2.6 Evolution of Feathers

For most of the past century, the origin of birds, the origin of feathers, and the origin of avian flight have been treated as a set of interrelated questions. Traditionally, specific hypotheses about each of these questions were closely associated with specific positions on the others. Advocates of the thecodont origin of birds argued for an aerodynamic origin of feathers and an arboreal, or gliding, origin of flight. Beginning with John Ostrom, many advocates of the theropod origin of birds argued for a thermoregulatory origin of feathers and a cursorial, or running, origin of flight. However, more progress can be made on all of these questions by treating them independently and by documenting the evolutionary patterns before making strong conclusions about the evolutionary process or the mechanisms that have led to these innovations.

For more than a century, the modern feathers of Archaeopteryx separated it from all the small dinosaurs of similar form. It turns out, however, that feathers are not unique to birds but rather evolved earlier in theropod dinosaurs. This new awareness started with the discovery of the first “feathered dinosaur,” the turkey-sized Caudipteryx with a well-preserved fan of vaned feathers on its tail and forelimbs, and the slim, chicken-sized Sinosauropteryx with filamentous downlike feathers (see Figure 2–11). Fossil feathers have now been found on more than a dozen nonavian theropod dinosaurs (Prum and Brush 2002; Norell and Xu 2005).

Ancient theropod feathers included downlike filamentous structures, or “dino-fuzz,” and advanced, vaned, essentially modern feather 144 structures. The relation of dino-fuzz to real feathers has been controversial; arguments range from their being unrelated structures to being precursors of feathers to being simplified feathers of flightless birds (Prum and Brush 2002; Lingham-Soliar 2003; Feduccia et al. 2005; see Box 2–2).

Less controversial are the well-preserved vaned feathers. Small Microraptor gui had front and hind wings that sported outer feathers with asymmetrical vanes, just as in the wings of modern flying birds (Xu et al. 2003; see Figure 2–11). Feathers clearly evolved in modern form in theropod dinosaurs and then diversified in form and function. It was long presumed that feathers are so perfectly adapted for flight that they must have evolved through selection from elongate scales for this aerodynamic capacity. However, feathers evolved not as modified, mature scales but as a novel epidermal structure (Prum 1999; Prum and Brush 2002; see also Chapter 4). Contrary to most speculation during the twentieth century, feathers did not originate in concert with the evolution of flight. Rather, avian flight evolved after the origin of complex, vaned feathers. The asymmetric feather vane evolved into its fully modern form in the ancestor of enantiornithines and ornithurine birds (Feo et al. 2015). We will return to the evolutionary origin of feathers in Chapter 4.

2.7 Evolution of Flight

How did avian flight evolve, and just how well could Archaeopteryx fly? What caused the forelimbs of reptilian ancestors to evolve into protowings in the first place? Two basic theories have been proposed: an arboreal, or gliding, theory and a cursorial, or running, theory. The arboreal theory proposes that the evolution of flight started with gliding and parachuting from elevated perches. This arboreal theory was favored for many years by opponents of the theropod origin of birds (Bock 1965; Feduccia 1980). The cursorial theory proposes that elongated forelimbs enhanced leaping ability in a small, bipedal theropod dinosaur that ran and jumped to catch prey. The cursorial theory has been frequently advocated by proponents of the theropod origin of birds (Ostrom 1997; Padian and Chiappe 1998). The arboreal versus cursorial theories are not clear alternatives. They pose a false dichotomy because the activities of the avian ancestors, including Archaeopteryx itself, may well have mixed these behaviors. The most important issue in the origin of flight is the evolution of the wing stroke that could produce the main s of powered flight: lift and thrust. A powered wing stroke required transformation of the wrist and shoulder from the skeletal wing precursors of theropod or other ancestors (Ostrom 1997). What were the behavioral steps that fostered this transformation? From the phylogeny of theropods, it is clear that many of the 146 anatomical and functional precursors of the avian flight stroke evolved for prey capture in entirely terrestrial theropods with a praying mantis– like forelimb movement and grasping hands (Padian and Chiappe 1998). This pattern provides strong evidence of the terrestrial context for many evolutionary events that together facilitated the evolution avian flight. However, the biggest challenge to the cursorial theory is that the aerodynamic force of lift that makes flight possible is easier to produce at higher speeds (Chapter 5). It is easier to produce high airspeeds over the limbs by gliding down from a high perch than it is by running along the ground. In an effort to expand the adaptive value of creating lift in a terrestrial context, Ken Dial (2003a) suggested that flapping their feathered forelimbs helped early terrestrial ancestors of birds climb steep inclines, such as tree trunks, to escape predators. Chickens and their relatives routinely improve foot traction and climbing ability through wing-assisted incline running (Figure 2–17). Incipient wings could have served avian ancestors in the same way, but the wingassisted incline running of modern birds requires a modern shoulder that Archaeopteryx and other early birds did not have. Continued improvement of such aerodynamic assistance could have favored evolutionary changes in wrist and shoulder structure that led to the powered stroke of the avian wing. We will return to the evolutionary origin of flight in Chapter 5.