crinoid ossicles in temperate waters, Friday harbor Labora- ogy 144:3–19. tories, Washington: implications for biodegradation of skele- Stewart, K. M., L. Lebranc, D. P. Matthiesen and J. West. 1999. tal carbonates. Palaios 8:278–288. Microfaunal remains from a modern east African raptor Simo˜es, M. G., A. C. Marques, L. H. C. Mello, and R. P. Ghilardi. roost: patterning and implications for fossil bone scatters. Pa- 2000a. The role of taphonomy in cladistic analysis: a case leobiology 25:483–503. study in Permian bivalves. Revista Espan˜ ola de Paleontologı´a Stoner, A. W., and M. Ray. 1996. Shell remains provide clues to 15:153–164. historical distribution and abundance patterns in a large sea- Simo˜es, M. G., M. Kowalewski, F. F. Torello, R. P. Ghilardi, and grass-associated gastropod (Stombus gigas). Marine Ecology L. H. C. Mello. 2000b. Early onset of modern-style shell beds Progress Series 135:101–108. 146 ANNA K. BEHRENSMEYER ET AL.

Strauss, D., and P. M. Sadler. 1989. Classical confidence intervals and inferences, hypotheses and models. In D. H. Erwin and and Bayesian probability estimates for ends of local taxon S. L. Wing, eds. Deep time: Paleobiology’s perspective. Paleo- ranges. Mathematical Geology 21:411–427. biology 26(Suppl. to No. 4):341–371. Sutcliffe, A. J. 1990. Rates of decay of mammalian remains in the Waisfeld, B. G., T. M. Sanchez, and M. G. Carrera. 1999. Biodiv- permafrost environment of the Canadian High Arctic. Pp. ersification patterns in the Early Ordovician of Argentina. Pa- 161–186 in C. R. Harrington, ed. Canada’s missing dimension: laios 14:198–214. science and history in the Canadian Arctic Islands, Vol. 1. Ca- Walker, K. R., and R. K. Bambach. 1971. The significance of fossil nadian Museum of Nature, Ottawa. assemblages from fine-grained sediments: time-averaged Tappan, M. J. 1994a. Savanna ecology and natural bone depo- communities. Geological Society of America Abstracts with sition: implications for early hominid site formation, hunting Programs 3:783–784. and scavenging. Current Anthropology 36:223–260. Walker, K. R., and W. W. Diehl. 1985. The role of marine cemen- ———. 1994b. Bone weathering in the tropical rain forest. Jour- tation in the preservation of lower Paleozoic assemblages. nal of Archaeological Science 21:667–673. Philosophical Transactions of the Royal Society of London B Tardy, Y., R. N’Kounkou, and J.-L. Probst. 1989. The global water 311:143–153. cycle and continental erosion during Phanerozoic time. Walker, S. E. 1988. Taphonomic significance of hermit crabs (An- American Journal of Science 289:455–483. omura: Paguridea): epifaunal hermit crab—infaunal gastro- Taylor, P. D. 1990. Preservation of soft-bodied and other organ- pod example. Palaeogeography, Palaeoclimatology, Palaeoe- isms by bioimmuration: a review. Palaeontology 33:1–17. cology 63:45–71. ———. 1994. Evolutionary paleoecology of symbioses between ———. 1989. Hermit crabs as taphonomic agents. Palaios 4:439– bryozoans and hermit crabs. Historical Biology 9:157–205. 452. Thayer, C. W. 1983. Sediment-mediated biological disturbance ———. 1995. Taphonomy of modern and fossil intertidal gas- and the evolution of marine benthos. Pp. 479–625 in M. J. S. tropod associations from Isla Santa Cruz and Isla Santa Fe, Tevesz and P. L. McCall, eds. Biotic interactions in recent and Galapagos Islands. Lethaia 28:371–382. fossil benthic communities. Plenum, New York. Walker, S. E., and J. R. Voight. 1994. Paleoecologic and tapho- Thomasson, J. R. 1991. Sediment-borne ‘‘seeds’’ from Sand nomic potential of deepsea gastropods. Palaios 9:48–58. Creek, northwestern Kansas: taphonomic significance and Walker, S. E., P. Van Cappellen, A. Roychoudhury, and C. Ko- paleoecological and paleoenvironmental implications. Pa- retsky. 1997. Preservation of experimentally deployed mol- laeogeography, Palaeoclimatology, Palaeoecology 85:213– luscan carbonate below the sediment-water interface. Geolog- 225. ical Society of America Abstracts with Programs 30:266. Traverse, A. 1990. Studies of pollen and spores in rivers and oth- Walker, S. E., K. Parsons-Hubbard, E. N. Powell, and C. E. Brett. er bodies of water, in terms of source-vegetation and sedi- 1998. Bioerosion or bioaccumulation? Shelf-slope trends for mentation, with special reference to Trinity River and Bay, epi- and endobionts on experimentally deployed gastropod Texas. Review of Palaeobotany Palynology 64:297–303. shells. Historical Biology 13:61–72. ———, ed. 1994. Sedimentation of organic particles. Cambridge Walter, M. R., J. J. Veevers, C. R. Calver, and K. Grey. 1995. Neo- University Press, Cambridge. proterozoic stratigraphy of the Centralian Superbasin, Aus- Trueman, C. N. 1999. Rare earth element geochemistry and ta- tralia. Precambrian Research 73:173–195. phonomy of terrestrial vertebrate assemblages. Palaios 14: Walters, L. J., and D. S. Wethey. 1991. Settlement, refuges, and 555–568. adult body form in colonial marine invertebrates: a field ex- Trueman, C. N., and M. J. Benton. 1997. A geochemical method periment. Biological Bulletin 180:112–118. to trace the taphonomic history of reworked bones in sedi- Warren, R. E. 1991. Ozarkian fresh-water mussels (Unionoidea) mentary settings. Geology 25:263–266. in the upper Eleven Point River, Missouri. American Mala- Underwood, C. J., and S. M. Bottrell. 1994. Diagenetic controls cological Bulletin 8:131–137. on multiphase pyritization of graptolites. Geological Maga- Wayne, R. K., J. A. Leonard, and A. Cooper. 1999. Full of sound zine 131:315–327. and fury: the recent history of ancient DNA. Annual Review Valentine, J. W. 1989. How good was the fossil record? Clues of Ecology and Systematics 30:457–477. from the Californian Pleistocene. Paleobiology 15:83–94. Webb, T., III. 1993. Constructing the past from late-Quaternary Valentine, J. W., D. Jablonski, and D. H. Erwin. 1999. Fossils, mol- pollen data: temporal resolution and a zoom lens space-time ecules, and embryos: new perspectives on the Cambrian ex- perspective. In Kidwell and Behrensmeyer 1993. plosion. Development 126:851–859. Wehmiller, J. F., L. L. York, and M. L. Bart. 1995. Amino acid Vaughan, A., and G. Nichols. 1995. Controls on the deposition racemization geochronology of reworked Quaternary mol- of charcoal: implications for sedimentary accumulations of lusks on US Atlantic coast beaches: implications for chron- fusain. Journal of Sedimentary Research A 65:129–135. ostratigraphy, taphonomy, and coastal sediment transport. Vermeij, G. J. 1977. The Mesozoic marine revolution: evidence Marine Geology 124:303–337. from snails, predators and grazers. Paleobiology 2:245–258. Westall, F., L. Boni, and E. Guerzoni. 1995. The experimental si- ———. 1987. Evolution and escalation: an ecological history of licification of microorganisms. Palaeontology 38:495–528. life. Princeton University Press, Princeton, N.J. Whittington, H. B., and S. Conway Morris, eds. 1985. Extraor- Voorhies, M. R. 1992. Ashfall: life and death at a Nebraska wa- dinary fossil biotas: their ecological and evolutionary signif- terhole ten million years ago. University of Nebraska State icance. Philosophical Transactions of the Royal Society of Museum, Museum Notes 81. London B 311:1–192. Vrba, E. S. 1995. On the connections between paleoclimate and Wilby, P. R., and D. E. G. Briggs. 1997. Taxonomic trends in the evolution. Pp. 24–45 in E. S. Vrba, G. H. Denton, T. C. Par- resolution of detail preserved in fossil phosphatized soft tis- tridge, and L. H. Burckle, eds. Paleoclimate and evolution, sues. Geobios Me´moire Spe´cial 20:493–502. with emphasis on human origins. Yale University Press, New Wilby, P. R., D. E. G. Briggs, P. Bernier, and C. Gaillard. 1996. Haven, Conn. Role of microbial mats in the fossilization of soft tissues. Ge- Wagner, P. J. 2000a. The quality of the fossil record and the ac- ology 24:787–790. curacy of phylogenetic inferences about sampling and diver- Wilf, P., and C. Labandeira. 1999. Response of plant-insect as- sity. Systematic Biology 49:65–86. sociations to Paleocene-Eocene warming. Science 284:2153– ———. 2000b. Phylogenetic analyses and the fossil record: tests 2156. TAPHONOMY AND PALEOBIOLOGY 147

Wilf, P., K. C. Beard, K. S. Davies-Vollum, and J. W. Norejko. Wing, S. L., and W. A. DiMichele. 1995. Conflict between local 1998. Portrait of a Late Paleocene (Early Clarkforkian) terres- and global changes in plant diversity through geologic time. trial ecosystem: Big Multi Quarry and associated strata, Was- Palaios 10:551–564. hakie Basin, southwestern Wyoming. Palaios 13:514–532. Wing, S. L, L. J. Hickey, and C. C. Swisher. 1993. Implications of Williams, M. E. 1994. Catastrophic versus noncatastrophic ex- an exceptional fossil flora for Late Cretaceous vegetation. Na- tinction of the dinosaurs: testing, falsifiability, and the burden ture 363:342–344. of proof. Journal of Paleontology 68:183–190. Wolff, R. G. 1975. Sampling and sample size in ecological anal- Wilson, M. V. H. 1988a. Taphonomic processes: information loss yses of fossil mammals. Paleobiology 1:195–204. and information gain. Geoscience Canada 15:131–148. Xiao, S., and A. H. Knoll. 1999. Fossil preservation in the Neo- ———. 1988b. Reconstruction of ancient lake environments us- proterozoic Doushantuo phosphorite Lagersta¨tte, South Chi- ing both autochthonous and allochthonous fossils. Palaeo- na. Lethaia 32:219–240. geography, Palaeoclimatology, Palaeoecology 62:609–623. Yang, H., and S. Yang. 1994. The Shanwang fossil biota in east- ———. 1988c. Predation as a source of fish fossils in Eocene lake ern China: a Miocene Konservat-Lagersta¨tte in lacustrine de- sediments. Palaios 2:497–504. posits. Lethaia 27:345–354. ———. 1993. Calibration of Eocene varves at Horsefly, British Zuschin, M., and J. Hohenegger. 1998. Subtropical coral-reef as- Columbia, Canada, and temporal distribution of specimens of sociated sedimentary facies characterized by molluscs (north- the Eocene fish Amyzon aggregatum Wilson. Kaupia (Darms- ern Bay of Safaga, Red Sea, Egypt). Facies 38:229–254. tadter Beitrage zur Naturgeschichte) 2:27–38. Zuschin, M., J. Hohenegger, and F. F. Steininger. 2000. A com- Wilson, M. V. H., and D. G. Barton. 1996. Seven centuries of parison of living and dead molluscs on coral reef associated taphonomic variation in Eocene freshwater fishes preserved hard substrata in the northern Red Sea—implications for the in varves: paleoenvironments and temporal averaging. Paleo- fossil record. Palaeogeography, Palaeoclimatology, Palaeoe- biology 22:535–542. cology 159:167–190.