Evolution of Grasping Among Anthropoids

Home , Evolution of primates, Oreopithecus

doi: 10.1111/j.1420-9101.2008.01582.x

Evolution of grasping among anthropoids

E. POUYDEBAT,* M. LAURIN, P. GORCE* & V. BELSà *Handibio, Universite´ du Sud Toulon-Var, La Garde, France Comparative Osteohistology, UMR CNRS 7179, Universite´ Pierre et Marie Curie (Paris 6), Paris, France àUMR 7179, MNHN, Paris, France

Keywords: Abstract behaviour; The prevailing hypothesis about grasping in primates stipulates an evolution grasping; from power towards precision grips in hominids. The evolution of grasping is hominids; far more complex, as shown by analysis of new morphometric and behavio- palaeobiology; ural data. The latter concern the modes of food grasping in 11 species (one phylogeny; platyrrhine, nine catarrhines and humans). We show that precision grip and precision grip; thumb-lateral behaviours are linked to carpus and thumb length, whereas primates; power grasping is linked to second and third digit length. No phylogenetic variance partitioning with PVR. signal was found in the behavioural characters when using squared-change parsimony and phylogenetic eigenvector regression, but such a signal was found in morphometric characters. Our ﬁndings shed new light on previously proposed models of the evolution of grasping. Inference models suggest that Australopithecus, Oreopithecus and Proconsul used a precision grip.

very old behaviour, as it occurs in anurans, crocodilians, Introduction squamates and several therian mammals (Gray, 1997; Grasping behaviour is a key activity in primates to obtain Iwaniuk & Whishaw, 2000). On the contrary, the food. The hand is used in numerous activities of manip- precision grip, in which an object is held between the ulation and locomotion and is linked to several func- distal surfaces of the thumb and the index finger, is tional adaptations (Godinot & Beard, 1993; Begun et al., usually viewed as a derived function, linked to tool use 1997; Godinot et al., 1997). In particular, the hand is and human morphological autapomorphies (Napier, involved in prehension, such as gripping of static foods 1956; Tuttle, 1965; Schultz, 1969; Susman, 1979, 1989; (fruits, leaves) and dynamic foods such as insects or other Marzke et al., 1992; Clark, 1993). The precision grip has prey (frogs, rodents, small antelopes). Some primates been considered the most important hand function of all such as chimpanzees (Pan troglodytes) and capuchins prehensile movements (Napier, 1980). (Cebus apella) use their hands to manipulate tools, to Our aim is to reconsider this simple model of grasping crack nuts, for example (Boesch & Boesch, 1990; Fraga- evolution in the light of morphometric data from szy et al., 2004), whereas gorillas (Gorilla gorilla) use their numerous species of primates and behavioural consider- hands to extract food from holes (Pouydebat et al., 2005). ations such as areas of contact between the fingers and The evolution of primates (humans included) is linked to the food grasped by extant primates. Therefore, the the development of those behaviours allowing organisms possible presence of a phylogenetic signal in the behavio- to exploit the resources in their environment. A general ural and relevant morphometric characters is investi- model of grasping in primates proposes an evolution from gated and the correlation between morphometric and a ‘power grip’ towards a ‘precision grip’, supposed to behavioural characters is also determined. We also have taken place in hominids; the precision grip has been present models that enable inference of behaviours from suggested to appear with Australopithecus afarensis (Mar- morphological characters, which we use to infer behav- zke, 1997) or with Homo (Napier, 1956, 1960). The power iours in three extinct primates: Proconsul africanus, grip is defined as a grasp with the palm, and is probably a Oreopithecus bambolii and Australopithecus afarensis, three species considered to have divergent grasping abilities.

Correspondence: Michel Laurin, Comparative Osteohistology, UMR CNRS Proconsul africanus and Australopithecus afarensis are from 7179, Universite´ Pierre et Marie Curie (Paris 6), Paris, France. Africa, which is probably the cradle of hominoid diver- Tel.: (33) 1 44 27 36 92; e-mail: [email protected] siﬁcation (Arnason et al., 2000; Folinsbee & Brooks,

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 1732 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1733

2007). Oreopithecus bambolii was found in Tuscany, Italy. object by using frame-by-frame analysis in the labora- It is included here because its prehensile behaviour has tory. We obtained a minimum of 90 min of observation been inferred in the literature (Moya´-Sola´ et al., 1999; of grasping behaviour for each chimpanzee, baboon, Susman, 2004). macaque and capuchin, and 180 min for each orang- utan, gorilla and gibbon. Material and methods Frame-by-frame analysis was performed with a Basler camera (Basler, Ahrensburg, Germany), recording 250 images per second. Each prehension technique was Quantification of areas of contact characterized by contacts between one or several lateral Subjects or ventral areas of a minimum of two digits or the The data represented in this study are based on a wide complete palm. From this analysis, we determined five variety of primates observed in captivity (Appendix S1): categories of object prehension. nine capuchins (C. apella), nine macaques (Macaca fuscata), nine baboons (Papio papio), three gibbons Size and nature of the objects (Hylobates lar), seven orang-utans (Pongo pygmaeus), three For all primates except humans, the objects were small gorillas (G. gorilla) and 14 chimpanzees (P. troglodytes). and scattered on the ground; the objects involved We also have observations from nine children, 2–5 years spherical cereals and fruits. In humans, the objects were of age (Homo sapiens), and nine adults (H. sapiens). Data spherical pearls. It was necessary to standardize the for three other species were collected from the literature diameter and the volume in order to calibrate these (Christel, 1993; Christel et al., 1998): black mangabey parameters according to the length of the hand of the (Cercocebus aterrimus), geladas (Theropithecus gelada) and species studied. In this paper, we always presented bonobos (Pan paniscus). These species represent a wide spherical objects to the animals and determined their array of body size, hand morphological traits and diameter. The diameter of the objects was calibrated anthropoid taxa. Indeed, capuchins do not possess an according to the length of the hand of the species. As we opposable thumb and none of the studied primate species knew the length of the hand of the smallest studied except humans has morphological traits usually associ- species (76.2 ± 5.3 mm for capuchin) and the diameter ated with precision grip. of the smallest object (3.0 ± 0.1 mm) grasped by this species, we deduced the diameter of objects for other Protocol of observations species as follows (D = diameter, L = length, all units The observations of grasping of small and large objects in mm): have been made in various groups of animals belonging to zoological gardens in France. All individuals observed D object for species x ¼L hand of the species x3:0=76:2 in any given species belong to a single group and the For example, to calculate the diameter of objects to be hierarchical position of each specimen was established. grasped by chimpanzees, we used the length of the The animals were observed without modification of their chimpanzee’s hand (235.0 mm) and that of the smallest social (within their group) or environmental (e.g. logs, hand (the capuchins’ hand: 76.2 mm) and the diameter rocks, ropes) context to maintain: (i) all behavioural of the smallest object (3.0 mm). In this example, Dxc interactions between the members of the group; (ii) all corresponds to the determined diameter of the small constraints in relation to the environment; and (iii) all object for chimpanzees (c). We calculated the following possibility of opportunistic manipulation (Parker & value for objects in chimpanzees: Dxc is equal to 9.0 mm Gibson, 1977). (235.0 mm · 3.0 ⁄ 76.2). We followed the same method All observations of the animals were made for to calculate the diameter of objects for each species 7 months (Pouydebat, 2004). The duration of observa- (Appendix S2). tion for each specimen was standardized following the usual methods suggested in comparative ethology Number of grasps (Lehner, 1996). A preliminary analysis was conducted by A total of 5549 grasps were recorded for the eight studied ‘ad libitum sampling’ (Altmann, 1974) that permits the species (Table 1). The percentage of each prehension individual recognition of all subjects for each species and category was calculated on the basis of the total number the identification of a wide variety of areas of fingers in of grasping observed in each species. contact with the presented objects. During the study, each individual was observed according to the method of Morphometric data ‘focal animal sampling’ (Altmann, 1974). We filmed the Morphometric data were obtained from hand skeletons animals during two sessions of 2 h each for chimpanzees, belonging to the collection of the Museúm National baboons, capuchins and macaques and six sessions of 2 h d’Histoire Naturelle (Paris). Our sample consisted of 17 each for orang-utans, gorillas and gibbons. Every 15 min, measurements of the hand of 26 taxa (Appendix S3). A sequences of grasping which lasted 5 min were analysed mean of 10 specimens per taxon (males and females) was to determine the area of the finger in contact with the measured.

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY 1734 E. POUYDEBAT ET AL.

Table 1 Use of the grasps from all categories in anthropoid species.

Category 1 Category 2 Category 3 Category 4 Category 5 Species N Mean Precision Thumb-distals Thumb-lateral Without thumb Power

Homo sapiens (adults) 600 50.0 ± 0.0 81 ± 3.5 19 ± 2.2 0 ± 0 0 ± 0 0 ± 0 Homo sapiens (child) 450 50.0 ± 0.0 59 ± 4.3 34 ± 1.4 0 ± 0 0 ± 0 7 ± 1.5 Pan troglodytes 804 57.4 ± 4.9 32 ± 2.6 7 ± 0.7 38 ± 2.2 22 ± 3.1 1 ± 0.5 Pan paniscus * * 15 ± 1.7 1 ± 0.2 45 ± 4.0 39 ± 3.2 0 ± 0 Gorilla gorilla 600 200.0 ± 6.7 67 ± 4.4 2 ± 1.1 17 ± 2.0 12 ± 1.9 2 ± 0.3 Pongo pygmaeus 624 89.1 ± 4.7 29 ± 1.3 3 ± 1.3 33 ± 3.1 34 ± 2.0 1 ± 0.4 Hylobates lar 546 182.0 ± 5.3 5 ± 1.2 0 ± 0 95 ± 2.1 0 ± 0 0 ± 0 Papio papio 581 64.6 ± 5.9 71 ± 5.2 3 ± 2.0 24 ± 2.8 0 ± 0 2 ± 0.6 Macaca fuscata 667 74.1 ± 6.2 76 ± 3.9 1 ± 0.3 22 ± 2.4 0 ± 0 1 ± 0.4 Theropithecus gelada * * 77 ± 3.3 2 ± 1.2 18 ± 1.1 0 ± 0 3 ± 1.2 Cercocebus aterrimus * * 70 ± 4.6 0 ± 0 28 ± 1.7 0 ± 0 2 ± 0.6 Cebus apella 677 75.2 ± 7.0 80 ± 4.1 13 ± 2.1 2 ± 0.4 4 ± 1.6 1 ± 0.3

N, Number of grasps of small objects. The mean number of observed grasps per individual for each species is also given. All other columns represent percentages. *Data derived from the literature (Christel, 1993; Christel et al., 1998).

Category 1: contact between the distal phalanx of the Categories of contacts thumb, the distal part of the index finger and Each of the six species of primates uses from 5 to 21 of 26 the object (precision). different modes of contacts between areas of digits and Category 2: contact between the distal phalanx of at the objects. This large number of contacts can be least three fingers and the object (thumb- classified into five main categories of grasping behaviour distals). (Table 1). In order to facilitate our comparison with the Category 3: contact between the distal part of the thumb, previous literature, each category is named (Fig. 1) as the lateral side of the middle and proximal suggested by Napier (1956) and Jones-Engel & Bard phalanges of the index finger and the object (1996): (thumb-lateral).

(a) (b)

Fig. 1 Definition of contacts (drawings modified from Schultz, 1969). (a) Precision (two digits). (b) Thumb-distal (3, 4 or 5 digits). (c) Thumb-lateral. (d) Without thumb. (e) Power. The arrows indicate the possible contact between the thumb and the other digits (a–c), and between the index and middle fingers (d). In the last prehension mode (e), the palm completely covers the object and the ventral surface of the digits is also implicated in the grasping action. These surfaces are presented in black and the arrows indicate possible modes of contact between the object and the digit surfaces. In the first example (a), the pad of the thumb and the pad or the extremity of the index can be simultaneously in contact with the object (photographs: E. Pouydebat).

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1735

Category 4: contact between one or several fingers, alternative trees (trees 2–5 in Appendix S4) that derive except the thumb, and the object (without from our main tree. Tree 2 differs from tree 1 in having thumb). older divergence dates for some nodes, especially among Category 5: contact between the palm, one or several hominoids (Fig. 2, grey). Tree 3 was produced by a fingers and the object (power). natural logarithmic transformation of the branch lengths of tree 1. Tree 4 was produced by setting all branch lengths to 1. Tree 5 is ultrametricized from tree 4. For the Data distribution and transformation behavioural data, the same trees were used, but taxa The length of the manus (LM) is used in the analyses with missing data were pruned from the trees. These below to remove some of the body size effect on the trees deviate increasingly from the starting tree that, we remaining characters. For all analyses, this character believe, includes plausible divergence times; thus, for all (LM) was log-transformed because body size usually analyses, we used the tree with the lowest designator follows a log-normal, rather than normal, distribution. possible (tree 1, and if not possible, tree 2, and so on). All All other morphometric characters were divided by LM these trees can be used to the extent that ‘the statistical before all the analyses below (Appendix S3). adequacy of any proposed branch lengths should be viewed as an empirical issue’ and ‘In general, any transformation of possible use for tip data…might also be tried Detection of phylogenetic signal for branch lengths’ (Garland et al., 1992: pp. 23–24). To determine whether or not the phylogeny needed to be A second set of tests of phylogenetic signal was incorporated into the analyses, and whether or not performed using phylogenetic eigenvector regression squared-change parsimony could be used to study char- (PVR) analysis (Diniz-Filho et al., 1998). This method acter evolution, we performed two types of tests of relies on a principal coordinate analysis of the phylo- phylogenetic signal. The first one consists of comparing the squared length of a character over the reference tree to the squared length of multiple (in this case 10 000) random trees. For quantitative characters optimized through squared-change parsimony, branch length data are critical. Thus, the most appropriate way to create random trees is to reshuffle the terminal taxa on a tree of fixed topology and branch lengths (Laurin, 2004; Laurin et al., 2004). However, squared-change parsimony optimization requires the same assumptions as independent contrast analysis; thus, we checked if these assumptions were met using the PDAP (Phenotypic Diversity Analysis Programs) module for MESQUITE (Midford et al., 2003). This module performs four relevant tests. The first three regress the absolute value of standardized contrast against: (i) their expected standard deviation (the square root of the sum of corrected branch lengths); (ii) the estimated value of the base node; and (iii) the corrected height of the base node. The fourth and last test is a regression of the estimated value of the base node against the corrected height of the base node. We performed all these tests (four) for all characters (21) for all trees (five). No corrections for multiple tests were made, which makes our procedure more stringent by rejecting trees which yield artefactual relationships which are significant when taken in isolation, but which might no longer be significant if such corrections were made. Further- more, it is not clear if such corrections should be made because these four tests evaluate different statistical Fig. 2 Time-calibrated supertree of primates used in this study. Tree 1 is in black; tree 2, with older divergence dates in Catarrhini, is in artefacts, and results for one character have no bearing grey. This figure was produced from a pdf file generated by on other characters. Deviations from the assumptions MESQUITE (Maddison & Maddison, 2006) and showing the tree in a were detected in several cases; this prevented analyses of stratigraphic time scale displayed by Stratigraphic Tools (Josse et al., the relevant characters on a given tree. To maximize the 2006). The numbers next to the tree indicate the age (Ma); those number of tests that could be performed and to test the next to the stage names indicate their duration (Ma). [A colour presence of a phylogenetic signal, we produced four version of this figure is available online.]

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY 1736 E. POUYDEBAT ET AL.

genetic distance matrix to extract eigenvectors that are using variance partitioning with a PVR analysis (Diniz- used in a standard linear regression against the character Filho et al., 1998; Desdevises et al., 2003). This method of interest. The eigenvectors represent the position of the incorporates the phylogeny into the analysis in the form taxa on the various principal coordinate axes. However, of principal coordinate axes, as explained in the section for n taxa, n – 1 principal coordinate axes are produced, on phylogenetic signal detection (above). We performed and they cannot all be used in a linear regression analysis some exploratory analyses (results not shown) in or there would be no degrees of freedom left. The axes PERMUTE (Casgrain, 2005) to choose the characters to we used were selected using a broken-stick model analyse through PVR. Correlation between these char- (Diniz-Filho et al., 1998), because it would have taken acters was then tested through variance partitioning with too long to test for a significant relationship between all PVR, using principal coordinate axes that were signi- axes and all characters separately (21 characters and up ficantly correlated with the dependent (behavioural) to 26 taxa yield 546 tests). Furthermore, none of the characters. These axes were selected by performing a behavioural characters exhibits a phylogenetic signal simple regression of all axes against the relevant according to the squared-change parsimony analysis (see behavioural characters, because in this case only three below), and that method is usually more powerful than characters are involved. PVR (Cubo et al., 2005). Thus, the broken-stick model was the only method applicable to all of our data. In all Linear regression models our analyses of phylogenetic signal, only the first two axes were used. They represent 60.7% of the phylo- Simple or multiple linear regressions were used to genetic variance. produce inference models of the behavioural characters. The phylogenetic distance matrix was obtained from These models are based on the morphometric characters tree 1 using the Stratigraphic Tools module (Josse et al., that are significantly correlated with the behavioural 2006). The principal coordinate analysis was performed characters according to the variance partitioning analysis in PROGICIEL R (Casgrain et al., 2004). Linear regressions with PVR (Appendix S5) or according to linear regres- were tested for statistical significance using 9999 per- sions, but they were constructed without incorporation mutations of the dependent variables (here, the mor- of the phylogeny. This is unavoidable because principal phometric or behavioural data) in PERMUTE! (Casgrain, coordinate axes have no absolute meaning (they differ 2005). A regression is significant (at a 0.05 threshold when a taxon is added or removed, or if the topology or value) if fewer than 5% of the data sets have an R2 branch length is changed), so incorporating them into value at least as large as the original data set (the predictive models would preclude their use in palaeobi- original, unpermuted set is included). The advantage of ological inference, which is self-defeating. These models using permutations to test the significance of the could not be used on some of the extinct taxa included in relationship is that this method requires far fewer our study because the relevant morphometric characters assumptions about the distribution of the data. Thus, are not known. To maximize the number of inferences contrary to the other method, this test could be applied which could be drawn about these taxa, we built to all characters. additional inference models for several combinations of these taxa and available osteological characters (Appen- dix S6). For this purpose, we used a forward selection Phylogenetically independent contrasts procedure in PERMUTE to select the characters, among We assessed correlations between the behavioural those that were significantly correlated with each other (dependent) and morphometric (independent) charac- according to the variance partitioning or the indepen- ters using phylogenetically independent contrasts dent contrast analyses. In two cases, to produce the (Felsenstein, 1985), whenever the assumptions of models, we had to extend character selection to other that method were met on at least one of our trees. These characters (because extinct taxa are incompletely tests were performed using the PDAP module for known); in such cases, we used a forward selection MESQUITE (Midford et al., 2003). As for the test a procedure in PERMUTE (with p to enter of 0.1) to build phylogenetic signal using squared-change parsimony, the model. we performed this test on the tree with the lowest designator (tree 1 if possible; if not, tree 2, etc.) that gave Results adequate contrast standardization for both characters analysed. Taxon-specific distribution of categories The platyrrhine and all catarrhines were able to modulate Variance partitioning with phylogenetic eigenvector their grasping behaviour for small food items. However, regression some clusters of species emerge for all categories of Correlations between the behavioural (dependent) and contact. These groups can be compared on the basis of morphometric (independent) characters were also tested the mean percentage (Table 1).

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1737

From these data, five main conclusions can be drawn. Character correlation assessed using phylogenetic First, great apes (except humans) and capuchins use independent contrasts grasping category 4 (without thumb), whereas humans and other primates never do. All primates except Only three of the behavioural characters were analysed humans use the lateral part of their index (category 3, using independent contrasts because for the others, the thumb-lateral), whereas capuchins hardly ever do and assumptions of that method were not met (Appendix S7). humans never do. Secondly, adult humans and capu- Some of the morphometric characters are clearly corre- chins are similar as regards a majority of grasping lated with behavioural data. The behavioural character categories. Thirdly, human children do not show clear ‘precision’, involving the contact of the distal phalanges similarities with other primate species for any of the of the thumb and the index, and the character ‘thumb- grasping categories. Fourthly, cercopithecids (i.e. maca- lateral’, involving contact of the distal phalanx of the ques and baboons), capuchins and humans use the thumb and the lateral side of the index, are correlated precision grip most often (Table 1). Fifthly, humans with the length of the carpus and the first ray (the thumb present some unique characteristics in their selection of and its metacarpal). Power grasping is linked with second grasping categories. Human adults are the only primates and third digit length. to use exclusively the tips of the fingers (precision and thumb distals). Human children and capuchins are more Character correlation assessed using variance similar behaviourally to human adults than to other partitioning with PVR primates in our sample, because they mainly use the tips of their fingers (93%). One key result from our data is The forward selection test in PERMUTE (with a p to enter that all species were able to grasp small objects with the of 0.1) resulted in only one or two morphometric precision grip corresponding to the contact between the characters being selected for each behavioural character tips of thumb and index finger. This applies even to (results not shown), which implies that grasping behav- capuchins, although their thumb is only pseudo-oppos- iour can be inferred fairly precisely using few morpho- able. metric data. No principal coordinate axis was selected. An additional (non-phylogenetic) test using simple linear regressions confirms that all the selected characters are Detection of phylogenetic signal correlated with behavioural characters and explain more Globally, about 40% of the characters display a phylo- than 50% of the variance in the behavioural characters genetic signal (Appendix S4). This proportion holds (results not shown). In addition, variance partitioning when using both tests, but is affected by taxonomic analyses of the characters which were correlated with at sampling; when only the 11 taxa for which behavioural least one of the phylogenetic principal coordinate axes data are available are tested, this proportion decreases to indicate that most of the explained variance is genuinely 20%. This may explain why no phylogenetic signal was explained by morphometry rather than by covariation found in the behavioural data using random taxon with the phylogeny (Table 2). reshuffling and squared-change parsimony or PVR analysis using the first two principal coordinate axes Linear regression models of inference selected by the broken-stick model (Appendix S4). However, when each axis was regressed separately Inference models obtained through linear regressions against the behavioural characters, axis 1 had a signif- include one or two morphometric characters (Appendix icant effect for precision, without thumb and power S5). Inferences were obtained from these models for grasp frequency. extant species in which morphometric data (but no

Table 2 Variance partitioning with PVR showing the correlation between behavioural (dependent) and the morphometric (independent) characters that were selected by the forward selection procedure using linear regressions with permutations (Appendix 5) and the phylogenetic principal coordinate axes (we included those that were correlated with behavioural characters based on several simple regressions with permutations).

Morphometric characters Phylogenetic axes

Dependent Portion of variance Variance explained by covariation Portion of variance Unexplained character Identity explained (probability) of morphometry and phylogeny Identity explained (probability) variance

Precision 2, 4 0.4364 (0.103) 0.3125 1 0.0528 (0.509) 0.1984 Without thumb 5 0.3544 (0.017) 0.3682 1 0.0268 (0.307) 0.2506 Power 15 0.3404 (0.061) 0.3132 1 0.0540 (0.488) 0.2923

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY 1738 E. POUYDEBAT ET AL.

Table 3 Inferred behaviour frequency (%) of extinct primate taxa Marjanovic´ & Laurin, 2007), and there are many other based on linear regression inference models (Appendix 5). pitfalls in such analyses (Shaul & Graur, 2002; Graur & Martin, 2004; Marjanovic´ & Laurin, 2007). Furthermore, Thumb- Thumb- Without Taxon Precision distals lateral thumb Power evolutionary rates are quite variable in primates (Gross- man et al., 2004; Steiper et al., 2004), which makes Australopithecus 58 2 35 )9 (0) 1 molecular dating more difficult. Oreopithecus 46 26 18 5 2 For all clades, we have adopted a compromise that uses Proconsul 76 7 28 13 1 minimal divergence ages from the fossil record in the As frequency of a behaviour cannot exceed the interval 0–100%, taxa where this record is reasonably abundant, and values outside this interval represent modelling errors and should be molecular ages for taxa such as Strepsirhini, for which interpreted as implying values close to the nearest bound (given in the fossil record is poor. When using molecular data, we parentheses). Note that as these are inferences, the sums of have tried to use studies that obtained ages compatible percentages on a given line do not necessarily add up to 100% (the with the fossil record. We assembled the tree in difference between the total and 100% represents modelling errors). MESQUITE (Maddison & Maddison, 2006) using the Stratigraphic Tools module (Josse et al., 2006). To facilitate comparisons with palaeontological literature, when- behavioural data) are available (Appendix S8), and for ever molecular dates fall close to geological stage three extinct species (Table 3). boundaries, we used the age of the boundary itself (Fig. 2); this also facilitates tree manipulation in Strati- Discussion graphic Tools (Josse et al., 2006). More information about individual divergence ages can be found in Appendix S9. Compilation of a time-calibrated tree Ontogenetic and taxonomic distribution of For many of the analyses performed below, a phylogeny grasping behaviours incorporating an estimate of branch lengths is needed. Thus, we compiled a time-calibrated tree (branch lengths The data presented above show that precision grasping reflect estimated evolutionary time). The topology fol- can be used by all arboreal and terrestrial primate species lows Goodman et al. (2005). Divergence time estimates in our study. Indeed, the precision grip was recently are much more contentious because the affinities of reported in capuchins (Spinozzi et al., 2004). However, several primate species based on fragmentary material according to that report, this grasping technique was less are poorly constrained (Ross et al., 1998), and because frequently used by immature individuals. In our study, divergence time estimates based on molecular data are all individuals were observed in their social groups, and often considerably older than the minimal times of we did not observe any obvious effect of ontogenetic age divergence based on fossils (see Marjanovic´ & Laurin, for this behaviour (data not shown). 2007; for a review). Estimates of divergence times based Comparisons between great apes and human children on molecular data also often differ substantially between refute the idea that they have similar sensorimotor studies; for instance, Arnason et al. (2000) estimated that organization (Parker & Gibson, 1977). In our study, this the divergence between strepsirhines and haplorhines at similarity is not great. Categories of grasping used for about 90 Ma (in the Turonian, in the Early Upper small objects differ strongly between human children and Cretaceous) and that the anthropoid radiation started at great apes. Great apes use precision grips and the distal 70 Ma, whereas Yoder & Yang (2004) estimated these phalanges of their digits less often than children. In events to have occurred at about 80 and 50 Ma, respec- addition, they use the lateral side of their index and the tively. Yoder & Yang (2004) furthermore estimate the grip without thumb, contrary to children. These differ- divergence between Lorisiformes and Lemuriformes at ences between great apes and children may be explained 70–75 Ma, but Roos et al. (2004) estimate it at about by the neural and morphological variability existing 60 Ma, although the oldest fossil in that clade dates from between humans and the other species, regardless of the Priabonian (Stucky & McKenna, 1993: p. 757), no their age. Finally, the comparison between great apes and more than 37.2 Ma, according to the geological time human adults does not show strong similarities. Human scale of Gradstein et al. (2004). Given the poor fossil adults use precision grips with small objects and the distal record of (crown group) strepsirhines (Stucky & phalanges of their digits to grasp large objects much more McKenna, 1993: p. 757), the paleontological date here often. is likely to seriously underestimate the actual divergence dates within Strepsirhini. Considerable differences be- Relationship between morphometry and grasping tween molecular dates are not unexpected because behaviour several methods can be used to obtain molecular dates. The choice of the calibration date also influences the The results of this study refute some well-established results (Brochu, 2004a, b; Poux & Douzery, 2004; ideas about the relationship between morphometry and

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1739

prehensile behaviour. Our data show that great apes precision and power grips with the same frequency to use precision grips less often than cercopithecoids when grasp small food items. We found that they opted more handling small objects and that the contrary pattern is often for precision grips (almost 80%). This difference observed when handling large objects (Pouydebat et al., between our results could be due to the population and 2006a). Furthermore, Pongo and Pan use the grip individual variability or the protocol of observation without the thumb, contrary to cercopithecoids. These which was not the same. Capuchins display a wide differences are partly explained by morphometrical data variety of prehensile abilities that confirm their capacity, such as the shorter thumb of great apes (Marzke et al., apparently atypical among New World monkeys, to use 1992), as shown by our analyses (Appendix S7, their hands dexterously during extractive foraging and precision and characters 3 and 5; Table 2, without object manipulation (Fragaszy et al., 1991; Fragaszy & thumb and character 5). Therefore, the length of the Boinski, 1995; Christel & Fragaszy, 2000; Pouydebat thumb, including the first metacarpal, is an important et al., 2006b), although they do not possess the true morphometrical character in the behaviour of precision opposable thumb typical of catarrhine primates. grasp, as previously suggested (Napier, 1956; Schultz, 1969; Susman, 1989). These morphometrical parame- Inferences about the grasping behaviour of extinct ters are also negatively correlated with the behavioural primates character ‘thumb-lateral’ (Appendix S7), involving the contact of the distal phalanx of the thumb and the The inference of grasping behaviour from morphological lateral side of the index. Therefore, the length of the analyses of the hands of fossils is a complex problem. first ray does not reflect precision grasping only. Some authors opposed the hand of extant apes to the Other morphometrical data are specifically correlated hand of humans and argued that extant apes are unable with a single behavioural character. For instance, to grasp objects with a precision grip or pad-to-pad the power grasping behaviour, involving the palm of gripping. However, many extant ape species use preci- the hand during the grasp, is strongly correlated with the sion gripping (Christel, 1993; Pouydebat, 2004) without lengths of the digits one to three (Appendix S7, power vs. meeting all the morphological criteria usually considered characters 7–9, 12, 15). The length of digits 2 and 3 is to be linked with precision gripping. We have shown that correlated only with power-grasping (among the behav- precision gripping can be performed by hands showing a iours studied here); short index and third digits seem to greater morphological diversity than previously thought. favour power-grasping. Our sampling of behavioural characters in anthropoids A few species show prehensile patterns which could largely restricts our discussion of their possible evolution not have been inferred from their hand morphology. For within this clade, as suggested by the extant phylogenetic instance, gorillas show a high percentage of use of the bracket principle (Witmer, 1995), which was extended precision grip in spite of their short thumb (Pouydebat into the context of continuous characters by Laurin et al. et al., 2006a). This reflects the fact that the length of the (2004). The absence of a phylogenetic signal in the thumb does not explain all the variance in use of the behavioural characters (Appendix S4, characters 17–21) precision grip. A quantitative, statistical approach is precludes tracing their history over the tree using required because of the complexity of the relationship optimization procedures because the results would not between morphometry and prehensile behaviour. be reliable (Laurin, 2004). This limitation may reflect the Finally, we wonder why carpus length is correlated relatively low number of taxa for which behavioural data with precision and thumb-lateral grips. It would be are available as analyses of our data with the same interesting to test if this can be explained by soft taxonomic sampling show a similar absence of a phylo- anatomy, such as muscle or tendon morphology, or by genetic signal in most morphometric characters (Appen- locomotory behaviour (such as arboreality). dix S4, central columns). Thus, it might be possible to reconstruct the evolution of these behavioural characters by sampling the same clade more densely. Unexpected similarities between capuchin and Another possibility is to use the linear regression human prehensile behaviours inference models that we have produced (Appendix S5) Our study reveals similarities in prehensile behaviours to infer the behaviour of extinct primate species (Table 3): between capuchins and humans. Capuchins use preci- A. afarensis, O. bambolii and P. africanus (Fig. 3). Thus, we sion and thumb-distal grips as often as human adults. can infer that the Plio-Pleistocene hominin A. afarensis Similar to humans, capuchin monkeys almost exclusively (3–4 Ma old) from Hadar (Ethiopia) exhibited frequent used the distal phalanges of their digits to grasp small precision behaviour (Table 3). The linear regression objects. These results can be compared with those of inference for thumb-distals behaviour in A. afarensis Spinozzi et al. (2004), who observed a wide variety of suggests the infrequent occurrence of this behaviour. grasping patterns in capuchins. These include various Results about precision behaviour are close to those forms of precision and power grips. Contrary to our obtained by Marzke (1997) and Panger et al. (2002); on results, Spinozzi et al. (2004) reported that capuchins use the contrary, several authors suggested that A. afarensis

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY 1740 E. POUYDEBAT ET AL.

Napier (1956, 1960) suggested that grasping objects with precision requires opposability of the thumb and favour- able relative lengths of digits I and II. Furthermore, some authors suggest that precision grasping is linked with brain organization and the development of cognitive processes (Napier, 1960; Jones-Engels & Bard, 1996). Under such hypotheses, hominoid fossils presenting morphological characters associated with precision grasping have been argued to be able to use tools (Susman, 1989; Marzke, 1997). The data presented in this paper refute this evolutionary scenario because species with highly different hand morphologies and brain structure use a precision grip. The evolution of grasping abilities in platyrrhines and catarrhines is much more complex than a simple trend from power to precision grasping. Indeed, Fig. 3 Time-calibrated supertree of extant and extinct hominoids several species (i.e. P. troglodytes, P. pygmaeus, M. fuscata, represented in this study. The numbers next to the tree indicate the P. papio and C. apella) use the tips of their thumb and age (Ma); those next to the stage names indicate their duration (Ma). index finger for grasping small objects, demonstrating For more information, see Fig. 2. [A colour version of this figure is that morphological criteria previously used for deducing available online.] grasping ability are not reliable (Susman, 1998). Fur- thermore, in our quantitative analysis based on the use power grasping most often (Bush et al., 1982; Stern & percentages of use of five simplified categories of grasp- Susman, 1983; Susman, 1991, 1994), which we infer to ing, capuchins are similar to humans (mainly adult have occurred infrequently. Australopithecus may share humans) even though they possess a pseudo-opposable with humans the absence of without-thumb behaviour, thumb (rather than a truly opposable thumb, capable of which may be a synapomorphy of Australopithecus and adduction and rotation of its carpo-metacarpus joint). Homo among hominoids (although cercopithecoids show This functional similarity of two species which diverged a convergent similarity). about 34 Ma (Fig. 2) is surprising. The Miocene ape Oreopithecus may have exhibited Our findings are in concordant with the evolutionary slightly less precision behaviour than Australopithecus, but model according to which a primitive power grasp was like the latter, it resorted to that behaviour much more subsequently modified into a derived precision grasp often than the power grip, which was rarely displayed. (Susman, 1979), although this transition occurred before This result is congruent with the assessment of Moya´- the appearance of hominoids. Our data suggest that both Sola´ et al. (1999). On the other hand, Susman (2004) behaviours were already present in the first anthropoids. suggested that O. bambolii emphasized the power grip It would be interesting to obtain some data on Tarsius and over the precision grip. Similar to most extant hominoids strepsirhines to determine when precision grasping and Proconsul, but unlike Homo and Australopithecus, appeared. Oreopithecus may have displayed the without-thumb All our results suggest that grasping has evolved in a behaviour (Table 3). more complex manner than previously realized. Our The linear regression inference models suggest that the observations show that precision grip is far more wide- Early Miocene (Burdigalian) stem-hominoid P. africanus spread than previously thought. This is coherent with the (16–18 Ma) from Kenya used the precision grip fre- findings that skilled forelimb movements are also present quently (Table 3) and thumb-distals grip more rarely. in other mammals (Ivanco et al., 1996; Whishaw et al., P. africanus resorted to thumb-lateral grip relatively 1998) and even in amphibians (Gray, 1997). These infrequently and used the without-thumb grip even less movements may not be homologous and represent frequently. These results about P. africanus are very convergent evolution of motor patterns that superficially different from those published in the literature, which resemble reaching (Bracha et al., 1990). However, the suggest that P. africanus did not use a precision grip similarities in reaching among different mammalian taxa (Napier & Davis, 1959; McHenry, 1983; Walker & Pick- suggest that the movements are homologous within ford, 1983; Begun et al., 1994). The widespread distribu- mammals. It would be interesting to apply comparative tion of the precision grip in the primate species sampled in methods to a far greater range of taxa to assess broader- this study support our palaeobiological inferences. scale evolutionary patterns of various grip patterns.

The evolution of grasping behaviour Acknowledgments The classical model proposes a late appearance of We would like to thank the Foundation Singer-Polignac precision grasping, often considered unique to hominids. for their ﬁnancial support, P. Piras for helping us with

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1741

some statistical analyses, D. Marjanovic´ for numerous Folinsbee, K. E. & Brooks, D. R. 2007. Miocene hominoid stylistic corrections, and all the staff of the Zoo of Beauval biogeography: pulses of dispersal and differentiation. J. Bio- and of the Monkey Valley (France). geogr. 34: 383–397. Fragaszy, D. M. & Boinski, S. 1995. Patterns of individual diet choice and efﬁciency of foraging in wedge-capped capuchin References monkeys (Cebus olivaceus). J. Comp. Psychol. 109: 339–348. Fragaszy, D. M., Baer, J. & Adams-Curtis, L. 1991. Behavioral Altmann, G. 1974. Observational study of behavior. Sampl. development and maternal care in tufted capuchins (Cebus Methods Behav. 49: 227–267. apella) and squirrel monkeys (Saimiri sciureus) from birth Arnason, U., Gullberg, A., Burguete, A. S. & Janke, A. 2000. through seven months. Dev. Psychobiol. 24: 375–93. Molecular estimates of primate divergences and new hypo- Fragaszy, D., Visalberghi, E. & Fedigan, L. 2004. The Complete theses for primate dispersal and the origin of modern humans. Capuchin. The Biology of the Genus Cebus. Cambridge University Hereditas 133: 217–228. Press, Cambridge, UK. Begun, D. R., Teaford, M. F. & Walker, A. 1994. Comparative Garland, T. Jr, Harvey, P. H. & Ives, A. R. 1992. Procedures for and functional anatomy of phalanges from the Proconsul the analysis of comparative data using phylogenetically Kaswanga Primate Site, Rusinga Island, Kenya. J. Hum. Evol. independent contrasts. Syst. Biol. 41: 18–32. 26: 89–165. Godinot, M. & Beard, K. C. 1993. A survey of fossil primate Begun, D. R., Ward, C. V. & Rose, M. D. 1997. Function, Phylogeny hands. In: Hands of Primates (H. Preuschoft & D. J. Chivers, . Plenum and Fossils: Miocene Hominoid Origins and Adaptations eds), pp. 335–378. Springer-Verlag, Vienna. Press, New York. Godinot, M., Lane`que, L., Verdier, R. & Apriletti, K. 1997. Boesch, C. & Boesch, H. 1990. Tool use and tool making in wild Quantitative morphology of Eocene primates compared to chimpanzees. Folia Primatol. 54: 86–99. living primates: examples of current studies. Geobios 30: 271– Bracha, V., Zhuravin, I. A. & Burges, J. 1990. The reaching 281. reaction in the rat: a part of the digging pattern? Behav. Brain Goodman, M., Grossman, L. I. & Wildman, D. E. 2005. Moving Res. 36: 53–64. primate genomics beyond the chimpanzee genome. Trends Brochu, C. A. 2004a. Calibration age and quartet divergence Genet. 21: 511–517. date estimation. Evolution 58: 1375–1382. Gradstein, F. M., Ogg, J. G. & Smith, A. G. (eds) 2004. A Geologic Brochu, C. A. 2004b. Patterns of calibration age sensitivity with Time Scale 2004. Cambridge University Press, Cambridge. quartet dating methods. J. Paleontol. 78: 7–30. Graur, D. & Martin, W. 2004. Reading the entrails of chickens: Bush, M., Lovejoy, C., Johanson, D. & Coppens, Y. 1982. molecular timescales of evolution and the illusion of precision. Hominid carpal, metacarpal, and phalangeal bones recovered Trends Genet. 20: 80–86. from the Hadar Formation: 1974–1977 collections. Am. J. Phys. Gray, L. A. 1997. Evolution of forelimb movement patterns Anthrop. 57: 651–677. for prey manipulation in anurans. J. Exp. Zool. 277: 417–424. Casgrain, P. 2005. [www document]. URL http://www. Permute! Grossman, L. I., Wildman, D. E. & Schmidt, T. R. 2004. bio.umontreal.ca/Casgrain/en/labo/permute/index.html. Accelerated evolution of the electron transport chain in Casgrain, P., Legendre, P. & Vaudor, A. 2004. The R Package for anthropoid primates. Trends Genet. 20: 578–585. [www document]. URL Multidimensional and Spatial Analysis Ivanco, T. L., Pellis, S. M. & Whishaw, I. Q. 1996. Skilled http://www.fas.umontreal.ca/BIOL/Casgrain/en/labo/R/v4/ forelimb movements in prey catching and reaching by rats progress.html. (Rattus norvegicus) and opossums (Monodelphis domestica): rela- Christel, M. 1993. Grasping techniques and hand preference in tions to anatomical differences in motor systems. Behav. Brain Hominoidea. In: Hands of Primates (H. Preuschoft & D. Chivers, Res. 79: 163–181. eds), pp. 91–108. Springer, Berlin. Iwaniuk, A. N. & Whishaw, I. Q. 2000. On the origin of skilled Christel, M. I. & Fragaszy, D. 2000. Manual function in Cebus forelimb movements. Trends Neurosci. 23: 372–376. Digital mobility, preshaping, and endurance in repet- apella. Jones-Engel, L. & Bard, K. A. 1996. Precision grips in young itive grasping. Int. J. Primatol. 21: 697–719. chimpanzees. Am. J. Primatol. 39: 1–15. Christel, M., Kitzel, S. & Niemitz, C. 1998. How precisely do Josse, S., Moreau, T. & Laurin, M. 2006. Stratigraphic Tools for bonobos (Pan paniscus) grasp small objects? Int. J. Primatol. 19: Mesquite. [Computer program]. URL http://www.mesquite 165–194. project.org/packages/stratigraphicTools/. Clark, J. D. 1993. Stone artifact assemblages from Members 1–3 Laurin, M. 2004. The evolution of body size, Cope’s rule and the Swartkrans Cave. In: Swartkrans: A Cave’s Chronicle of Early origin of amniotes. Syst. Biol. 53: 594–622. Man (C. K. Brain, ed.), pp. 167–194. Transvaal Museum, Laurin, M., Girondot, M. & Loth, M.-M. 2004. The evolution of Pretoria. long bone microanatomy and lifestyle in lissamphibians. Cubo, J., Ponton, F., Laurin, M., de Margerie, E. & Castanet, J. Paleobiology 30: 589–613. 2005. Phylogenetic signal in bone microstructure of saurop- Lehner, P. N. 1996. Handbook of Ethological Methods. Cambridge sids. Syst. Biol. 54: 562–574. University Press, Cambridge, UK. Desdevises, Y., Legendre, P., Azouzi, L. & Morand, S. 2003. Maddison, W. P. & Maddison, D. R. 2006. Mesquite: a modular Quantifying phylogenetically structured environmental vari- system for evolutionary analysis, Version 1.1. [Computer ation. Evolution 57: 2467–2652. program]. URL http://www.mesquiteproject.org. Diniz-Filho, J. A. F., de Sant’Ana, C. E. R. & Bini, L. M. 1998. An Marjanovic´, D. & Laurin, M. 2007. Fossils, molecules, divergence eigenvector method for estimating phylogenetic inertia. times, and the origin of lissamphibians. Syst. Biol. 56: 369–388. Evolution 52: 1247–1262. Marzke, M. W. 1997. Precision grips, hand morphology and Felsenstein, J. 1985. Phylogenies and the comparative method. tools. Am. J. Phys. Anthrop. 102: 91–110. Am. Nat. 125: 1–15.

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY 1742 E. POUYDEBAT ET AL.

Marzke, M. W., Wullstein, K. L. & Viegas, S. F. 1992. Evolution Stern, J. & Susman, R. 1983. The locomotor anatomy of of the power grip and its morphological correlates in homi- Australopithecus afarensis, Am. J. Phys. Anthrop. 60: 279–317. nids. Am. J. Phys. Anthrop. 89: 283–298. Stucky, R. K. & McKenna, M. C. 1993. Mammalia. In: The Fossil McHenry, H. M. 1983. The capitate of Australopithecus afarensis Record 2 (J. M. Benton, ed.), pp. 739–771. Chapman & Hall, and A. africanus. Am. J. Phys. Anthrop. 62: 187–198. London. Midford, P., Garland, T. J. & Maddison, W. P. 2003. PDAP Package Susman, R. L. 1979. Comparative and functional morphology of for Mesquite. [Computer program]. URL http://mesquiteproject. hominoid fingers. Am. J. Phys. Anthrop. 50: 215–236. org/pdap_mesquite/index.html. Susman, R. L. 1989. New hominid fossils from the Swartkrans Moya´-Sola´, S., Ko¨ hler, M. & Rook, L. 1999. Evidence of formation: postcranial specimens. Am. J. Phys. Anthrop. 79: hominid-like precision grip capability in the hand of the 451–474. Miocene ape Oreopithecus. Proc. Natl. Acad. Sci. USA 96: 313– Susman, R. L. 1991. Who made the Oldowan tools? Fossil 317. evidence for tool behavior in Plio-Pleistocene hominids. Napier, J. R. 1956. The prehensile movements of the human J. Anthrop. Res. 47: 129–149. hand. J. Bone Joint Surg. 38B: 902–913. Susman, R. L. 1994. Fossil evidence for early hominid tool use. Napier, J. R. 1960. Studies of the hands of living primates. Proc. Science 265: 1570–1573. Zool. Soc. London 134: 647–657. Susman, R. L. 1998. Hand function and tool behavior in early Napier,J.R.1980.Hands.PrincetonUniversityPress,Princeton,NJ. hominids. J. Hum. Evol. 35: 23–46. Napier, J. & Davis, P. 1959. The forelimb skeleton and Susman, R. L. 2004. Oreopithecus bambolii: an unlikely case of associated remains of Proconsul africanus. Fossil Mam. Afr. 16: hominidlike grip capability in a Miocene ape. J. Hum. Evol. 46: 1–70. 105–117. Panger, M., Brooks, A. & Richmond, B. 2002. Wood, older than Tuttle, R. H. 1965. A Study of the Chimpanzee Hand with Comments the Oldowan? Rethinking the emergence of hominin tool use. on Hominoid Evolution. University of California, Berkeley, CA, Evol. Anthrop. 11: 235–245. 526 p. Parker, S. T. & Gibson, K. R. 1977. Object manipulation, tool Walker, A. C. & Pickford, M. 1983. New postcranial fossils of use, and sensorimotor intelligence as feeding adaptations in Proconsul africanus and Proconsul nyanzae. In: New Interpretations cebus monkeys and great apes. J. Hum. Evol. 6: 623–641. of Ape and Human Ancestry (R. L. Ciochon & R. S. Corruccini, Poux, C. & Douzery, E. J. P. 2004. Primate phylogeny, eds), pp. 325–351. Plenum Press, New York. evolutionary rate variations, and divergence times: a contri- Whishaw, I. Q., Sarna, J. R. & Pellis, S. M. 1998. Rodent-typical bution from the nuclear gene IRBP. Am. J. Phys. Anthrop. 124: and species-specific limb use in eating: evidence for specialized 1–16. paw use from a comparative analysis of ten species. Behav. Pouydebat, E. 2004. La pre´hension chez les Primates: approches Brain Res. 96: 79–91. e´thologique, biomećanique et morphome´trique. Ph.D. Thesis, at the Witmer, L. M. 1995. The extant phylogenetic bracket and the Museúm National d’Histoire Naturelle, Paris, 300 p. importance of reconstructing soft tissues in fossils. In: Pouydebat, E., Berge, C., Gorce, P. & Coppens, Y. 2005. Use and Functional Morphology in Vertebrate Paleontology, Vol. 1 (J. J. manufacture of tools to extract food by captive Gorilla Thomason, ed.), pp. 19–33. Cambridge University Press, New gorilla gorilla: experimental approach. Folia Primatol. 76: 180– York. 183. Yoder, A. D. & Yang, Z. 2004. Divergence dates for Malagasy Pouydebat, E., Berge, C., Gorce, P. & Coppens, Y. 2006a. La lemurs estimated from multiple gene loci: geological and pre´hension chez les Primates: prećision, outils et perspectives evolutionary context. Mol. Ecol. 13: 757–773. e´volutives. C. R. Palevol. 5: 597–602. Pouydebat, E., Gorce, P., Coppens, Y. & Bels, V. 2006b. Substrate Received 28 March 2008; revised 18 June 2008; accepted 23 June 2008 optimisation in nut cracking by capuchin monkeys. Am. J. Primatol. 68: 1017–1024. Roos, C., Schmitz, J. & Zischler, H. 2004. Primate jumping genes Supporting information elucidate strepsirrhine phylogeny. Proc. Natl Acad. Sci. USA Additional supporting information may be found in the 101: 10650–10654. online version of this article: Ross, C., Williams, B. & Kay, R. F. 1998. Phylogenetic analysis of Appendix S1 Animals observed. anthropoid relationships. J. Hum. Evol. 35: 221–306. Schultz, A. H. 1969. The Life of Primates. Weidenfeld and Appendix S2 Diameter of the objects for the studied Nicolson, London, 341 p. species. Shaul, S. & Graur, D. 2002. Playing chicken (Gallus gallus): Appendix S3 Matrix of morphometric (1–16) and methodological inconsistencies of molecular divergence date behavioural (17–21) characters used in this study. estimates due to secondary calibration points. Gene 300: 59–61. Appendix S4 Phylogenetic signal in the characters. Spinozzi, G., Truppa, V. & Lagana, T. 2004. Grasping behavior in Appendix S5 General linear regression inference mod- tufted capuchin monkeys (Cebus apella): grip types and manual els for the behavioural characters. laterality for picking up a small food item. Am. J. Phys. Anthrop. Appendix S6 Specific linear regression inference mod- 125: 30–41. els for the behavioural characters. Steiper, M. E., Young, N. M. & Sukarna, T. Y. 2004. Genomic Appendix S7 data support the hominoid slowdown and an Early Oligocene Correlation between behavioural (depen- estimate for the hominoid-cercopithecoid divergence. Proc. dent) and morphometric (independent) characters Natl Acad. Sci. USA 101: 17021–17026. assessed using independent contrasts.

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY Evolution of grasping among anthropoids 1743

Appendix S8 Inferred behavioural character values Please note: Wiley-Blackwell are not responsible for for extant primate taxa for which such data are unavailable the content or functionality of any supporting materials but for which relevant morphometric data are known. supplied by the authors. Any queries (other than missing Appendix S9 Notes about the divergence dates used to material) should be directed to the corresponding author compile the time-calibrated supertree. for the article.

ª 2008 THE AUTHORS. J. EVOL. BIOL. 21 (2008) 1732–1743 JOURNAL COMPILATION ª 2008 EUROPEAN SOCIETY FOR EVOLUTIONARY BIOLOGY