The interrelationship of the premaxilla and the mandibular symphysis in anthropoid primates.

摘要

The premaxilla and mandibular symphysis have been traditionally studied in isolation, despite their anatomical vicinity and functionally complementary roles. The latter are particularly seen in the accommodation of incisors and the dissipation of masticatory forces. The diversity in the morphology of both premaxilla and mandibular symphysis has been viewed in the past as phylogenetically informative, but recent research has highlighted such a large degree of inherent variability, that its diagnostic validity is being seriously challenged. 50 inter-landmark distances, reflecting the upper and lower anterior masticatory complex, were collected from 26 species of extant anthropoid primates. Morphological correlations in the upper and lower masticatory complex were investigated using Principal Components Analysis. Discriminant Function Analysis (DFA) was employed to determine the extent to which the morphology of this region reflects diet. The contribution of phylogenetic inertia to morphological and dietary correlations was evaluated with Comparative Analysis by Independent Contrasts (CAIC). All analyses were run on both raw and size-adjusted data, except CAIC which was only carried out on size-adjusted data. Intraspeciflcally, no conclusive correlations were observed between the premaxillary and symphyseal regions. In DFA based on size-adjusted data of the full sample, distinct premaxillary and symphyseal morphologies were found to correspond to diet across phylogenetically diverse taxa. However, when DFA was run on size-adjusted data on each of the large taxonomic units in the sample (hominoids, cercopithecoids, ceboids) interpretation of the results was hampered by the inherent lack of dietary diversity within small taxonomic units and the strong effect of phylogeny. Dietary predictions on fossil taxa were in accordance with published accounts based on other lines of evidence. When the phylogenetic constraints were controlled for, only a small number of correlations between the areas of interest were identified. Small sample sizes, however, rendered the CAIC results inconclusive.