The smallest of the largest: a new look at possible dwarfing in sauropod dinosaurs

Franz Baron Nopcsa recognised that the Late Cretaceous titanosaurid sauropod from Transylvania, Magyarosaurus dacus, was much smaller than the better known sauropods from the Upper Jurassic of the United States and even from Late Cretaceous deposits elsewhere in the world. In keeping with his biogeographic interpretations, Nopcsa viewed this difference as a consequence of body-size effects via island habitation. We present a preliminary re-evaluation of Nopcsa's claims and their heterochronic consequences using two approaches, viz. (1) regression analysis of humeral data as a means of establishing patterns in body size among titanosaurids, and (2) optimisation of humeral data onto titanosaurid cladograms to evaluate evolutionary trends within the clade. Our regression analysis is based on twenty species distributed among fourteen genera of neosauropods and uses length and mid-shaft mediolateral width of the humerus of presumed fully adult forms, and of growth samples that consist of postnatal to adult individuals. Linear regression analysis suggests that, among adult neosauropods, M. dacus appears to be represented by the smallest individuals; that M. dacus humeri appear to be more similar to those of subadults than to adults of other taxa; and that this juvenile morphology may constitute dwarfing in M. dacus by paedomorphosis. In order for these regressions to reflect their evolutionary context more fully, we also present optimisation analyses of humeral form within Titanosauroidea. Although many aspects of the phylogeny of this clade are relatively poorly resolved, preliminary results of our analyses are consistent with the conclusion that M. dacus was a heterochronic dwarf.