One Overview of Sauropod Phylogeny and Evolution Jeffrey A. Wilson SAUROPOD STUDIES FROM OWEN TO long bones” and “the toes being terminated by THE PRESENT strong claws” (Owen 1842:102), but this assess- ment was based on limited anatomical evidence This year marks the one hundred sixty-fourth (Owen 1875:27). Key data emerged with the dis- anniversary of Richard Owen’s (1841) description covery of abundant Cetiosaurus bones in of the first sauropod—Cetiosaurus, the “whale Oxfordshire by John Phillips. Thomas Huxley lizard”—on the basis of vertebrae and limb ele- examined this “splendid series of remains” ments from localities across England. Although before the publication of Phillips’ (1871) mono- these remains “had been examined by Cuvier graph and was the first to place Cetiosaurus within and pronounced to be cetaceous” (Buckland Dinosauria (Iguanodontidae [Huxley, 1869:35]). 1841:96), Owen (1841:458–459) demonstrated Phillips (1871) interpreted Cetiosaurus as a plant- the saurian affinities of Cetiosaurus on the basis eating dinosaur and hypothesized that its limb of several features, including the absence of epi- bones were “suited for walking.” He could not physes (growth plates) on caudal vertebrae (fig. rule out the possibility that it was amphibious, 1.1). He differentiated Cetiosaurus from other however, concluding that it was a “marsh-loving extinct saurians on the basis of its large size and or riverside animal.” Owen (1875:27) later acqui- characteristics of its vertebrae (see Upchurch esced, referring Cetiosaurus to the Dinosauria and Martin 2003:215). Owen (1841:462) con- because of its four sacral vertebrae. He admitted cluded his initial description with this assess- that it may have had some terrestrial capabilities ment: “The vertebræ, as well as the bones of the but concluded that Cetiosaurus was an estuarine extremities, prove its marine habits . the sur- or marine animal based on its “organ of swim- passing bulk and strength of the Cetiosaurus ming,” the tail (Owen 1875:41). were probably assigned to it with carnivorous These early interpretations, based on some- habits, that it might keep in check the what limited samples, were followed by the Crocodilians and Plesiosauri.” He regarded discovery of abundant sauropod skeletons in Cetiosaurus as a crocodilian by the “form of the western North America and eastern Africa during 15 extended across the Atlantic surrounded the posture of sauropods. American scientists favored an upright, columnar posture, whereas their German colleagues deemed a lacertilian pose more appropriate (Holland 1910; Desmond 1975). A second question, less con- troversial but farther-reaching, emerged from the study of these two large collections of sauro- pod material—How should sauropod diversity be classified? TRADITIONAL CLASSIFICATION When Marsh (1878) coined the suborder Sauropoda, it included only a single family, FIGURE 1.1. Sagittally sectioned posterior caudal vertebra Atlantosauridae. Several of the features Marsh of Cetiosaurus oxoniensis (OUM-J13697) with label in Owen’s hand. This sectioned vertebra was used to demon- (1878:412) listed in that initial diagnosis of strate the lack of epiphyses at either end of the caudal cen- Sauropoda are now well-corroborated synapo- trum. Scale equals 5 cm. morphies for the group or for more exclusive sauropod subgroups that were not identified the late nineteenth and early twentieth centuries. at the time of Marsh’s writing. Marsh invented O.C. Marsh and E.D. Cope described numerous new families to accommodate the increasing new and well represented sauropod genera sauropod diversity revealed by new discoveries from the Morrison Formation of the western worldwide (e.g., Atlantosauridae, Morosauridae, United States, including the first complete Diplodocidae, Pleurocoelidae, Titanosauridae). sauropod skull (Diplodocus [Marsh 1884]), The formal familial diagnoses for these groups reconstructions of the skeletons of Brontosaurus (Marsh 1884, 1895) also recognized features by Marsh (1883; fig. 1.2) and Camarasaurus by currently considered synapomorphies for sauro- Cope (Osborn and Mook, 1921:pl. 82; fig. 1.2), pod subclades. These diagnoses, however, did and the first mount of a complete sauropod not resolve how these groups were interrelated; skeleton (Diplodocus [Anonymous 1905]). These Marsh’s ranked classifications did not function discoveries provided the first examples of onto- as hypotheses of evolutionary descent. genetic variation and phylogenetic diversity in On the basis of his burgeoning Tendaguru sauropods. Later, German expeditions to East collection, Janensch (1929a) produced a very dif- Africa (present-day Tanzania) produced sauro- ferent classification of Sauropoda that employed pod material rivaling that from North America. higher level groupings. He recognized two prin- Janensch and others led field crews at cipal sauropod subgroups, one with broad, later- Tendaguru, where they collected more than ally facing nares and spatulate tooth crowns and 235,000 kg of fossils (Maier 2003:105) that rep- the other with elevated, dorsally facing nares and resented many new genera described over the narrow tooth crowns. Janensch named these two course of 50 years (e.g., Janensch, 1914, 1929a, families Bothrosauropodidae and Homalosauro- 1935–36, 1950, 1961). The abundance and podidae, and recognized three and four subfam- diversity of sauropod remains unearthed in ilies within each, respectively. Huene (1956) fol- North America and Africa not only answered lowed this dichotomous scheme, raising many of the queries posed by early sauropod Janensch’s subfamilies to familial rank and researchers (e.g., dinosaurian affinities and ter- Janensch’s families to “family-group” rank. In restrial habits of sauropods) but also posed new contrast to that of Marsh, Janensch’s classifica- ones. One of the major controversies that tion could be interpreted as an evolutionary 16 OVERVIEW OF SAUROPOD PHYLOGENY AND EVOLUTION FIGURE 1.2. First reconstruction of sauropod dinosaurs. Top: “Brontosaurus” (=Apatosaurus) excelsus from Marsh (1883). From a lithograph later published in Ostrom and McIntosh (1966). Bottom: Camarasaurus supremus, as drawn by Ryder in 1877 under the direction of Cope. Later published in Osborn and Mook (1921:pl. 82). hypothesis that involved divergence between two derived feature characterizing Diplodocus-like lineages differing in tooth morphology. taxa (i.e., Diplodocoidea) and titanosaurs. A dichotomous scheme for higher-level clas- Salgado et al. (1997) were the first to depart sification of sauropods based on tooth form and from this traditional dichotomy by providing narial position became widely accepted, despite character evidence linking narrow-crowned nomenclatural differences (Brachiosauridae titanosaurs to the broad-crowned Brachiosaurus, versus Titanosauridae [Romer 1956, 1966]; rather than to the other narrow-crowned group Camarasauridae versus Atlantosauridae [Steel (Diplodocoidea). This result was corroborated 1970]). Other traditional classifications of by Wilson and Sereno (1998). In a subsequent sauropods, however, follow Marsh in recogniz- analysis, Upchurch (1998) produced a topol- ing taxa of equivalent rank (usually families) ogy that agreed in many ways with those of with no higher-level hierarchical information Salgado et al. (1997) and Wilson and Sereno (e.g., McIntosh 1990). Bonaparte (1986a) also (1998) but also explored the relationships of utilized serially ranked families, but he genera not treated by either. These three analy- regarded Late Jurassic and younger sauropod ses agree on several topological points, includ- families (“Neosauropoda”) as advanced relative ing the separation of early-appearing genera to older forms (“Eosauropoda”). (e.g., Vulcanodon, Shunosaurus, Barapasaurus, Numerical methods for assessing phyloge- Omeisaurus) from a derived clade called netic relationships in sauropod dinosaurs were Neosauropoda (Bonaparte 1986a), the identifi- first introduced by Gauthier (1986) in his analy- cation of the two constituent neosauropod lin- sis of saurischian dinosaurs. His character eages Diplodocoidea (e.g., Apatosaurus) and choice reflected those cited by previous authors Macronaria (e.g., Camarasaurus), and the posi- (e.g., Romer 1956; Steel 1970) and his topology tioning of the titanosaur lineage within consequently conformed to the traditional Macronaria (fig. 1.3). dichotomy. Since then, more than a dozen Despite points of agreement, other topologi- cladistic analyses focusing on Sauropoda or its cal differences persist. The most significant of subgroups have appeared (Russell and Zheng these centers on the phylogenetic affinities of 1993; Calvo and Salgado 1995; Upchurch 1995, two groups of Asian sauropods: the Chinese 1998; Salgado et al. 1997; Wilson and Sereno “euhelopodids” (Shunosaurus,” Omeisaurus, 1998; Sanz et al. 1999; Curry 2001; Curry Mamenchisaurus, Euhelopus) and the Mongolian Rogers and Forster, 2001; Wilson 2002; Calvo nemegtosaurids (Nemegtosaurus, Quaesitosaurus). and González Riga 2003; González Riga 2003; Upchurch (1995) proposed “Euhelopodidae” as Upchurch et al. 2004). Together these analyses a clade that evolved while China was geograph- have scored 1,964 characters in 229 sauropod ically isolated from Europe from Middle taxa, resulting in a variety of phylogenetic Jurassic until Early Cretaceous times (Russell hypotheses that are discussed briefly below. 1993; Z. Luo 1999; Barrett et al. 2002; Upchurch et al. 2002; Zhou et al. 2003). It CLADISTIC HYPOTHESES