The effects of ecology and evolutionary history on robust capuchin morphological diversity.

Recent molecular work has confirmed the long-standing morphological hypothesis that capuchins are comprised of two distinct clades, the gracile (untufted) capuchins (genus Cebus, Erxleben, 1777) and the robust (tufted) capuchins (genus Sapajus Kerr, 1792). In the past, the robust group was treated as a single, undifferentiated and cosmopolitan species, with data from all populations lumped together in morphological and ecological studies, obscuring morphological differences that might exist across this radiation. Genetic evidence suggests that the modern radiation of robust capuchins began diversifying ∼2.5 Ma, with significant subsequent geographic expansion into new habitat types. In this study we use a morphological sample of gracile and robust capuchin craniofacial and postcranial characters to examine how ecology and evolutionary history have contributed to morphological diversity within the robust capuchins. We predicted that if ecology is driving robust capuchin variation, three distinct robust morphotypes would be identified: (1) the Atlantic Forest species (Sapajus xanthosternos, S. robustus, and S. nigritus), (2) the Amazonian rainforest species (S. apella, S. cay and S. macrocephalus), and (3) the Cerrado-Caatinga species (S. libidinosus). Alternatively, if diversification time between species pairs predicts degree of morphological difference, we predicted that the recently diverged S. apella, S. macrocephalus, S. libidinosus, and S. cay would be morphologically comparable, with greater variation among the more ancient lineages of S. nigritus, S. xanthosternos, and S. robustus. Our analyses suggest that S. libidinosus has the most derived craniofacial and postcranial features, indicative of inhabiting a more terrestrial niche that includes a dependence on tool use for the extraction of imbedded foods. We also suggest that the cranial robusticity of S. macrocephalus and S. apella are indicative of recent competition with sympatric gracile capuchin species, resulting in character displacement.

Wild bearded capuchin monkeys (Sapajus libidinosus) strategically place nuts in a stable position during nut-cracking.

Humans can use hand tools smoothly and effectively in varying circumstances; in other words, skillfully. A few other species of primates crack encased foods using hammer tools and anvils. Are they skilled? Positioning the food on the anvil so that it does not fall off when struck is a component of skilled cracking. We discovered that bearded capuchin monkeys deliberately place palm nuts in a relatively stable position on the anvil before striking them. In the first experiment, we marked the meridians of palm nuts where they stopped when rolled on a flat surface ("Stop meridian"). We videotaped monkeys as they cracked these nuts on an anvil. In playback we coded the position of the Stop meridian prior to each strike. Monkeys typically knocked the nuts on the anvil a few times before releasing them in a pit. They positioned the nuts so that the Stop meridian was within 30 degrees of vertical with respect to gravity more often than expected, and the nuts rarely moved after the monkeys released them. In the second experiment, 14 blindfolded people (7 men) asked to position marked nuts on an anvil as if to crack them reliably placed them with the Stop meridian in the same position as the monkeys did. In the third experiment, two people judged that palm nuts are most bilaterally symmetric along a meridian on, or close to, the Stop meridian. Thus the monkeys reliably placed the more symmetrical side of the nuts against the side of the pit, and the nuts reliably remained stationary when released. Monkeys apparently used information gained from knocking the nut to achieve this position. Thus, monkeys place the nuts skillfully, strategically managing the fit between the variable nuts and pits in the anvil, and skilled placement depends upon information generated by manual action.

Fallback foraging as a way of life: using dietary toughness to compare the fallback signal among capuchins and implications for interpreting morphological variation.

The genus Cebus is one of the best extant models for examining the role of fallback foods in primate evolution. Cebus includes the tufted capuchins, which exhibit skeletal features for the exploitation of hard and tough foods. Paradoxically, these seemingly "specialized" taxa belong to the most ubiquitous group of closely related primates in South America, thriving in a range of different habitats. This appears to be a consequence of their ability to exploit obdurate fallback foods. Here we compare the toughness of foods exploited by two tufted capuchin species at two ecologically distinct sites; C. apella in a tropical rainforest, and C. libidinosus in a cerrado forest. We include dietary data for one untufted species (C. olivaceus) to assess the degree of difference between the tufted species. These data, along with information on skeletal morphology, are used to address whether or not a fallback foraging species exhibits a given suite of morphological and behavioral attributes, regardless of habitat. Both tufted species ingest and masticate a number of exceedingly tough plant tissues that appear to be used as fallback resources, however, C. libidinosus has the toughest diet both in terms of median and maximal values. Morphologically, C. libidinosus is intermediate in absolute symphyseal and mandibular measurements, and in measures of postcranial robusticity, but exhibits a higher intermembral index than C. apella. We propose that this incongruence between dietary toughness and skeletal morphology is the consequence of C. libidinosus' use of tools while on the ground for the exploitation of fallback foods.