Science Literature

In the past few years, studies of fossil feathers have yielded some surprising and unexpected results. How do graphic artists know what colours to use when illustrating extinct birds? The answer has been: we don't know - we use artistic licence. The new research has not yet provided different answers to this question, but methodologies are in place which should allow more definitive statements to be made in the future.

This Middle Eocene fossil feather had iridescent colours (source here)

Pigment colours are derived from melanin. These are produced within organelles called melanosomes that have a shape not dissimilar to bacteria. Since researchers were expecting to find bacteria rather than melanosomes, bacteria are what they reported - until last year.

"Melanins are synthesized in a special class of pigment cells (melanocytes) and are packaged within organelles (melanosomes) that vary in morphology between tissues, colours and organisms; the chemically inert structure is not fully understood. Wuttke (1983) interpreted the aligned oblate bodies that constitute the fossil feathers of the Messel Oil Shale as lithified bacteria. This interpretation was extrapolated to fossil feathers from other localities. Our recent investigation of a colour-banded feather from the Cretaceous Crato Formation of Brazil led us to interpret the oblate bodies as eumelanosomes, which contain black melanin (Vinther et al. 2008)."

Examples like these demonstrate the importance of multiple working hypotheses in science to avoid the danger of confirmation bias. If there is only one hypothesis on the table, there is a strong temptation to 'fit' the data to the hypothesis (rather than test the hypothesis using data). An example of this phenomenon was pointed out here. These considerations are significant for the application of design thinking within science. At very least, ID provides information-based alternatives to naturalistic hypotheses to account for biological complexity - as is discussed below.

The melanosomes have implications for plumage colouration. Modern-day examples show an association with black, brown, red and buff colours. The new research reports the "laminar nanoscale organization of melanosomes in feather barbules" that is associated in modern birds with iridescence. The melanosome layer lies underneath a very thin covering of keratin.

"These nanostructures produce a structural colour by interference among light waves scattered by the keratin layer and the underlying layer of melanosomes. [. . .] Feathers with this type of colour-producing nanostructure generally appear black with a glossy or oily iridescent sheen. Depending on the thickness of the keratin layer (from approx. 100 to over 300 nm), the iridescent colour varies in reflectance from saturated ultraviolet or blue to an oily appearance. The most distinctive feature of these nanostructures is the highly uniform superficial layer of closely packed melanosomes. The melanosomes may be rod shaped as in some passerines, galliform birds and the fossils described here or flattened as seen in some ducks and swifts."

The "dense external layer of melanosomes" in the fossil feather is now perceived as diagnostic of structural colour. The original hue of the feather cannot be determined because the outer keratin layer has been degraded. However, coauthor Julia Clark says that the reported research is just "proof of concept" and many more insights into colouration lie ahead.

The authors think that "[t]his type of thin-film nanostructure has evolved numerous times in passeriform and non-passeriform birds". The appearance of this phenomenon in diverse groups points, in their minds, to evolutionary convergence. However, it could be argued (as it is here) that structural colour has a high level of complex specified information.

Some types of colour may point to irreducible complexity. If so, it would seem desirable to adopt a multiple working hypotheses approach and consider design explanations alongside convergence. It is not without significance that the fossil feather documented in this paper is from the Messel Shale (considered to be 40 Ma). It demonstrates the kind of stasis we would expect from a complex specified system. The pattern of abrupt appearance of mature structures is not unusual and the implication is that complexity emerges rapidly (not gradually). The claim for convergence should be understood as an inference from theory, as yet unvalidated by testing.

Inevitably, evolutionary theorists have sought to relate this work to dinosaur 'feathers' and to the 'consensus' view that birds evolved from feathered dinosaurs. Zimmer comments on the ambitions of palaeontologists to extend the methodology to dinosaur feathers. He might have said that such detailed work could reveal fundamental differences between dino-feathers and bird feathers, but he did not. He might have acknowledged the controversial aspects of the dino-bird scenario (see here) but he did not. Rather, he wrote: "It is possible that dinosaurs evolved these colors before they evolved the ability to fly". I see in this an indication that not only the origin of feathers, but also the origin of structural colour, needs to be pushed back earlier and earlier, in order to allow time for such complexity to originate by natural processes.

Structural coloration in a fossil feather
Jakob Vinther, Derek E. G. Briggs, Julia Clarke, Gerald Mayr and Richard O. Prum
Biology Letters, published online before print August 26, 2009, doi:10.1098/rsbl.2009.0524

Abstract: Investigation of feathers from the famous Middle Eocene Messel Oil Shale near Darmstadt, Germany shows that they are preserved as arrays of fossilized melanosomes, the surrounding beta-keratin having degraded. The majority of feathers are preserved as aligned rod-shaped eumelanosomes. In some, however, the barbules of the open pennaceous, distal portion of the feather vane are preserved as a continuous external layer of closely packed melanosomes enclosing loosely aligned melanosomes. This arrangement is similar to the single thin-film nanostructure that generates an iridescent, structurally coloured sheen on the surface of black feathers in many lineages of living birds. This is, to our knowledge, the first evidence of preservation of a colour-producing nanostructure in a fossil feather and confirms the potential for determining colour differences in ancient birds and other dinosaurs.

See also:

Ancient Bird's Feathers Had Iridescent Glow, livescience.com (August 26 2009)

Zimmer, C. First Trace of Color Found in Fossil Bird Feathers, New York Times (August 31, 2009)