Site Map      Contact

   You are here: Home > Research > Melanosomes > Taphonomy of the Jehol beds
   Melanosomes home
   Melanin and melanosomes
   Melanosomes in birds
   Melanosome preservation
   Colour in Jehol birds
   Colour in Jehol dinosaurs
   The Jehol biota
   Jehol taphonomy
   Methods of study
   Other fossil melanosomes
   Feathers and bird origins
   Feathered ornithischians
Taphonomy of the Jehol beds

The remarkable fossils of the Jehol Biota display a range of tissues. Preserved remains include hard tissues such as cuticles of plants and invertebrates and vertebrate skeletons, as well as soft tissues such as stomach contents, colour patterns and twigs with leaves and flowers still attached. The taphonomy of these fossils requires careful study, and is still only partially understood.

In understanding the preservation of the Jehol fossils, it is likely that one model will not be sufficient. After all, the sediments belong to three formations, and span 11 million years. Further, they are exposed over a huge area, many thousands of kilometres wide. Nevertheless, there are some general points: the fossils are nearly all found in fine-grained sediments deposited in ancient lakes and there are often volcanic sediments (ashes, lavas) associated. Further, like most sites of exceptional fossil preservation, it can be assumed that oxygen was absent from the lake-bed sediments when fossils were preserved: this then limited scavenging and decay processes and allowed mineralization of soft tissues.

Four themes can be explored in a little more detail: a preliminary taphonomic model; the initial stages of the taphonomic history, including potential killing mechanisms and mode of entry to the depositional environment; the origin of the opisthotonic posture in fossil vertebrates, where the neck is bent back, as if in a death spasm; and the mode of preservation of the feathers.

Preliminary taphonomic model

The general taphonomic history links preservation of the biota to episodes of volcanic activity (Zhou et al. 2003). The most productive horizons are beneath ash tuff falls, which would have entombed most of the organisms present in the water column - the tuff layers are strongly correlated with mass mortality events. Animals and plants that lived in the water, as well as carcasses that were washed in, or fell in (insects, birds, pterosaurs) were presumably buried rapidly in the fine-grained sediments.

The most productive fossil beds lie below ash-fall deposits, so these are presumably mass mortality beds where the living organisms were suffocated, and organic remains were encased instantly, in the absence of oxygen and hence of decomposers and scavengers. The tuffs also appear to have sealed the fossil-rich layers; this, or more plausibly, unfavourable bottom-water conditions, could have inhibited later burrowing or digging from above, and so the fossils remain in a highly articulated state.

Leng & Yang (2003) noticed that pyrite microcrystallines and framboids occur extensively inside and on the surface of plant fossils, indicating dysoxic aqueous conditions with free oxygen levels less than 30 Ķmol per litre for the microenvironment where these framboids were formed. They also suggested rapid tissue degradation that occurred during the earliest stages of fossilization. They further proposed a 'fossil envelope' model to accommodate the different geochemical conditions between the microenvironment surrounding the fossil material and the macroenvironment of the background lake bottom water. This is therefore a classic Konservat Lagerstätte, a deposit known for exceptional preservation, where the soft parts are preserved as impressions or casts.

The preservation of soft tissues includes dermal structures, such as scales, feathers and hair, sometimes showing colour banding (Fig. 1 below; probably not the original colours, but indicating the presence and absence of different original colours). In addition to dermal structures, many Jehol fossils retain a two-dimensional, dark-coloured body outline, often considered to be 'organic' or 'carbonaceous' in composition (Fig. 1).

[left] Figure 1. The dark, organic body outline, perhaps representing decayed muscle and internal organs around two specimens of the bird Confuciusornis sanctus from the Yixian Formation, also showing light and dark colouring on the long tail plumes and other feathers of the presumed male (left) and female (right). (Courtesy of IVPP.)

Figure 2. The teleost fish Lycoptera, showing the skeleton, a light brown region representing the body outline, and a black spot representing the carbonaceous remains of the retina, and probably rich in melanosomes.

Figure 3. A small larval cryptobranchoid salamander showing the skull and backbone, not fully ossified yet, with a clear body outline and gills outlined in light brown material. The tadpole-like tail fin even shows gentle folds. (Courtesy of IVPP.)

Specimens often display a lighter-coloured periphery associated with this dark-coloured film (Figures 2, 3 above; Wang & Evans, 2006), a combination and arrangement strikingly similar to that found in some other localities. Evans & Wang (2007) reported the impressions of skin cells and square scales in a lizard specimen, and our initial analyses of Jehol material have also revealed examples of subcircular squameous skin cells between the feathers of several taxa of birds and dinosaurs. The fish Lycoptera shows a darkened patch corresponding possibly to the swim bladder, and black eye spots (Figure 2 above).

The absence of more labile soft tissues, those that rot away first, such as muscles and internal organs, is a surprise. This is a major contrast with some other Mesozoic and Cenozoic sites of exceptional preservation which apparently share similar sedimentary conditions, such as the Santana and Crato formations of Brazil (Early Cretaceous), Las Hoyas, Spain (Early Cretaceous), Messel in Germany (Eocene), and the Libros locality in Spain (Miocene). The virtual absence of these labile tissues in the Jehol Biota is considered real. Similarly, evidence for fossilized microbes is also absent.

Death and burial: the initial stages of taphonomic history

Jehol specimens were rarely entombed in ash beds. Specimens are found in finely laminated sediments, probably more or less where they fell. The sediment layers are so thin, being generally only a millimetre or so thick, that the carcasses could not have been transported by the gentle currents that deposited the sediment. Specimens of birds lie flat, with their largest surface area parallel to bedding, suggesting that they settled vertically through the water column and came to rest on the sediment–water interface (Figure 1).

Most skeletons of Jehol dinosaurs, birds, and mammals are usually more or less complete. Others may lack the head or feet, or other portions, showing that they were broken up between death and burial. This, and the widespread loss of labile tissues such as muscle and internal organs suggests that death and burial did not occur instantaneously in Jehol times.

Origin of the opisthotonic posture in fossil vertebrates

Many specimens of birds and dinosaurs are preserved with the head bent sharply backwards, and this has been interpreted variously as evidence that the animals died in agony, or more often that the ligaments down the back of the neck somehow tightened up after death. This is a typical mode of preservation in the Jehol beds, and Jehol specimens have been instrumental in recent debates.

Figure 4. The 'Berlin' specimen of Archaeopteryx, collected in 1877, showing the opisthotonic posture, with the head folded sharply back over the shoulder region.

The 'head-back' preservation position (Figure 4) is called opisthotonus, formally a position where the backbone is extreme dorsally hyperextended, with the skull and neck recurved over the back, and with strong extension of the tail (Marshall Faux & Padian, 2007). These authors argue that opisthotonus arose at the time of death, not afterwards, and they attribute it to poisoning of the central nervous system, a phenomenon seen most often in animals with high metabolic rates, such as birds and dinosaurs.

Rigor mortis, the stiffening of the body after death, typically persists for 24-48 hours, after which the muscles are once again flaccid. Any disturbance of the specimen after that period, such as by scavenging or current transport, is likely to obliterate evidence of an opisthotonic posture acquired at the time of death.

Marshall Faux & Padian (2007) argue that rapid burial of recently deceased carcasses is the key to retaining an opisthotonic posture. They consider this likely to have happened not only in the Jehol Biota, but also in other cases, including the classic specimens of Archaeopteryx from the Solnhofen lagoon in southern Germany (Figure 4). Their argument for rapid burial of animals that stiffened into an opsithotonic posture at death is weakened by evidence that the Solnhofen organisms were overgrown by a microbial mat after deposition; clearly rapid burial is inconsistent with this. Marshall Faux & Padian's (2007) case is probably strongest for the Jehol Biota, in which an association of fossiliferous horizons and event beds (volcanic ash horizons) has been noted above.

But why would vertebrates apparently so often die in agony? Various potential causes of an opisthotonic posture are identified by Marshall Faux & Padian (2007), including asphyxiation and environmental toxins, specifically, gases derived from volcanic eruptions and cyanobacterial blooms. This might seem to be a possibility in the Jehol beds because volcanic ash layers are commonly associated with the fossil beds. Perhaps then gases explelled into the atmosphere by erupting volcanoes poisoned the air and killed the animals in agony.

Neither model for opisthotonus is entirely convincing: there are few modern examples where one can observe either a killing event, or accumulation of carcasses, where all animals have died in agony with their heads back, nor are there convincing examples of carcasses tightening and the heads pulling back some time after death. In the case of the Jehol beds, careful mapping of the occurrence of opisthotonus bed-by-bed and correlation, or not, with volcanic beds is required.

Taphonomy of feathers

The taphonomy of the Jehol feathers has wider implications in the current debate about the integumentary structures of early birds and dinosaurs. These have now been shown (Zhang et al. 2010) to preserve remarkable detail at the finest levels of resolution under the scanning electron microscope (SEM), including melanosomes, the organelles that provide colour to feathers.

At higher magnification, the rachis or central quill, is nearly always absent, and yet it is composed of keratin like the rest of the feather, and relativel thick layers of keratin at that. The vanes of feathers are dark in colour in backscattered electron images (Fig. 5b), suggesting that they are preserved as carbon; energy dispersive X-ray analysis confirms this. The carbon is clearly the degraded remains of the original tissues; i.e. these have not been replicated in authigenic minerals. There is, however, some replication of the rachis in calcium phosphate (e.g. Confuciusornis; Figures 5c, d) and the claw sheath of the same sample is also phosphatized.

XXXFigure 10 from PGA paper Figure 10. Photographs illustrating aspects of the taphonomy of Liaoning feathers. (a) Typical example of an isolated feather from Liaoning (IVPP V15210); note that the rachis, the central 'quill', is not preserved. (b) Back-scatter electron SEM image of a typical feather, in which dark tones are indicative of carbon preservation (confirmed by X-ray analysis); the central rachis contains the same silicate assemblage as the rock matrix, and the small carbonaceous particles are plant fragments. (c) Light micrograph of a Confuciusornis feather, and (d) back-scatter electron SEM image of the same feather showing the light-coloured phosphatized rachis in a largely silicate matrix; feather barbs are less distinct than in (b), indicating less carbon. Scale bars in millimetres.

Feathers occur in various states of preservation. These range from essentially pristine, with the barbs and barbules of each vane in life position, to a more 'bedraggled' appearance, in which groups of barbs are bundled along the length of the rachis and individual bundles are separated from each other. Inside each bundle the barbs occur juxtaposed or overlapping. This is more likely to be preservational damage than the original appearance of the feather in life.

Literature cited

  • Benton, M.J., Zhou Z., Orr, P.J., Zhang F., & Kearns, S.L. 2008. The remarkable fossils from the Early Cretaceous Jehol Biota of China and how they have changed our knowledge of Mesozoic life. Proceedings of the Geologists' Association 119, 209-228. pdf
  • Evans, S.E. & Wang, Y. 2007. A juvenile lizard specimen with well-preserved skin impressions from the Upper Jurassic/ Lower Cretaceous of Daohugou, Inner Mongolia, China. Naturwissenschaften 94, 431–439.
  • Leng, Q. & Yang, H. 2003. Pyrite framboids associated with the Mesozoic Jehol Biota in northeastern China: Implications for microenvironment during early fossilization. Progress in Natural Science 13, 206–212.
  • Marshall Faux, C. & Padian, K. 2007. The opisthotonic posture of vertebrate skeletons: post-mortem contraction or death throes? Paleobiology 33, 201-226.
  • Wang, Y. & Evans, S.E. 2006. A new short-bodied salamander from the Upper Jurassic/ Lower Cretaceous of China. Acta Palaeontologica Polonica 51, 127-130.
  • Zhang, F., Kearns, S.L, Orr, P.J., Benton, M.J., Zhou, Z., Johnson, D., Xu, X., and Wang, X. 2010. Fossilized melanosomes and the colour of Cretaceous dinosaurs and birds. Nature 463, 1075-1078 (doi:nature08740.3d). pdf.
  • Zhou, Z., Barrett, P.M. & Hilton, J. 2003. An exceptionally preserved Lower Cretaceous ecosystem. Nature 421, 807-814.

Dicynodon Illustration courtesy of John Sibbick.
Design by ParanoidFish Website & Graphic Design & EikonWorks.
Dept. of Earth Sciences, University of Bristol, Wills Memorial Building, Queen's Road, Bristol, UK BS8 1RJ
Tel: +44 117 9545400  Fax: +44 117 9253385  Email:  Web: