Past studies of the morphology of the claw-bearing phalanges among mammalian taxa has revealed a correspondence of ungual phalanx shape and the broad form of locomotor behavior carried out by those taxa. Further research has provided evidence for convergence in the morphology of ungual phalanges among distantly related taxa that engage in similar locomotion. This evolutionary parallelism is proposed to have been motivated by conserved, shared morphogenic pathways. Such evidence would argue that, despite phylogenetic divergence, we would expect to observe: 1) similar biomechanical properties in ungual cross-sectional geometry among unrelated taxa performing similar forms of locomotion; and 2) phylogenetic modeling of the evolution of ungual phalanx morphology should demonstrate evidence for selection and convergence. In this study, we test these hypotheses using linear dimensions and cross-sectional geometric properties (CGSP) obtained from 42 mammalian taxa. This sample represents a breadth of taxa that have previously been classified into five broad locomotor guilds (arboreal, cursorial, fossorial, generalist, scansorial). CSGP (Ix, Iy, J) were obtained using a Stratec pQCT scanner, and linear dimensions of the phalangeal length, mid-length mediolateral width and dorsopalmar height, and extensor and flexor lever arm lengths were taken using digital Mitutoyo sliding calipers. MANOVAs of linear and CSGP traits demonstrate significant differences in locomotor groups (Pillai's Trace = 0.7206; p=0.0007). Discriminant function analysis shows this is motivated by J and the proportion of mediolateral width to dorsopalmar height of the phalangeal mid-length. We then subjected these properties to phylogenetic models assessing evolutionary models of Brownian motion (using R packages ‘bayou’ and ‘OUwie’). Results present little evidence for shared selective regime changes among taxa. One notable model results indicates selective shifts that occurred among taxa that evolved from an ancestral state of mediolaterally narrow phalanges to broader phalanges (as would be associated with fossorial behaviors). Thus, while convergences in shape and CSGP that relate to locomotor behavior are observed among taxa, phylogenetic comparative methods support selection as occurring only in species that evolved from a narrow to broader ungual phalanx morphology. Further analyses using OU models were used to assess phylogenetic models of convergence.
Read full abstract