Back

Three-dimensionally preserved 'Stage IIIb' fossil down feather supports developmental modularity in feather evolution

Roy, A.; Miller, C. V.; Pittman, M.; Kaye, T. G.; Peretti, A.

2020-08-27 paleontology
10.1101/2020.08.26.268060 bioRxiv
Show abstract

We describe a unique three-dimensionally preserved fossil down feather from the Late Cretaceous of Myanmar. Its morphology is highly congruent with Stage IIIb of the widely accepted Prum and Brush model of feather evolution-development. This makes the new specimen the first evidence of this developmental stage in the fossil record. The Stage IIIb diagnosis is robustly supported by the absence of a central rachis and by its paired barbules emanating from radially positioned barbs that are attached to a short calamus. Prum and Brushs model hypothesises a bifurcation in the evolution-development pathway at Stage III. Stage IIIa involves rachis development and branching into barbs. Stage IIIb involves branching of the barbs from the calamus and then further branching of the barbules from the barbs. These two pathways then converge into Stage IIIa+b where feathers produce a rachis, barbs and barbules in nested order, finally leading to Stage IV. Evolution-development studies on the morphogenesis of feathers have unequivocally shown that such feather branching can be controlled by BMP, Noggin, Shh and several other proteins. Therefore, molecular crosstalk can convert a barb into a rachis and vice versa. The topology of this down feather, consistent with specific patterns of modular protein-protein signalling already observed, provides the first definitive evidence that such signalling was responsible for the evolution of a diverse inventory of feather morphologies in non-avialan dinosaurs and early birds since the middle Jurassic.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.