Gregory S. Paul. 2002. The Johns Hopkins University Press, Baltimore and London. ix + 460 pp., ISBN 0-8018-6763-0. Hardcover, $49.95.—Among the spate of recent books on the supposed origin of birds from theropod dinosaurs is that of artist and freelance “dinosaurologist” Gregory Paul, whose past work includes the controversial Predatory Dinosaurs of the World (1988), which featured a good dose of “feathered, hot-blooded dinosaurs” presaging their prevalence in the popular press over the past few years. This latest and equally controversial treatise argues that certain of the birdlike theropods are actually secondarily flightless birds, possibly closer to modern birds than is Archaeopteryx. Paul’s attempted massive documentation includes 460 pages, with six appendices and dozens of stylized “Paulian” illustrations.

Paul is a creative artist, and his illustrations of dinosaurs represent his personalized interpretation of their anatomies and lifestyles. For example, in Predatory Dinosaurs of the World he illustrated the giant sauropod Mamenchisaurus reared up on its hindlimbs, a feat almost inconceivable for that creature given its size and lack of complex epiphyses on its long bones. Although Predatory Dinosaurs, his first book, was published in 1988, before the discovery of so-called feathered dinosaurs, his illustrations depicted theropods from the late Triassic coelurosaur Coelophysis to the late Cretaceous Velociraptor adorned with feathers. Even in the late Triassic, the 235-my-old protodinosaur Lagosuchus sports feathers on the back and head. Speaking of the small-sized Lagosuchus, Paul notes, “This size squeeze probably marked the evolution of a fully avian–mammalian physiology” (p. 240). The implication, of course, is that all dinosaurs were fully endothermic and feathered. And, that same general theme carries over into his current magnum opus.

Not surprisingly, Paul finds no evidence for avian cranial kinesis or birdlike feathers in the skull of Archaeopteryx, contrary to detailed work of many careful workers, and he treats Caudipteryx simply as a ”feathered dinosaur,” despite conclusive evidence that it is a secondarily flightless bird, a Mesozoic kiwi (Jones et al. 2000). He denies the presence of feather-like appendages in the late Triassic Longisquama (now certain), and readily accepts evidence for a furlike pelage on pterosaurs, generally not accepted by paleontologists. He accepts the “short, fibrous material” preserved on the early Cretaceous compsognathid Sinosauropteryx as “another source of evidence for dinosaur feathers” (p. 66), but treats the ornithomimid dinosaurs known as “alvarezsaurids” as birds (p. 266). On page 119, there is a restoration of a lemur-like Ornitholestes adorned with a fuzz-like protofeather covering. Yet there is still no evidence linking the dino-fuzz preserved in Chinese Cretaceous deposits to feathers.

Birds are considered dromaeosaur derivatives, and Archaeopteryx is illustrated as a terrestrial creature, hyperextending its second toes. Yet Archaeopteryx did not have a dromaeosaur-like hypertrophied second sickle claw. If dromaeosaurs can be unequivocally identified by salient, derived features such as extremely elongate prezygapophyses and chevrons and specialized pedal digit II (Xu et al. 2003), then one must ask just how Archaeopteryx fits into the scheme, with typical Mesozoic “avian” teeth, no specialized digit II, and certainly no elongate prezygapophyses. How can one imagine such a creature being derived from a dromaeosaurid ancestor defined by these salient features? Yet despite a profile nearly identical in outline (including wing shape) to that of a coucal or magpie (Feduccia 2001), Paul can state,

However, Paul’s main theme is “Were some dinosaurs also neoflightless birds?” (chapter 11), and he argues that dromaeosaurs, troodonts, oviraptorosaurs, therizinosaurs, and ornithomimids all have characters expected in neoflightless birds. His arguments, however, are narrative and he rejects cladistics as a methodology. “Cladistics will be a useful but limited tool for further investigating the neoflightless dinosaur hypothesis” (p. 254), and “The limits of cladistic methodology mean that no matter what the results of contemporary cladistic analyses or investigations of flight heritage are, in the end, only an improved set of transitional fossils will reveal the actual situation” (p. 255). Paul summarizes, “Protarchaeopteryx, dromaeosaurs, troodonts, caudipterygians, oviraptosaurs, and therizinosaurs provide telling evidence of having descended from fliers more advanced than Archaeopteryx” (p. 257).

Much of the remainder of the book is dedicated to phylogenies, the Mesozoic, extinctions, and the Cenozoic. That discussion is followed by a series of appendices with character lists and other associated information on anatomy, locomotion, etc. The discussion is full of convoluted statements that are difficult to grasp, such as “the global archosaur fauna was not healthy,” and “A nonlinear, chaotic response to these environmental perturbations may have exaggerated what should have been survivable events…” (p. 301). In the discussion on the Cenozoic, Paul states that “The skull of Diatryma appears similar to that of the terror bird Phorusrhacos… but is even more similar to the shoebill stork… and that of the possibly herbivorous dromornithid Bullockornis…” (p. 308). First, the skulls of those birds are totally and dramatically dissimilar, and second, does this mean that the shoebill (Balaeniceps) is extinct? Such statements do not lend much confidence to Paul’s anatomical comparisons of archosaurs and theories of origins based on paleontological morphology.

Despite its faults and the perception of a Disney-like fantasia produced by his freelance creative artwork, Paul’s thesis, that certain birdlike terrestrial dinosaurs may be secondarily flightless birds deserves attention, and recent analyses provide evidence that Caudipteryx and oviraptosaurids may well be birds (Jones et al. 2000, Maryanska et al. 2002). Too, S. Czerkas (2002) has theorized that all the Dromaeosauridae may be a lineage of secondarily flightless birds, but derived from a predinosaur, a basal archosaur, not part of the theropod assemblage. In addition, the recently described four-winged dinosaur from China (Xu et al. 2003) appears to be much more birdlike than dromaeosaurlike, and its supposedly diagnostic dromaeosaur tail (also like that of a ramphorhynchoid pterosaur) and claws bear little close resemblance to those of the typical dromaeosaurs such as Deinonychus and Velociraptor. Are these early Cretaceous Chinese fossils actually remnants of the early avian radiation, which also produced secondarily flightless forms such as Caudipteryx? We must keep an open mind to bizarre possibilities, and certainly Paul’s book will encourage all interested in this topic to begin thinking out of the box.