Morphometric analysis of humerus and femur shape in Morrison sauropods: implications for functional morphology and paleobiology

Paleobiology, Summer 2004 by Bonnan, Matthew F

Abstract.-Morphometric analyses of sauropod limbs have the potential to illuminate functional aspects of sauropod locomotion and paleobiology. However, analyses of sauropod limb dimensions typically show few discernible morphological trends because of large size differences among the individuals in a sample. For sauropods, combined analyses of both limb dimension and shape may be more desirable. Numerous humeri and femora from Apatosaurus, Diplodocus, and Camarasaurus provide an opportunity to explore and compare limb morphology in contemporaneous, sympatric sauropods. Thin-plate splines were used to analyze landmark-based shape differences in combination with traditional morphometrics. The aims of the analysis were (1) to determine if humerus and femur shape were significantly different among the genera; (2) to determine where shape changes occurred; and (3) to infer the basic functional implications of the shape differences using an Extant Phylogenetic Bracket approach. Few differences were detected among the genera using traditional morphometric analyses, and linear regression revealed a predominantly isometric relationship between most measurement variables and element size. Thin-plate splines revealed significant shape differences among the taxa. Apatosaurus humeri and femora were the most robust, with expanded regions for muscle insertion and more distally placed deltopectoral and caudofemoral landmarks. Diplodocus humeri and femora were gracile, with more proximally located landmarks of muscular insertion. Camarasaurus humeri were surprisingly gracile, with a less extensive deltopectoral crest, but had more robust femora similar to those of Apatosaurus. Few differences distinguished juvenile from adult specimens. These data suggest some locomotor differences were present among the three genera.

Introduction

Both qualitative and quantitative analyses of bone shape continue to play a central role in the evaluation of dinosaur evolution and paleobiology (Chapman and Weishampel 1997a,b). Appendicular synapomorphies related to locomotor adaptations form a majority of the characters on which dinosaur evolution and phylogeny have been established (Benton 1997; Sereno 1999), and the relationship between limb morphology and locomotor performance has long been explored in many saurischian and ornithischian taxa. Because locomotion is the result of a complex series of interactions among bones, their soft tissues, and a substrate, simplification of these factors for the purposes of discovering broad evolutionary or functional trends has been the goal of most morphometric studies on dinosaurs. Recognition of a link between limb proportions and general locomotor performance (Alexander and Jayes 1983; Alexander 9 199 89; Hildebrand and Goslow 2001) has spawned a proliferation of morphometric studies that use limb dimensions, limb segment lengths, and ratios to elucidate locomotor categories and trends among major dinosaur clades (e.g., Coombs 1978; Alexander 9 199 89; Thulborn 9 199 89; Lehman 1990; Weishampel and Chapman 1990; Foster 1995; Carrano 1997, 1998, 9 1999; Chapman 1997; Curtice et al. 1997; Gatesy and Middleton 1997; Farlow et al. 2000; Jones et al. 2000).

Few morphometric studies have focused on sauropod dinosaur limb elements. Analysis of sauropod limbs inherently suffers from difficulties related to size, size differences among individuals, and poor taxonomic samples, all of which may account for the general paucity of morphometric studies on these dinosaurs. Sauropod limbs have the disadvantage of being large, massive, and fragile, and this in itself poses problems for data collection (e.g., one cannot easily access all sides of a limb bone). However, interpretation of morphometric signals is by far the most problematic. For example, differences in the size of limb elements between juvenile or subadult sauropods and those of adults, or between taxa of differing size, are so great that size alone accounts for almost all variation in a particular sample. This is reflected in the results from the few previous preliminary (Foster 1995; Curtice et al. 1997; Wilhite and Curtice 1998; Wilhite 1999) and published (Christiansen 1997) morphometric analyses of sauropod limbs. These show collectively that most differences in limb dimensions among sauropod taxa and individuals are size-related with weak allometric signals. Despite these difficulties, morphometric analyses of sauropod limbs have the potential to illuminate evolutionary and functional trends. In spite of their size, the broad geographic distribution of several North American Late Jurassic taxa (e.g., Apatosaurus, Diplodocus, Camarasaurus) (McIntosh 1990; Dodson 1990; Mclntosh et al. 1997) suggests that at least some of these dinosaurs were quite vagile animals. Quantitative assessment of bone shape may augment functional interpretations of locomotion or identify morphological differences among contemporaneous taxa. Moreover, many synapomorphies rooted in appendicular skeleton morphology are used to chart sauropod evolutionary history (Upchurch 1995, 1998; Wilson and Sereno 1998; Wilson 2002).