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

Paleobiology, Summer 2004 by Bonnan, Matthew F

Specimens and Measurement.-The humeri and femora of Apatosaurus, Diplodocus, and Camarasaurus were examined at 14 North American collections (Fig. 1) ( see Appendix 1 for repository information; measurements available in Bonnan 2001). Humeri and femora were selected for analysis because these appendicular elements are the most abundant in North American collections and provide the largest sample size for statistical tests. Other sauropod limb elements (especially ulnae, radii, and fibulae) have exceedingly simple morphologies with few discernible landmarks, characteristics that make them more difficult to digitize for shape analysis (Borvnan 2001). Tibiae were difficult to digitize because their morphology does not translate well to a twodimensional analysis (see Bonnan 2001 for more details). As discussed above, these three taxa were selected because of their probable sympatric and paleoecological associations in Morrison Formation assemblages (Dodson 1990). Contemporaneous sauropod taxa of the Tendagaru provide another well-represented sample of appendicular elements (Dodson 1990). However, examination of these taxa is beyond the scope of the present analysis and will constitute a separate study (Bonnan unpublished data).

Taxonomic identifications of most humeri and femora were based on their association with other skeletal elements or, in some cases (e.g., Apatosaurus louisae [CM 3018]), were made from nearly complete specimens. No unidentified or questionable elements were included in the analysis. For isolated humeri and femora, the element's generic identity was established by using identifications, figures, and descriptions of complete or nearly complete skeletons published by Ostrom and McIntosh (1966) and McIntosh (1990). In no case was there a contradiction between the isolated element's identification and the basic descriptions of that element from more complete material. For a comprehensive review of Morrison sauropod appendicular material, see Wilhite 2003.

Standardized measurements were taken on the humeri and femora (Fig. 2) with measuring tape to the nearest centimeter (±1 cm) and are reported here in millimeters ( see Appendix 2 for variable abbreviations). A subsample of the humeri and femora was photographed for thin-plate splines digitization. Controlled photography of each element in a specific orientation can be difficult because many elements will be fixed in skeletal mounts and some will be damaged. As much as possible, selection of the elements to be digitized was based on the following criteria: (1) a photograph or series of photographs could be taken of the element in which the morphological features of interest lay on a contour parallel to the plane of the camera lens; and (2) a standardized, photographic position of the element lying naturally on the floor was possible without extensive propping or elevation of certain morphological aspects relative to others. Humeri were photographed lying on a flat surface, their anterior faces exposed, whereas femora were photographed lying on a flat surface, their posterior faces exposed (Fig. 3). Photographs were taken with a standard 35mm camera mounted on a tripod over the bone of interest at a predetermined, standardized height of 1850 mm. The tripod was leveled by using a built in bubble level, and a separate level was used on the camera itself, to ensure as little distortion of the image as possible. Each element was then centered in the lens to decrease distortion from lens curvature or parallax. The resulting photographs were then scanned on a flatbed scanner, cropped, scaled to a height of 5.08 cm, and digitized (see below for digitization details).