REPORTS petition between major clades are often vague, Superiority, Competition, difficult to test conclusively, and prone to over- simplification (4). Rather than focusing on such and Opportunism in the Evolutionary imprecise terms, it is illuminating to examine macroevolutionary patterns. Here, we compared evolutionary rates and relative morphospace oc- Radiation of Dinosaurs cupation in dinosaurs and crurotarsans, in an effort to shed light on their evolutionary dynam- † Stephen L. Brusatte,* Michael J. Benton, Marcello Ruta, Graeme T. Lloyd ics and to assess long-standing perceptions such as “superiority.” The rise and diversification of the dinosaurs in the Late Triassic, from 230 to 200 million We used a new phylogeny of Triassic archo- years ago, is a classic example of an evolutionary radiation with supposed competitive saurs (Fig. 1) and a data set consisting of 64 replacement. A comparison of evolutionary rates and morphological disparity of basal dinosaurs taxa and 437 discrete skeletal characters (13)to “ ” and their chief competitors, the crurotarsan archosaurs, shows that dinosaurs exhibited lower calculate numerical measures of evolutionary rates disparity and an indistinguishable rate of character evolution. The radiation of Triassic archosaurs [patristic dissimilarity per branch and patristic as a whole is characterized by declining evolutionary rates and increasing disparity, suggesting a dissimilarity divided by branch duration (14)] and decoupling of character evolution from body plan variety. The results strongly suggest that disparity (morphospace occupation) (13). Note “ ” historical contingency, rather than prolonged competition or general superiority, was the primary that evolutionary rates analysis approximates factor in the rise of dinosaurs. the amount of morphologic evolution separat- ing species, whereas disparity analysis approx- he rise of the dinosaurs in the Late Triassic (after the CNEE) followed by larger theropods imates the amount of morphologic difference and Early Jurassic (230 to 190 million years and armored herbivore groups in the Early Ju- between species (14–17). These are related but Tago) is a classic example of an evolutionary rassic [after extinction of carnivorous crurotarsans separate measures of morphological evolution radiation. During that time, the clade Dinosauria at or near the Triassic-Jurassic boundary (TJEE)]. that together give insights into patterns of mac- expanded from a single lineage to many dozens of This two-step model has been supported by re- roevolutionary change within and between clades. lineages and from one ecological and morphologi- cent studies of theropods, which became larger Disparity analysis does not depend on a specific cal type to many, and the range of body sizes ex- andmorecommonaftertheTJEE(7), and or- phylogenetic hypothesis, but evolutionary rates panded to include truly gigantic forms (1, 2). nithischians, which are now known to have been analysis does. Through this expansion in diversity and disparity, rare in the Late Triassic after the reassignment There is no clear evidence for differences in dinosaurs became the preeminent vertebrates on of many supposed ornithischian fossils to non- overall evolutionary rates between dinosaurs and land, occupying many ecological roles—especially dinosaurian groups (9). crurotarsans during the Triassic as a whole. Dino- those at medium to large size—in terrestrial eco- The critical interval to consider is the Late saurs exhibit higher mean rates than crurotarsans systems worldwide. Triassic, especially the Norian and Rhaetian for all measures (Fig. 2, A and B, and fig. S2, A The expansion of Dinosauria has long been (Fig. 1), a 28-million-year span between the and B)—as does the entire dinosaur “total group,” seen as an example of a “competitive adaptive CNEE and TJEE. The key “competitors” of Ornithodira (sister taxon to Crurotarsi)—but these radiation” in which one group supplants another the early dinosaurs were the crurotarsans, the differences are generally not significant (tables (3, 4). The dinosaurs were said to have outcom- “crocodile-line” archosaurs, which show a range S1 and S28). A pruned analysis of equal sample peted other terrestrial tetrapods (notably basal of morphologies and adaptations during this sizes for the two clades returns the same result archosaurs, rhynchosaurs, and nonmammalian time: long-snouted fish- and flesh-eating phyto- (table S2), as does an analysis restricted to Norian synapsids) by virtue of their upright or erect saurs, armored herbivorous aetosaurs, and large to taxa (table S3). There is limited evidence for sig- posture, which gave them advantages of speed giant carnivorous “rauisuchians.” The crurotarsans nificantly higher rates in Carnian dinosaurs, but this and maneuverability (5), or because they were even replicated many dinosaurian body plans may be due to small sample size (table S3). Tem- endothermic (possessing fully warm-blooded (large terrestrial predators; small swift predators; poral trends do not show a coupled increase in physiology) (6). The alternative, opportunistic mid- to large-bodied low-browsing herbivores; dinosaur rates and decrease in crurotarsan rates, as model (3) proposes that dinosaurs diversified agile bipedal herbivores). Several new discoveries might be expected under some models of “com- in the Norian, after a Carnian-Norian extinction show striking convergences between crurotarsans petition” (Fig.2,EandF,andfig.S2,EandF). event (CNEE) 228 million years ago that saw and dinosaurs (10), and many Triassic crurotarsans Relative to crurotarsans, dinosaurs exhibit a signif- the demise of rhynchosaurs, dicynodonts, and were previously erroneously identified as dino- icantly higher rate of evolution of the appendicular chiniquodontids; dinosaurian clades were then saur ancestors (11) or even as true dinosaurs (12). skeleton, but not of the cranial or axial skeleton added through the Late Triassic and Early Jurassic Such morphological convergence suggests that di- (table S9). However, there are no significant dif- until they reached their full diversity. nosaurs and crurotarsans were exploiting similar ferences between rates for different regions of the Most previous studies have treated the rise resources in the Late Triassic. In some Norian dinosaur skeleton (tables S16 and S17). of the dinosaurs as a single event, whether com- faunas, crurotarsans were numerically more abun- Perhaps counterintuitively, the disparity study petitive or opportunistic (3, 5–8). However, phy- dant than dinosaurs (3) and seem to have ex- shows that crurotarsans occupied a larger amount logenies and diversity trends suggest that this ploited a wider range of body plans. However, of morphospace than did dinosaurs and ornitho- was a two-step process, with the diversification by the end of the Triassic all crurotarsans were dirans as a whole (Fig. 3, A and B, fig. S3, A and of herbivorous sauropodomorphs in the Norian extinct, save a few lineages of crocodylomorphs. B, and tables S21 and S29). Rarefaction curves The key question is why the major dino- show that these results are not biased by sample Department of Earth Sciences, University of Bristol, Bristol saur lineages survived the TJEE, ushering in size (fig. S4). The same pattern holds within the BS8 1RJ, UK. the 135-million-year “age of dinosaurs,” while Carnian and Norian (table S22), and there are no *Present address: Division of Paleontology, American most crurotarsan groups went extinct. One com- coupled temporal trends (Fig. 3, E and F, and fig. Museum of Natural History, New York, NY 10024, USA, mon explanation is that dinosaurs outcompeted S3, E and F). Dinosaurs and crurotarsans occupy and Department of Earth and Environmental Sciences, Columbia University, Palisades, NY 10964, USA. crurotarsans in the Late Triassic, and notions of adjacent areas of morphospace (Fig. 1), which is †To whom correspondence should be addressed. E-mail: general dinosaurian “superiority” have long per- expected because the analysis is based on cladis- [email protected] vaded the literature (5, 6). Hypotheses of com- tic characters. www.sciencemag.org SCIENCE VOL 321 12 SEPTEMBER 2008 1485 REPORTS Crurotarsans convergent with dinosaurs (popo- wide range of organisms, time scales, and clade logically selective (26),andsotherelativepro- sauroids, “rauisuchids,” and basal crocodylo- dimensions. portions (or relative success) of two groups dur- morphs) occupy an intermediate area between For the first 30 million years of their history, ing normal times may reverse during a sudden the majority of crurotarsans and dinosaurs. The dinosaurs lived alongside and shared niches with crisis. Nonetheless, our results are consistent with higher disparity of crurotarsans is borne out by another major clade (Crurotarsi) that occupied at least two explanations: (i) Crurotarsans died visual examination of Fig. 1, which shows a much more morphospace and evolved at indistinguish- out by chance, despite their larger range of mor- larger morphospace than that for dinosaurs. Un- able rates. These patterns seriously contrast with phospace and similar evolutionary rates to dino- expectedly, this larger crurotarsan morphospace general notions of dinosaurian “superiority” and saurs; (ii) dinosaurs prevailed because of one or is associated with significantly higher rates of the long-standing view that dinosaurs were pre- several key adaptations. The second suggestion homoplasy (table S18), which suggests that char- ordained for success