The University of Manchester Research

Experimental analysis of soft-tissue fossilization

DOI:10.1111/pala.12360

Document VersionAccepted author manuscript

Link to publication record in Manchester Research Explorer

Citation for published version (APA):Purnell, M. A., Donoghue, P. J. C., Gabbott, S. E., McNamara, M. E., Murdock, D. J. E., & Sansom, R. S. (2018).Experimental analysis of soft-tissue fossilization: opening the black box. Palaeontology, 61(3), 317-323.https://doi.org/10.1111/pala.12360

Published in:Palaeontology

Citing this paperPlease note that where the full-text provided on Manchester Research Explorer is the Author Accepted Manuscriptor Proof version this may differ from the final Published version. If citing, it is advised that you check and use thepublisher's definitive version.

General rightsCopyright and moral rights for the publications made accessible in the Research Explorer are retained by theauthors and/or other copyright owners and it is a condition of accessing publications that users recognise andabide by the legal requirements associated with these rights.

Takedown policyIf you believe that this document breaches copyright please refer to the University of Manchester’s TakedownProcedures [http://man.ac.uk/04Y6Bo] or contact uml.scholarlycommunications@manchester.ac.uk providingrelevant details, so we can investigate your claim.

Download date:22. Jul. 2020

https://doi.org/10.1111/pala.12360https://www.research.manchester.ac.uk/portal/en/publications/experimental-analysis-of-softtissue-fossilization(a69eafdc-1ad6-4eeb-be86-9882677a4bcf).html/portal/robert.sansom.htmlhttps://www.research.manchester.ac.uk/portal/en/publications/experimental-analysis-of-softtissue-fossilization(a69eafdc-1ad6-4eeb-be86-9882677a4bcf).htmlhttps://doi.org/10.1111/pala.12360

Experimental analysis of soft-tissue fossilization – opening

the black box

Journal: Palaeontology

Manuscript ID PALA-11-17-4114-RC.R1

Manuscript Type: Review

Date Submitted by the Author: n/a

Complete List of Authors: Purnell, Mark; University of Leicester, School of Geography, Geology and the Environment Donoghue, Philip; University of Bristol School of Earth Sciences Gabbott, Sarah; University of Leicester, School of Geography, Geology and the Environment McNamara, Maria; University College Cork, School of Biological, Earth and

Environmental Sciences Murdock, Duncan; Oxford University Museum of Natural History; University of Leicester, School of Geography, Geology and the Environment Sansom, Robert; University of Manchester, School of Earth and Environmental Sciences

Key words: decay, fossilization, taphonomy, experiment, exceptional preservation

Palaeontology

Palaeontology

Experimental analysis of soft-tissue fossilization – opening the black box

Mark A. Purnell1*, Philip J. C. Donoghue2, Sarah E. Gabbott1, Maria McNamara3, Duncan

J. E. Murdock1,4, Robert S. Sansom5

1University of Leicester, School of Geography, Geology and the Environment, University

Road, Leicester, LE1 7RH, UK. mark.purnell@leicester.ac.uk

2University of Bristol, School of Earth Sciences, Life Sciences Building, Tyndall Avenue,

Bristol, BS8 1TQ, UK

3University College Cork, School of Biological, Earth and Environmental Science, Distillery

Fields, North Mall, Cork, T23 TK30, Ireland

4Oxford University Museum of Natural History, Parks Road Oxford, OX1 3PW, UK

5University of Manchester, School of Earth and Environmental Sciences, Manchester, M13

9PT, UK

*corresponding author

Page 1 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

ABSTRACT Taphonomic experiments provide important insights into fossils that preserve the remains of

decay-prone soft tissues – tissues that are usually degraded and lost prior to fossilization. These

fossils are among the most scientifically valuable evidence of ancient life on Earth, giving us a view

into the past that is much less biased and incomplete than the picture provided by skeletal remains

alone. Although the value of taphonomic experiments is beyond doubt, a lack of clarity regarding

their purpose and limitations, and ambiguity in the use of terminology, are hampering progress.

Here we distinguish between processes that promote information retention and those that promote

information loss in order to clarify the distinction between fossilization and preservation.

Recognising distinct processes of decay, mineralization and maturation, the sequence in which

they act, and the potential for interactions, has important consequences for analysis of fossils, and

for the design of taphonomic experiments. The purpose of well-designed taphonomic experiments

is generally to understand decay, maturation, and preservation individually, thus limiting the

number of variables involved. Much work remains to be done, but these methodologically

reductionist foundations will allow researchers to build towards more complex taphonomic

experiments and a more holistic understanding and analysis of the interactions between decay,

maturation and preservation in the fossilization of non-biomineralized remains. Our focus must

remain on the key issue of understanding what exceptionally preserved fossils reveal about the

history of biodiversity and evolution, rather than on debating the scope and value of an

experimental approach.

Page 2 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Arguably the most scientifically valuable evidence of ancient life on Earth comes from fossils

preserving remains of the decay-prone soft tissues (e.g. integument and muscle) that are usually

degraded and lost prior to fossilization. These examples of ‘exceptional preservation’ and the fossil

biotas from which they are recovered represent invaluable fossil archives, giving us a view into the

past that is much less biased and incomplete than the partial picture provided by skeletal remains

alone. Recent technological and methodological advances have allowed the acquisition of

progressively more detailed anatomical and chemical data on fossil soft tissues and, as a result,

reports of high fidelity preservation of anatomy (at micro- and macro- scales) and biomolecules are

expanding the known limits of morphological and chemical fossil preservation. One component of

current advances in the field is a reinvigoration of experimental investigations into the taphonomy

of non-biomineralized organisms and tissues: laboratory-based analyses of post-mortem decay,

maturation and mineralization, and the implications for processes of fossilization of soft tissue

remains and biomolecules (e.g. Raff et al. 2008; Sansom et al. 2010; Sansom et al. 2011;

Cunningham et al. 2012a; Cunningham et al. 2012b; McNamara et al. 2013; Murdock et al. 2014;

Colleary et al. 2015; Naimark et al. 2016). This type of taphonomic experiment, focused on non-

biomineralized remains, is the subject of this contribution; throughout, all references to ‘taphonomic

experiments’ do not include those designed to address questions of skeletal taphonomy. We use

the terms ‘soft tissues’ and ‘non-biomineralized tissues’ interchangeably, and to include sclerotized

tissues.

The application of experimental taphonomy to exceptional preservation has developed over

several decades (see Briggs and McMahon 2016 for a recent review), and has made major

contributions to our understanding of how non-biomineralized tissues become fossilized.

Experiments have provided significant insights, for example, into preservation of soft tissues

through maturation and stabilization of organic compounds (e.g. Gupta et al. 2006; Gupta et al.

2009), processes of microbially mediated authigenic mineralization (e.g. Sagemann et al. 1999),

microbial pseudomorphing of soft tissues (e.g. Raff et al. 2008; Raff et al. 2013), and how non-

random patterns of anatomical decay can introduce systematic biases into the interpretation of

exceptionally preserved fossils (e.g. Sansom et al. 2010; Murdock et al. 2014).

Page 3 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

The value of experimental and analytical approaches applied to the study of exceptionally

preserved fossils is beyond doubt, but we identify two issues that are hampering further progress.

First, a lack of clarity regarding the purpose and limits of experimental approaches, and second,

ambiguity in the use of terminology. More precise use of language and greater clarity regarding the

rationale for conducting taphonomic experiments will allow researchers to focus on the key issue of

what exceptionally preserved fossils reveal about the history of biodiversity and evolution, rather

than the scope and value of an experimental approach.

DECAY, MATURATION, PRESERVATION AND FOSSILIZATION

Clarity regarding ‘fossilization’ and ‘preservation’ are clearly crucial to taphonomic analysis,

otherwise we risk confusing processes with results, yet the terms are used interchangeably by

some authors and to mean distinct and different things by others.

Fossilization is one outcome of the range of processes that affect an organism after death (Figure

1). These processes cumulatively result in both loss and retention of information, and can balance

out in different ways, with the most obvious alternative outcome to fossilization being non-

fossilization - the loss of features or complete absence from the fossil record. Every part of every

organism ends up somewhere on a spectrum from partial to complete non-fossilization. We focus

here on decay, maturation and mineralization as distinct processes resulting in information

retention (preservation) and information loss.

Decay is the post-mortem process by which original biomolecules, tissues and structures are

degraded and lost through abiotic processes (such as chemical thermodynamics) and biotic

processes, such as autolysis and microbially mediated decomposition. For many researchers, the

antithesis of decay is preservation, but preservation is not the same thing as fossilization (see

below).

Preservation refers to the processes that directly result in retention of information (Figure 1);

processes by which inorganic and/or organic chemical activity replicates the form or converts the

remains of non-biomineralized tissues into minerals (mineralization) and organic compounds that

Page 4 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

are stable over geological timescales (maturation). This conversion can be via replacement or

replication by minerals (mineralization), chemical transformation through maturation of organic

compounds, or a combination. Different types of preservation may occur at different stages of

decay and maturation. This definition of preservation is neither new nor out of step with its

widespread use in the taphonomic literature; it serves merely to clarify the distinction between

preservation and fossilization.

Mineralization is replacement or replication of non-biomineralized tissues by minerals. It can

occur at any stage post-mortem, pre- or post-burial, early or late, and different modes of

mineralization can act at different times (and under different conditions) in the taphonomic history

of a fossil. Different modes of mineralization can occur in different parts within the same carcass,

linked to differences in microenvironments (McNamara et al. 2009). Although many exceptionally

preserved fossils have been mineralized, mineralization is not a requirement for exceptional

fossilization.

Maturation of organic remains occurs post-mortem, mostly post-burial, primarily in the diagenetic

realm, and can involve processes resulting in degradation and loss of biological information and/or

processes resulting in stabilization of organic compounds such that they can survive over

geological timescales. Maturation thus includes elements of both loss and retention of information,

with preservation of organic remains through maturation involving in situ polymerization of more

labile compounds (e.g. Gupta et al. 2006; Gupta et al. 2009; McNamara et al. 2016).

Secondary transformations might be considered as an additional stage in the post-mortem history

of a fossil. Like maturation, they can involve either loss or retention of information. Mineral

replacements from early stages of preservation can be lost or altered by subsequent chemical

activity under different conditions of diagenesis (iron sulfides converted to iron oxides, for

example). And organic remains stabilized through low temperature polymerization might be

oxidized, depolymerized or otherwise mobilized and lost during later, higher grades of maturation.

Information loss through weathering will also be a factor, but we do not consider this further here.

Page 5 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Focussing on these processes of decay, mineralization and maturation clarifies the distinction

between fossilization and preservation: fossilization of an organism’s remains is one outcome of

the balance and interactions between processes of information loss (decay and maturation) and

preservation (information retention via maturation and mineralization). This balance can be

expressed as two very simple equations:

1. processes of information loss > processes of information retention = no fossil

2. processes of information loss < processes of information retention = fossil

The balance and outcome reflected in equation 1 are far more prevalent than those of equation 2.

Original biological tissues, or components thereof, are either lost or transformed into materials that

are stable over geological timescales, and the existence of a fossil, even the most exquisitely

preserved, does not imply survival of its biological information without alteration and loss. In these

terms, ‘exceptional preservation’ is shorthand for part of the process that results in exceptional

fossilization; we should not ignore the unexceptional aspects of information loss.

THE RATIONALE FOR CONDUCTING TAPHONOMIC EXPERIMENTS

The goal of the vast majority of taphonomic experimental analyses is not ‘experimental

fossilization’: their aim is not to replicate the fossilization process in the laboratory or field. This is

because fossilization involves many variables, including both known and unknown unknowns, and

multiple confounding variables mean that - if our focus is on understanding the processes of

fossilization - the results of such fossil replication experiments, whether in general or for specific

Lagerstätten, are unlikely to tell us much of interest. While their success could be evaluated in

terms of whether they produce something that might look more or less like a fossil, the

experiments can reveal little if anything about the various processes involved in fossilization. This

is not to say that observations of what happens when organisms decay under natural conditions

have no place: these observations can provide useful constraints on the design of taphonomic

experiments, or guidance on what a fossil jellyfish might look like, for example. But crude

experiments that attempt to replicate fossilization without controlling variables are, in effect,

Page 6 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

treating fossilization as a Black Box (Figure 2). To understand the processes that control

information loss and information retention, the processes which ultimately produce the outcome of

fossilization, we need to see inside the box.

This is the focus of robust taphonomic experiments: experimental decay, experimental maturation,

and experimental preservation. The purpose of well-designed taphonomic experiments is generally

to understand these processes individually (thus limiting the number of variables involved and

making experiments tractable). In particular, robust taphonomic experiments investigate how

specific variables, e.g. environmental pH, availability of ions, diffusion rate, burial temperature, etc.,

have potentially affected the loss and retention of anatomical information and biased exceptionally

preserved biotas.

A simplifying initial assumption of this reductionist approach is that there are no direct causal links

between processes of decay, maturation and preservation. However, it is important to note that

this assumption applies only to the design of taphonomic experiments and does not preclude

finding evidence of links between the various processes of decay, preservation and maturation,

either in fossil data or in experimental results (a good example being taphonomic experiments

demonstrating that certain forms of mineralization require decay to establish the geochemical

gradients across which they operate (Sagemann et al. 1999)). In general, analysing and

understanding each of these processes is a prerequisite for clear understanding of potential

interactions between them.

Recognising the distinction between processes, the sequence in which they act (Figure 1), and the

potential for interactions, has important consequences for analysis of fossils; missing this point has

led to some unjustified criticism of experimental approaches. There is no reason to assume, for

example, that sequences of decay should provide a guide to sequences of preservation: decay

commences first (Figure 1) and provides a timeline and pattern of morphological modification and

loss — the products of decay, in the form of incomplete carcasses and decay-modified characters,

are the substrate upon which the processes of maturation and mineralization act. It is the timing

and interplay between processes of loss and retention that govern what anatomical structures,

Page 7 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

tissues and biomolecules are ultimately fossilized, and this can be unravelled only if we understand

the taphonomic patterns resulting from decay. Similarly, claims that decay experiments are not

applicable to the fossil record because either the environment or the experimental taxon used is

not a good analogue for what occurred in deep time are missing the point of controlled taphonomic

experiments. The same is true of criticisms that taphonomic experimental models are not

informative because fossilization in each deposit, taxon and even specimen reflects a suite of

unique conditions. Were these criticisms to be levelled at experiments designed to transform

carcasses into fossils, we would agree, but they carry little weight as arguments against the validity

of taphonomic experiments as an approach to modelling the general parameters that control

decay, maturation and mineralization and their respective roles in fossilization across

environments and taxa.

TAPHONOMIC EXPERIMENTS AND COMPARATIVE ANATOMY OF EXCEPTIONALLY

PRESERVED FOSSILS

It is important to point out here that taphonomic experiments can be designed to address a range

of different questions. For some, the goal is to understand the environmental conditions in which

organisms decayed and were ultimately preserved (e.g. Plotnick 1986; Kidwell and Baumiller 1990;

Hellawell and Orr 2012), or the degree to which a fossil biota is diminished in diversity by the loss

of soft bodied organisms. In such cases taphonomic signatures are a proxy for palaeoenvironment

or for faunal completeness, and the loss of anatomical information translates into gains in

geological data.

For much recent work, however, the purpose of taphonomic experiments is to provide better data

upon which to base interpretations of the anatomy of fossil organisms and, in turn, more accurate

reconstructions of phylogenetic relationships and evolutionary patterns. These endeavours rely on

comparative anatomical analysis, and when applied to exceptionally preserved non-biomineralized

fossils, this analysis is predicated on the assumption that differences between taxa are not simply

the result of random taphonomic processes. This is a crucial point that is often overlooked,

particularly in the most fundamental elements of comparative anatomy: the individuation of body

Page 8 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

parts and characters, and determination of the suite of characters present in a taxon (see Rieppel

and Kearney 2002 for discussion of individuation). Comparative analysis is possible only if the

similarities and differences in characters reflect original anatomy and, critically, if taphonomic

factors can be detected and taken into account. Furthermore, because comparative analysis

ultimately requires comparison with extant organisms, investigators must distinguish differences

that arise because of evolutionary history from those that simply reflect the incompleteness of the

fossil (Donoghue and Purnell 2009). This decision can be based either on intuitions and

assumptions, or on the crucial evidence generated by taphonomic experiments. We advocate the

latter approach.

TAPHONOMY, EXCEPTIONAL PRESERVATION, AND EXPERIMENTS - A WAY FORWARD

The processes of decay, maturation, and mineralization are controlled by diverse factors that vary

both spatially and temporally in how they act. A focus of future studies must be on deconvolving

the relative impact of these processes on our reading of the fossil record of exceptionally

preserved organisms and particular Lagerstätten. In many exceptionally preserved fossils, the

suite of characters present does not correspond to what might be predicted from a simplistic

application of experimental decay results (i.e. the fossils are not simply a collection of the most

decay resistant body parts). This is because the effects of preservational processes, sometimes

highly selective with regard to tissue types, are superimposed upon the results of decay; it is false

reasoning to suggest that fossilization of more than just decay resistant remains in itself indicates

that patterns and sequences of character transformation and loss observed in experimental decay

deviate from those that occurred in particular fossils. To use a specific example, invertebrate

nervous tissues decay rapidly under controlled experimental conditions (e.g. Murdock et al. 2014,

Sansom et al. 2015), yet mounting evidence supports the interpretation that some exceptionally

preserved biotas include specimens with fossilized nervous tissues (e.g. Ma et al. 2012; Strausfeld

et al. 2016). There is no conflict here: the fossils do not falsify the experiments, and the

experiments do not falsify the anatomical interpretations of the fossils. Together they highlight a

gap in our understanding of how nervous tissues become fossilized (Murdock et al. 2014). We

argue that decay experiments are the best starting point for understanding the biases and filters of

Page 9 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

fossilization because it is upon decayed remains that the processes of maturation and preservation

must operate.

Further work is also required to better understand the processes of mineralization and authigenic

mineral replication of original tissue. Despite experimental studies of the processes of replication of

soft tissues in calcium phosphate (Briggs and Kear 1993a, b; Briggs and Wilby 1996) and in pyrite

(Grimes et al. 2001; Grimes et al. 2002) other authigenic minerals have received little attention (but

see Martin et al. 2004; McCoy et al. 2015). Soft tissues may be preserved via multiple pathways in

a single fossil (e.g. Butterfield 2002; McNamara et al. 2009) but the controlling factors have yet to

be elucidated experimentally. Future studies focussing on these preservational processes will be

especially critical for attempts to extract tissue-specific and taxonomic signatures from fossil

specimens that preserve different tissues in different minerals. Similarly, greater understanding of

how microbial communities are mediating both decay and mineralization is likely to yield significant

new insights into preservational biases.

We have attempted here to clarify the essential elements and the terminology that form the

conceptual framework of taphonomic experiments, and the value of considering decay, maturation,

mineralization, and preservation as distinct but interacting processes. Greater emphasis on the

rationale for conducting particular experiments will enhance experimental design, allowing

taphonomists to construct more tightly constrained models of taphonomic processes. This

reductionist approach allows us to see beyond the Black Box view of fossilization, and from these

foundations we can build towards a more holistic understanding of the roles of — and interactions

between — decay, maturation and preservation in the fossilization of non-biomineralized remains.

Contributor statement. MAP, SEG and DJEM outlined the framework for discussions, and these

concepts were subsequently developed by all the authors. MAP produced the initial draft of the

manuscript; all authors contributed to its subsequent development and are listed alphabetically.

Page 10 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Acknowledgements. This paper arose from discussions during the informal workshop on

experimental and analytical taphonomy (Friends of the the Rotten) held in association with the

2016 Annual Meeting of the Palaeontological Association in Lyon. The stimulating environment

created by the participants in that workshop is gratefully acknowledged, as are the organisers of

the Annual Meeting for logistical support. MAP, SEG and DJEM funded by NERC grant

NE/K004557/1; PCJD by Royal Society Wolfson Merit Award and NERC NE/P013678/1, DJEM by

a Leverhulme Trust Early Career Fellowship, MMN by a European Research Council Starting

Grant H2020-20140-ERC-StG-637691-ANICOLEVO, RSS by NERC fellowship NE/I020253/2 and

BBSRC grant BB/N015827/1. The manuscript benefitted from constructive reviews provided by

Alex Liu and Paddy Orr, for which we are grateful.

Page 11 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Figure 1. A. Terminology for processes involved in fossilization. B. The sequence of action, effects and potential timespan over which processes act. Processes are not continuous; intervals when processes (horizontal bars) are operating more intensely are shown by more intense colours; the relative timing and duration of periods of more intense loss and retention of information (via mineralization and maturation) are schematic. Decay starts before other processes but the potential timespan over which it operates is shorter. It is irreversible, and the rate of decay is not constant. Mineralization can start before maturation, but not before decay has commenced, and can occur at any subsequent point, although not continuously (multiple phases of mineralization are possible). Mineralization that post-dates decay can only preserve information previously retained through either mineralization or maturation. Maturation, both information loss and retention, can start before or after mineralization and can occur at any subsequent point. That maturation is a process that promotes both information loss and information retention does not imply separation of these processes in fossilization. The diagram does not attempt to show

Page 12 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

interactions between processes, and should not be taken to imply, for example, that information loss and information retention through maturation occur simultaneously (although maturation can affect different tissues differently). Late stage information loss through weathering processes is not shown. ‘Information’, in the context of this figure, refers to primary anatomical, microanatomical and biochemical data, not information regarding the geological processes of fossilization and palaeoenvironment.

Page 13 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Figure 2. Cartoon illustrating the difference between experiments that attempt to replicate

fossilization, treating the process as a black box, and those that focus on the processes of decay,

maturation or mineralization. The black box approach reveals little about the processes of

information loss and information retention, the cumulative effects and interactions of which

ultimately results in a fossil (or, more often, not). Well-designed taphonomic experiments do not

attempt to replicate fossilization and do not result in a fossil, but provide reproducible data

concerning the processes of fossilization. An ideal taphonomic experiment, with all variables

identified and controlled for, is rarely attainable in practice.

REFERENCES

BRIGGS, D. E. G. and KEAR, A. J. 1993a. Decay and preservation of polychaetes: taphonomic

thresholds in soft-bodied organisms. Paleobiology, 19, 107-135.

--- 1993b. Fossilization of soft tissue in the laboratory. Science, 259, 1439-1442.

BRIGGS, D. E. G. and MCMAHON, S. 2016. The role of experiments in investigating the taphonomy

of exceptional preservation. Palaeontology, 59, 1-11.

BRIGGS, D. E. G. and WILBY, P. R. 1996. Authigenic mineralization of soft-bodied fossils: the

calcium carbonate - calcium phosphate switch. Journal of the Geological Society, 153, 665-

668.

BUTTERFIELD, N. J. 2002. Leanchoilia guts and the interpretation of three-dimensional structures

in Burgess Shale-type deposits. Paleobiology, 28, 155-171.

Page 14 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

COLLEARY, C., DOLOCAN, A., GARDNER, J., SINGH, S., WUTTKE, M., RABENSTEIN, R., HABERSETZER,

J., SCHAAL, S., FESEHA, M., CLEMENS, M., JACOBS, B. F., CURRANO, E. D., JACOBS, L. L.,

SYLVESTERSEN, R. L., GABBOTT, S. E. and VINTHER, J. 2015. Chemical, experimental, and

morphological evidence for diagenetically altered melanin in exceptionally preserved

fossils. Proc Natl Acad Sci U S A, 112, 12592-7.

CUNNINGHAM, J. A., THOMAS, C. W., BENGTSON, S., KEARNS, S. L., XIAO, S., MARONE, F.,

STAMPANONI, M. and DONOGHUE, P. C. 2012a. Distinguishing geology from biology in the

Ediacaran Doushantuo biota relaxes constraints on the timing of the origin of bilaterians.

Proc Biol Sci, 279, 2369-76.

CUNNINGHAM, J. A., THOMAS, C. W., BENGTSON, S., MARONE, F., STAMPANONI, M., TURNER, F.

R., BAILEY, J. V., RAFF, R. A., RAFF, E. C. and DONOGHUE, P. C. 2012b. Experimental

taphonomy of giant sulphur bacteria: implications for the interpretation of the embryo-like

Ediacaran Doushantuo fossils. Proc Biol Sci, 279, 1857-64.

DONOGHUE, P. C. J. and PURNELL, M. A. 2009. Distinguishing heat from light in debate over

controversial fossils. Bioessays, 31, 178-189.

GRIMES, S. T., BROCK, F., RICKARD, D., DAVIES, K. L., EDWARDS, D., BRIGGS, D. E. G. and PARKES,

R. J. 2001. Understanding fossilization: experimental pyritization of plants. Geology, 29,

123-126.

GRIMES, S. T., DAVIES, K. L., BUTLER, I. B., BROCK, F., EDWARDS, D., RICKARD, D., BRIGGS, D. E. G.

and PARKES, R. J. 2002. Fossil plants from the Eocene London Clay: the use of pyrite

textures to determine the mechanism of pyritization. Journal of the Geological Society,

159, 493-501.

GUPTA, N. S., CODY, G. D., TETLIE, O. E., BRIGGS, D. E. G. and SUMMONS, R. E. 2009. Rapid

incorporation of lipids into macromolecules during experimental decay of invertebrates:

Initiation of geopolymer formation. Organic Geochemistry, 40, 589-594.

GUPTA, N. S., MICHELS, R., BRIGGS, D. E. G., EVERSHED, R. P. and PANCOST, R. D. 2006. The

organic preservation of fossil arthropods: an experimental study. Proceedings of the Royal

Society B-Biological Sciences, 273, 2777-2783.

HELLAWELL, J. and ORR, P. J. 2012. Deciphering taphonomic processes in the Eocene Green River

Formation of Wyoming. Palaeobiodiversity and Palaeoenvironments, 92, 353-365.

KIDWELL, S. M. and BAUMILLER, T. 1990. Experimental disintegration of regular echinoids - roles

of temperature, oxygen, and decay thresholds. Paleobiology, 16, 247-271.

MA, X., HOU, X., EDGECOMBE, G. D. and STRAUSFELD, N. J. 2012. Complex brain and optic lobes in

an early Cambrian arthropod. Nature, 490, 258-61.

MARTIN, D., BRIGGS, D. E. G. and PARKES, R. J. 2004. Experimental attachment of sediment

particles to invertebrate eggs and the preservation of soft-bodied fossils. Journal of the

Geological Society, 161, 735-738.

MCCOY, V. E., YOUNG, R. T. and BRIGGS, D. E. G. 2015. Sediment permeability and the

preservation of soft-tissues in concretions: An experimental study. Palaios, 30, 608-612.

MCNAMARA, M. E., BRIGGS, D. E., ORR, P. J., FIELD, D. J. and WANG, Z. 2013. Experimental

maturation of feathers: implications for reconstructions of fossil feather colour. Biol Lett, 9,

20130184.

MCNAMARA, M. E., ORR, P. J., KEARNS, S. L., ALCALA, L., ANADON, P. and PENALVER MOLLA, E.

2009. Soft-tissue preservation in Miocene frogs from Libros, Spain: insights into the genesis

of decay microenvironments. Palaios, 24, 104-117.

MCNAMARA, M. E., VAN DONGEN, B. E., LOCKYER, N. P., BULL, I. D. and ORR, P. J. 2016.

Fossilization of melanosomes via sulfurization. Palaeontology, 59, 337-350.

MURDOCK, D., GABBOTT, S. E., MAYER, G. and PURNELL, M. A. 2014. Decay of velvet worms

(Onychophora), and bias in the fossil record of lobopodians. BMC Evol Biol, 14, 222.

Page 15 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

NAIMARK, E., KALININA, M., SHOKUROV, A., BOEVA, N., MARKOV, A., ZAYTSEVA, L. and GABBOTT,

S. 2016. Decaying in different clays: implications for soft-tissue preservation.

Palaeontology, 59, 583-595.

PLOTNICK, R. E. 1986. Taphonomy of a modern shrimp: implications for the arthropod fossil

record. Palaios, 1, 286-293.

RAFF, E. C., ANDREWS, M. E., TURNER, F. R., TOH, E., NELSON, D. E. and RAFF, R. A. 2013.

Contingent interactions among biofilm-forming bacteria determine preservation or decay

in the first steps toward fossilization of marine embryos. Evol Dev, 15, 243-56.

RAFF, E. C., SCHOLLAERT, K. L., NELSON, D. E., DONOGHUE, P. C. J., THOMAS, C.-W., TURNER, F. R.,

STEIN, B. D., DONG, X., BENGTSON, S., HULDTGREN, T., STAMPANONI, M., CHONGYU, Y.

and RAFF, R. A. 2008. Embryo fossilization is a biological process mediated by microbial

biofilms. Proceedings of the National Academy of Sciences, 105, 19360-19365.

RIEPPEL, O. and KEARNEY, M. 2002. Similarity. Biological Journal of the Linnean Society, 75, 59-82.

SAGEMANN, J., BALE, S. J., BRIGGS, D. E. G. and PARKES, R. J. 1999. Controls on the formation of

authigenic minerals in association with decaying organic matter; an experimental

approach. Geochimica et Cosmochimica Acta, 63, 1083-1095.

SANSOM, R. S., GABBOTT, S. E. and PURNELL, M. A. 2010. Non-random decay of chordate

characters causes bias in fossil interpretation. Nature, 463, 797-800.

SANSOM, R. S., GABBOTT, S. E. and PURNELL, M. A. 2011. Decay of vertebrate characters in

hagfish and lamprey (Cyclostomata) and the implications for the vertebrate fossil record.

Proceedings of The Royal Society B-Biological Sciences, 278, 1150-1157.

STRAUSFELD, N. J., MA, X. and EDGECOMBE, G. D. 2016. Fossils and the Evolution of the Arthropod

Brain. Curr Biol, 26, R989-R1000.

Page 16 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Figure 1. A. Terminology for processes involved in fossilization. B. The sequence of action, effects and potential timespan over which processes act. Processes are not continuous; intervals when processes

(horizontal bars) are operating more intensely are shown by more intense colours; the relative timing and

duration of periods of more intense loss and retention of information (via mineralization and maturation) are schematic. Decay starts before other processes but the potential timespan over which it operates is shorter. It is irreversible, and the rate of decay is not constant. Mineralization can start before maturation, but not before decay has commenced, and can occur at any subsequent point, although not continuously (multiple phases of mineralization are possible). Mineralization that post-dates decay can only preserve information

previously retained through either mineralization or maturation. Maturation, both information loss and retention, can start before or after mineralization and can occur at any subsequent point. That maturation is a process that promotes both information loss and information retention does not imply separation of these processes in fossilization. The diagram does not attempt to show interactions between processes, and should

not be taken to imply, for example, that information loss and information retention through maturation

Page 17 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

occur simultaneously (although maturation can affect different tissues differently). Late stage information loss through weathering processes is not shown. ‘Information’, in the context of this figure, refers to primary

anatomical, microanatomical and biochemical data, not information regarding the geological processes of fossilization and palaeoenvironment.

197x238mm (300 x 300 DPI)

Page 18 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Figure 2. Cartoon illustrating the difference between experiments that attempt to replicate fossilization, treating the process as a black box, and those that focus on the processes of decay, maturation or

mineralization. The black box approach reveals little about the processes of information loss and information retention, the cumulative effects and interactions of which ultimately results in a fossil (or, more often, not). Well-designed taphonomic experiments do not attempt to replicate fossilization and do not result in a fossil, but provide reproducible data concerning the processes of fossilization. An ideal taphonomic experiment,

with all variables identified and controlled for, is rarely attainable in practice.

79x56mm (300 x 300 DPI)

Page 19 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

16-Jan-2018 BCC All Referees Dear Mark, Your manuscript ID PALA-11-17-4114-RC entitled 'Experimental analysis of soft-tissue fossilization – opening the black box' has been reviewed. The comments of the referees are included at the bottom of this letter. Sorry this has taken longer than I would have hoped but the Christmas shutdown got in the way as I am sure you understand. The referees have recommended publication after minor revision. Please act on their comments and revise the manuscript accordingly. There seems little of substance that will cause you difficulties I think. When resubmitting your revision you must indicate how you have responded to each item raised by myself and the referees by cutting and pasting the remarks below and specifying against each comment what action has been taken. If you disagree with any specific action point and have made no change please explain your reasons. To resubmit your paper, log into https://mc.manuscriptcentral.com/pala and enter your Author Centre, where you will find your manuscript title listed under 'Manuscripts with Decisions'. Under 'Actions', click on 'Create a Revision'. Your manuscript number has been updated to denote a revision. IMPORTANT: Your original files are available to you when you upload the revised manuscript. Please delete any that are redundant before completing the submission. If your revision deadline expires before you are able to submit, please contact the Publications Officer Sally Thomas (editor@sallyandnik.co.uk) to arrange an extension. Because we aim at timely publication of papers submitted to Palaeontology, your revised manuscript should be uploaded as soon as possible. If you cannot submit it in a reasonable amount of time, we may be obliged to regard the paper as a new submission, which means that it will be sent out to referees again. Once again, thank you for submitting your paper to Palaeontology. I look forward to receiving the revised version. Yours sincerely, Dr. Andrew Smith Editor, Palaeontology a.smith@nhm.ac.uk Referees' Comments to Author: Referee: 1 Comments to the Author This manuscript is a useful contribution to research into experimental taphonomy, and provides clear definitions of the relevant processes involved in fossilization. I found it informative and enjoyable to review, and have only a small number of minor suggestions:

Page 20 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Preservation: Defining preservation as a “process by which… activity converts the remains of non-biomineralized tissues into minerals and organic compounds” may be arguably too restrictive, in that it may not be broad enough to cover replication processes. Replication is listed under the definition of mineralization, which is itself listed as a component of preservation in Table 1, so I think this should be clarified. Good point. Our wording was more restrictive than we intended. We have changed the text to read ‘processes by which inorganic and/or organic chemical activity replicates the form or converts the remains of non-biomineralized tissues into minerals’ p.9, l. 52: This section of the manuscript seems to be a response to the Parry et al. 2017 paper noted below (I appreciate citing this could be seen as accusatory, but for future researchers not familiar with the debate, it might be worth including reference to it somewhere). The referee is incorrect; this is not a response to Parry et al. (our paper was written before we were even aware of their manuscript. In fact, bizarrely, their paper is in part a response to what they thought (wrongly) our paper would be about). We would strongly prefer not to refer directly to Parry et al., precisely to avoid continuation of the accusatory approach that Parry et al. employ. See our response to reviewer 2 point 9. I felt that in the recent Palass Annual Address, one of the most compelling aspects of the talk was the case study of how to explain the preservation of only less-decay-resistant characters in a specimen (and other perceived to be ‘problematic’ permutations). The box-based, diagrammatic approach used was very effective, and the authors might like to consider including this here as an additional figure (I do not insist on this, merely suggest it as I found it very clear). The box-based diagrammatic approach presented in the Annual Address is taken from another, near-finalized paper that looks explicitly at the relationship between decay and preservation in particular exceptionally preserved fossils. We cannot employ it here. We have attempted to better convey the interactions of information loss and gain in our new Figure 1. Table 1: Rather than ‘Affects’, I would suggest that the listed bullet points are ‘components’ or ‘processes’. It might also be worth explaining the distribution of crosses and dashes in the lowest section – I was trying to match it up to the columns above for a while, before eventually realising that wasn’t the point. Table1 has been redrafted as Figure 1. This addresses the reviewers concerns p. 3, l. 38: Remove ‘our’ done p. 3, l. 47: Remove the spare ‘)’ done p. 5, l. 38: The recent Parry et al., 2017 paper in Bioessays also shows some nice examples of this differential mineralization within individual specimens. See our response to reviewer 2 point 9. p. 17, l. 35: Note the typo in ‘ultimately’ corrected Alex Liu University of Cambridge Referee: 2

Page 21 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Comments to the Author The manuscript has two purposes. It defines a series of terms that are widely, and (it is argued) loosely, used by taphonomists (and palaeobiologists in general), and details how these relate to each other. The second purpose is to argue that recent criticisms of the relevance of decay experiments to understanding fossilisation is because critics have misunderstood their purpose. In summary the argument presented is that critics have expectations of the results from experimental studies that the experiments were never designed to (and cannot) deliver in the first place. The paper provides a useful definition of certain key terms, and challenges the reader to think about the purpose and limitations of decay experiments. It is a timely contribution to a ‘hot’ discussion topic at the moment (see also Parry et al 2018 Bioessays 40, 1700167 – to which this paper should probably refer (Are there any obvious references omitted? YES), so appropriate for Palaeontology. It is relevant to both the specialist (discussion of experiments), but a useful introduction of concepts to the uninitiated. Review is relatively straightforward: minor revisions are required. My comments and suggestions are listed below and correspond to numbering on the annotated pdf. Some relate to typographical errors which the authors should, and will want to, correct. Of the remainder I would emphasise point 4 below, and those that relate to the content of the table. I think there are possible modifications and clarifications that could be made to it. Patrick Orr 1. I know fossilisation is inclusive of preservation but would “preservation and fossilization” be better than fossilization – given one key argument is that most experimental approaches are not intended to simulate fossilization (“the number of variables involved in fossilization makes replication of the entire process inherently intractable”). This would also (at the outset of the paper and in one sentence) make it clear that the two terms are not synonyms. We can see what the reviewer is getting at here, but we do not think referring to both preservation and fossilization in the title would clarify. The importance of distinguishing between fossilization and preservation is in the abstract, so we do deal with at the outset of the paper. 2. commonly – but incorrectly of course – eb = subset of KL – Seilacher’s definition of latter includes biotas comprising solely articulated multi-element skeletons. I’d clarify this to avoid accidentally perpetuating this. We have deleted the parenthetical reference to Konservat Lagerstätten. 3. is non-random the best descriptor? The corollary, that a random pattern of decay will not introduce a systematic bias, isn’t correct. However, unlikely (and depending on n) it is possible that a pattern could be defined that was random but also systematic. It might be sufficient to say how the patterns of anatomical decay identified via experimentation can introduce systematic biases… We disagree with the reviewer on this point; if a pattern of decay introduces systematic bias, it’s not a random pattern.

Page 22 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

In any case, even if the reviewer were correct, our statement that ‘non-random patterns of anatomical decay can introduce systematic bias’ is unequivocally correct. We have not made any change here. 4 I think it would be extremely useful to include a summary diagram along the lines of that sketched at the back of the attached pdf that shows visually how the terms relate to each other – fossilisation as the inclusive term, preservation and decay as competing sources of ‘retention’ and ‘gain’, the 2 alternative pathways for preservation. I could see this as a powerpoint slide and textbook diagram in the future. We struggled to understand elements of the sketch provide by the reviewer. We have incorporated two simple equations into the text, and modified Figure 1 (which was Table 1). Combined, we think these edits address the reviewer’s point. 5 is a process integral to the cell really abiotic? It is essentially a biochemical process programmed to occur after cell death? The reviewer has misunderstood us here, and we have rewritten the text to make our meaning clearer. 6 I think this refers to preservation so perhaps “…the latter is not the same…” would be more specific We have changed the text to reduce ambiguity. “…preservation is not the same as fossilization” 7 via (as it is a mechanism/process)rather than through? done 8 I don’t think the results will necessarily be ambiguous – they could be very easily to interpret definitively – i.e. unambiguous – I think the risk is that they will be imprecise or not that relevant Text modified to clarify our meaning: “This is because fossilization involves many variables, including both known and unknown unknowns, and multiple confounding variables mean that - if our focus is on understanding the processes of fossilization - the results of such fossil replication experiments, whether in general or for specific Lagerstätten, are unlikely to tell us much of interest. While their success could be evaluated in terms of whether they produce something that might look more or less like a fossil, the experiments can reveal little if anything about the various processes involved in fossilization.” 9 Here, and below, there is reference to “criticisms” and “claims” by others, but no publications are specified. For anyone looking to use this paper by way of an introduction to the debate, pointers to where alternative opinions are expressed would be useful. It would allow them to consider the contrasting opinions ‘in the round’. This is why I’ve indicated that I think its general relevance could be enhanced We have tried hard with this manuscript to avoid the accusatory tone of some recent papers and do not wish to further inflame the debate. If we pick specific examples, these authors may feel they have been unfairly singled out, and that they are being provoked. This would not further the science. Furthermore, we feel that it would do little to enhance the general relevance of our paper to refer to papers that attack a simplistic straw man misrepresentation of robust decay experiments. This isn’t a difference of opinion, its misrepresentation (possibly reflecting misunderstanding) of the purpose and scope of experiments. We prefer to address this by articulating clearly what experimental taphonomy

Page 23 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

is about – the purpose of our paper - rather than respond to specific claims concerning what it is not about. Nevertheless, we have changed the text to address the reviewers underlying point about disagreement. The opening paragraph of the section TAPHONOMY, EXCEPTIONAL PRESERVATION, AND EXPERIMENTS - A WAY FORWARD now includes the following: “This is because the effects of preservational processes, sometimes highly selective with regard to tissue types, are superimposed upon the results of decay; it is false reasoning to suggest that fossilization of more than just decay resistant remains in itself indicates that patterns and sequences of character transformation and loss observed in experimental decay deviate from those that occurred in particular fossils. To use a specific example, invertebrate nervous tissues decay rapidly under controlled experimental conditions (e.g. Murdock et al. 2014, Sansom et al. 2015), yet mounting evidence supports the interpretation that some exceptionally preserved biotas include specimens with fossilized nervous tissues (e.g. Ma et al. 2012; Strausfeld et al. 2016). There is no conflict here: the fossils do not falsify the experiments, and the experiments do not falsify the anatomical interpretations of the fossil. Together they highlight a gap in our understanding of how nervous tissues become fossilized (Murdock et al. 2014).” 10 It is not “the results of decay that are the substrate” – the result of decay could be non-fossilisation – i.e. no substrate. It is the products generated during decay – I wouldn’t use substrate – implication of mineral templating? I’d go with materials We have modified the text in response to the reviewer’s comments, but we still think that ‘substrate’ conveys our meaning better than materials. Text now reads “…the products of decay, in the form of incomplete carcasses and decay-modified characters, are the substrate upon which the processes of maturation and mineralization act.” 11 somewhere earlier I’d use a sentence to define what decay patterns are – see also point 3 (where I suggested removing patterns from the actual text). There the original text mentions what they are but doesn’t define them., That spot, however, could be a place to comment as follows: “how the patterns of anatomical decay identified via experimentation can introduce systematic biases into the interpretation of exceptionally preserved fossils. It is the effect of these systematic biases on anatomical characters and character complexes that allow predictive decay patterns to be defined.” …or something along those lines This point seems to be labelled as a second point 10 in the annotated pdf. We have changed the text to read “patterns resulting from decay” rather than patterns of decay. We have also checked and edited the manuscript to ensure that our use of ‘pattern’, as distinct from the process of decay, is as unambiguous as possible. 12 by “differences in characters” do you mean different characters are present in e.g. two taxa, or that a character present exists in a different state in the two taxa being compared. Potentially, it could of course be both. Amend to “similarities and differences in characters and character states….”? It is both. That this is the way the reviewer has understood it (as indicated by his suggested rewording) suggests we don't need to change this. It remains unchanged. 13 what does spatially refer to – palaeogeography? inside a carcass? Both.

Page 24 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

14 temporally – is this referring to the duration of the decay history – could also refer to changes over time through geological recors – e.g. phosphatization window. perhaps some qualifying statements on what both ‘spatially’ and ‘temporally’ refer to would be useful here It refers to both. This applies also to comment 13. Our use of ‘vary both spatially and temporally’ is meant in an inclusive sense. We don't think it is ambiguous, so we have not changed the text here. 15 Yes, and as the authors know it is actually even more complicated than that especially at the scale of an organism. Preservation (often of specific tissues) can precede decay of other tissues (think Myoscolex for example); i.e. decay and preservation are more decoupled than the text as phrased might imply to readers less familiar with the topic. It is not just because I’d add a sentence at end “Superimposed on this general relationship, preservation via mineralisation is often highly selective; for example, for certain tissues or even parts thereof (Myoscolex, Lean. mid-gut fglands, E-K layer in Libros frogs, Luke Parry’s Lebanon worms) and certain loci in body (Unwin pterosaur skin). We have modified the text here to reduce ambiguity regarding patterns (see comment 11) and address the reviewers point. See also 16 and 9. Paragraph now reads: ‘This is because the effects of preservational processes, sometimes highly selective with regard to tissue types, are superimposed upon the results of decay; it is false reasoning to suggest that fossilization of more than just decay resistant remains in itself indicates that patterns and sequences of character transformation and loss observed in experimental decay deviate from those that occurred in particular fossils. To use a specific example, invertebrate nervous tissues decay rapidly under controlled experimental conditions (e.g. Murdock et al. 2014, Sansom et al. 2015), yet mounting evidence supports the interpretation that some exceptionally preserved biotas include specimens with fossilized nervous tissues (e.g. Ma et al. 2012; Strausfeld et al. 2016). There is no conflict here: the fossils do not falsify the experiments, and the experiments do not falsify the anatomical interpretations of the fossils. Together they highlight a gap in our understanding of how nervous tissues become fossilized (Murdock et al. 2014).. 16 Hmm, this is correct, but it is not equally true for maturation and preservation. I think it might be worth flagging that point. We have modified this paragraph extensively to address the reviewers point (although we have not used his suggested wording). See response to 15 and 9. “Mineralisation is often strongly decoupled from the relative recalcitrance of the tissues (in fact the most decay resistant are not those preserved via mineralization most frequently). Preservation involving mineralisation will often produce a suite of characters in a fossil that are markedly different from what would be predicted from the decay of an analogue. This does not negate the value of decay experiments – in fact just the opposite. Where such discrepancies arise the null hypothesis should be that factors in addition to retarding the decay process and maturation are involved.” “Maturation involves altering the biochemistry of the organic remains, thus what is preserved is more likely to relate largely, but not exclusively*, to relative recalcitrance and thus will be strongly sympathetic to the patterns and sequences of decay recovered via experimentation.” * One point I’m not sure of (and whether it has been tested or not I don’t know) is whether certain tissues show any propensity for maturation i.e. mature selectively? Bog body studies might be relevant)

Page 25 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

18 One could argue that ‘soft’ is as misused a term as preservation and fossilization. Would non-biomineralized be (slightly) better – at least there is a (slightly) better defined cut-off threshold than ‘soft’. For example, if you are going to cite the Butterfield (2002 Leanchoila paper) would those heavily sclerotized and thick arthropod cuticles really be soft? Strictly speaking, we agree with the reviewer on this point, and we have added a statement to the introduction to clarify this: ‘We use the terms ‘soft tissues’ and ‘non-biomineralized tissues’ interchangeably.’ ‘Soft tissues’ is common in the taphonomic literature, and doesn't seem to cause much confusion. 19 we need not just better understanding of the interactions between these – we need a better understanding of the individual role of each. You could amend this as follows: “and from these foundations we can build towards a more holistic understanding of the roles of each of, and interactions between, decay, maturation and preservation” I’d delete “and analysis” – the understanding depends on that so it is implicit Edited. Now reads ‘and from these foundations we can build towards a more holistic understanding of the roles of — and interactions between — decay, maturation and preservation in the fossilization of non-biomineralized remains.’ 20 various typos in this figure caption 21 I don’t think any info. is ever gained via taphonomy – it is retained or lost Response to points 20 and 21. A shabby caption. Corrections made. 22 ‘at surface or post-burial – I’d omit this – decay occurs while carcasses float Changed to pre- or post-burial 23 maturation is shown schematically in the diagram at bottom, and stated here and in the text to be initiated only after burial – I am not so sure – why can’t it occur at sediment surface – might be worth checking this out (of course, post-burial could imply a mm below surface, but I don’t think that is the sense that is intended). Notably, one of the pathways/processes listed in the table as maturation loss is oxidation – which occurs in other than the post-burial environment – even in the water column. We have modified the text to take the reviewers point into account, and now state that maturation occurs post-mortem, mostly post-burial, primarily in the diagenetic realm. 24 Should this be Effects not Affects? - these are the effects (i.e. the results) of decay – not what decay affects i.e. impacts on. Yes. Silly error corrected. 25 I know this is schematic, but I’m not convinced as to what it adds, and it could perhaps be misleading. For example are maturation (loss) and maturation (retention) always synchronous, or nearly so. I thought for a while that the pattern for each process mapped onto the columns above, but I don’t think that is intended. Could you be persuaded to remove it and state in the caption ‘Decay will be initiated immediately after death. The other processes, maturation (loss and retention) and mineralization will occur at any stage of the preservation process and can occur on more than one occasion” and get rid of this bit?

Page 26 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

We have revised this part of what was a table and incorporated it into the new figure. Hopefully this improves clarity of what we are trying to convey. 26 Should authigenesis really be distinguished from diagenesis? Is authigenesis (neomorphic minerals) not part of diagenesis? See also point 28 re terminology. Changed to ‘Authigenic mineralization and templating of soft tissues’ 27 I don’t think this is right – it is not “polymerization of breakdown products of decay” – the breakdown products are of the original tissues and generated during decay. it is not the products of decay that are being broken down wording changed to “Polymerization of original biomolecules and breakdown products formed during decay” 28 The post-burial setting is divided by the authors into the pre-diagenetic and diagenetic realms, and distinguished from the ‘surface’. “Processes occurring post mortem, at surface or post-burial (in prediagenetic and diagenetic realms)” I don’t think the terms pre-diagenetic and diagenetic necessarily map onto ‘surface’ and ‘post-burial’ as this implies/indicates. For example, syn-sedimentary cementation – which is a diagenetic process, can occur at the sea-floor (think hardgrounds pebble flat conglomerates), i.e. pre-burial. Also “Processes occurring post mortem, at surface or post-burial” – these are separated as if three stages – strictly speaking those at the surface and post-burial are also post-mortem not distinct from it. A three-fold distinction would be “Processes occurring before deposition, at surface or post-burial…” This isn’t intended to be three stages or a three-fold distinction. ‘Pre-diagenetic’ and ‘diagenetic’ are not intended to map onto ‘surface’ and ‘post-burial’. The reviewer has perhaps overthought this. We have replaced ‘at surface’ with ‘pre-burial’ which helps avoid ambiguity. 29 NBT = ? Error. deleted Referee: 3 Comments to the Author Thank you for paying such close attention to the journal style. The following notes concern technical issues that will need to be addressed. If you have any queries, particularly technical aspects of figures, please contact me directly (Sally Thomas, editor@palass.org). There are guidelines on journal style available on www.palass.org. General notes • Please respond directly to all referee comments, including the technical comments below. It is particularly important that you explain your reasoning if you have not followed any of the suggestions made in the reports. • Some abstracting databases have a limit on the number of words accepted. Please reduce your abstract to no more than 250 words. Done

Page 27 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Figure • Please supply your figure at a resolution of 600 dpi; either PNG, or TIF format with LZW compression. Please do not use JPG compression at any stage of figure construction. • Please note that the figure should be 166 mm wide if it is intended to be full page width, or 110 mm wide if two-thirds page width. It would be helpful if you could indicate which is intended by noting the mm width in the file name. Done. Submitted as illustrator files. • Please use single quotes throughout (you currently have ‘fossil’ and “Black Box”; journal style would be ‘Black Box’). Quotes have been removed from ‘black box’ in figure. • As this is being considered as a Frontiers in Palaeontology review paper (rather than a Rapid Communication) the figure would be printed in colour, with the fee covered by PalAss. Table • Tables in journal style have a very restricted style, and I would suggest referring to this as a figure. We have reformatted this as figure 1. • You currently have the definition of ‘Information’ both as a footnote and in the caption. It would probably work as either, but it shouldn’t be both. If this was a table it should definitely be a footnote (and the caption should be a single sentence description), but as this is a figure it fits well at the end of the caption. done • Assuming that the ‘Affects’ label refers to all four boxes in the row, could I suggest setting this as a vertical label (with text rotated) to the left of the current box, using a column that could be blank next to the top three rows? ‘Relative timing’ could be styled in the same way even though it is clearer as this row is not split. Addressed in new figure 1. Sally Thomas editor@palass.org

Page 28 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Experimental analysis of soft-tissue fossilization – opening the black box

Mark A. Purnell1*, Philip J. C. Donoghue2, Sarah E. Gabbott1, Maria McNamara3, Duncan J. E. Murdock1,4, Robert S. Sansom5

1University of Leicester, School of Geography, Geology and the Environment, University Road, Leicester, LE1 7RH, UK. mark.purnell@leicester.ac.uk

2University of Bristol, School of Earth Sciences, Life Sciences Building, Tyndall Avenue, Bristol, BS8 1TQ, UK

3University College Cork, School of Biological, Earth and Environmental Science, Distillery Fields, North Mall, Cork, T23 TK30, Ireland

4Oxford University Museum of Natural History, Parks Road Oxford, OX1 3PW, UK

5University of Manchester, School of Earth and Environmental Sciences, Manchester, M13 9PT, UK

*corresponding author

Deleted: of

Deleted: current address: Deleted:

Page 29 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

ABSTRACT Taphonomic experiments provide important insights into fossils that preserve the remains of

decay-prone soft tissues – tissues that are usually degraded and lost prior to fossilization. These

fossils are among the most scientifically valuable evidence of ancient life on Earth, giving us a view

into the past that is much less biased and incomplete than the picture provided by skeletal remains

alone. Although the value of taphonomic experiments is beyond doubt, a lack of clarity regarding

their purpose and limitations, and ambiguity in the use of terminology, are hampering progress.

Here we distinguish between processes that promote information retention and those that promote

information loss in order to clarify the distinction between fossilization and preservation.

Recognising distinct processes of decay, mineralization and maturation, the sequence in which

they act, and the potential for interactions, has important consequences for analysis of fossils, and

for the design of taphonomic experiments. The purpose of well-designed taphonomic experiments

is generally to understand decay, maturation, and preservation individually, thus limiting the

number of variables involved. Much work remains to be done, but these methodologically

reductionist foundations will allow researchers to build towards more complex taphonomic

experiments and a more holistic understanding and analysis of the interactions between decay,

maturation and preservation in the fossilization of non-biomineralized remains. Our focus must

remain on the key issue of understanding what exceptionally preserved fossils reveal about the

history of biodiversity and evolution, rather than on debating the scope and value of an

experimental approach.

Formatted: Font:Helvetica, 11 pt

Deleted: Taphonomicexperimentsprovideimportantinsightsintofossilsthatpreservetheremainsofdecay-pronesofttissues–tissuesthatareusuallydegradedandlostpriortofossilization.ThesefossilsareamongthemostscientificallyvaluableevidenceofancientlifeonEarth,givingusaviewintothepastthatismuchlessbiasedandincompletethanthepictureprovidedbyskeletalremainsalone.Althoughthevalueoftaphonomicexperimentsisbeyonddoubt,alackofclarityregardingtheirpurposeandlimitations,andambiguityintheuseofterminology,arehamperingprogress.Herewedistinguishbetweenprocessesthatpromoteinformationretentionandthosethatpromoteinformationlossinordertoclarifythedistinctionbetweenfossilizationandpreservation.Recognisingdistinctprocessesofdecay,mineralizationandmaturation,thesequenceinwhichtheyact,andthepotentialforinteractions,hasimportantconsequencesforanalysisoffossils,andforthedesignoftaphonomicexperiments.Thegoalofthevastmajorityoftaphonomicexperimentsisnot‘experimentalfossilization’becausethenumberofvariablesinvolvedinfossilizationmakesreplicationoftheentireprocessinherentlyintractable.Thepurposeofwell-designedtaphonomicexperimentsisgenerallytounderstanddecay,maturation,andpreservationindividually,thuslimitingthenumberofvariablesinvolved.Muchworkremainstobedone,butthesemethodologicallyreductionistfoundationswillallowresearcherstobuildtowardsmorecomplextaphonomicexperimentsandamoreholisticunderstandingandanalysisoftheinteractionsbetweendecay,maturationandpreservationinthefossilizationofnon-biomineralizedremains.Ourfocusmustremainonthekeyissueofwhatexceptionallypreservedfossilsrevealaboutthehistoryofbiodiversityandevolution,ratherthandebatingthescopeandvalueofanexperimentalapproach.

Formatted: Highlight

Page 30 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Arguably the most scientifically valuable evidence of ancient life on Earth comes from fossils

preserving remains of the decay-prone soft tissues (e.g. integument and muscle) that are usually

degraded and lost prior to fossilization. These examples of ‘exceptional preservation’ and the fossil

biotas from which they are recovered represent invaluable fossil archives, giving us a view into the

past that is much less biased and incomplete than the partial picture provided by skeletal remains

alone. Recent technological and methodological advances have allowed the acquisition of

progressively more detailed anatomical and chemical data on fossil soft tissues and, as a result,

reports of high fidelity preservation of anatomy (at micro- and macro- scales) and biomolecules are

expanding the known limits of morphological and chemical fossil preservation. One component of

current advances in the field is a reinvigoration of experimental investigations into the taphonomy

of non-biomineralized organisms and tissues: laboratory-based analyses of post-mortem decay,

maturation and mineralization, and the implications for processes of fossilization of soft tissue

remains and biomolecules (e.g. Raff et al. 2008; Sansom et al. 2010; Sansom et al. 2011;

Cunningham et al. 2012a; Cunningham et al. 2012b; McNamara et al. 2013; Murdock et al. 2014;

Colleary et al. 2015; Naimark et al. 2016). This type of taphonomic experiment, focused on non-

biomineralized remains, is the subject of this contribution; throughout, all references to ‘taphonomic

experiments’ do not include those designed to address questions of skeletal taphonomy. We use

the terms ‘soft tissues’ and ‘non-biomineralized tissues’ interchangeably, and to include sclerotized

tissues.

The application of experimental taphonomy to exceptional preservation has developed over

several decades (see Briggs and McMahon 2016 for a recent review), and has made major

contributions to our understanding of how non-biomineralized tissues become fossilized.

Experiments have provided significant insights, for example, into preservation of soft tissues

through maturation and stabilization of organic compounds (e.g. Gupta et al. 2006; Gupta et al.

2009), processes of microbially mediated authigenic mineralization (e.g. Sagemann et al. 1999),

microbial pseudomorphing of soft tissues (e.g. Raff et al. 2008; Raff et al. 2013), and how non-

Deleted: – commonly referred to as ‘Konservat-Lagerstätten’ -

Deleted: our

Deleted: )

Page 31 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

random patterns of anatomical decay can introduce systematic biases into the interpretation of

exceptionally preserved fossils (e.g. Sansom et al. 2010; Murdock et al. 2014).

The value of experimental and analytical approaches applied to the study of exceptionally

preserved fossils is beyond doubt, but we identify two issues that are hampering further progress.

First, a lack of clarity regarding the purpose and limits of experimental approaches, and second,

ambiguity in the use of terminology. More precise use of language and greater clarity regarding the

rationale for conducting taphonomic experiments will allow researchers to focus on the key issue of

what exceptionally preserved fossils reveal about the history of biodiversity and evolution, rather

than the scope and value of an experimental approach.

DECAY, MATURATION, PRESERVATION AND FOSSILIZATION

Clarity regarding ‘fossilization’ and ‘preservation’ are clearly crucial to taphonomic analysis,

otherwise we risk confusing processes with results, yet the terms are used interchangeably by

some authors and to mean distinct and different things by others.

Fossilization is one outcome of the range of processes that affect an organism after death (Figure

1). These processes cumulatively result in both loss and retention of information, and can balance

out in different ways, with the most obvious alternative outcome to fossilization being non-

fossilization - the loss of features or complete absence from the fossil record. Every part of every

organism ends up somewhere on a spectrum from partial to complete non-fossilization. We focus

here on decay, maturation and mineralization as distinct processes resulting in information

retention (preservation) and information loss.

Decay is the post-mortem process by which original biomolecules, tissues and structures are

degraded and lost through abiotic processes (such as chemical thermodynamics) and biotic

processes, such as autolysis and microbially mediated decomposition. For many researchers, the

antithesis of decay is preservation, but preservation is not the same thing as fossilization (see

below).

Deleted: Table

Deleted: ,Deleted: and abiotic processes (such as autolysis and chemical thermodynamics)Deleted: this

Page 32 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Preservation refers to the processes that directly result in retention of information (Figure 1);

processes by which inorganic and/or organic chemical activity replicates the form or converts the

remains of non-biomineralized tissues into minerals (mineralization) and organic compounds that

are stable over geological timescales (maturation). This conversion can be via replacement or

replication by minerals (mineralization), chemical transformation through maturation of organic

compounds, or a combination. Different types of preservation may occur at different stages of

decay and maturation. This definition of preservation is neither new nor out of step with its

widespread use in the taphonomic literature; it serves merely to clarify the distinction between

preservation and fossilization.

Mineralization is replacement or replication of non-biomineralized tissues by minerals. It can

occur at any stage post-mortem, pre- or post-burial, early or late, and different modes of

mineralization can act at different times (and under different conditions) in the taphonomic history

of a fossil. Different modes of mineralization can occur in different parts within the same carcass,

linked to differences in microenvironments (McNamara et al. 2009). Although many exceptionally

preserved fossils have been mineralized, mineralization is not a requirement for exceptional

fossilization.

Maturation of organic remains occurs post-mortem, mostly post-burial, primarily in the diagenetic

realm, and can involve processes resulting in degradation and loss of biological information and/or

processes resulting in stabilization of organic compounds such that they can survive over

geological timescales. Maturation thus includes elements of both loss and retention of information,

with preservation of organic remains through maturation involving in situ polymerization of more

labile compounds (e.g. Gupta et al. 2006; Gupta et al. 2009; McNamara et al. 2016).

Secondary transformations might be considered as an additional stage in the post-mortem history

of a fossil. Like maturation, they can involve either loss or retention of information. Mineral

replacements from early stages of preservation can be lost or altered by subsequent chemical

activity under different conditions of diagenesis (iron sulfides converted to iron oxides, for

Deleted: Table

Deleted: in

Deleted: tDeleted: through

Page 33 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

example). And organic remains stabilized through low temperature polymerization might be

oxidized, depolymerized or otherwise mobilized and lost during later, higher grades of maturation.

Information loss through weathering will also be a factor, but we do not consider this further here.

Focussing on these processes of decay, mineralization and maturation clarifies the distinction

between fossilization and preservation: fossilization of an organism’s remains is one outcome of

the balance and interactions between processes of information loss (decay and maturation) and

preservation (information retention via maturation and mineralization). This balance can be

expressed as two very simple equations:

1. processes of information loss > processes of information retention = no fossil

2. processes of information loss < processes of information retention = fossil

The balance and outcome reflected in equation 1 are far more prevalent than those of equation 2.

Original biological tissues, or components thereof, are either lost or transformed into materials that

are stable over geological timescales, and the existence of a fossil, even the most exquisitely

preserved, does not imply survival of its biological information without alteration and loss. In these

terms, ‘exceptional preservation’ is shorthand for part of the process that results in exceptional

fossilization; we should not ignore the unexceptional aspects of information loss.

THE RATIONALE FOR CONDUCTING TAPHONOMIC EXPERIMENTS

The goal of the vast majority of taphonomic experimental analyses is not ‘experimental

fossilization’: their aim is not to replicate the fossilization process in the laboratory or field. This is

because fossilization involves many variables, including both known and unknown unknowns, and

multiple confounding variables mean that - if our focus is on understanding the processes of

fossilization - the results of such fossil replication experiments, whether in general or for specific

Lagerstätten, are unlikely to tell us much of interest. While their success could be evaluated in

terms of whether they produce something that might look more or less like a fossil, the

experiments can reveal little if anything about the various processes involved in fossilization. This

Deleted: through

Deleted: too

Deleted: , making experimental replication of the entire process inherently intractable, even if one ignores the issue of the geological timescales over which some processes of fossilization operate. The full complexity of the natural environmental conditions experienced by any particular fossil organism from the moment of its death to discovery cannot be reconstructed in an experiment. MDeleted: to replicate fossilizationDeleted: will Deleted: be ambiguous

Page 34 of 45

Palaeontology

Palaeontology

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

is not to say that observations of what happens when organisms decay under natural conditions

have no place: these observations can provide useful constraints on the design of taphonomic

experiments, or guidance on what a fossil jellyfish might look like, for example. But crude

experiments that attempt to replicate fossilization without controlling variables are, in effect,

treating fossilization as a Black Box (Figure 2). To understand the processes that control

information loss and information retention, the processes which ultimately produce the outcome of

fossilization, we need to see inside the box.

This is the focus of robust taphonomic experiments: experimental decay, experimental maturation,

and experimental preservation. The purpose of well-designed taphonomic experiments is generally

to understand these processes individually (thus limiting the number of variables involved and

making experiments tractable). In particular, robust taphonomic experiments investigate how

specific variables, e.g. environmental pH, availability of ions, diffusion rate, burial temperature, etc.,

have potentially affected the loss and retention of anatomical information and biased exceptionally

preserved biotas.

A simplifying initial assumption of this reductionist approach is that there are no direct causal links

between processes of decay, maturation and preservation. However, it is important to note that

this assumption applies only to the design of taphonomic experiments and does not preclude

finding evidence of links between the various processes of decay, preservation and maturation,

either in fossil data or in experimental results (a good example being taphonomic experiments

demonstrating that certain forms of mineralization require decay to establish the geochemical

gradients across which they operate (Sagemann