functional and adaptive significance of primate pads and claws: evidence from new world anthropoids

Functional and Adaptive Significance of Primate Padsand Claws: Evidence From New World Anthropoids

MARK W. HAMRICK*Department of Anthropology and Division of Biomedical Sciences,Kent State University, Kent, Ohio

KEY WORDS ungulae; tegulae; primate origins; epidermal ridges;hand morphology

ABSTRACT This study tests predicted morphoclines in fingertip morphol-ogy among four small-bodied (,1 kg) New World monkeys (Saimiri sciureus,Leontopithecus rosalia, Callithrix jacchus, and Saguinus oedipus) in order totest previous functional and adaptive explanations for the evolution offlattened nails, expanded apical pads, and grasping extremities within theOrder Primates. Small-bodied platyrrhines which frequently forage amongsmall-diameter substrates are expected to possess 1) relatively expandedapical pads, 2) well-developed epidermal ridges, 3) distally broad terminalphalanges, and 4) reduced flexor and extensor tubercles compared to thosespecies which use large-diameter arboreal supports more frequently for theirlocomotor and postural behaviors. Results show that as the frequency ofsmall-branch foraging increases among taxa within this sample, relativedistal phalanx breadth also increases but distal phalanx length, height, andflexor tubercle size decrease. Moreover, epidermal ridge development becomesmore pronounced as the frequency of small-branch foraging increases.Terminal phalanx breadth and epidermal ridge complexity are both positivelycorrelated with apical pad size. The large, flexible apical pad increasesstability of the hand and foot on small-diameter arboreal supports because thepad can contact the substrate in several planes which, in turn, enables the padto resist disruptive forces from different directions by friction and interlocking(Hildebrand, 1995). The observed morphoclines demonstrate that a gradientin form from claw- to nail-like tegulae exists among these taxa. Thus, thedistinction between claw- and nail-bearing platyrrhines is essentially arbi-trary. These observations corroborate Cartmill’s (1972) functional and adap-tive model for the loss of claws in primates: namely, expanded apical pads arerequired for habitual locomotor and postural behaviors on small-diametersupports whereas claws are more useful for positional behaviors on large-diameter substrates. Finally, results from this study support previous sugges-tions that the keeled tegulae of callitrichines represent a derived posturaladaptation rather than a primitive retention from an ancestral eutheriancondition. Am J Phys Anthropol 106:113–127, 1998. r 1998 Wiley-Liss, Inc.

The evolution of flattened nails on alldigits of the hand and foot distinguishesprimates from other eutherian orders (Clark,1936, 1959). Cartmill (1972, 1974) sug-gested that, although claws are superior toflattened nails for locomotor and posturalbehaviors on large-diameter vertical arbo-real supports, claws are not as useful as

expanded apical pads and flattened nails forlocomotion on small-diameter arboreal sub-strates. Thus, he explained the evolutionary

*Correspondence to: Dr. Mark W. Hamrick, Department ofAnthropology, Box 5190, Kent State University, Kent, OH 44242USA. E-mail: [email protected]

Received 11 August 1997; accepted 19 February 1998.

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 106:113–127 (1998)

r 1998 WILEY-LISS, INC.

replacement of claws with flattened nails asan adaptive transformation enabling earlyprimates to forage successfully for insectsamong the slender terminal branches oftrees and shrubs (Cartmill, 1972, 1974).New World primates of the family Callitrichi-dae, however, possess claws on all digitsexcept the hallux, but certain callitrichidssuch as Saguinus oedipus spend much oftheir time foraging for fruits and insectswithin the ‘‘terminal branch niche’’ (Garber,1980; Garber and Sussman, 1984). Claws,therefore, do not appear to hinder the move-ments of these primates on small-diametersubstrates, leading Cartmill (1985, p. 83) toconclude that ‘‘the clawlessness of more typi-cal primates . . . thus presents a persistenttheoretical problem’’ (see also Dagosto, 1988).

Claws and nails have been characterizedas discrete character states, although bothLeGros Clark (1936) and Thorndike (1968)showed that ‘‘a gradient in form from claw tonail, rather than a clear dichotomy betweenthe two, is seen among living primates’’(Thorndike, 1968, p. 251). Evidence from thenails and claws of birds (Bock and Miller,1959; Feduccia, 1993), rodents (Haffner,1996), carnivores (VanValkenburgh, 1987;Bryant et al., 1996), and other mammals(Macleod and Rose, 1993) reveals that inter-specific variation in fingertip morphology isfunctionally related to interspecific varia-tion in locomotor, postural, and manipula-tory behavior. Hence, morphoclines in apicalpad morphology and distal phalanx architec-ture can be used to test functional andadaptive hypotheses concerning the evolu-tion of primate cheiridial form. The objectiveof this study is to test predicted morpho-clines in apical pad and distal phalanx mor-phology among small-bodied New Worldmonkeys which differ significantly from oneanother in their foraging behaviors in orderto test Cartmill’s (1972, 1974, 1985) func-tional and adaptive model for the evolutionof flattened nails, expanded apical pads, andgrasping extremities in primates.

PREDICTIONSApical pad morphology

The apical pads of primates bearing flat-tened ungulae are relatively expanded andexhibit well-developed papillary ridges (Mar-

tin, 1990). The papillary ridges serve notonly to improve the frictional properties ofthe digital skin but also to increase theskin’s sensitivity to tactile stimuli (Cauna,1954; Napier, 1980; Loesch and Martin,1984). The frictional properties of the ridgesderive primarily from their ability to pro-duce surface interlocking (Whipple, 1904;Cartmill, 1974). The importance of well-developed friction pads for small-branch for-agers is clearly illustrated by Cartmill (1985),who showed that grasping digits allow aclimbing mammal to exert torques oppositeto those produced by gravity when pitchingover on small-diameter substrates. Theridges are also important for tactile sensitiv-ity, since pressure on the elevated ridgesapplies a mechanical stimulus to tactilereceptors called Meissner’s corpuscles, rap-idly adapting mechanoreceptors located un-derneath the basins between the papillaryridges which signal stimulus onset and off-set (Cauna, 1954, 1956; Burgess and Perl,1973; Halata, 1975).

Cartmill’s model predicts that New Worldprimates such as squirrel monkeys (Saimirisciureus), which frequently (.80%) forageand feed among small-diameter (,3 cm)arboreal supports of variable orientation(Fleagle et al., 1981; Boinski, 1989; Fon-taine, 1990), should have relatively largeapical pads bearing numerous papillaryridges in order to increase the interlockingproperties of their grasping fingers. In con-trast, small-bodied monkeys such as com-mon marmosets (Callithrix jacchus), whichforage for exudates by clinging to large-diameter (.15 cm) vertical supports andtravel primarily by leaping between suchsupports (Lacher et al., 1984; Sussman andKinzey, 1984; Garber, 1992), would be ex-pected to have relatively small apical padsand poorly developed papillary ridges. Thecotton-topped tamarin S. oedipus, which for-ages on large- and small-diameter supports(Garber, 1980; Garber and Sussman, 1984),and the golden lion tamarin Leontopithecusrosalia, which forages frequently on small-and medium-sized (3–12 cm) horizontal sup-ports (Peres, 1989; Garber, 1992; Stafford etal., 1996), would both be expected to possessapical pads and papillary ridges that are

114 M.W. HAMRICK

more well developed than those of Callithrixbut less so than those of Saimiri.

Terminal phalanx shape

The falculae of most arboreal mammals(e.g., Tupaia, Distoechurus) are character-ized by both a superficial corneous layerderived from a basal matrix of germinal cellsand a deep corneous layer derived from aterminal matrix of germinal cells (Clark,1936). The evolution of flattened ungulae inthe majority of primates is associated withthe loss of the terminal matrix and deepcorneous layer. A number of extant primatesbearing claw-like tegulae do, however, re-tain both a terminal matrix and deep corne-ous layer (Clark, 1936). Clark (1936) showedthat, among the platyrrhines he included forstudy, C. jacchus had the most well-devel-oped deep layer, which comprises about 17%of the total claw depth. The deep layer of S.oedipus represents about 10% of the totalclaw depth and Cebus albifrons possessesonly a rudimentary deep layer (Thorndike,1968). The studies of both Clark (1936) andThorndike (1968) illustrate clear morpho-clines in nail and claw ultrastructure amongplatyrrhines and their analyses are not re-peated here. The following predictions relatevariation in morphology of the platyrrhineterminal phalanx to the aforementionedvariations in platyrrhine substrate use andnail and claw form.

Arboreal mammals bearing well-devel-oped falculae typically possess terminal pha-langes with enlarged flexor and extensortubercles, which increase the lever arms forthe long flexors and extensors, respectively,acting at the distal interphalangeal joints(Clark, 1936; Rosenberger, 1977). The largelever arm for the long flexor tendon permitspowerful interlocking of the claw with thesubstrate, required for clinging and climb-ing behaviors on large-diameter vertical ar-boreal supports where the central anglesubtended by the hands and feet is less than180° (Cartmill, 1979). The large lever armfor the long extensor permits powerful dorsi-flexion of the claw during quadrupedal walk-ing, running, and bounding on these large-diameter substrates. The distal phalanx ofarboreal mammals bearing falculae is alsocompressed radioulnarly and high dorsopal-

marly in order to accommodate the keeledcorneous sheath (Clark, 1936).

The flexor and extensor tubercles are bothreduced in arboreal primates that utilizegrasping extremities for locomotor and pos-tural behaviors on small-diameter sub-strates (Clark, 1936). Arboreal mammalsbearing grasping extremities can wrap theirfingers and toes around the substrate, effec-tively subtending a central angle exceeding180°. The adduction force applied to thesupport by the digit is exerted by all of thephalangeal segments across all of the inter-phalangeal joints, not just the distal inter-phalangeal joints (Preuschoft, 1970, 1973).Thus, compared to clawed arborealists thatpossess a very well-developed lever arm forthe long flexor tendon at the distal interpha-langeal joint, clawless mammals exhibit arelatively reduced flexor tubercle on thedistal phalanx. Finally, the distal phalanx ofarboreal mammals with flattened ungulaeis, in contrast to that of clawed mammals,expanded radioulnarly and shortened dorso-palmarly in order to accommodate a flat-tened superfical corneous layer and broadapical pad.

C. jacchus, which spends the overwhelm-ing majority of its feeding time clinging tolarge-diameter vertical arboreal supports(Table 1), would be expected to possess rela-tively large flexor tubercles and radioul-narly compressed and dorsopalmarly ex-panded terminal phalanges related to itsclaw-like tegulae. S. sciureus, which is moreof a small-branch forager, would be expectedto have reduced flexor and extensor tu-bercles and a more radioulnarly expandedand dorsopalmarly compressed distal pha-lanx in order to support both a relativelybroad nail and apical pad. S. oedipus, whichforages on small- and large-diameter sup-ports, and L. rosalia, which forages on large-and medium-diameter branches, would bothbe expected to possess flexor tubercles thatare somewhat intermediate in relative sizebetween those of Saimiri and Callithrix.Callitrichids do not use their keeled tegulaeduring quadrupedal walking and running(Rosenberger, 1977; Rosenberger andStafford, 1994) and they would therefore notbe expected to possess very large extensortubercles used for claw dorsiflexion. L. rosa-

115PLATYRRHINE PAD AND CLAW MORPHOLOGY

lia frequently forages for cryptic insect preyconcealed in bark and under leaves by prob-ing with its elongate fingers (Garber, 1992;Dietz et al., 1997). This taxon would, likeother arboreal mammalian extractive forag-ers (e.g., Daubentonia, Dactylopsila, Dacty-lonax), be expected to possess relativelyelongate terminal phalanges related to theuse of its fingers as probes (Godinot, 1992).

MATERIALS AND METHODSSample

Morphoclines in apical pad and terminalphalanx morphology were examined in atotal of 36 skeletal specimens and eightcadaver specimens of S. sciureus, S. oedipus,L. rosalia, and C. jacchus (Table 1). Thesetaxa were chosen for analysis because theyare similarly sized (,1 kg) and their chei-ridial dimensions can therefore be comparedwithin a ‘‘narrow allometric’’ context (sensuSmith, 1980). The skeletal specimens stud-ied are housed in the National Museum ofNatural History, Smithsonian Institution.Cadaver specimens of Saimiri and Leontopi-thecus are also from the National Museum ofNatural History, whereas specimens ofSaguinus and Callithrix are from the Univer-sity of Tennessee at Knoxville. Body weights,diets, and substrate preferences for thesetaxa are shown in Table 1.

Histology

Apical pad morphology was examined inall four species by severing the pad from thevolar surface of the third manual distalphalanx in one male and one female fluid-preserved specimen representing each spe-cies. A tissue sample was removed from the

middle of each pad where the papillaryridges run proximodistally in order to en-sure that similar regions of the apical padwere examined in each individual. Thesetissue specimens were then washed in tapwater for 24 hours and placed in a 70%ethanol solution. Specimens were later dehy-drated using a graded series of ethanolsolutions, cleared, and embedded in paraffinfollowing the general procedure describedby Humason (1972). Once specimens wereembedded in paraffin blocks they were sec-tioned at 10 µm in the transverse planeusing a rotary microtome. Thin sectionswere then mounted on slides and stainedusing Weigert’s hematoxylin counterstainedwith picroponceau in order to visualize struc-ture of the dermis and epidermis (Humason,1972). Stained sections were viewed andphotographed through a high-powered Olym-pus stereo light microscope.

Osteometrics

The terminal phalanx from the third fin-ger of one hand representing each skeletalspecimen was traced in lateral and palmarviews using a Zeiss stereo microscope withcamera lucida attachment. Ten points werethen digitized on these outlines using Sig-maScany digitizing software in order tocalculate the following linear distancesshown in Figure 1: terminal phalanx length(L), from the most distal point on the proxi-mal articular surface to the distal tip of thebone; height of the phalangeal body (MSH),from the dorsal and palmar borders of thephalanx at approximately midshaft; exten-sor lever arm (ETUB), from the most distalpoint on the proximal articular surface to

TABLE 1. Sample sizes (n 5 number of skeletal specimens, number of cadaver specimens) and adaptive profilesfor the four New World monkey species included for study

Species (n) Body massa Dietb Substrate preferenceb

Saimiri sciureus (10, 2) 825 g. insects (70%) horizontal, obliquefruit ,3 cm diameter

Saguinus oedipus (7, 2) 485 g. insects (40%) horizontal, oblique/verticalgum,c fruit ,3 cm, .15 cmc

Leontopithecus rosalia (9, 2) 600 g. insects (50–70%) horizontalnectar, fruit 3–15 cm diameter

Callithrix jacchus (10, 2) 275 g. gum (,70%) verticalinsects .15 cm diameter

a Pooled-sex values from Fleagle (1988).b Data from Fleagle et al. (1981), Fleagle (1988), Garber (1992), Stafford et al. (1996), and Dietz et al. (1997).c Frequencies of exudate feeding associated with the use of large-diameter vertical supports increase during the dry season,approaching 50% compared to only 20–30% in the wet season (Garber, 1992).

116 M.W. HAMRICK

the dorsal border of the extensor process;flexor lever arm (FTUB), from the mostdistal point on the proximal articular sur-face to the palmar border of the extensorprocess; apical tuft breadth (TUFW), fromthe radial and lateral borders of the distalend at its widest diameter; base breadth(BASW), from the radial and lateral bordersof the phalangeal base. These linear dimen-sions were chosen for analysis because theycapture functionally relevant aspects of dis-tal phalanx shape discussed earlier. The rawlinear dimensions were expressed as log–shape ratios for statistical analysis follow-ing the procedure described by Mosimannand James (1979), Falsetti et al. (1993), andJungers et al. (1996).

Statistical analyses

Nonparametric statistical tests were usedfor interspecific comparisons of log–shaperatio values in order to eliminate distribu-tional and dispersion assumptions withinthis sample. Kruskal-Wallis tests were firstused to investigate interspecific differencesin these shape ratio values and Mann-Whitney U tests were then used for pairwisebetween-species comparisons, following So-kal and Rohlf (1981) and Zar (1984). Amultivariate discriminant analysis was also

performed on the logged ratio values inorder to test for interspecific differencesusing the multivariate dataset. A multivari-ate analysis of variance (MANOVA) was thepreferred technique for this analysis, sincethe groups were defined a priori (Neff andMarcus, 1980). Log–shape variables werealso tested for correlations with size bygenerating Spearman-rank correlations be-tween the log–shape ratios and the log–size(geometric mean) variable.

RESULTSApical pad morphology

The epidermal structures referred to hereare named according to the terminologyutilized by Martin (1990, p. 502). Apicalpads of C. jacchus differ from those of theother three taxa primarily in being mark-edly compressed radioulnarly relative totheir dorsopalmar height (Fig. 2). The volarapical pad skin of both Callithrix and L.rosalia exhibits undulating papillary ridgeswhich typically correspond to intermediateridges (5 glandular ridges of Halata, 1975)that penetrate into the dermis (Fig. 3a,b).The intermediate ridges of Callithrix formelongate, finger-like papillae on the mostvolar tip of the pad, whereas they are rela-tively reduced on the radial and ulnar sidesof the pad. Many sections of Callithrix volarskin exhibit intermediate ridges that arewidely spaced and sparsely distributed. Le-ontopithecus also has sparsely distributedintermediate ridges and the basins betweenthe papillary ridges are rather low and flat(Fig. 3b). Epidermal ridge development ap-pears to be correlated, for the most part,with the relative size of the apical pad inthese two species.

The volar apical pad skin of S. oedipus andS. sciureus, like that of both Callithrix andLeontopithecus, exhibits undulating papil-lary ridges (Fig. 3c,d). The intermediateridges of Saguinus are, however, more well-developed than those of both Callithrix andLeontopithecus. Furthermore, the apical padof Saguinus is expanded radioulnarly com-pared to those of both Leontopithecus andCallithrix (Fig. 2). Specimens of Saimiriexhibit well-developed papillary ridges andelongate intermediate ridges that are compa-rable in morphology to those of Saguinus

Fig. 1. Terminal phalanx from the third manualdigit of Callithrix jacchus in lateral (left column) andpalmar (right column) views showing the ten pointsdigitized on the camera lucida tracings (top row) and thesix linear distances calculated between the ten digitizedpoints (bottom row).


(Fig. 3d). The basins between the papillaryridges correspond to deeply projecting limit-ing ridges (5 adhesive ridges of Halata,1975) which, although elongate, do not pro-trude so deeply into the dermis as the inter-mediate ridges. The intermediate ridges ofSaimiri are considerably more well devel-oped than those of the other three taxastudied. Moreover, apical pads of Saimiriare much broader radioulnarly than those ofthe other three platyrrhines examined(Fig. 2).

Terminal phalanx shape

Kruskal-Wallis tests show that terminalphalanges of the four platyrrhine taxa in-cluded for analysis differ significantly fromone another in relative size of the flexortubercle, breadth and height of the phalan-geal body, breadth of the phalangeal base,and length of the terminal phalanx (Figs. 4,5; Tables 2, 3). The log–shape variables arenot significantly correlated with log–size(Table 4), indicating that these ratios arediscriminating between taxa primarily onthe basis of shape differences. Pairwise com-parisons and examination of the raw datashow that Saimiri has the relatively small-est flexor tubercle of all the taxa, whereasCallithrix has the relatively largest flexortubercle (Figs. 4; Tables 2, 3). Saguinus andLeontopithecus possess flexor tubercles thatare somewhat intermediate in size betweenthose of Saimiri and Callithrix and, in fact,do not differ significantly from one anotherin this measurement (Tables 2, 3). Saimirialso has a very dorsopalmarly compressedand bilaterally expanded phalangeal body,whereas that of Callithrix is very high dorso-palmarly and compressed radioulnarly (Figs.4, 5a,b). The phalangeal body of Saguinusdiffers significantly from that of Callithrixin being more dorsopalmarly compressedand radioulnarly expanded but differs fromthat of Saimiri in being more expandeddorsopalmarly and compressed radioulnarly(Table 3; Figs. 4, 5a,b).

The morphoclines observed among thesetaxa in relative breadth of the distal pha-lanx are also observed in relative breadth ofphalangeal base. Saimiri has the broadest(radioulnarly) base of all the taxa and Calli-thrix has the relatively narrowest phalan-geal base (Table 2; Figs. 4, 5c). Saguinus

Fig. 2. Tracings of dorsopalmar sections through theapical pad of the third finger in A: Callithrix jacchus, B:Saguinus oedipus, and C: Saimiri sciureus showing theradioulnarly compressed pad of Callithrix and moreradioulnarly expanded pads of Saguinus and Saimiri.The epidermis is shaded and the dorsal surface of eachsection faces toward the top of the page whereas thepalmar surface of each section faces toward the bottomof the page. Scale bars 5 1 mm.

118 M.W. HAMRICK

differs significantly from Saimiri in having aslightly narrower base but differs signifi-cantly from Callithrix (and Leontopithecus)in having a relatively wider base (Table 3;Fig. 5). Finally, these four small-bodied plat-yrrhines also differ significantly from oneanother in relative length of their thirddistal phalanx. Leontopithecus differs signifi-cantly from the other taxa in having arelatively elongate terminal phalanx,whereas Saimiri has the relatively shortestterminal phalanx of all the taxa studied(Tables 2, 3; Figs. 4, 5d). Callithrix andSaguinus, which do not differ significantlyfrom one another in this respect, possessterminal phalanges that are intermediate inlength between those of Saimiri and Leonto-pithecus (Tables 2, 3; Fig. 5). Leontopithecushas been noted previously as having veryelongate forelimb elements (e.g., Dykyj, 1982;

Rosenberger and Stafford, 1994; Aronsen,1997) and it is therefore not surprising thatthis taxon has very elongate terminal pha-langes related to its extractive foraginghabits.

The morphoclines observed among thefour taxa in the univariate analyses are alsoobserved in the multivariate analysis. TheMANOVA is highly significant (Wilkes-Lambda 5 10.76, P , .001) and classified97% of the cases correctly. The first discrimi-nant axis shows a strong negative correla-tion with breadth of the phalangeal base andstrong positive correlation with height of thephalangeal body (Table 5). Thus, Saimiri,which has a broad phalangeal base anddorsopalmarly compressed phalangeal body(Table 2; Fig. 4), has low scores on this axis(Fig. 6), whereas Callithrix and Leontopithe-cus, which both have a radioulnarly com-

Fig. 3. Photomicrographs of dorsopalmar sections through the apical pad of the third finger in A:Callithrix jacchus (3140, scale bar 5 .25 mm), B: Leontopithecus rosalia (3140; scale bar 5 .25 mm), C:Saguinus oedipus (3120; scale bar 5 .50 mm), and D: Saimiri sciureus (380; scale bar 5 .50 mm). Notethe undulating stratum corneum (sc) of each species and the deeply projecting intermediate ridges (ir) ofSaguinus and Saimiri. The arrows in D point to the limiting ridges of Saimiri.


pressed phalangeal base and dorsopalmarlyexpanded phalangeal body (Table 2; Fig. 4),have high scores on this axis (Fig. 6). Sagui-nus, which exhibits a pattern of distal pha-lanx morphology that is somewhat interme-diate in form between that of Saimiri andCallithrix, has intermediate scores on thisaxis (Fig. 6). The second discriminant axisshows a strong negative correlation withphalangeal length (Table 5). This axis sepa-rates Leontopithecus, which has a relativelyelongate third manual distal phalanx (Table2; Fig. 5), from the other three species,which tend to have shorter terminal phalan-ges compared to those of the golden liontamarin (Fig. 6).

DISCUSSIONEvolution of platyrrhine pads and claws

Garber (1980) showed that the keeledtegulae of S. oedipus do not hinder themovements of this species on small-diam-

eter substrates. This study reveals thatSaguinus possesses apical pads and epider-mal ridges that are relatively well developedcompared to those of other callitrichids (e.g.,Callithrix, Leontopithecus). The proficiencyof Saguinus on slender arboreal supportscan therefore be explained by the fact thatits apical pads and terminal phalanges arequite similar in morphology to those of othersmall-branch foragers (e.g., Saimiri). Theresults of this study, therefore, corroborateCartmill’s (1972, 1974, 1985) functional andadaptive model for the origin of expandedapical pads, flattened nails, and graspingextremities within the Order Primates. Spe-cifically, results presented here indicate thatrelatively expanded apical pads and broad,flat ungulae characterize those species thatfrequently forage within the ‘‘small-branchniche’’ of the arboreal milieu. Conversely,claw-like tegulae and reduced apical padsare more useful than flattened ungulae andexpanded apical pads for habitual locomotorand postural behaviors on large-diameterarboreal supports.

The morphoclines in terminal phalanxand apical pad morphology observed amongthe callitrichines examined in this studyraise the question of what pattern might bephylogenetically primitive for callitrichinesand which patterns might therefore repre-sent derived conditions. Garber concludedthat, ‘‘based on the dental anatomy, dietarypatterns, and positional behavior of extanttamarins and marmosets, Saguinus appearsto be the most ecologically generalized’’ (Gar-ber, 1992, p. 479) of the callitrichines. C.jacchus is clearly a more dedicated gumni-vore than Saguinus sp. and Leontopithecusis a specialized extractive forager. The mostparsimonious transformation scheme, inlight of Garber’s (1992) behavioral data andrecent phylogenetic evidence (e.g., Schneideret al., 1993; Schneider and Rosenberger,1996), is that the well-developed apical padsand broad terminal phalanges of S. sciureusapproximate the primitive platyrrhine condi-tion. The more keeled tegulae of S. oedipusare therefore derived from Saimiri-like un-gulae in relation to the evolution of seasonalexudate feeding. Thus, the morphology of S.oedipus tegulae approximates the conditionprimitive for callitrichines. The morphology

Fig. 4. Lateral (left column) and dorsal (right col-umn) views of the third manual terminal phalanx inCallithrix jacchus (top row), Saguinus oedipus (middlerow), and Saimiri sciureus (bottom row).

120 M.W. HAMRICK

of Callithrix and Leontopithecus tegulae rep-resent derived character states, related tothe evolution of habitual gumnivory andextractive foraging behaviors in these twotaxa, respectively.

Hershkovitz (1970) and Napier (1980) sug-gested that the claws of callitrichids are amechanical obstruction to their fingertipsthat interfere with both the use of theirfingertips as tactile organs and with fullphalangeal flexure during single-handed(i.e., prehensile) grips. Lemelin (1994) andLemelin and Grafton (in press) have, how-ever, recently shown that Saguinus is ascompetent as Saimiri at single-handed pre-

TABLE 2. Summary statistics for log–shape ratiovalues showing the mean and (standard deviation)

for each variable for each species*

Saimirisciureus

Saguinusoedipus

Leontopithecusrosalia

Callithrixjacchus

ETUB 2.57 2.62 2.67 2.61(.07) (.12) (.06) (.08)

FTUB 20.4 2.19 2.17 .07(.09) (.10) (.21) (.05)

MSH 2.15 .02 .14 .17(.05) (.07) (.07) (.08)

TUFW 2.17 2.50 2.58 2.70(.04) (.11) (.16) (.15)

BASW .29 .12 .01 2.07(.07) (.05) (.07) (.08)

L 1.01 1.16 1.28 1.14(.06) (.04) (.06) (.04)

* Measurement abbreviations are explained in the text.

Fig. 5. Univariate plots of log-shape ratio values for a: relative midshaft height, b: relative tuftbreadth, c: relative base breadth, and d: relative phalangeal length in four New World anthropoids. Thehorizontal bars represent the range of values for each species and the vertical bars represent the medianvalues for each species. P-values are shown for Mann-Whitney comparisons between species.


hension of small food items such as berriesand crickets. Results presented in this studyagain show that similarities in prehensilecapabilities between Saguinus and Saimirican be explained at least in part by the factthat both Saguinus and Saimiri have rela-tively broad apical pads and well-developedepidermal ridges. Given the fact that Calli-thrix has relatively reduced apical pads it

would be interesting to compare the prehen-sile capabilities of this taxon with those ofSaguinus. If expanded apical pads are aprerequisite for tactile sensitivity and pre-hensility, as some workers have suggested(see below), then we might expect Callithrixto exhibit less prowess at one-handed pre-hension of small food items compared toSaguinus.

This study also demonstrates that termssuch as ‘‘claws’’ and ‘‘nails’’ do not accuratelyreflect the variation in apical pad and termi-nal phalanx morphology seen among thefour monkeys studied. It is obvious fromFigure 5 that there is no single bony feature

TABLE 3. Results of between-species pairwise comparisons (Mann-Whitney U-tests) for the log–shape variablesincluded for analysisa

FTUB(H 5 24.42,P , .001)

MSH(H 5 26.42,P , .001)

TUFW(H 5 24.05,P , .001)

BASW(H 5 27.24,P , .001)

L(H 5 26.27,P , .001)

Ss vs. So ** *** *** *** ***Ss vs. Lr * *** *** *** ***Ss vs. Cj *** *** *** *** ***So vs. Lr ns ** ns ** ***So vs. Cj *** ** ** *** nsLr vs. Cj ** ns ns * ***a Results of Kruskal-Wallis tests are shown in parentheses. No pairwise tests were executed for ETUBL because there was nosignificant difference among the species sampled in this log–shape variable (H 5 6.74, P 5 .08).Ss 5 Saimiri sciureus, So 5 Saguinus oedipus, Lr 5 Leontopithecus rosalia, Cj 5 Callithrix jacchus.* P , .05.** P , .01.*** P , .001.ns 5 P . .05.

TABLE 4. Spearman’s rank correlation coefficients(rho) between the log–shape variables included in the

discriminant analysis and the log–size variable*

Correlationwith log–size

ETUB 2.11(.49)

FTUB 2.13(.41)

MSH 2.05(.77)

TUFW 2.03(.84)

BASW .12(.47)

L .27(.11)

* Rho values are shown above P-values, which are inparentheses.

TABLE 5. Correlations (loadings) between dependentvariables and factor scores for each discriminant axis

of the discriminant analysis (MANOVA) shownin Figure 6

Axis 1 Axis 2 Axis 3

MSH .53 .27 2.25BASW 2.52 2.49 .48TUFW 2.44 2.40 2.38L .43 2.85 .10FTUB .33 .73 .15ETUB 2.09 .22 .08

Fig. 6. Bivariate plot of the first two discriminantaxes for a discriminant analysis of third terminal pha-lanx log-shape ratio values in four New World anthro-poids. The centroids enclose the range of values for eachspecies. Ss 5 Saimiri sciureus, So 5 Saguinus oedipus,Cj 5 Callithrix jacchus, Lr 5 Leontopithecus rosalia.

122 M.W. HAMRICK

that can be used to divide this sample intoclaw- and nail-bearing forms. That is, anydistinction would be an arbitrary one.Rather, as Thorndike (1968) noted for pri-mates as a group, a gradient in form fromnail- to claw-like tegulae exists among thesesmall-bodied New World primates. Thesedata also reveal that differences in anthro-poid apical pad and terminal phalanx mor-phology are tightly correlated with subtledifferences in platyrrhine substrate prefer-ences. Recent morphofunctional studies ofplatyrrhine postcranial morphology (e.g.,Davis, 1996, and references therein) haveshown that postcranial morphology and posi-tional behavior do not always form an exclu-sive one-to-one relationship within thisgroup. The present study serves as an ex-ample of platyrrhine postcranial morphol-ogy closely matching that expected under apredictive functional and adaptive model.This is, however, a relatively narrow com-parison among only four taxa. Future analy-ses should be directed at investigating simi-lar patterns in other monophyletic groups ofarboreal mammals that exhibit diversity intheir substrate preferences and cheiridialmorphology (e.g., burramyid marsupials,lorisiform primates).

Rosenberger (1977) and Szalay (1981) ar-gued that the keeled tegulae of callitrichidsare not homologous with those of otherclawed arboreal mammals such as squirrelsand tree shrews. They noted that callitrich-ids, unlike Sciurus and Tupaia, lack sesa-moids at their distal interphalangeal jointsand also have reduced extensor tubercles ontheir distal phalanges. Quantitative resultsfrom this study show that the taxa includedfor analysis do indeed have reduced exten-sor tubercles (Table 2). Moreover, these taxado not differ significantly from one anotherin the relative size of their extensor tu-bercles. For example, both Saimiri, whichbears flattened ungulae, and Callithrix,which possesses keeled tegulae, have exten-sor tubercles that are similar in relativesize. The large extensor tubercles of squir-rels and tree shrews enable them to force-fully retract their claws from arboreal sup-ports during quadrupedal locomotion. Theclaws of callitrichids, however, do not con-tact arboreal substrates during locomotion

(Rosenberger and Stafford, 1994). The re-sults presented in this study therefore sup-port both Garber’s (1992) hypothesis thatcallitrichid tegulae are a derived posturaladaptation related to exudate feeding andRosenberger’s (1977) hypothesis that callit-richid tegulae are neither morphologicallyhomologous nor functionally analogous tothose of clawed arborealists such as Tupaia.

The functional significance of primatevolar skin morphology

Results from this study suggest that well-developed papillary ridges characterizesmall-branch foragers (e.g., Saguinus,Saimiri), whereas these ridges are not sowell developed in taxa that habitually uselarge-diameter substrates (e.g., Callithrix).What functional advantage(s) do these epi-dermal ridges confer on a small-branch for-ager? The first hypothesis is that the ridgesenhance tactile sensitivity by sending elon-gate intermediate ridges into the dermis(Schlaginhaufen, 1905; Clark, 1936; Loeschand Martin, 1984). The intermediate ridgesserve as levers that transmit a mechanicalstimulus from the papillary ridge to nerveendings located within the dermal cryptsalongside the intermediate ridges (Cauna,1954). The papillary ridges are, therefore,sensors that enable an arboreal primate torapidly assess the size, orientation, flexibil-ity, and movement of the substrate. Thesecond hypothesis is that the primary role ofthe papillary ridges is to enhance pad fric-tion by surface interlocking (Whipple, 1904;Cartmill, 1974, 1985). The dorsally project-ing intermediate ridges simply correspondto the size of the palmarly protruding papil-lary ridges, a fact that led Napier (1980) toliken the papillary ridges to icebergs. Thus,selection favoring well-developed papillaryridges for surface interlocking would pro-duce correlated evolution of well-developedintermediate ridges.

Comparative evidence suggests that thelatter hypothesis is preferable. Humans havevery keen tactile sensitivity but have rela-tively flat papillary ridges (Fig. 7). Ourintermediate ridges are numerous but, rela-tive to the overall size of our apical pads,they are not very elongate and do not projectvery deeply into the dermis (Fig. 7). Our


Fig. 7. Photomicrographs of dorsopalmar sections through the pollical apical pad in Didelphisvirginiana (top; 3120; scale bar 5 .25 mm) and Homo sapiens (bottom; 380; scale bar 5 .50 mm). Notethat in Didelphis the stratum corneum (sc) bears undulating papillary ridges which give rise to deeplypenetrating intermediate ridges (ir) whereas in Homo the intermediate ridges (ir; arrows) are reducedrelative to the papillary ridges at the skin’s surface.

124 M.W. HAMRICK

enhanced tactile sensitivity derives fromnumerous rapidly adapting (Meissner’s cor-puscles) and slowly adapting (Merkel end-ings) mechanoreceptors located within thedermis close to the intermediate and lateralridges (Winkelmann, 1963; Kandel et al.,1991). Hence, well-developed papillary ridgesare not necessarily a prerequisite for tactilesensitivity. This hypothesis is corroboratedfurther by the morphology of raccoon gla-brous volar skin. The raccoon’s palmar padsare richly innervated by slowly adaptingmechanoreceptors located between undulat-ing limiting ridges (Munger and Pubols,1972). The papillary ridges and associatedintermediate ridges are, however, poorly de-veloped in raccoon digital skin (see Mungerand Pubols, 1972).

Additional evidence to support this me-chanical argument is provided by the marsu-pial Didelphis. This climbing opossum has adivergent grasping hallux like primates andalso exhibits well-developed papillary ridgesassociated with elongate intermediate ridgeson its thumb and finger pads (Fig. 7). Didel-phis, however, possesses arm and hand so-matic sensory projections within its neocor-tex that are poorly developed compared tothose of both Procyon and Homo (Welkerand Seidenstein, 1959; Johnson, 1977; Kan-del et al., 1991). In other words, the brain ofDidelphis, in contrast to that of both Homoand Procyon, is relatively unspecialized fortactile reception. The well-developed papil-lary and intermediate ridges of Saimiri,Saguinus, and Didelphis, and the morepoorly developed papillary and intermediateridges of Homo and Procyon, suggest thatwell-developed papillary ridges evolved inthe majority of primates in order to facilitatethe mechanical, rather than sensory, proper-ties of the apical pads.

A scenario for primate claw loss

If keeled tegulae do not hinder the move-ments of arboreal mammals on small-diam-eter arboreal substrates, then why wouldevolution favor the loss of claws on all digitsof the primate hand and foot? I propose that1) the habitual use of small-diameter sup-ports by the earliest primates favored theevolution of relatively large and broad (radi-oulnarly) apical pads on the fingers and toes

for the mechanical reasons discussed below,2) the overall size and breadth of the apicalpads are increased by increasing the breadthof the terminal phalanges, so that selectionfor large, broad apical pads favored theevolution of broad terminal phalanges in theearliest primates, 3) broad, flat ungulae onthe hands and feet are a correlated by-product of having broad terminal phalangeson the fingers and toes, that is, the three-dimensional shape of the superficial anddeep corneous layers conform to the shape ofthe terminal phalanx, and 4) primates re-tained papillary ridges on their fingers andtoes from a clawed arboreal ancestor butthese ridges were more well-developed inprimates due to an overall increase in apicalpad size.

This adaptive hypothesis is supported bythe fact that Anolis lizards living in a small-branch environment have relatively broaderapical pads on their toes than their conspecif-ics living in a large-branch environment(Losos et al., 1997). A functional explanationunderlying this phenomenon is offered byHildebrand (1995), who suggested that alarge, flexible apical pad increases stabilityof the digit on small-diameter arboreal sup-ports because the pad can contact the sub-strate in several planes. This in turn enablesthe pad to resist, via friction, adhesion, orsurface interlocking, disruptive forces fromdifferent directions. Cartmill (1974) hasshown that increasing the size of the apicalpad does not increase the pad’s frictionalproperties (Amonton’s first law). A largerpad bearing interlocking or adhesive asperi-ties (e.g., setae, scales, or papillary ridges)does, however, increase stability of the padon the support because the number of con-tact points between these aperities and thesubstrate is increased. This mechanicalmodel explains why both reptiles and mam-mals living in a small-branch environmenthave enlarged apical pads on their fingersand toes, even though both have quite differ-ent patterns of volar skin structure andinnervation.

CONCLUSIONS

This study was undertaken in order totest predicted morphoclines in hand apicalpad and terminal phalanx morphology


among four small-bodied (,1 kg) New Worldmonkeys. Results show that as the fre-quency of small-branch foraging increasesamong taxa within this sample relative dis-tal phalanx breadth increases but distalphalanx length, height, and flexor tuberclesize decrease. Moreover, epidermal ridgedevelopment becomes more pronounced asthe frequency of small-branch foraging in-creases. Terminal phalanx breadth and epi-dermal ridge complexity are both positivelycorrelated with apical pad size in this sam-ple of New World monkeys. Apical pad sizein turn appears to be positively correlatedwith stability of the digit on small-diametersupports. The platyrrhines studied exhibitcontinuous interspecific variation in thesehand features, revealing that there is noclear morphological basis for distinguishingbetween claw- and nail-bearing New Worldmonkeys. These observations corroborateCartmill’s (1972, 1974, 1985) functional andadaptive model for the loss of claws inprimates. That is, relatively expanded api-cal pads and broad, flat ungulae character-ize those species that frequently foragewithin the ‘‘small-branch niche’’ of the arbo-real milieu, whereas keeled tegulae andreduced apical pads are observed in specieswhich exhibit a preference for positionalbehaviors on medium- and large-diameterarboreal supports. Moreover, these resultssupport previous suggestions that the keeledtegulae of callitrichines are a derived pos-tural adaptation related to the evolution ofseasonal exudate feeding in the last commonancestor of the Callitrichinae.

ACKNOWLEDGMENTS

I am grateful to Dr. R.W. Thorington, Jr.and Ms. Linda Gordon at the National Mu-seum of Natural History and Dr. A. Kramerand Ms. L. Flournoy at the University ofTennessee for access to skeletal and cadaverspecimens in their care. Dr. Kathleen K.Smith, Duke University Medical Center, pro-vided much of the expertise needed for thehistology work presented here; her assis-tance is greatly appreciated. Morphometricdata were collected in the Duke UniversityMorphometrics Laboratory under the direc-tion of Dr. John Mercer. Dr. L. Digby pro-vided helpful information on New World

monkey behavior and ecology and Drs. D.Schmitt and M. Cartmill provided helpfulinsights on the problem of primate claw loss.Drs. P. Lemelin, M. Norconk, M. Dagosto,and two anonymous reviewers provided com-ments and suggestions that improved thequality of the manuscript. I also thank P.Lemelin and B. Grafton for providing mewith an unpublished version of their manu-script on manual prehensility in Saimiriand Saguinus.

LITERATURE CITED

Aronsen GP (1997) The functional anatomy of theforelimb in Callimico goeldii and Leontopithecus rosa-lia: A study integrating kinematic and morphologicaldata. Am. J. Phys. Anthropol. Suppl. 24:68.

Bock WJ, and Miller W (1959) The scansorial foot of thewoodpeckers with comments on the evolution of perch-ing and climbing feet in birds. Am. Mus. Nat. Hist.Novitates 1931:1–45.

Boinski S (1989) The positional behavior and substrateuse of squirrel monkeys: Ecological implications. J.Hum. Evol. 18:659–677.

Bryant HN, Russell A, Laroiya R, and Powell G (1996)Claw retraction and protraction in the carnivora:Skeletal microvariation in the phalanges of the Feli-dae. J. Morph. 229:289–308.

Burgess A, and Perl C (1973) Cutaneous mechanorecep-tors and nociceptors. In A Iggo (ed): Handbook ofSensory Physiology, Vol. II. New York: Springer-Verlag, pp 29–78.

Cartmill M (1972) Arboreal adaptations and the originof the order Primates. In R Tuttle (ed): Functional andEvolutionary Biology of Primates. Chicago: Aldine-Atherton, pp 97–122.

Cartmill M (1974) Pads and claws in arboreal locomo-tion. In FA Jenkins Jr (ed): Primate Locomotion. NewYork: Academic Press, pp 45–83.

Cartmill M (1979) The volar skin of primates: Itsfrictional characteristics and their functional signifi-cance. Am. J. Phys. Anthropol. 50:497–510.

Cartmill M (1985) Climbing. In M Hildebrand, DBramble, K Liem, and D Wake (eds): FunctionalVertebrate Morphology. Cambridge: Harvard Univer-sity Press, pp 73–88.

Cauna N (1954) Nature and functions of the papillaryridges of the digital skin. J. Anat. 119:449–468.

Cauna N (1956) Nerve supply and nerve endings inMeissner’s corpuscles. Am. J. Anat. 99:315–350.

Clark WE LeGros (1936) The problem of the claw inprimates. Proc. Zool. Soc. Lond. 1936:1–24.

Clark WE LeGros (1959) The Antecedents of Man.Edinburgh: Edinburgh University Press.

Dagosto M (1988) Implications of postcranial evidencefor the origin of euprimates. J. Hum. Evol. 17:35–56.

Davis LC (1996) Functional and phylogenetic implica-tions of ankle morphology in Goeldi’s monkey (Cal-limico goeldii). In M Norconk, A Rosenberger, and PGarber (eds): Adaptive Radiations of Neotropical Pri-mates. New York: Plenum Press, pp 133–156.

Dietz J, Peres CA, and Pinder L (1997) Foraging ecologyand use of space in wild golden lion tamarins (Leonto-pithecus rosalia). Am. J. Primatol. 41:289–305.

Dykyj D (1982) Allometry of trunk and limbs in NewWorld monkeys. Doctoral Dissertation, City Univer-sity of New York.

126 M.W. HAMRICK

Falsetti AB, Jungers WL, and Cole TM (1993) Morpho-metrics of the callitrichid forelimb: A case study in sizeand shape. Int. J. Primatol. 14:551–572.

Feduccia A (1993) Evidence from claw geometry indicat-ing arboreal habits of Archaeopteryx. Science 259:790–793.

Fleagle JG, Mittermeier R, and SkopecA(1981) Differen-tial habitat use by Cebus apella and Saimiri sciureusin central Surinam. Primates 22:361–367.

Fontaine R (1990) Positional behavior in Saimiri boli-viensis and Ateles geoffroyi. Am. J. Phys. Anthropol.82:485–508.

Garber PA (1980) Locomotor behavior and feeding ecol-ogy of the Panamanian tamarin (Saguinus oedipusgeoffroyi). Int. J. Primatol. 1:185–201.

Garber PA (1992) Vertical clinging, small body size, andthe evolution of feeding adaptations in the Callitrichi-nae. Am. J. Phys. Anthropol. 88:469–482.

Garber PA, and Sussman R (1984) Ecological distinc-tions between sympatric species of Saguinus andSciurus. Am. J. Phys. Anthropol. 65:135–146.

Haffner M (1996) A tendon-locking mechanism in twoclimbing rodents, Muscardinus avellanarius and Mi-cromys minutus (Mammalia, Rodentia). J. Morph.229:219–227.

Halata M (1975) The mechanoreceptors of the mamma-lian skin: Ultrastructure and morphological classifica-tion. Adv. Anat. Embryol. Cell Bio. 50:3–75.

Hershkovitz P (1970) Notes on Tertiary platyrrhinemonkeys and a description of a new genus from thelate Miocene of Columbia. Folia Primatol. 12:1–37.

Hildebrand M (1995) Analysis of Vertebrate Structure.New York: John Wiley & Sons, Inc.

Humason GL (1972) Animal Tissue Techniques. SanFrancisco: W.H. Freeman.

Johnson JI Jr. (1977) Central nervous system of marsu-pials. In D Hunsaker II (ed): The Biology of Marsupi-als. New York: Academic Press, pp 159–278.

Jungers WL, Falsetti AB, and Wall C (1995) Shape,relative size, and size-adjustments in morphometrics.Yrbk. Phys. Anthropol. 38:137–161.

Kandel ER, Schwartz JH, and Jessell TM (1991) Prin-ciples of Neural Science. Norwalk:Appleton and Lange.

Lacher T, Bouchardet G, Alves C, and Magalhaes-CastroB (1984) Parasitism of trees by marmosets in a centralBrazilian gallery forest. Biotropica 16:202–209.

Lemelin P (1994) Manual grasping behavior in Sagui-nus midas and its relevance for early primate evolu-tion. Am. J. Phys. Anthropol. 18:128–129.

Lemelin P, and Grafton B (in press) Grasping perfor-mance in Saguinus midas and the evolution of handprehensility in primates. In E Strasser and J Fleagle(eds): Primate Locomotion: Recent Advances. NewYork: Plenum Press.

Loesch DZ, and Martin NG (1984) Finger ridge patternsand tactile sensitivity. Ann. Hum. Biol. 11:113–124.

Losos JB, Warheit KI, and Schoener TW (1997) Adaptivedifferentiation following experimental island coloniza-tion in Anolis lizards. Nature 387:70–73.

MaCleod N, and Rose KD (1993) Inferring locomotorbehavior in paleogene mammals via eigenshape analy-sis. Am. J. Sci. 293A:300–355.

Martin RD (1990) Primate Origins and Evolution. NewJersey: Princeton University Press.

Mosimann JE, and James FC (1979) New statisticalmethods for allometry with applications to Floridared-winged blackbirds. Evolution 23:444–459.

Munger BL, and Pubols LM (1972) The sensorineuralorganization of the digital skin of the raccoon. BrainBehav. Evol. 5:367–393.

Napier JR (1980) Hands. New Jersey: Princeton Univer-sity Press.

Neff NA, and Marcus LF (1980) A Survey of MultivariateMethods for Systematics. New York: American Mu-seum of Natural History.

Peres CA (1989) Exudate-eating by wild golden liontamarins, Leontopithecus rosalia. Biotropica 21:287–288.

Preuschoft H (1970) Statische untersuchungen an denfussen der primaten. Teil II: statik des ganzen fusses.Z. Anat. Entw. Gesch. 131:156–192.

Preuschoft H (1973) Functional anatomy of the upperextremity. In G Bourne (ed): The Chimpanzee. Balti-more: University Park Press, pp 34–120.

Rosenberger A (1977) Xenothrix and ceboid phylogeny. J.Hum. Evol. 6:461–481.

Rosenberger A, and Stafford B (1994) Locomotion incaptive Leontopithecus and Callimico: A multimediastudy. Am. J. Phys. Anthropol. 94:379–394.

Schlaginhaufen O (1905) Das hautleistebsystem derprimatenplanta unter mitberucksichtigung der palma.Morph. Jahrb. 34:1–125.

Schneider H, and Rosenberger AL (1996) Molecules,morphology, and platyrrhine systematics. In M Nor-conk, A Rosenberger, and P Garber (eds): AdaptiveRadiations of Neotropical Primates. New York: Ple-num Press, pp 3–19.

Schneider H, Schneider MPC, Sampaio MIC, HaradaML, Stanhope M, and Goodman M (1993) Molecularphylogeny of the New World monkeys (Platyrrhini,Primates), Mol. Phyl. Evol. 2:225–242.

Smith RJ (1980) Rethinking allometry. J. Theor. Biol.87:97–111.

Sokal R, and Rohlf F (1981) Biometry. New York: W.H.Freeman.

Stafford BJ, Rosenberger AL, Baker AJ, Beck BB, DietzJM, and Kleiman DG (1996) Locomotion of golden liontamarins (Leontopithecus rosalia). In M Norconk, ARosenberger, and P Garber (eds): Adaptive Radiationsof Neotropical Primates. New York: Plenum Press, pp111–132.

Sussman RW, and Kinzey WG (1984) The ecological roleof the Callitrichidae: A review. Am. J. Phys. Anthropol.64:419–449.

Szalay FS (1981) Functional analysis and the practice ofthe phylogenetic method as reflected by some mamma-lian studies. Am. Zool. 21:37–45.

Thorndike E (1968) A microscopic study of the marmosetclaw and nail. Am. J. Phys. Anthropol. 28:247–262.

Van Valkenburgh B (1987) Skeletal indicators of locomo-tor behavior in living and extinct carnivores. J. Vert.Paleont. 7:162–182.

Welker WI, and Seidenstein S (1959) Somatic sensoryrepresentation in the cerebral cortex of the raccoon. J.Comp. Neur. 111:469–501.

Whipple IL (1904) The ventral surface of the mamma-lian cheiridium, with special reference to conditionsfound in man. Z. Morph. Anthropol. 7:261–368.

Winkelmann RK (1963) Nerve endings in the skin ofprimates. In J Buettner-Janusch (ed): Evolutionaryand Genetic Biology of Primates, Vol. 1. New York:Academic Press, pp 229–259.

Zar JH (1984) Biostatistical Analysis. Englewood Cliffs,NJ: Prentice-Hall.


functional and adaptive significance of primate pads and claws: evidence from new world anthropoids

Documents