Relative growth of the limbs and trunk in sifakas: Heterochronic, ecological, and functional considerations

Limb, trunk, and body weight measurements were obtained for growth series of Milne‐Edwards's diademed sifaka, Propithecus diadema edwardsi , and the golden‐crowned sifaka, Propithecus tattersalli . Similar measures were obtained also for primarily adults of two subspecies of the western sifaka: Propithecus verreauxi coquereli , Coquerel's sifaka, and Propithecus verreauxi verreauxi , Verreaux's sifaka. Ontogenetic series for the larger‐bodied P. d. edwardsi and the smaller‐bodied P. tattersalli were compared to evaluate whether species‐level differences in body proportions result from the differential extension of common patterns of relative growth. In bivariate plots, both subspecies of P. verreauxi were included to examine whether these taxa also lie along a growth trajectory common to all sifakas. Analyses of the data indicate that postcranial proportions for sifakas are ontogenetically scaled, much as demonstrated previously with cranial dimensions for all three species (Ravosa, 1992). As such, P. d. edwardsi apparently develops larger overall size primarily by growing at a faster rate, but not for a longer duration of time, than P. tattersalli and P. verreauxi ; this is similar to results based on cranial data. A consideration of Malagasy lemur ecology suggests that regional differences in forage quality and resource availability have strongly influenced the evolutionary development of body‐size variation in sifakas. On one hand, the rainforest environment of P. d. edwardsi imposes greater selective pressures for larger body size than the dry‐forest environment of P. tattersalli and P. v. coquereli , or the semi‐arid climate of P. v. verreauxi . On the other hand, as progressively smaller‐bodied adult sifakas are located in the east, west, and northwest, this apparently supports suggestions that adult body size is set by dry‐season constraints on food quality and distribution (i. e., smaller taxa are located in more seasonal habitats such as the west and northeast). Moreover, the fact that body‐size differentiation occurs primarily via differences in growth rate is also due apparently to differences in resource seasonality (and juvenile mortality risk in turn) between the eastern rainforest and the more temperate northeast and west. Most scaling coefficients for both arm and leg growth range from slight negative allometry to slight positive allometry. Given the low intermembral index for sifakas, which is also an adaptation for propulsive hindlimb‐dominated jumping, this suggests that differences in adult limb proportions are largely set prenatally rather than being achieved via higher rates of postnatal hindlimb growth. Our analyses further indicate that the larger‐bodied P. d. edwardsi has a higher adult intermembral index than P. tattersalli and P. verreauxi , thus supporting the allometric argument regarding the interspecific scaling of limb proportions in arboreal primates which employ vertical postures (Cartmill, 1974,1985; Jungers, 1978, 1985). Lastly, additional analyses indicate that P. d. edwardsi exhibits significant sexual dimorphism where adult females are larger than adult males in about one‐third of all adult comparisons, whereas P. tattersalli exhibits significantsexual dimorphism in about one‐fifth of all adult comparisons. Among western sifakas, adult P. v. coquereli exhibit significant sex dimorphism in aboutone‐third of all comparisons, whereas adult P. v. uerreauxi show no significant differences between the sexes. Given that all taxa are ontogenetically scaled,this suggests that sexual dimorphism develops via such processes as well. Interestingly, our data indicate that sex dimorphism is allometric, with larger‐bodied taxa like P. d. edwardsi being more dimorphic.