Journal of Human Evolution xxx (2013) 1e9 Contents lists available at SciVerse ScienceDirect Journal of Human Evolution journal homepage: www.elsevier.com/locate/jhevol The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes Melissa Tallman a, b, *, Sergio Almécija c, d, e, Samantha L. Reber e, David M. Alba f, Salvador Moyà-Solà g a Department of Biomedical Sciences, Grand Valley State University, Padnos Hall, Allendale, MI 49401, USA City University of New York and NYCEP at the Department of Vertebrate Paleontology, American Museum of Natural History, 79th St and Central Park West, New York, NY 10024, USA c Department of Anatomical Sciences, Stony Brook University Medicine, Stony Brook, NY 11794, USA d Department of Vertebrate Paleontology, American Museum of Natural History and NYCEP, 79th St and Central Park West, New York, NY 10024, USA e Forensic & Investigative Sciences School, University of Central Lancashire, Preston, Lancashire PR1 2HE, United Kingdom f Institut Català de Paleontologia Miquel Crusafont, Universitat Autònoma de Barcelona, Edifici ICP, Campus de la UAB s/n, 08193 Cerdanyola del Vallès, Barcelona, Spain g ICREA at Institut Català de Paleontologia Miquel Crusafont and Unitat d’Antropologia Biològica (Dept. BABVE), Universitat Autònoma de Barcelona, Edifici ICP, Campus de la UAB s/n, 08193 Cerdanyola del Vallès, Barcelona, Spain b a r t i c l e i n f o a b s t r a c t Article history: Received 1 May 2012 Accepted 25 July 2012 Available online xxx IPS18800 is a partial skeleton attributed to the fossil great ape Hispanopithecus laietanus, and dated to 9.6 Ma (millions of years ago). Previous studies on the postcranial anatomy of this taxon have shown that it displayed a derived, extant great ape-like orthograde body plan with suspensory adaptations, uniquely coupled with adaptations for above-branch pronograde locomotion. Here, for the first time, we describe and analyze in detail the distal tibia of the IPS18800 skeleton of Hispanopithecus with the aid of threedimensional geometric morphometrics based on 53 landmarks and semilandmarks collected on a broad sample of extant catarrhines and fossil hominoids. Results of principal components and canonical variate analyses reveal that the distal tibia of Hispanopithecus occupies a unique position in the morphospace, similar in some respects to pronograde monkeys, and in other respects to extant apes. The IPS18800 distal tibia combines adaptations for above branch quadrupedalism, such as a keeled trochlear surface and strong intercollicular groove, with adaptations for vertical climbing, such as an anteroposteriorly flattened shaft, enlarged fibular facet and a tibial stop. These results on the distal tibia agree with those from other anatomical regions, indicating that this taxon displayed a locomotor repertoire unlike any extant ape, combining vertical climbing and clambering with above-branch quadrupedalism. Ó 2013 Elsevier Ltd. All rights reserved. Keywords: Functional morphology Geometric morphometrics Postcrania Hominoid evolution Introduction The partial skeleton (IPS18800; Moyà-Solà and Köhler, 1996) and associated face (IPS18000; Moyà-Solà and Köhler, 1993, 1995) of the fossil great ape Hispanopithecus laietanus, from the late Miocene (9.6 Ma [millions of years ago]) site of Can Llobateres 2 (Vallès-Penedès Basin, Catalonia, Spain; Agustí et al., 1996; Alba et al., 2011; Casanovas-Vilar et al., 2011), indicate that this taxon is the oldest known fossil hominoid combining an orthograde body plan with below-branch suspensory adaptations (e.g., Moyà-Solà and Köhler, 1996; Almécija et al., 2007; Alba et al., 2012). However, as noted by the latter authors, the Hispanopithecus postcranium is not completely modern ape-like, but still retains some pronograde adaptations for above-branch quadrupedalism in the hand. * Corresponding author. E-mail address: tallmame@gvsu.edu (M. Tallman). This study will further investigate these conflicting functional signals by examining the hindlimb. Here we describe in detail the preserved left distal tibia of the IPS18800 skeleton, and provide morphometric analyses within a broad comparative context. Our results permit us to draw additional morphofunctional inferences for this taxon, which confirm previous assertions that it displayed a mosaic locomotor repertoire unknown among extant catarrhines. Description The distal tibia IPS18800 (Figure 1) is 74.0 mm long, mediolaterally wide, and anteroposteriorly compressed (Figure 1). The interosseus crest on its lateral aspect is well developed. It terminates at a fibular facet that is approximately 5.4 mm long and 11.7 mm wide, and which makes a slightly obtuse angle with the trochlear surface (Figure 1B). The anterior border of the fibular facet is defined by the interosseous crest, while the posterior border is 0047-2484/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jhevol.2012.07.009 Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 2 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Figure 1. The IPS18000 distal tibia. (A) anterior view; (B) lateral view; (C) posterior view; (D) medial view; (E) distal view; (F) proximal view. poorly developed. The medial malleolus is robust and has a strong distal projection. The tibialis posterior groove of the medial malleolus is deep, with strong crests on either side (Figure 1C). The maximum distal projection of the medial malleolus occurs anteriorly and it has a well-developed intercollicular groove for the posterior tibiotalar ligament (Figure 1B, D, E). The articular surface on the malleolus faces laterally with a moderate extension onto the anterior surface. In anterior view, the medial malleolus makes a right angle with the trochlear surface (Figure 1A). The trochlear surface is quadrangular and conspicuously defined by marked anterior and posterior lips. A rounded and anteroposteriorly concave median keel connects the anterior and posterior surfaces, dividing the trochlear surface into a large medial section and a smaller lateral section. The lateral portion of the trochlear surface slopes proximally away from the medial portion. The medial portion slightly slopes anteriorly creating a secondary facet on the trochlear surface. The median keel also defines the maximum anteroposterior diameter of the trochlear surface (Figure 1E). kindly provided by Jeremy DeSilva. All data were collected by S.L.R. to eliminate the effects of interobserver error. Specimens were registered with respect to one another using a Generalized Procrustes Analysis (GPA) in Morphologika2 (O’Higgins and Jones, 2006). A GPA minimizes the sums of squared distances between the landmark configurations of each specimen by centering all landmark configurations on a common origin (the centroid), Materials and methods We conducted numerical analyses of the distal tibia using threedimensional geometric morphometrics (3D-GM; see Rohlf and Slice, 1990) on the basis of coordinates collected with the Landmark Editor (Wiley et al., 2005) from original tibiae scanned using either the NextEngine or Konica Minolta Vivid 500 surface laser scanners. Eight landmarks and 45 semilandmarks (Figure 2B) were placed to define the distal articular surface and contour of the medial malleolus, following Harcourt-Smith et al. (2008). The semilandmarks were placed as a grid and anchored by nine type II and III landmarks (landmarks 5e13; Bookstein, 1991) on the outline of the distal tibial articular surface. Semilandmarks were slid into their most analogous positions by minimizing the Procrustes distance between individuals using in-house software written for MATLABÒ (2010, The Mathworks). Landmarks along the edge were slid in a single direction whereas landmarks in the middle of the grid were slid in two directions, with respect to the landmarks on either side, both vertically and horizontally. Landmarks 6, 8e10 and 12 were also slid along the edges, thus only landmarks 5, 7, 11, and 13 in the grid remain as homologous landmarks (Figure 2B). A similar landmark configuration, but lacking the semilandmarks, was utilized by Turley et al. (2011). Data were collected on a broad modern comparative sample of non-human catarrhine primates. All extant individuals were wildshot adults displaying full epiphyseal closure. Where possible, the left side of each element was digitized in order to minimize random differences due to slight bilateral asymmetry. The comparative fossils were scanned from the original specimens (Table 1) and Figure 2. (A) Linear measurements collected on the IPS18000 distal tibia. (B) Landmarks collected on the IPS18000 distal tibia. Large white circles indicate homologous type 1, 2, or 3 landmarks. Large black circles indicate landmarks that were originally laid as homologous type 2 or 3 landmarks but were subsequently slid and treated as semilandmarks. Small gray circles represent semilandmarks. Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Table 1 Sample of individuals used in this study. Taxon Colobus guereza Gorilla gorilla Hylobates sp. Macaca arctoides Macaca fascicularis Macaca fuscata Macaca mulatta Nasalis larvatus Pan troglodytes Papio hamadryas Pongo pygmaeus Hispanopithecus laietanus (IPS 18800) Dendropithecus or Proconsul africanus (KNM-LG 583) Proconsul nyanzae (KNM-RU 1939) Proconsul heseloni (KNM-RU 2036 BA, KNM-RU 3589) Proconsul major (NAP I’58) Sivapithecus indicus (YGSP 1656) TOTAL n 8 36 19 9 11 2 2 15 44 12 16 1 1 1 2 1 1 181 3 differentiating individuals. Secondly, a CVA not only maximizes the differences between groups, but it does so relative to the variability within the groups. Thus, unlike a PCA, a CVA is not merely a rotation of the data in multidimensional space. Last, the axes in a CVA are scaled differently than in a PCA because of the way that it maximizes differences between groups relative to the variability within groups, which can change the orientation of individuals within a morphospace (Zelditch, 2004). For the purposes of visualization in both analyses, a 3D model of the IPS18800 tibia was morphed to match a configuration representing the maximum and minimum of each PCA and CVA axis using Landmark Editor (Wiley et al., 2005). These landmark configurations were calculated by multiplying the loadings on each axis by the position of the axis to be visualized, and adding it to the consensus configuration (Polly, 2008). In addition to 3D analyses, and in order to allow direct comparison of the IPS18800 distal tibia with other published fossils and extant taxa, we also provide standard measurements after DeSilva et al. (2010) (Figure 2A). Results rotating them about this point, and adjusting them for size (Rohlf and Slice, 1990). We performed two standard multivariate analyses on the Procrustes aligned coordinates of the sample. First, a principal components analysis (PCA) with the extant species means and fossil specimens was used to explore shape differences among the taxa. A minimum spanning tree (MST) based on Procrustes distances was placed over the PCA in order to visualize which specimens are closest to one another in multidimensional shape space. Procrustes distance is calculated as the square root of the sum of squares difference between two landmark configurations (Bookstein, 1991). In this case, it is equivalent to the Euclidean distance between two vectors of Procrustes aligned coordinates, as Euclidean distance in n-dimensions is also calculated as the square root of the sum of squares difference between two vectors (Borgefors, 1984). Using the means has the effect of weighting the fossils equally to the extant taxa and allowing shape differences among them to contribute equally to the morphospace. Secondly, a canonical variate analysis (CVA) was conducted on the Procrustes aligned coordinates. This is mathematically possible as the number of individuals in the analysis is greater than the number of observations (Mitteroecker and Gunz, 2009). Each extant genus was given its own group while the fossils were left ungrouped and were classified by the analysis. Using only the extant genera in this analysis has the effect of situating the fossils within a morphospace dictated solely by the extant groups. When the number of landmarks is large in comparison to the number of individuals, the CVA will naturally find differences between groups which are sometimes not significant (Mitteroecker and Gunz, 2009; Mitteroecker and Bookstein, 2011). Thus, we tested for significant differences between the group means using MorphoJ (Klingenberg, 2011) to compute permutation tests of 10,000 replicates of both Mahalanobis and Procrustes distances between groups. Permutation tests in MorphoJ test whether the differences between known groups are greater than what is expected in random groups drawn from the same sample. We also tested the efficacy of our model by crossvalidation, where each individual is left out of the analysis in turn and then classified by the remaining sample, using SPSS v17 (Chicago, IL). Canonical variate and PCA analyses are complimentary and do not necessarily yield the same results. First, a CVA works to maximize the differences between groups chosen a priori, while a PCA works to maximize the differences between individuals. In some datasets, the shape variables that are most important for differentiating groups are different from those that are most important for Principal components analysis The results of a PCA of the means and MST are presented in Figure 3 (also see Table 2). Principal component 1 divides all extant hominoids from cercopithecoids. This analysis shows that in the trochlea of hominoids, the anteroposterior borders are asymmetrical and the surface is less keeled, and that the medial malleolus is positioned such that its maximum distal protrusion is towards the center of the malleolus. Cercopithecoids have a more symmetrical, keeled distal surface, with a malleolus that occupies most of its medial aspect such that its maximum distal protrusion occurs anteriorly. Principal component 2 is driven by the proximodistal height of the medial malleolus relative to the mediolateral breadth of the distal articular surface. Individuals or means towards the positive values have more projecting malleoli (YGSP 1656, Gorilla), whereas those towards the negative values have shorter malleoli (Pongo). All fossil individuals fall among the extant cercopithecoids and share nearest-neighbor relationships with monkeys or other fossils. Specifically, IPS18800 is most similar to the mean of Macaca fascicularis in both two-dimensional and multi-dimensional space (Figure 3). Among the fossils, the malleolus and trochlear surface of IPS18800 is most similar in overall shape to KNM-RU 1939, attributed to Proconsul nyanzae (Table 3). Canonical variate analysis Figure 4 (see also Table 4) illustrates the results of the CVA. There were significant differences between the means of all groups (Table 5). Canonical axis 1 (CA1) is driven by the projection of the medial malleolus and its angle with the distal articular surface of the tibia. Groups with more positive values (Pongo and cercopithecoids) have shorter, less distally-projecting medial malleoli that make a more obtuse angle with the distal articular surface. Groups with more negative values (Pan, Hylobates) have taller, more distally projecting malleoli that make a more acute angle where they meet the distal articular surface. Positive values on CA2 separate taxa (Gorilla and, to a lesser degree, Pongo) that have wide and short medial malleoli and relatively flat articular surfaces, from groups with the most negative values (Papio), which have longer, slightly narrower medial malleoli and distal articular surfaces that are strongly keeled. Additionally, more positive values on CA2 are correlated with a more oblique orientation of the medial malleolus relative to the anteroposterior axis of the distal articular surface. Finally, CA3 is driven by the length and size of the medial malleolus Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 4 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Figure 3. PCA of the Procrustes aligned, slid coordinates for the distal tibia with an MST based on Procrustes distances. All points are labeled in the graphs. Models represent the maximum and minimum of each axis and were created by warping the scan of IPS18000 to those configurations. They do not represent an actual specimen in this analysis and are purely for visualization. Models are shown in distal (top, right), anterior (bottom, right) and oblique (left) views. relative to the size and proportions of the distal articular surface, as well as the degree of anteroposterior concavity. Groups with positive values (cercopithecoids, Gorilla) have anteroposteriorly narrow and concave articular surfaces with relative large medial Table 2 Eigenvalues and % variance explained by each PC axis for the PCA of the fossils and extant means for the first 12 functions. Function PC PC PC PC PC PC PC PC PC PC PC PC 1 2 3 4 5 6 7 8 9 10 11 12 Eigenvalue % of total variance explained Cumulative % 0.001557 0.001055 0.000957 0.000609 0.000559 0.000526 0.000312 0.000277 0.000148 0.000143 0.000126 0.000112 23.268060 15.758650 14.301280 9.097608 8.356015 7.859143 4.661328 4.133042 2.204811 2.143596 1.885762 1.676140 23.26806 39.02671 53.32799 62.42560 70.78162 78.64076 83.30209 87.43513 89.63994 91.78354 93.66930 95.34544 malleoli, which are proximodistally long and occupy the whole medial aspect. Pongo occupies the most negative region of the morphospace due to its flatter, wider trochlear surface, and relatively smaller and more posteriorly restricted medial malleolus. The results of the classification for fossil taxa are reported in Tables 6 and 7, and 86% of individuals were correctly classified in cross-validation tests. IPS18000 is classified either as Macaca or Hylobates, although its position in the CVA plot indicates that its morphology is inconsistent with any extant group (Figure 4). Mahalanobis distances for individuals in each extant group are reported in Table 8. Discussion and conclusions Morphofunctional interpretation The distal tibia of IPS18800 morphologically resembles that of extant apes in several respects (especially Pongo and Hylobates), but in other respects it displays closer morphological affinities with extant cercopithecoids (particularly Nasalis and Macaca) (see Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 5 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Table 3 Pairwise Procrustes distances between fossils and extant means. Values for Hispanopithecus laeitanus are bold. RU 2036 BA IPS 18800 LG 583 RU 1939 RU 3589 YGSP 1656 Col. M. fusc. M. fasc. M. mul. Pap. M. arct. Nasalis Hylo. Pan Gor. Pongo Symph. NAP I’58 RU 2036 BA IPS 18800 LG 583 RU 1939 RU 3589 YGSP 1656 Colobus M. fuscata M. fascicularis M. mulatta Papio M. arctoides Nasalis Hylobates Pan Gorilla Pongo 0.11 0 0.13 0.12 0 0.14 0.10 0.12 0 0.12 0.09 0.10 0.09 0 0.13 0.11 0.12 0.12 0.09 0 0.12 0.11 0.13 0.14 0.09 0.09 0 0.11 0.10 0.11 0.12 0.10 0.13 0.11 0 Figure 5). The strong distal projection of the medial malleolus of IPS18800 most closely resembles the morphology of arboreal cercopithecoids, and functions to lend some stability to the medial aspect of the ankle (Lewis, 1980). IPS18800 also has a welldeveloped intercollicular groove, which is most similar in shape to Nasalis (Figure 5). This indicates some limits on dorsiflexion by the posterior tibiotalar ligament, which is consistent with abovebranch arboreal quadrupedalism (DeSilva, 2008). In contrast, the shape of the articular facet on the medial malleolus is more like that of the extant great apes, suggesting that the ankle was less restricted in the parasagittal plane than in monkeys and also suitable for other positional behaviors. Moreover, the medial malleolus of IPS18800 lacks the morphology seen in cercopithecoids and leaping platyrrhines, in which a large and rounded medial malleolar surface is present to stabilize the ankle during flexion and extension as it makes contact with a deep malleolar facet on the talus (Harrison, 1989; Davis, 1996; DeSilva et al., 2010). Also similar to extant great apes, the medial malleolus of IPS18800 is relatively mediolaterally thick, which has been functionally linked to absorption of forces generated by the foot during vertical climbing in apes (Wunderlich, 1999). The deep posterior groove for the tibialis posterior tendon with a prominent crest is associated with a strong grasping ability of the foot during climbing and arboreal quadrupedalism (Lewis, 1980). Finally, the tibial shaft is mediolaterally compressed, which is more great ape-like as compared with monkeys. This form has been linked to the ability to dorsiflex and invert the foot during bouts of vertical climbing (DeSilva, 2008). This trait is seemingly at odds with a well-developed intercollicular groove, but perhaps due to the combination of great-ape features and monkey-like features, Hispanopithecus could have used a different kind of vertical climbing than that seen in extant great apes which required opposing forces working at the ankle joint. The shape of the trochlear surface is similar to Nasalis and Macaca, displaying a strong sagittal keel with distinct medial lateral depressions, although less extreme than in Papio. This morphology has been interpreted to increase stability at the ankle during movement in the parasagittal plane (Harrison, 1989). The facet on the anterior margin of the distal articular surface of the tibia is for contacting the ‘tibial stop’ on the talus, a concave groove on the distalmost edge of the talar trochlea. Among catarrhines, this facet has been reported in humans, gorillas, orangutans and baboons (Thompson, 1889; Trinkaus, 1975), being variably present in individuals from the genera Hylobates, Macaca, Pongo, and Gorilla in our sample. Although this feature has been shown to vary widely 0.13 0.11 0.11 0.10 0.11 0.12 0.12 0.10 0 0.12 0.09 0.09 0.10 0.08 0.10 0.09 0.06 0.08 0 0.16 0.11 0.12 0.11 0.11 0.11 0.11 0.10 0.07 0.07 0 0.12 0.10 0.12 0.11 0.08 0.11 0.09 0.08 0.11 0.07 0.11 0 0.09 0.08 0.10 0.10 0.10 0.11 0.12 0.08 0.08 0.07 0.10 0.09 0 0.13 0.09 0.10 0.10 0.10 0.11 0.10 0.06 0.08 0.05 0.07 0.09 0.08 0 0.14 0.12 0.09 0.11 0.09 0.13 0.11 0.11 0.11 0.08 0.11 0.11 0.12 0.09 0 0.13 0.13 0.11 0.12 0.10 0.13 0.12 0.11 0.10 0.09 0.12 0.11 0.11 0.09 0.06 0 0.15 0.14 0.14 0.14 0.12 0.14 0.11 0.13 0.13 0.11 0.13 0.13 0.14 0.11 0.10 0.07 0 0.15 0.13 0.14 0.14 0.15 0.15 0.16 0.13 0.11 0.12 0.12 0.16 0.12 0.10 0.13 0.12 0.13 0 0.14 0.12 0.10 0.12 0.11 0.12 0.11 0.11 0.09 0.08 0.09 0.12 0.11 0.08 0.07 0.07 0.09 0.10 among individuals within a single genus (Davis, 1996), its presence in H. laietanus is suggestive of hyperdorsiflexion ability probably related to vertical climbing (Thompson, 1889). The fibular facet of the IPS18800 tibia is larger than in most of the extant cercopithecoids, Pan and Hylobates, but smaller than in Pongo and Gorilla. The presence of a well-defined fibular facet is generally correlated with a weight-bearing role of the fibula, associated with an ankle joint that allows for extensive inversion and eversion, also related to vertical climbing (Lewis, 1980). In a broad study of factors that influence extant catarrhine tibial morphology, Turley et al. (2011) found that substrate preference was the most important predictor of distal tibial shape. Several traits in IPS18800 are indicative of arboreal substrate preference, including more rounded borders of the trochlear surface and a convex proximal border of the medial malleolus where it meets the trochlear surface. Several of the above-mentioned qualitative morphologic features of the distal tibia, in particular the distal projection of the medial malleolus and contour of the trochlear surface, are some of the key features that drive shape change on the principal component and canonical variate axes (Figures 3e4). Hence, the combination of ape-like and monkey-like qualitative features in the IPS18800 tibia is further evidenced by the intermediate position of this specimen in our quantitative analyses. Both in the PCA and CVA, IPS18800 falls in the area of the morphospace that is close to the consensus configuration (the shape at 0,0) of all taxa in the analysis, indicating that the former is less similar to any individual taxon than it is to the average made by all taxa. In the CVA analyses, IPS18800 is classified as either Macaca or Hylobates (Table 6), although it is more distant from both group centroids than what would be expected from a group member. Among Miocene apes, IPS18800 is most similar in shape to RU 1939, attributed to Proconsul nyanzae and inferred to be a pronograde, above-branch arboreal quadruped (Rafferty et al., 1995; DeSilva, 2008). In contrast, the H. laietanus specimen is most dissimilar in shape to Proconsul major (NAP I’58) and Sivapithecus indicus (YGSP 1656), both of which have been interpreted as large-bodied above-branch arboreal quadupeds that engaged in some great ape-like vertical climbing on the basis of the tibia (DeSilva, 2008; DeSilva et al., 2010) (Table 2). It is notable that, among extant apes, IPS18800 is closer to Hylobates than to great apes, as there is evidence that these taxa employ different modes of vertical climbing. Hylobatids tend to hold their bodies further away from vertical substrates, and have more eccentric footfall patterns with longer strides (Isler, 2003, 2005). Hylobatids also lack the typical great-ape Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 6 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Figure 4. CVA of the Procrustes aligned, slid coordinates for the distal tibia. Extant taxa are represented as per the key in the graph, and fossil individuals are labeled individually. Models represent the maximum and minimum of each axis and were created by warping the scan of IPS18000 to those configurations. They do not represent an actual specimen in this analysis and are purely for visualization. Models are shown in distal (top) and anterior (bottom) views. specializations in the ankle related to vertical climbing. This may or may not be size related, as similarly-sized atelines do possess adaptations for stability in the distal tibia that are convergent on the great ape morphology (DeSilva, 2008). Hispanopithecus positional behavior The morphology of the distal tibia IPS18800 of H. laietanus, as revealed by both qualitative and quantitative analyses, is indicative Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 7 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Table 4 Eigenvalues and % variance explained by each CA axis for the CVA. Function 1 2 3 4 5 6 7 8 Eigenvalue % of variance Cumulative % Canonical correlation 31.355 30.389 19.205 7.043 5.965 5.086 3.733 1.838 30.0 29.0 18.4 6.7 5.7 4.9 3.6 1.8 30.0 59.0 77.4 84.1 89.8 94.7 98.2 100.0 0.984 0.984 0.975 0.936 0.925 0.914 0.888 0.805 Table 8 Minimum, maximum and average squared Mahalanobis distances for individuals in each extant group. Group Colobus Gorilla Hylobates Macaca Nasalis Pan Papio Pongo of a taxon with a mosaic locomotor repertoire unlike that of any extant hominoid or cercopithecoid, combining above-branch quadrupedalism with orthograde behaviors such as vertical climbing and clambering. These results are consistent with Min Max Average 3.574 1.572 1.995 2.683 2.753 0.734 2.236 2.597 15.175 18.602 13.473 16.474 19.573 16.637 17.914 28.396 8.088 8.301 7.322 7.839 8.643 6.693 7.327 8.049 previous studies based on other anatomical regions. The morphology of the thoracic and lumbar vertebrae enable us to infer the possession of a wide and shallow thorax associated with an orthograde body plan (Moyà-Solà and Köhler, 1996; Köhler et al., Table 5 Results of permutation test of Procrustes distances and Mahalanobis distances between groups. Colobus Gorilla Hylobates Macaca Nasalis Pan Papio Pongo Colobus Gorilla Hylobates Macaca Nasalis Pan Papio Pongo 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.0046 <0.0001 <0.0001 0 <0.0001 <0.0001 <0.0001 <0.0001 0.0144 <0.0001 <0.0001 0.0029 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0 Procrustes distance values are above the zeros and Mahalanobis distances are below the zeros. Table 6 Classification and predicted group results from a CVA of the Procrustes aligned landmarks of the distal tibiae of Miocene hominoids by genus. Accession number Taxon NAP I’58 KNM-RU 2036 BA IPS 18800 KNM-LG 583 Proconsul major Proconsul heseloni Hispanopithecus laeitanus Dendropithecus or Proconsul africanus Proconsul nyanzae Proconsul heseloni Sivapithecus indicus KNM-RU 1939 KNM-RU 3589 YGSP 1656 D2 from group 1 Predicted group 2 D2 from group 2 Papio Nasalis Macaca Nasalis 78.658 20.775 74.347 118.668 Colobus Macaca Hylobates Gorilla 114.062 111.685 75.191 256.914 Hylobates Macaca Pongo 59.145 185.429 51.528 Pan Nasalis Macaca 110.265 210.770 159.876 Predicted group 1 100% of individuals were correctly classified initially and 86% of individuals were correctly classified during cross-validation. D2 indicates squared Mahalanobis distances from predicted group centroids. The overall average d2 for the extant taxa is 7.581, with a maximum of 28.396 and a minimum of 0.734 (see Table 8 for more details). Table 7 Classification and cross-validation counts for a CVA of the Procrustes aligned landmarks of the distal tibia. Group Original Count Cross-validated Count Gorilla Colobus Macaca Papio Nasalis Hylobates Symphalangus Pan Pongo Gorilla Colobus Macaca Papio Nasalis Hylobates Symphalangus Pan Pongo Predicted group membership Total Gorilla Colobus Macaca Papio Nasalis Hylobates Symph. Pan Pongo 36 0 0 0 0 0 0 0 0 35 0 0 0 0 0 1 0 0 0 8 0 0 0 0 0 0 0 0 6 2 1 0 0 0 0 0 0 0 22 0 0 0 0 0 0 0 1 16 0 1 0 0 0 2 0 0 0 12 0 0 0 0 0 0 1 1 11 0 0 0 0 0 0 0 0 0 15 0 0 0 0 1 0 3 0 13 0 0 0 1 0 0 0 0 0 19 0 0 0 0 0 0 0 0 13 0 3 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 2 1 0 0 0 0 0 0 0 0 44 0 0 0 0 0 0 3 0 39 0 0 0 0 0 0 0 0 0 16 0 0 0 0 1 0 0 1 13 36 8 22 12 15 16 3 44 16 36 8 22 12 15 16 3 44 16 Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 8 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Figure 5. Comparison of the distal tibia in posterior (top) and distal (bottom) views in the fossils and representative individuals from each extant taxon. Distal surface is painted orange and the fibular facet is painted purple. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 2001; Susanna et al., 2011). In turn, the inferred high intermembral index (Moyà-Solà and Köhler, 1996; Köhler et al., 2001), the femoral morphology (Moyà-Solà and Köhler, 1996; Köhler et al., 2002; Pina et al., 2012a, b), the configuration of the proximal ulna (Alba et al., 2012) and the elongated hand with long and very curved phalanges (Almécija et al., 2007; Deane and Begun, 2008; Alba et al., 2010) are indicative of forelimb-dominated, below-branch suspensory behaviors. In spite of these orthograde and suspensory-related features, Hispanopithecus has been interpreted to retain adaptations for above-branch quadrupedalism at several anatomical regions (Alba et al., 2012), and particularly at the metacarpals and the metacarpophalangeal region (Moyà-Solà and Köhler, 1996; Almécija et al., 2007, 2009; Alba et al., 2010; Alba, 2012). Our results for the distal tibia reinforce the contention that Hispanopithecus displayed a unique combination of positional behaviors, thus strengthening the view that, during great-ape evolution, palmigrady was gradually abandoned as suspensory behaviors became more adaptively significant (Almécija et al., 2007, 2009; Alba et al., 2010, 2012). Acknowledgments We would like to thank Steve Frost and Kevin Turley for scans of some extant taxa, as well as Eileen Westwig (AMNH) for access to collections. We are especially grateful to Jeremy DeSilva for providing us with the scans of original fossil taxa and his helpful suggestions, to David Pilbeam for allowing us to include the S. indicus (YGSP 1656) specimen in our analyses, and to the associate editor and three anonymous reviewers for their constructive comments. This work is funded by: NSF IIS 11-16921; Fulbright Commission 2009 BFUL 00049 to S.A.; Generalitat de Catalunya, 2009 BP-A 00226 to S.A., and 2009 SGR 754 GRC; Spanish Ministerio de Economía y Competitividad, CGL2011-28681, CGL201127343 and RYC-2009-04533 to D.M.A. This is NYCEP morphometrics group contribution 66. References Agustí, J., Köhler, M., Moyà-Solà, S., Cabrera, L., Garcés, M., Parés, J.M., 1996. Can Llobateres: the pattern and timing of the Vallesian hominoid radiation reconsidered. J. Hum. Evol. 31, 143e155. Alba, D.M., 2012. Fossil apes from the Vallès-Penedès Basin. Evol. Anthropol. 21, 254e269. Alba, D.M., Almécija, S., Moyà-Solà, S., 2010. Locomotor inferences in Pierolapithecus and Hispanopithecus: reply to Deane and Begun (2008). J. Hum. Evol. 59, 143e149. Alba, D.M., Casanovas-Vilar, I., Robles, J.M., Marmi, J., Moyà-Solà, S., 2011. New paleontological excavations at the Late Miocene site of Can Llobateres 1 (VallèsPenedès Basin, NE Iberian Peninsula): preliminary results of the 2010 campaign. In: Pérez-García, A. (Ed.), Viajando a Mundos Pretéritos. Ayuntamiento de Morella. Morella, pp. 35e43. Alba, D.M., Almécija, S., Casanovas-Vilar, I., Méndez, J.M., Moyà-Solà, S., 2012. A partial skeleton of the fossil great ape Hispanopithecus laietanus from Can Feu and the mosaic evolution of crown-hominoid positional behaviors. Plos One 7, e39617. Almécija, S., Alba, D.M., Moyà-Solà, S., Köhler, M., 2007. Orang-like manual adaptations in the fossil hominoid Hispanopithecus laietanus: first steps towards great ape suspensory behaviours. Proc. R. Soc. B 274, 2375e2384. Almécija, A., Alba, D.M., Moyà-Solà, S., 2009. Pierolapithecus and the functional morphology of Miocene ape hand phalanges: paleobiological and evolutionary implications. J. Hum. Evol. 57, 284e297. Bookstein, F.L., 1991. Morphometric Tools for Landmark Data: Geometry and Biology. Cambridge University Press, New York. Borgefors, G., 1984. Distance transformations in arbitrary dimensions. Comput. Vis. Graph. 27, 321e345. Casanovas-Vilar, I., Alba, D.M., Garcés, M., Robles, J.M., Moyà-Solà, S., 2011. Updated chronology for the Miocene hominoid radiation in Western Eurasia. Proc. Natl. Acad. Sci. 108, 5554e5559. Davis, L., 1996. Functional and phylogenetic implications of ankle morphology in Goeldi’s monkey. In: Norconk, M.A. (Ed.), Adaptive Radiations of Neotropical Primates. Plenum Press, New York, pp. 133e156. Deane, A.S., Begun, D.R., 2008. Broken fingers: retesting locomotor hypotheses for fossil hominoids using fragmentary proximal phalanges and high-resolution polynomial curve fitting (HR-PCF). J. Hum. Evol. 55, 691e701. DeSilva, J.M., 2008. Vertical climbing adaptations in the anthropoid ankle and midfoot: implications for locomotion in Miocene catarrhines and PlioPleistocene hominins. Ph.D. Dissertation, University of Michigan. DeSilva, J.M., Morgan, M.E., Barry, J.C., Pilbeam, D., 2010. A hominoid distal tibia from the Miocene of Pakistan. J. Hum. Evol. 58, 147e154. Harcourt-Smith, W.E., Tallman, M., Frost, S.R., Wiley, D.F., Rohlf, F.J., Delson, E., 2008. Analysis of selected hominoid joint surfaces using laser scanning and geometric morphometrics: a preliminary report. In: Sargis, E.J., Dagosto, M. (Eds.), Mammalian Evolutionary Morphology. A Tribute to Frederick S. Szalay. Springer, Dordrecht, pp. 373e383. Harrison, T., 1989. New postcranial remains of Victoriapithecus from the middle Miocene of Kenya. J. Hum. Evol. 18, 3e54. Isler, K., 2003. Characteristics of vertical climbing in gibbons. Evol. Anthropol. 11, 49e52. Isler, K., 2005. 3D-Kinematics of vertical climbing in hominoids. Am. J. Phys. Anthropol. 126, 66e81. Klingenberg, C.P., 2011. MorphoJ: an integrated software package for geometric morphometrics. Mol. Ecol. Resour. 11, 353e357. Köhler, M., Moyà-Solà, S., Alba, D.M., 2001. Eurasian hominoid evolution in the light of recent Dryopithecus findings. In: Hominoid Evolution and Climatic Change in Europe, vol. 2. Cambridge University Press, Cambridge, pp. 192e212. Köhler, M., Alba, D.M., Moyà-Solà, S., MacLatchy, L., 2002. Taxonomic affinities of the Eppelsheim femur. Am. J. Phys. Anthropol. 119, 297e304. Lewis, O.J., 1980. The joints of the evolving foot. Part I. The ankle joint. J. Anat. 130, 527e543. Mitteroecker, P., Bookstein, F., 2011. Linear discrimination, ordination, and the visualization of selection gradients in modern morphometrics. Evol. Biol. 38, 100e114. Mitteroecker, P., Gunz, P., 2009. Advances in geometric morphometrics. Evol. Biol. 36, 235e247. Moyà-Solà, S., Köhler, M., 1993. Recent discoveries of Dryopithecus shed new light on evolution of great apes. Nature 365, 543e545. Moyà-Solà, S., Köhler, M., 1995. New partial cranium of Dryopithecus Lartet, 1863 (Hominoidea, Primates) from the Upper Miocene of can Llobateres, Barcelona, Spain. J. Hum. Evol. 29, 101e139. Moyà-Solà, S., Köhler, M., 1996. A Dryopithecus skeleton and the origins of great-ape locomotion. Nature 379, 156e159. O’Higgins, P., Jones, N., 2006. Tools for statistical shape analysis. Hull York Medical School,. http://sites.google.com/site/hymsfme/resources. Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009 M. Tallman et al. / Journal of Human Evolution xxx (2013) 1e9 Pina, M., Alba, D.M., Almécija, S., Fortuny, J., Moyà-Solà, S., 2012a. Locomotor inferences in Hispanopithecus laietanus on the basis of its femoral neck cortical thickness. Am. J. Phys. Anthropol. 147 (Suppl. 54), 237. Pina, M., Alba, D.M., Almécija, S., Fortuny, J., Moyà-Solà, S., 2012b. Brief communication: paleobiological inferences on the locomotor repertoire of extinct hominoids based on femoral neck cortical thickness: the fossil great ape Hispanopithecus laietanus as a test-case study. Am. J. Phys. Anthropol. 149, 142e148. Polly, P.D., 2008. Adaptive zones and the pinniped ankle: a three-dimensional quantitative analysis. In: Sargis, E.J., Dagosto, M. (Eds.), Mammalian Evolutionary Morphology: A Tribute to Frederick S. Szalay. Springer, Dordrecht, pp. 167e198. Rafferty, K.L., Walker, A., Ruff, C.B., Rose, M.D., Andrews, P.J., 1995. Postcranial estimates of body weight in Proconsul, with a note on a distal tibia of P. major from Napak, Uganda. Am. J. Phys. Anthropol. 97, 391e402. Rohlf, F.J., Slice, D., 1990. Extensions of the Procrustes method for the optimal superimposition of landmarks. Syst. Biol. 39, 40e59. 9 Susanna, I., Alba, D.M., Almécija, S., Moyà-Solà, S., 2011. Vertebral remains of the Late Miocene ape Hispanopithecus laietanus (Primates: Hominidae): functional morphology and paleobiological implications. J. Vert. Paleontol, 202. 71st Annual Meeting SVP 2011. Thompson, A., 1889. The influence of posture on the form of the articular surfaces of the tibia and astragalus in the different races of man and the higher apes. J. Anat. 23, 616e639. Trinkaus, E., 1975. Squatting among the Neanderthals: a problem in the behavioral interpretation of skeletal morphology. J. Archaeol. Sci. 2, 327e351. Turley, K., Guthrie, E.H., Frost, S.R., 2011. Geometric morphometric analysis of tibial shape and presentation among catarrhine taxa. Anat. Rec. 294, 217e230. Wiley, D.F., Amenta, N., Alcantara, D.A., Ghosh, D., Kil, Y.J., Delson, E., HarcourtSmith, W.E.H., Rohlf, F.J., St. John, K., Hamann, B., 2005. Evolutionary morphing. Proc. IEEE Visualizations, 431e438. Wunderlich, R., 1999. Pedal form and plantar pressure distribution in anthropoid primates. Ph.D. Dissertation, State University of New York at Stony Brook. Zelditch, M., 2004. Geometric Morphometrics for Biologists: A Primer. Elsevier, London. Please cite this article in press as: Tallman, M., et al., The distal tibia of Hispanopithecus laietanus: More evidence for mosaic evolution in Miocene apes, Journal of Human Evolution (2013), http://dx.doi.org/10.1016/j.jhevol.2012.07.009