Remarks On Scenarios Of The Acquisition Of Nails

In the initial formulation of the visual predation theory, Cartmill (1972, 1974a,b) explained the acquisition of the opposable hallux by the ancestral primate and the subsequent loss of claws in several groups of primates by the invasion of the fine branch milieu. It is usually now accepted that the possession of nails on the extremities is also a part of the primate morphotypic condition. Recent comparative and functional studies of arboreal primates and marsupials went further in the study of the convergences between these two groups and led to an increase in our knowledge of the arboreal adaptation of some of them (Rasmussen, 1990; Lemelin, 1999; Larson et al., 2000; Hamrick, 1998, 2001). These inquiries have shown the importance of long phalanges to grasp fine branches. They revealed that the peculiarities of primate "hindlimb domination" and diagonal sequence in footfall exist also in the highly arboreal marsupials, which suggests that these peculiarities are also adaptations for locomotion on fine branches (Schmitt and Lemelin, 2002). These discoveries are interpreted in favor of the fine branch milieu as the adaptive shift explaining the origin of primate appendicular characteristics (Hamrick, 1998, 2001; Larson et al., 2000; Lemelin, 1999; Rasmussen, 1990; Schmitt and Lemelin, 2002). However, are these findings sufficient to confirm the fine branch milieu hypothesis concerning primate origins? Consideration of both marsupials and primates leads me to embrace a more complex scenario.

Concerning marsupials, why did the arboreal ones fail to evolve nails on their non-hallucal digits, if habitual grasping of fine branches is the key adap-tational shift explaining the transformation of claws into nails? A marked reduction of claws has been reported in the Australian Tarsipes, Cercartetus, and Burramys (Cartmill, 1974b). A detailed study of their terminal phalanges would be important. The American Marmosa and Caluromys retain claws. Are those claws reduced? Are their terminal phalanges broadened? Is their arboreal adaptation too recent to have allowed the replacement of claws by nails? American Marmosidae are known since 12.5 MYA (Flynn et al., 1997;

Goin, 1997). The author has not found information concerning fossil caluromyids (a family in Nowak, 1999). However, Burramys is found in the Late Oligocene-Middle Miocene Ngama fauna, and Burramyidae, as well as Pseudocheiridae and Petauridae, are reported since the Late Oligocene-Early Miocene of Australia (Rich, 1991). It is also well known that Didelphidae, in a restricted sense including tarsal characters and opposable hallux (i.e., having arboreal adaptations), are known at least since the Early Eocene (Szalay, 1982). In fact, Ameridelphians, and possibly Metatheria, are thought to have possessed arboreal characters. In this context, it is very strange that there are not many marsupials having reached the "ultimate" fine branch adaptation of nails, if the last indeed are due to fine branch locomotion. Marsupial analogs suggest that claw loss may have been a very long process.

A real effort to understand the replacement of claws by nails was made by Cartmill (1974b). He suggested that, if a mammal, whose first toe has become divergent enough to oppose the other four toes in grasping slender branches, continues to emphasize prolonged cautious locomotion among slender branches, its "first digit will become proportionately more powerful, and claw grip will be proportionately enfeebled"—something he reported as true of marsupials such as Marmosa, Didelphis, and Petaurus (idem, p. 71). What makes little sense is that the process was not conducted to complete claw loss in a larger number of marsupials, particularly given how ancient their arboreal adaptations are. One would also guess in such a scenario that a similar opposition of the thumb to the other digits of the manus would occur (or another type such as digits one and two opposed to the three others). However, it seems that opposability in the hand has rarely been developed in arboreal "didelphid-like" marsupials, whereas it did in larger forms such as Phascolarctos. A better understanding of hand grasping in small marsupials might help understand why they did not evolve opposable hands and nails on their extremities. Some degree of hallucial opposition is also known in other mammals, including some rodents, tree shrews, and fossil plesiadapiforms. It is suggested that a "didel-phid-like" adaptation, with opposable hallux, claws on other digits and non-opposable thumbs represents a successful arboreal adaptation, allowing the possible use of fine branches if the digits become long enough. It would represent a step toward but not a complete convergence with primate appendicular characteristics. It seems that opposition in the hand and the presence of nails are less common; could these two properties be somehow related? Some primates have very little thumb opposition, and among them are the callitrichids, which have reacquired claws. Further study is needed to clarify this relationship. In fact, nails could be acquired on hands and feet as a result of selection on feet only, or hands only, due to the developmental modules that link structures between the limbs (Hallgrimsson et al., 2002). Their role in removing skin parasites and social grooming should also be explored.

Concerning primates, when we first discussed the proportions of the primitive morphotype hand, we proposed that a relatively long hand within the forearm, as found in living claw climbers, vertical clingers and leapers, and Notharctus, was probably primitive for the order (Jouffroy et al., 1991). Concerning the intrinsic proportions of the hand, we hypothesized that hands with long digits close to those of galagines and Tarsius (i.e., with third digit amounting to 62-65% of total hand length), would make a good hypothetical morphotype. Notharctus is close, with similar proportions of its third digit, but with a different position on the diagram due to its very short metacarpals (Jouffroy et al., 1991). Subsequent reconstructions of proportions for other available Middle Eocene primates underscored a group of long-digited fossil species, having proportions similar to those of Galago and differing from Tarsius only in their longer carpus and shorter metacarpus. This group of long-digited fossils was proposed as giving a good approximation of the primitive primate morphotype for hand proportions (Godinot, 1992). In these studies, cheirogaleids are at some distance from this group, being closer to other lemurs and a group of platyrrhines. If these speculations about the ancestral primate hand were valid, the cheirogaleid hand would not be a good analog for the most primitive primate hands. Since our work, a much more complete Notharctus hand was described (Hamrick and Alexander, 1996), and its describers concurred with us that a long hand with short metacarpus and long digits probably was morphotypic for primates. Possibly more significantly, Hamrick (2001; this volume) shows that such digital and metacarpal proportions are manifested early in primate ontogeny and are thus likely to reflect a strong phylogenetic signal. This suggests that the proportions of Middle Eocene primate hands are very significant, and much closer to the primitive primate morphotype than those of most living primates.

What is the adaptive significance of hands with very long digits? The hands of Tarsius are especially good at catching insects, and apparently not very good for grasping small branches. Likewise, galagines are well-known for their ability to catch insects flying away from a branch. The working hypothesis is that primitive primate hands, long and with very long digits, were especially well adapted for catching insects. In a preliminary functional study of Eocene primate hands, the author speculated about the function of distal phalanges having retained from ancestral claw-bearing a high proximal part, but having acquired a broadened distal part for nail bearing (Godinot, 1991). The author, in this regard, could come up only with stabilization as the major functional difference between a nail-bearing and a claw-bearing digit. The mediolateral stabilization of the tip of the digits is transmitted proximally through the broad proximal expansion of primate distal phalanges so that the whole hand and foot must have had increased control of items being grasped. This in turn would be beneficial for both stabilizing insect prey in the hand, and more importantly stabilizing the feet on the support and controlling lower limb movements during insect catching. Galago, for example, is known to jump and catch an insect by sweeping the air while remaining attached to a branch by its feet, which allows it to retract back on the branch and eat the prey. For such an acrobatic behavior, a firm stabilization of the feet is certainly important. This stabilization is greater than would be provided by a weakly opposable hallux. It demands that the hallux and the opposed digits achieve a powerful grasp so that forces can be transmitted from the digits to the limbs appropriately. Thus, the hypothesis developed in 1991 states that primate morphotypic appendicular characters, nails, the powerful opposable hallux, and correlated postcranial characters are better explained by insect catching in the arboreal milieu than by invasion of the fine branch niche. This modest modification of the visual predation theory in its initial formulation has the advantage of making it simpler, the postcranial characters being an integral part of the behavioral and functional complex implied by the visual predation hypothesis. This view is in complete accord with the fact that the recently studied arboreal marsupials do not show the transition to nails: they do not have the insect predation specialization, despite the fact that they snatch insects (Cartmill, 1974b; Nowak, 1999; Rasmussen, 1990). It appears that primate locomotor characteristics were acquired in at least two steps: (1) a "didelphid-like" step with opposable hallux and long phalanges allowing the grasping of small branches, with the correlative gait characteristics found in both groups (a "Ptilocercus-like" step with incipient hallucial grasping may illustrate either a preceding step or a different scenario) and (2) a second step reached only by primates, with the insect-catching specialization implying long hands, powerful hallucial grasping, nails, and other correlated tarsal and long bones characters (for examples of more detailed scenarios, see Dagosto, this volume).

The visual predation theory has been challenged by Sussman (1991, 1995), who proposed that angiosperm feeding played a major role in the acquisition of primate characteristics. When Szalay (1968) interpreted the evolution of the earliest primate dentitions by a shift toward a frugivorous diet, he was mainly concerned with early plesiadapiforms. Since then, functional studies of dentitions have been made (Kay, 1975; Strait, 1993). Williams and Covert (1994) found Teilhardina americana just at the limit of predominantly insect eaters. The author suspects that the earliest primate dentitions (Altiatlasius, Teilhardina belgica, Donrussellia) would appear more insectivorous; however, a real quantification needs to be done. Even if the history of the group is one of mixed-feeding, small mammals need a source of protein, which is often constituted by small arthropods. Thus, during the post-Cretaceous radiation of small mammals, there must have been strong competition among all the small species requiring some insects. It is very possible that this competition was the most stringent, and therefore, provided the driving force for evolutionary change. Plesiadapiform insect-catching was through the incisors, in animals having claws and relying on olfaction (possibly also on audition, as they had big bullae). This is in sharp contrast with primates, which appear to have specialized on insect snatching via hand capture, relying on visual and auditory cues. This adaptive contrast could be the consequence of competition; however, it could also be linked with very different types of forests. The lesser development of high canopy in the Paleocene of North America and Europe may have favored claw-climbing species. On the other hand, we do not know which taxa were competing with primates during the Paleocene in Asia; as the diversity of plesiadapiforms was apparently not great in Asia, we may suspect other groups. To conclude, even if mixed-feeding was their actual dietary adaptation, primates may have nevertheless acquired their characteristics as the result of a specialization for the capture of their insect prey by audiovisually directed predation with hand capture.

0 0

Post a comment