Alabama’s Wealth of Fossil Dinosaur Feathers

Two months ago, I wrote a short piece for Nature News about a collection of eleven feathers locked in Cretaceous amber that had been discovered in western Canada. These bits of ancient plumage were not like the flattened, carbonized feathers which surround the bodies of dinosaurs found in China’s Cretaceous ashfall beds – they were the actual tissues which covered the bodies of 70 million year old theropod dinosaurs. Which feathers belonged to early birds and which belonged to their non-avian dinosaur relatives is uncertain, but the fact that such intricate structures were so perfectly preserved is wonderful.

But there was something I missed when I wrote that article. A few months earlier, in the journal PALAIOS, Auburn University paleontologists Terrell Knight, Sean Bingham, Ronald Lewis, and Charles Savrda described a slightly larger collection of fourteen fossil feathers found in the remnants of a roughly 84 million year old estuary now preserved in Alabama’s Ingersoll shale. While the preservation of these feathers is not as spectacular as those found in the Canadian amber, they still can give us a peek at the colorful world of feathered dinosaurs along the prehistoric coast of Alabama.

All fourteen of the recovered specimens were contour feathers. Most of them formed the outer layer of plumage on the body, although the researchers also categorized one as a wing feather and another as a tail feather. Each was found within a fossil-rich layer of clay in what was once part of a tidal channel. This setting for a feather deposit is unusual. Many other assemblages of fossil feathers have been found in the remnants of prehistoric lakes or in collections of amber.

How the feathers wound up in the primeval estuary is anyone’s guess. As Knight and co-authors point out, feathers might be lost when a plumage-covered theropod preens or molts, when a carnivore catches or scavenges a feathery snack, or when the bloated body of the organism floats in water and gradually drops body parts due to decay. There is nothing about these fossils that allow researchers to distinguish between these possibilities.

Not all of the feathers were entirely intact. Some were complete, while others appear to have been damaged prior to burial. But it’s the more intricate aspects of these feathers that make them noteworthy. In many of the specimens, the goldish mineral pyrite has entirely replaced parts of the feathers and, as a result, the feathers retained their three-dimensional structure. Knight and co-authors also hypothesize that the carbonized residue from the breakdown of the original feather material also aided in the preservation process.

This fine level of detail in the feathers allowed Knight and collaborators to zoom in on the feathers to see if any indications of fossil colors might be found. During the past few years paleontologists and bird specialists have discovered that many fossil feathers preserve the shape of tiny organelles called melanosomes. In living birds – and other organisms – these tiny, tiny structures contain melanin and impart color to things such as feathers. Structure corresponds to color, and when melanosomes are preserved in fossil feathers they can be compared to their counterparts in modern birds to reconstruct colors.

Previously, the elongated melanosomes were thought to be fossilized bacteria. Knight and co-authors bring up the point that distinguishing between true melanosomes and similarly-shaped bacteria is still a difficult task, but conclude that the morphology and alignment of the tiny structures in the feathers they examined more closely resemble the melanin-carrying bodies. Frustratingly, though, Knight and co-authors do not divulge much detail about the color of each feather. They simply say that the feathers in their sample had colors “ranging from gray and brownish gray to black”, and that the feathers seemed to be just one color rather than patterned.

But what animals did the feathers come from? Numerous fluffy and fuzzy non-avian dinosaurs found in China have shown that feathers were a widespread trait among coelurosaurian theropod dinosaurs – the group which contains the famous tyrannosaurs, sickle-clawed deinonychosaurs, bizarre therizinosaurs, ostrich-like ornithomimosaurs, and birds, among other lineages. Feathers were a trait birds inherited from their non-avian dinosaur ancestors.

Both birds and non-avian coelurosaurs were around in Cretaceous Alabama. Knight and collaborators propose that multiple species are represented in the feather sample, but knowing which feather corresponds to which animal is tough. The problem is made even worse by the fact that many of the potential candidates from these estuary and nearshore environments are known from a paltry collection of remains. What happened to the animals between death and burial played a large role – many of the known fossils are partial skeletons that gradually fell apart as their gas-filled bodies floated in waterways on their way to the coast. “Bloat and float” is the bane of paleontologists working on land-dwelling animals known only from bodies deposited in the prehistoric coastal shallows.

Still, the new paper presents a tentative list of animals the feathers might have belonged to. The most obvious candidate for many of the feathers is Ichthyornis – a toothed bird found in the area and often reconstructed as the prehistoric equivalent to a seagull. Another possibility is an archaic bird named Halimornis, although this bird might have been washed out from a more inland location. Based upon the type of feathers and the presumed plumage of prehistoric birds, Ichthyornis and Halimornis could have carried at least thirteen of the fourteen feathers. The outlier – a specimen said to be a large tail feather – might indicate the presence of the toothed, loon-like bird Hesperornis, but there are some non-avian alternatives.

A few bits and pieces of non-avian coelurosaurs indicate that they existed in the right place and during the right time frame to make them reasonable candidates for the owners of the feathers. Knight and co-authors suggest that the large tail feather could have belonged to one of the sickle-clawed dromaeosaurid dinosaurs – think Deinonychus or Velociraptor – and some of the other feathers might have been left by an ornithomimid dinosaur. No ornithomimid dinosaur has yet been found with preserved feathers, but, given that they were coelurosaurs and every other coelurosaurian lineage had feather-covered representatives, the hypothesis cannot be discounted. Strangely, though, one of the best-known coelurosaurs from prehistoric Alabama may be the worst candidate for the feathers. Appalachiosaurus was a tyrannosauroid (incorrectly identified as a tyrannosaurid in the paper), and the only tyrannosauroid found with plumage so far – Dilong paradoxus – had simple, wispy protofeathers. Appalachiosaurus may have had feathers, but, based upon Dilong and a lack of additional information, tyrannosauroids are thought to have only retained simple coatings of dinofuzz unlike the feathers found at the site in question.

Intricately-preserved feathers like those found in the Alabama clay and Canadian amber are wonderfully frustrating fossils. They are delicate remnants of rarely-preserved tissues that may allow us to better understand the palette of some Cretaceous dinosaurs, but attributing them to specific animals is extremely difficult. Without knowing the full feather covering of avian or non-avian theropods from hatching to adulthood, how can we appropriately match feathers to their owners? We may never be able to do so. Even though these ancient feathers have added a little more to our understanding of life in Cretaceous Alabama, they have also brought attention to new mysteries that will require future fossil discoveries to resolve.

Top Image: Laughing gulls along the shore of an estuary in Cape May, New Jersey. These avian dinosaurs leave their feathers in this environment, just as their avian and non-avian Cretaceous relatives did among the wetlands of prehistoric Alabama. Photo by the author.

References:

Carr, T., Williamson, T., & Schwimmer, D. (2005). A new genus and species of tyrannosauroid from the Late Cretaceous (Middle Campanian) Demopolis Formation of Alabama Journal of Vertebrate Paleontology, 25 (1), 119-143 DOI: 10.1671/0272-4634(2005)025[0119:ANGASO]2.0.CO;2

KNIGHT, T., BINGHAM, P., LEWIS, R., & SAVRDA, C. (2011). FEATHERS OF THE INGERSOLL SHALE, EUTAW FORMATION (UPPER CRETACEOUS), EASTERN ALABAMA: THE LARGEST COLLECTION OF FEATHERS FROM NORTH AMERICAN MESOZOIC ROCKS PALAIOS, 26 (6), 364-376 DOI: 10.2110/palo.2010.p10-091r

Li, Q., Gao, K., Vinther, J., Shawkey, M., Clarke, J., D’Alba, L., Meng, Q., Briggs, D., & Prum, R. (2010). Plumage Color Patterns of an Extinct Dinosaur Science, 327 (5971), 1369-1372 DOI: 10.1126/science.1186290

[My thanks to University of Utah paleontology graduate student Joshua Lively for bringing this paper to my attention during this week’s paleobiology seminar.]