The taphonomy of soft tissues and the evolution of feathers

  • Evan Saitta

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)


Understanding how fossils form is imperative if one wishes to interpret them. For palaeontology to contribute to biology regarding the behaviour, physiology, ecology, and evolution of extinct organisms, interdisciplinary approaches must account for physical and chemical changes of organismal remains through geologic processes.
One example of this interdisciplinary requirement is with regard to feather evolution. Feathers serve many key functions (e.g., thermoregulation, signalling, camouflage, water repellence, flight) and are a key adaptation with respect to the diversity and longevity of the avian clade and some non-avian dinosaurs. Feathers are also complex integumentary structures consisting of organic and inorganic components (e.g., keratin protein, pigments, and calcium phosphate) and, although not preserved as often as skeletal material, are represented by a relatively large sample of exceptionally preserved fossils. This thesis attempts to understand the taphonomy of keratinous structures and soft tissues more generally, which can help us to better interpret fossils, especially when modern analogues are lacking.
Following an introduction to molecular taphonomy, a background behind the analytical methods used, and a primer on the application of molecular taphonomy to the study of feather evolution, part one of this thesis, containing chapters one and two, examines fossil feather preservation. In chapter one, a Late Cretaceous Shuvuuia deserti dinosaur feather fibre, once thought to consist of keratin protein based on antibody staining, is reanalysed using microscopic, elemental, and molecular analyses and found to be inorganic, inconsistent with protein, and largely calcium phosphate. The fact that Shuvuuia feathers are preserved as simple phosphatic fibres potentially indicates that the feathers of this taxon were more complex in life than originally thought, with a calcified rachis whose inorganic molecules are the sole tissue component preserved and barbs whose organics were lost. The analyses revealed that cyanoacrylate glue coated the fibre, deriving from the fossil’s preparation. Cyanoacrylates are known to adsorb antibodies, offering a mechanism to explain the earlier antibody stains for keratin protein as false positives. Applying immunohistochemistry to fossils is therefore ill-advised given risks of false positives. Furthermore, the direct chemical analyses of this fossil indicate that protein is lost while other components can survive.
In chapter two, the more heavily-studied mode of integumentary structure preservation is examined using photographs of carbonaceously-preserved feathers from China. Previous work has concluded that such organic fossils of keratinous structures consist solely of stable, yet diagenetically-altered, melanin pigment and melanosomes. Here, decay processes that separated individual feathers from the rest of the plumage reveal that some paravians had primitive, extinct contour feather morphologies. Other primitive traits of dinosaur feathers are examined with a discussion of how these extinct morphologies might have altered feather function. In addition, this chapter highlights the importance of taphonomy when interpreting fossils that lack appropriate modern analogues (e.g., evo-devo models do not predict some fossil feather morphologies).
While keratinous structures can be hypothesised to show differential stability of their underlying components through phosphatic or pigment-based preservation using direct observations of fossils, the loss of keratin protein can also be experimentally investigated. Part two of this thesis, containing chapters three through five, focuses on diagenesis. In chapter three, differential survival between tissues or the components of a tissue is examined through the development of a novel experimental procedure. Thermal maturation has been used for decades to simulate molecular stability through diagenesis, but the breakdown products observed are not always present in fossils. To account for molecular mobility and stability, specimens are compacted into sediment prior to maturation, allowing pore spaces to act as a filter whereby stable organics remain in situ while unstable organic breakdown products can escape through the sediment. In this chapter, experiments on feathers and lizards produce macrostructures and ultrastructures resembling those of exceptionally-preserved fossils, showing simultaneous loss of proteins and labile lipids while melanosomes remain. Sediment-encased maturation may assist in addressing a wide variety of taphonomic hypotheses and emphasises the importance of diagenetic stability on the preservation potential of organismal remains.
One question that arises is whether the pattern of preferential loss of proteins and labile lipids presented here represents an accurate understanding, given that claims of dinosaur bone proteins, cells, and vessels abound in the literature. In chapter four, this thesis looks beyond keratin and critically examines the organics within dinosaur fossil bone. Chemical and biological analyses of Late Cretaceous dinosaur bone, an open system, collected in a manner to minimize contamination suggest that no endogenous protein or cellular structures survive although some kerogen from endogenous lipids might be present. Instead, the bone provides a habitat for a unique and thriving subterranean microbiome, complicating claims of original bone organics and demonstrating how even recent biological processes must be considered when studying taphonomy.
It is important to determine what chemical signature protein would yield after tens-of-millions-of-years without complications due to the loss of original organics or introduction of exogenous contamination. To address this, closed systems must be examined. In chapter five, chemical analyses of Late Cretaceous sauropod dinosaur eggshell calcite, a closed system, reveal endogenous amino acids consistent with complete peptide sequence loss (i.e., only the four most thermally-stable amino acids were detected and all are fully racemised, while no non-contaminant peptide sequences could be detected). These results represent well-supported protein-derived material from non-avian dinosaurs. More importantly, the complete hydrolysis of closed-system proteins suggests that preservation of Mesozoic peptide sequences is unlikely.
Fossils are the only available data on extinct organisms for which there are no appropriate modern analogues, and correct interpretation of this data must consider taphonomy. This thesis concludes with a summary of the current state of knowledge on keratinous structure fossilisation, a judgement on the validity of previous dinosaur ‘soft tissue’ claims, and suggestions for future molecular taphonomy research. When analysing fossil keratinous structures, pigments and calcium phosphate are likely the only surviving components in specimens of appreciable age or thermal maturation. However, despite the limited preservation potential of proteins, their thermally-stable amino acids, in addition to stable biominerals, pigments, and certain lipids, will likely offer great potential for molecular taphonomists to study life in the deep past. Stable biominerals have always provided the vast majority of palaeontological data, while fossil pigments and lipids can provide unique palaeobiological insight such as colour patterning and, possibly in the future, sex determinations of fossil specimens, respectively. Although biomolecules prone to extreme diagenetic degradation lose information, even greatly-altered fossil biomolecules, such as closed-system eggshell amino acids, may at the very least provide material for new stable isotope approaches.
Date of Award23 Jan 2019
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorJakob Vinther (Supervisor)

Cite this