Refining locomotory style in the fossil record through use of 3D Geometric Morphometrics and muscle attachment sites in the fossil species Australopithecus afarensis

Determining muscular anatomy for fossil species can be problematic. Whilst bony material does fossilise, the soft tissue elements are rarely, if ever, preserved to any useful degree and so determination of musculature is often speculative. This has meant that the exact movements of fossil species can be difficult to determine. A prime example of this problem can be seen can be seen in the controversy surrounding the locomotion of Australopithecus afarensis. It has been debated as to whether this fossil species walked in a way similar to modern humans (a striding biped), in a bent-hip-bent-knee style, or in some other novel way. Until relatively recently, it has not been possible to determine the way in which a hypothetical muscular system may have behaved. However, with recent advances in computing, computer modelling of locomotory behaviour has been made possible.
This study used Geometric Morphometric study of the sacrum, Os Coxa, femur and tibia to determine how similar the shape of these bones were between the fossil specimen A.L. 288-1 (Australopithecus afarensis) and extant apes, including Pan paniscus, Pan troglodytes, Gorilla sp., Pongo pygmaeus and Homo sapiens.
Where there was similarity in the shape of the bones, it was hypothesized that there was similarity in the attached musculature, as the species in question are closely related. This was then used to determine a theoretical musculature for the lower limb of Australopithecus afarensis. The properties of individual muscles for Australopithecus afarensis were extrapolated from the properties of the muscles of the species with the most closely shaped area of bone in the extant species, resulting in a muscular pattern most similar to modern humans at the hip, and then increasing in ‘ape-like’ properties distally to the knee and ankle.
This theorised musculature was then input into the forward dynamic modelling program, GaitSym, to establish whether such a muscular configuration would result in a locomotory pattern that conformed to an existing theory or to one that had not yet been proposed. The model proposed unfortunately travelled backwards, likely due to the constraints of the modelling program, however he method of this motion indicated a bipedal ‘shuffle’ as proposed by Hunt (1994).