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

Paleobiology, Summer 2004 by Bonnan, Matthew F

Paleobiological Inferences.-Morphometric data in this study are used to infer basic muscle changes associated with anatomical landmarks, or with regions between landmarks, in Apatosaurus, Diplodocus, and Camamsaurus. The successful application of any morphometric method for inferring paleobiological constraints on fossil vertebrates is ultimately tied to the interpretation of bone morphology in light of known or probable soft tissue parameters. Functional inferences reported here follow the Extant Phylogenetic Bracket method (EPB) (Witmer 1995), rooting soft tissue comparisons in an established phylogenetic framework of extant outgroup taxa. A detailed functional analysis is beyond the scope of this study, although dissection of Alligator mississippiensis and Columba livia specimens was performed to confirm the gross correlations between landmarks and soft tissue reported here (see Bonnan 2001). To simplify the present analysis, only generalized inferences of muscle function derived from Level I soft tissue inferences (Witmer 1995), or one-to-one correlations of landmarks and their soft tissues are considered. In other words, the functional inferences that are described here are based on unequivocal, shared landmark-tissue associations by the extant bracketing taxa (crocodilians, avians) and a further removed extant outgroup (lepidosaurs). For example, it is recognized that the deltopectoral crest of the humerus is a landmark correlated with deltoid and pectoral musculature in all these taxa (Romer 1956; Hildebrand and Goslow 2001). Therefore, differences in deltopectoral crest shape among the sauropod taxa are inferred to be associated with differences in deltopectoral muscle placement and distribution along this landmark.

Functional inferences drawn, from EPB associations are two levels removed from the osteology (as per the inference pyramid [Witmer 1995]). Moreover, for simplicity, gross muscle function is being inferred without a detailed consideration of other soft tissue parameters (e.g., cartilage, ligaments). As with all complex systems, however, simplifying the number of variables under consideration can be beneficial for inferring functional morphology provided that certain assumptions and limitations are acknowledged a priori and respected throughout the analysis. This study assumes: (1) that sauropod humeri and femora were loaded vertically and followed a parasagittal arc of movement during locomotion, according to simple biomechanical principles and comparisons with other graviportal vertebrates (Bonnan 2001, 2003); (2) that long-axis rotation and abduction of the humerus and femur were very restricted, judging from the shapes of their articular surfaces and those of their respective girdles (Bonnan 2001, 2003); (3) that the action of most muscles pulling on the humerus or femur would be restricted mostly to extension or flexion in a parasagittal plane; and (4) that because the humerus and femur were loaded vertically and restricted largely to a parasagittal plane of movement, missing data on joint cartilage morphology and other soft tissues will not affect the generalized interpretations of the column-like movements reported here (see Bonnan 2003 for a more detailed consideration of this issue). The functional inferences presented in the discussion are constrained by these four assumptions.