Commentary 1 , 2 , 3 5 2 , 3 , 4 : Accommodating the cladogram

Although most fields of science are constantly struggling with which methodologies to use, the field of systematics, and especially paleontology, has adopted phylogenetic systematics (cladistic methodology) to the exclusion of other approaches. Despite a barrage of cautions and criticism, cladistics reigns. Considered a triumph of the field of paleontology and the crown jewel of cladistic methodology is the proof of the dinosaurian origin of birds. Among the most important features thought to link theropod dinosaurs and birds is a hand reduced to three fingers, a tridactyl hand. We know that the grasping raking hand of theropod dinosaurs is composed of digits 1-2-3, that is, the thumb and the next two fingers, because a variety of late Triassic forms, show digits 4 and 5 having undergone dramatic reduction and remaining only in vestigial form (Fig. 1). Yet virtually all credible developmental studies have concluded that the bird hand is composed of digits 2-3-4. Here, in a provocative theoretical paper, Wagner and Gauthier (1) accept the developmental evidence that birds have a 2-3-4 hand but propose a frame-shift hypothesis, by which the developmental properties responsible for digits (D)1–D3 are shifted onto embryonic precartilaginous condensations (C)2–C4. To put it another way, although these authors accept that the homology of the primordial condensations are correctly identified as C2, C3, and C4, they propose that subsequent anatomical differentiation reflects a frame shift in digit primoidia (anlagen) in later ontogeny such that C2 becomes D1, C3 becomes D2, and C4 becomes D3. By this proposed shift, birds can still be nested in the same clade or monophyletic grouping as maniraptoran theropod dinosaurs (the putative bird ancestors), as concluded by cladistic hypotheses, which date back to a paper by Gauthier (2), generally conceded to be the pivotal paper for this theory. As Neil Shubin (ref. 3, p. 263) notes, the phylogenetic interpretation of Gauthier (2) unambiguously supports the view of nonavian theropod digital reduction. The extension of this view to birds involves homoplasy (convergence) either in developmental patterns (in the I-II-III interpretation) or in numerous other skeletal characters (in the II-III-IV interpretation). But all recent developmental biologists have concluded that birds have a 2-3-4 hand (4–6). The developmental evidence overwhelmingly suppports the 2-3-4 theory of the wing skeleton in birds (ref. 6; Fig. 2). Too, phalangeal formulae, used to ally birds and theropods, are highly variant among amniotes. They are also developmentally plastic, as indicated by Zho and Niswander (7), who studied a bone morphogenetic protein that mediates programmed cell death (apoptosis) and showed that blockage of this factor in the avian limb results in hands missing only the most distal phalanges, thus providing a mechanism for symmetrical phalangeal reduction. Some obvious objections to the frame-shift hypothesis are that (i) there is no evidence for any substantial morphological change in theropod hands that would indicate any kind of shift throughout their evolution; (ii) the similarity between the hands of theropods and Archaeopteryx (which is quite distinctive) are overemphasized in drawings, and the semilunate wrist bone, considered a definitive bird–dinosaur link, is thought by many not to be homologous (8); and (iii) in bird development, the foreand hindlimbs exhibit the same highly conserved developmental pattern (ref. 5; Fig. 3), so if there is a frame shift, it would have to occur in the forelimb but not in the hindlimb. Although a common stem ancestor for birds and dinosaurs (and an arboreal origin of flight) was accepted for most of the 20th century, in the early 1970s Yale University’s John Ostrom resurrected Huxley’s dinosaurian origin of birds hypothesis (and the ground-up or cursorial origin of flight) based on Ostrom’s discovery of the Lower Cretaceous superficially birdlike Deinonychus and the proposal that such dinosaurs were hot-blooded or endothermic. It is interesting to note that the modern version of the dinosaurian origin of birds originally had nothing to do with cladistic methodology. It sprang from overall similarity (phenetics) and an evolutionary scenario by which hot-blooded dinosaurs became clothed with feathers for insulation and somehow sprouted wing feathers, and flight originated from the ground up—the cursorial theory for avian flight. For many of today’s paleontologists, birds are simply living dinosaurs. “The smallest dinosaur is the bee hummingbird . . . found only in Cuba” (ref. 9, p. 25). Despite the popularity of the dinosaurian origin of birds, many ornithologists and physiologists, in particular, have had tremendous difficulty with the theory (8, 10, 11) because of a huge and growing body of contrary evidence and the fact that a ground-up origin of avian flight is considered a near biophysical impossibility (12). Aside from criticism concerning the cursorial origin of avian flight, there are problems related to the geologic, temporal occurrence of putative dinosaurian ancestors, which occur some 30 to 80 million years after the appearance of the earliest known bird Archaeopteryx, and these forms become more and more superficially birdlike as one approaches the latest Cretaceous. There is also the fact that virtually all of the anatomical features used to ally birds and dinosaurs have been disputed. There have been a number of embarrassing cladistic traps, the most recent relating to pterosaur phylogenetics. A cladistic analysis performed by Kevin Padian in 1984 (13) indicated that pterosaurs were a sister group of dinosaurs and therefore must have evolved from small active bipedal terrestrial predecessors from the ground up (14). Padian’s studies relied on the basal pterosaur Dimorphodon, interpreting it as digitigrade and theropodlike, with the hindlimbs obligately held in an upright bipedal posture. These assertions have recently been shown to be erroneous. Analysis of the rhamphorhynchoid Sordes revealed the presence of a membrane that extended between the hindlimbs, negating any erect bipedal posture (15), and the hairlike fibers found on the specimens were supportive elements (rods) of the membranes, not fur. More recently, the discovery of the threedimensionally preserved articulated foot of the basal pterosaur Dimorphodon shows a flat-footed stance and confirms obligate quadupedality and a plantigrade stance as primitive features for the group (16). Quadrupedality in numerous pterosaurs has also been confirmed by the discovery of myriad trackways, none of which show any signs of bipedality. Interestingly, H. G. Seeley, who opposed Huxley’s dinosaurian origin of birds, correctly