A phylogenetic analysis of Diplodocoidea (Saurischia: Sauropoda) JOHN A. WHITLOCK* Museum of Paleontology and Department of Geological Sciences, University of Michigan, 1109 Geddes Avenue, Ann Arbor, Michigan 48109-1079, USA Received 25 November 2009; revised 4 March 2010; accepted for publication 5 March 2010 Diplodocoidea includes some of the first well-known sauropod dinosaurs, including such late 19 th century and early 20 th century discoveries as Apatosaurus, Diplodocus, and Dicraeosaurus. As a consequence of their long history of study, the basic set of suprageneric diplodocoid interrelationships is well resolved, and the diagnostic features of each genus are well established. However, intergeneric relationships are less resolved, including the relationships of putatively basal taxa like Amphicoelias and Haplocanthosaurus, the flagellicaudatan Suuwassea, and the highly specialized rebbachisaurids. For the rebbachisaurids, this uncertainty is coupled with a recent surge in the discovery of new taxa. Comparative cladistic methods demonstrate that character and taxon sampling need to be improved before greater phylogenetic resolution can be expected. Here, I present a new phylogenetic analysis that resolves many of the outstanding questions regarding the relationships within Diplodocoidea and examines palaeobiogeographical trends within the group. Suuwassea is recovered as a basal dicraeosaurid (the only Laurasian member of the group), and two distinct clades of rebbachisaurids are identified: a group closely allied with Nigersaurus and a clade associated with Limaysaurus. Amphicoelias, Amazonsaurus, and Haplocanthosaurus are provisionally placed as successively less-derived taxa at the base of Diplodocoidea. A North American origin for Diplodocoidea and Flagellicaudata is hypothesized based on the geographical and temporal distribution of those taxa. Rebbachisaurid taxa demonstrate a South American/African vicariance pattern, but the timing of the event pre-dates the proposed final rifting of those continents by c. 40 million years; the meaning of this discrepancy is uncertain. © 2011 The Linnean Society of London, Zoological Journal of the Linnean Society, 2011, 161, 872–915. doi: 10.1111/j.1096-3642.2010.00665.x ADDITIONAL KEYWORDS: character sampling – comparative cladistics – data matrix – data recycling – palaeobiogeography. INTRODUCTION Diplodocoid sauropods such as Diplodocus and Apa- tosaurus, with their long necks and tails and immense size, are amongst the most iconic and easily recognized dinosaurs (Fig. 1G), with a rich palaeon- tological history dating back into the late 19 th century. In particular, diplodocoids have been identified by their unique cranial anatomy: an antorbitally elon- gate skull with a broad, variably rectangular muzzle, filled only at its distal extreme with long, narrow- crowned teeth (Fig. 1A, B). This morphology varies both taxonomically and ontogenetically; the narrow- est snouts are known from juveniles (Whitlock, Wilson & Lamanna, 2010), although adult variation also occurs, from the round-snouted Dicraeosaurus to the extremely square snout of Nigersaurus, with its nearly linear, orthogonally orientated tooth row (Sereno et al., 1999). This variation in shape has been suggested to correspond with specialized feeding behaviour (Calvo, 1994), including low-browsing (Barrett & Upchurch, 1994, 2005; Christiansen, 2000; Upchurch & Barrett, 2000; Sereno et al., 2007; Whit- lock et al., 2010). This hypothesis is supported by *E-mail: jawhitl@umich.edu Zoological Journal of the Linnean Society, 2011, 161, 872–915. With 9 figures © 2011 The Linnean Society of London, Zoological Journal of the Linnean Society, 2011, 161, 872–915 872