Technical Comments

Comment on “Grasping Primate Origins”

Science  02 May 2003:
Vol. 300, Issue 5620, pp. 741b
DOI: 10.1126/science.1081587

Bloch and Boyer (1) recently described a new specimen of Carpolestes simpsoni, which they claimed furnishes evidence that primates (2) and carpolestids are sister taxa. Although we do not dispute the great significance of this specimen for understanding the paleobiology of carpolestids, we have three significant concerns with the Bloch and Boyer analysis and interpretation of the new fossil material.

First, it is unclear to us why Bloch and Boyer (1) restricted their phylogenetic analysis to postcranial characters only. Their specimen includes an undescribed “virtually complete skull and left dentary” (1), and additional crania of carpolestids have been figured in previous publications (35). Furthermore, the dentition of C. simpsoni has been described (5) and the phylogenetic relationships of carpolestids and other plesiadapiforms have been thoroughly studied in a recent series of dental cladistic analyses (6, 7).

The inclusion of only postcranial characters in the analysis by Bloch and Boyer (1) is highly problematic because craniodental data do not support their phylogenetic conclusions. Indeed, cladistic studies of cranial morphology in plesiadapiforms and extant euarchonta (8) favor the hypothesis that primates and scandentians are sister taxa (9, 10). Similarly, dental evidence strongly indicates that carpolestids and plesiadapids comprise a clade that excludes primates and other euarchontans (5, 6, 11). All carpolestids and plesiadapids share a highly derived configuration of the anterior dentition, in which the upper central incisors are enlarged and bear multiple cusps, the lower central incisors are procumbent with lanceolate crowns, the canines are greatly reduced in size, and the lower first premolar is lost (5, 6, 11). By contrast, no element of this anterior dental complex is known to be shared by the earliest primates (12). It is largely because of the detailed dental similarities between carpolestids and plesiadapids that most researchers over the past 60 years have recognized these two families as sister taxa (11).

Although their phylogeny based on the postcranium represents a substantial departure from the findings of previous craniodental studies, Bloch and Boyer (1) make no attempt to reconcile these conflicting sources of information. This issue is crucial, because they acknowledge the possibility that all or some of the derived postcranial similarities shared by primates and carpolestids could have arisen in parallel. The potential for functional convergence is demonstrated by the fact that a divergent clawless hallux (one of the primary characters linking primates and carpolestids) is also found in some extant marsupials (13, 14) and rodents (15). Until steps are taken to address the discrepancy between postcranial and craniodental data sets, the phylogenetic relationships of carpolestids and primates relative to other plesiadapiform and euarchontan clades will be far from resolved.

In addition to casting doubt on their phylogenetic results, the exclusion of a substantial body of relevant craniodental data from the analysis of Bloch and Boyer undercuts their conclusions regarding primate origins. In their discussion, Bloch and Boyer claim that the small and divergent orbits of C. simpsoni are “inconsistent with the visual predation hypothesis” (1). However, the anatomy of C. simpsoni cannot provide evidence concerning the stem primate morphotype unless carpolestids and primates are sister taxa. If, as indicated by craniodental data, carpolestids and plesiadapids form a clade that excludes primates and other euarchontans, then the grasping extremities of primates and carpolestids must represent parallelisms. If this conclusion is correct, then reconstructions of carpolestid ecology have no direct bearing on theories of primate origins except as an additional comparative example of the independent acquisition of grasping hands and feet in a nonprimate lineage.

Second, we point out that Bloch and Boyer were in error when they claimed that the visual-predation hypothesis requires that forward-facing eyes and grasping hands and feet evolved simultaneously for the purpose of visually directed predation. As originally formulated (16), Cartmill's thesis interprets the prehensile, clawless extremities of primates as adaptations for locomotion on slender arboreal supports. Although the visual-predation hypothesis implies that the last common ancestor of living primates used its hands for prey capture much as extant insectivorous primates do, it identifies primate-like grasping extremities as having originally evolved for locomotor purposes. If Bloch and Boyer are correct in their assertion that grasping extremities appeared in the primate stem lineage before forward-facing eyes, they have made a significant contribution to understanding the ecological steps involved in primate origins, but they have not weakened the case for the visual-predation hypothesis.

Third, by claiming that C. simpsoni had “low-crowned molar teeth adapted for eating fruit,” Bloch and Boyer suggested that their proposed phylogeny is inconsistent with the visual-predation hypothesis, presumably because their phylogeny might therefore imply that primates evolved from a frugivorous common ancestor with carpolestids. This conclusion, however, is not supported by functional studies of carpolestid dentitions (3, 17). These analyses indicate that the most primitive members of the Carpolestidae included a significant proportion of insects in their diet and that all carpolestids were at least partly faunivorous. Dental evidence thus favors the conclusion that primates and C. simpsoni evolved independently from more faunivorous ancestral groups. Accordingly, the fossil material described by Bloch and Boyer (1) provides no evidence to reject the hypothesis that the last common ancestor of living primates was a nocturnal visual predator that inhabited a fine-branch arboreal niche (16, 18).

References and Notes

Navigate This Article