The Neanderthal lower arm

Journal of Human Evolution 61 (2011) 396e410

Contents lists available at ScienceDirect

Journal of Human Evolution journal homepage: www.elsevier.com/locate/jhevol

The Neanderthal lower arm Isabelle De Groote a, b a b

University College London, Department of Anthropology, Gower Street, London, WC1E 6BT, United Kingdom The Natural History Museum, Palaeontology Department, London, SW7 5BD, United Kingdom

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 November 2010 Accepted 23 May 2011

Neanderthal forearms have been described as being very powerful. Different individual features in the lower arm bones have been described to distinguish Neanderthals from modern humans. In this study, the overall morphology of the radius and ulna is considered, and morphological differences among Neanderthals, Upper Paleolithic Homo sapiens and recent H. sapiens are described. Comparisons among populations were made using a combination of 3D geometric morphometrics and standard multivariate methods. Comparative material included all available complete radii and ulnae from Neanderthals, early H. sapiens and archaeological and recent human populations, representing a wide geographical and lifestyle range. There are few differences among the populations when features are considered individually. Neanderthals and early H. sapiens fell within the range of modern human variation. When the suite of measurements and shapes were analyzed, differences and similarities became apparent. The Neanderthal radius is more laterally curved, has a more medially placed radial tuberosity, a longer radial neck, a more antero-posteriorly ovoid head and a well-developed proximal interosseous crest. The Neanderthal ulna has a more anterior facing trochlear notch, a lower M. brachialis insertion, larger relative midshaft size and a more medio-lateral and antero-posterior sinusoidal shaft. The Neanderthal lower arm morphology reflects a strong cold-adapted short forearm. The forearms of H. sapiens are less powerful in pronation and supination. Many differences between Neanderthals and H. sapiens can be explained as a secondary consequence of the hyper-polar body proportions of the Neanderthals, but also as retentions of the primitive condition of other hominoids. Ó 2011 Elsevier Ltd. All rights reserved.

Keywords: Homo neanderthalensis Homo sapiens Radius Ulna Curvature Geometric morphometrics

Introduction From the well-pronounced muscle attachment sites on their upper limb bones, it is suggested that Neanderthals had very powerful forearms (Trinkaus and Churchill, 1988). There are several features in the lower arm bones that distinguish Neanderthals from modern humans (Fischer, 1906; Patte, 1955; Trinkaus and Churchill, 1988; Aiello and Dean, 1990; Vandermeersch and Trinkaus, 1995; Pearson and Grine, 1997). The Neanderthal radius has been described as more laterally curved than that of humans (Fischer, 1906; Patte, 1955; Vandermeersch and Trinkaus, 1995; Carretero et al., 1999; Czarnetzki, 2000). Increased curvature of the radius results in a greater distance between the ulna and the radius, which increases the distance between the insertions of the M. pronator quadratus and the M. pronator teres. Increased curvature moves the lines of action further away from the axis of pronation and supination, and therefore

E-mail address: [email protected]. 0047-2484/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.jhevol.2011.05.007

enhances the power arms of these muscles. African apes are less curved than other mammals (Swartz, 1990). Swartz (1990) suggests that this is due to the long bones of primates being longer than those of other mammals and that this produces larger bending stresses during normal locomotion. Experimental work has demonstrated the need for normally functioning muscles in order for normal bone curvature to develop (Lanyon, 1980). Higher degrees of radial curvature in anthropoids have been explained as the result of an increase in size and functional importance of the supinator musculature, but in gibbons curvature was not affected by differential muscle mass (Swartz, 1990). Compared with humans, however, apes have a higher degree of lateral curvature (Aiello and Dean, 1990). The higher degree of lateral curvature in African apes (Martin and Saller, 1959; Knussman, 1967 in Swartz, 1990) and a more lateral insertion of the M. pronator teres increases the lever advantage of the lower arm (Aiello and Dean, 1990). The lateral subtense of the radius of the Neanderthals is remarkable (Fischer, 1906; Botez, 1926 in Patte, 1955; Vandermeersch and Trinkaus, 1995; Carretero et al., 1999; Czarnetzki, 2000). The supinator crest of the radius is strongly

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

developed. Neanderthals also possess a more medially positioned radial tuberosity (Trinkaus and Churchill, 1988). This position is a measure of the lever advantage of the M. biceps brachii and the range of action over which this muscle can operate as a supinator. In apes, the radial tuberosity is also positioned more medially and gives them a greater mechanical advantage of the M. biceps brachii in supination (Aiello and Dean, 1990). If the radial tuberosity is placed more antero-laterally, as it is in modern humans, then the power advantage is lost during the final phases of supination (Trinkaus and Churchill, 1988; Aiello and Dean, 1990; Pearson and Grine, 1997). The radial subtense, supinator crest and position of the radial tuberosity may indicate that Neanderthals closely resemble earlier hominins in the morphology and strength of the radius, and that the Neanderthal forearm and elbow was especially strong during pronation and supination (Trinkaus and Churchill, 1988). Neanderthals have been described as having a pronounced posterior subtense in the ulna (Fischer, 1906). In comparison with the African apes, hominins, including modern humans, have a more anterior facing trochlear notch (Drapeau, 2004, 2008). The Neanderthal proximal ulna, however, has been described as having an even more anterior facing trochlear notch than modern humans (Trinkaus and Churchill, 1988). Trinkaus and Churchill (1988) propose that this would not have limited the range of movement but was rather an expression of different habitual behavior, such as the increased use of forearms with the elbow flexed. A more anterior facing trochlear notch was also observed in the Australopithecus afarensis ulna, A.L. 438-1a, from Hadar, Ethiopia (Drapeau et al., 2005). It is unknown what kind of habitual behavior results in this morphology. The M. pronator quadratus crest in Neanderthals is very pronounced and also

397

suggests a more muscular forearm, although the interosseous crest is poorly developed and the shaft is relatively narrow (Trinkaus and Churchill, 1988; Aiello and Dean, 1990). In addition to reportedly having more laterally curved radii and posteriorly curved ulnae, Neanderthals have a suite of characteristics that, when considered independently, may occur in modern human populations, but that, as a suite, set apart the Neanderthals as a group that is distinct from modern humans (Boule and Vallois, 1952; Trinkaus, 1983; Hublin, 1989; Stringer, 1992; Hublin et al., 1996). Many post-cranial characters have been interpreted as the result of the Neanderthal hyper-polar body shape and muscular hypertrophy (Patte, 1955; Vl
cek, 1961; Rak and Arensburg, 1987; Tompkins and Trinkaus, 1987; Holliday and Trinkaus, 1991; Ruff and Walker, 1993; Ruff et al., 1993; Walker and Leakey, 1993; Ruff, 1994; Trinkaus et al., 1994, 1998; Vandermeersch and Trinkaus, 1995; Pearson and Grine, 1997; Churchill, 1998; Trinkaus and Ruff, 1999; Pearson, 2000a; Holliday and Ruff, 2001; Shackelford and Trinkaus, 2002; Majó et al., 2003; Weaver, 2003; Thompson and Nelson, 2005; Shackelford, 2007; De Groote, 2011). Some characteristic Neanderthal post-cranial features may be primitive retentions in Neanderthals (Trinkaus, 1981, 1983), whereas others may be autapomorphic traits or phenotypic adaptations to a particular environmental or functional environment (Howell, 1957; Pearson, 2000a,b; Pearson and Lieberman, 2004; Churchill, 2005, 2006; Pearson et al., 2006; Trinkaus, 2006; Weaver, 2009; De Groote, 2011). The aim of this study is to describe Neanderthal forearm morphology, particularly the morphological differences in the radius and ulna of Neanderthals and modern humans, and to understand the functional relevance of these differences.

Table 1 List of modern humans in the sample. Population African American Alaskan Aleutian Isl. Alaskan Point Hope Andaman Arizona Australian Bantu Belgian Medieval Belgian Mesolithic Belgian Neolithic Chinese Colorado Native Czech Medieval Danish Medieval Danish Neolithic Egyptian English Medieval English Urban French Medieval French Neolithic Greenland Inuit Kazach Medieval Khoikhoi Lapland Natufian New Mexico Ohio Peru Pygmy Russian Eskimo Russian Mesolithic Siberia South Dakota Tasmanian Tierra del Fuego N ¼ Radius/Ulna.

N

Absolute latitude

Collection

Location

12/14 7/10 12/13 11/11 18/19 7/10 1/0 18/20 1/0 23/15 4/7 2/3 34/33 10/11 19/10 5/5 16/12 19/20 5/4 3/0 14/13 0/7 9/8 22/15 11/9 8/8 14/12 3/6 1/3 14/14 7/6 14/14 14/12 2/2 1/1

n/a 71 68 11 36 30 7 50 50 50 35 43 49 55 55 26 54 51 49 48 69 47 28 67 32 31 40 11 7 66 58 66 45 42 54

African-Americans Terry Collection Aleutian Islands Collection Alaskan Inuit College of Surgeons Collection Canyon del Muertos College of Surgeons Collection Republic of Congo Spy and Gutschoven Abri des Autours Furfooz, Maurenne, Hastière, Dinant Chinese Cemetary, Karluk Quad Alaska Montezuma County, Colorado Moravian Empire Collection Sankt Bendtskirke, Ringsted Korshoj Adby, Guldhoj, Borreby Egyptian Dynasty Scarborough Spitalfields 18th-19thC Villebourg, St. Gabriel Valée du Petit Morin Tuqutut, Ilutalik, Uunartoq, Ilorsuit Southern Volga Region Oxford Collection Russian Saami Mallaha Aztec Ruins Madissonville, Ohio Ancon (Lima) Lituri Central Africa Siberian Peninsula, Ekveni Vasilievski Sibstey, Salehard Siberia Campbell County, Ohae Reservoir Tasmania Tierra del Fuego, Argentina

Smithsonian, Washington Peabody, Harvard NHM, New York NHM, London NHM, New York NHM, London RBINS, Brussels RBINS, Brussels RBINS, Brussels RBINS, Brussels Smithsonian, Washington Peabody, Harvard NHM, Prague University, Copenhagen University Copenhagen NHM, Paris NHM, London NHM, London NHM, Paris NHM, Paris University, Copenhagen St. Petersburg NHM, London Moscow State Univ. University, Tel Aviv NHM, New York Peabody, Harvard NHM, Paris RBINS, Brussels Moscow State University St.-Petersburg Moscow State University Moscow State University NHM, London, Brussels NHM, Vienna

398

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Materials and methods To investigate the differences in lower arm morphology, a sample of 21 fossil radii (nine Neanderthal and 12 early H. sapiens) and 349 recent H. sapiens radii (34 populations) were analyzed, as well as 21 fossil ulnae (eight Neanderthal, 13 early H. sapiens) and 344 recent H. sapiens ulnae (29 populations). The sample was divided into three groups: recent H. sapiens, early H. sapiens and Neanderthals. Recent H. sapiens constituted the largest group (Table 1). This group consists of samples drawn from geographically and behaviorally diverse populations to capture as much of modern H. sapiens variation as possible (Table 2). Due to large differences in sample sizes, the population means were used for the Canonical Linear Discriminant Function Analysis. Rather than attempting to measure individual curvatures using traditional techniques (which are not always straight-forward and reliable), the diaphyses of the radius and the ulna and the shape of the epiphyses were quantified using 3D geometric morphometrics. Several authors, including O’Higgins (2000) and authors in Slice (2005), have summarized the analytical approaches used in this work. The application of 3D geometric morphometrics to the quantification of long bone curvature and its advantages over linear measurements has been demonstrated by De Groote et al. (2010). The semi-landmarks make it possible to include point and outline information in a single analysis and to consider the curves or epiphyses separately or as part of the overall morphology of the bone. Landmarks and semi-landmarks were collected with a Microscribe 3DX digitizer (Immersion Corporation), Microsoft Excel, Microsoft Utility Software v.4.0 (MUS 4.0) and a laptop computer. A dataset with a homologous configuration of landmarks and semi-landmarks was created by reducing the automatically

Table 2 List of Neanderthals and early Homo sapiens. Neanderthal Original Europe La Quina 5a La Ferassie 1b La Ferrassie 2b La Chapelle aux Saintsb

Levant Kebarac Shanidar 1d Shanidar 5d

Cast Europe Le Moustiere Neandertalf a b c d e f g h i

Radius

Left Right Right Right

Right Left

Left Right

Ulna

Left Right Right Right

Early Homo sapiens Original Europe Chanceladef Combe Capellef Abri Pataud 2a Western Asia

Left Left Right

Sungirg Dolni Vestonice 13h Dolni Vestonice 14h Dolni Vestonice 16h Levant Ein Gev 1c Qafzeh 9c Ohalo 21c Skhul IVc

Right

Cast Europe St. Germain la Rivièrea Western Asia Kostienki 14i

Musée National du Prehistoire. Musee de l’Homme Paris. Tel Aviv University. Smithsonian Institution, Washington. Museum für Vor- und Frühgeschichte in Berlin. Rheinisches Museum in BonnMusee du Perigeux. Laboratory for Reconstruction, Moscow. Dolni Vestonice. Kunstcamera St Petersburg.

Radius

Ulna

Right Left

Right Right Left

Right Right Right Right

Right Right Right Right

Right Right Right

Left Left Right Left

Right

Left

Left

Right

collected semi-landmarks and equidistantly spacing them along the diaphyseal surface (De Groote et al., 2010). Each subset of configurations (separate curves and epiphyses) was exported into Morphologika 2 (O’Higgins and Jones, 1998) and analyzed using General Procrustes Analysis and Principal Component Analysis. The principal component (PC) scores were used as data in subsequent univariate and multivariate analyses, and combined with other variables (Bookstein, 1991; Adams et al., 2004; Gunz et al., 2005; Slice, 2005; De Groote, 2011). Allometric effects were explored by a Pearson’s correlation between the PC scores and centroid size (the square root of the sum of squared distances from each landmark to the centroid). Linear measurements were calculated from the landmark configurations (Table 3; Appendix). Some of these were standardized for length to correct for differences in body proportions. Ratios were multiplied by 100 to facilitate the comparisons. All subsequent statistical analyses were performed using SPSS v.15 on all individuals. First, an ANOVA was used on the PC scores and on the linear measurements to determine the effect of group membership (Neanderthal, early H. sapiens, recent H. sapiens). Post-hoc tests were performed to identify significant differences between the group means. For this, a Hochberg’s GT2 was used as it is particularly suited for analyses of groups of very different sample sizes (Field, 2000).

Table 3 Definitions of the linear measurements. Radius Maximal length (Martin no 1)

Maximum length measured from the most superior point on the articular surface on the head to the most distal point on the styloid process. Neck-shaft angle Also collo-diaphyseal angle. The angle described by the (Martin no 7) shaft-axis (going through the middle of the shaft) and the neck-axis (going through the middle of the neck) Head shape ratio Antero-posterior head diameter/medio-lateral head diameter * 100 Head size relative to Antero-posterior head diameter þ medio-lateral head length diameter/maximum length * 100 (formerly head-robusticity) Head size relative to Antero-posterior head diameter þ medio-lateral shaft size head diameter/antero-posterior mid-shaft diameter þ medio-lateral mid-shaft diameter * 100 Shaft shape mid-shaft Antero-posterior mid-shaft diameter/medio-lateral mid-shaft diameter * 100 Relative shaft size Antero-posterior mid-shaft diameter þ medio-lateral mid-shaft diameter/maximum length * 100 (formerly mid-shaft robusticity) Ulna Maximum length (Martin no 1)

Maximum length measured from the most superior point on the olecranon process to the most distal point on the articular surface (not styloid process). Olecranon shaft ratio Size of the proximal articulation: olecranon size/length * 100. Olecranon size is the depth from the tip of the olecranon to the posterior surface of the trochlear notch. Coronoideolecranon Height olecranon/height coronoid * 100. This height is ratio the distance from the middle of the trochlear notch to the tip of the coronoid or olecranon. Brachialis ratio The position of the brachialis tuberosity: distance (Solan, 1992) from the proximal extremity to the most distal point of the brachialis tuberosity. Relative shaft size Antero-posterior mid-shaft diameter þ medio-lateral mid-shaft diameter/maximum length * 100 (formerly mid-shaft robusticity) Relative head size Antero-posterior head diameter þ medio-lateral head diameter/maximum length * 100 (formerly head-robusticity) Head orientation The angle at the olecranon when a triangle is formed angle (Martin, 15a) between the anterior point at 80% shaft length surface, the tip of the olecranon and the coronoid

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

399

Second, canonical linear discriminant functions were calculated using the principal component scores (Dytham, 1999; Weaver, 2002, 2003) and linear measurements for Neanderthals, early H. sapiens, and recent H. sapiens. Due to the difference in sample sizes, the recent H. sapiens dataset was reduced by using the mean for each population. This resulted in 34 population means used for the radius and 29 for the ulna in the discriminant analysis (Table 1). The use of indices and 3D coordinate analyses increase the likelihood of colinearity. Although the subsets of 3D landmark coordinates are selected not to overlap, the use of the length to standardize certain linear measurements for example, warrant caution when the results are interpreted. To understand the overall morphology, it is important to use some measurements that may be correlated. It is for this reason that when the results were interpreted, they were done so with caution within the perspective of previously published results. To determine the relationship between the individual variables, Pearson’s Moment Correlation Coefficient for the recent H. sapiens sample using population means was calculated. Because Neanderthals have been described as showing hyper-polar adaptations in their morphology, the relationship between all variables and principal components for recent H. sapiens and absolute latitude of each population was also considered. Results The changes along each principal component were visualized using Morphologika 2 (O’Higgins, 2000). The visualizations correspond to the most extreme positive and negative individuals on the scale for each PC. The curves combine two landmarks at the end of the curves with eight proportional semi-landmarks on the shaft surface. Radius The first three principal components of the medial surface of the radius explain 46.1%, 13.2% and 8.94%, respectively, of the variation (total 68.2%). Subsequent PCs explain minimal amounts of the variation and are not considered further. PC1 reflects the variation in the degree to which the radius bends laterally on the medial surface (lateral curvature) (Fig. 1a). PC2 is related to the medial expansion of the proximal interosseous crest and the direction of the distal end of the medial surface (Fig. 1b). PC3 is the sinusoidal shape of the shaft in the antero-posterior plane (Fig. 1c). The first three PCs of the lateral surface of the radius explain 40.6%, 20.9% and 9.43%, respectively, of the variation (total 70.9%). Subsequent PCs explain minimal amounts of the variation and are not considered further. PC1 reflects differences in the degree to which the radius bends laterally on the lateral surface (lateral curvature) (Fig. 2a). PC2 is influenced by the apex of curvature and the direction of the distal end of the lateral surface (Fig. 2b). PC3 relates to the sinusoidal shape of the lateral curve in the anteroposterior plane (Fig. 2c). Only the first PC is correlated with centroid size (r ¼ 0.365, N ¼ 55, P < 0.04). The first two PCs of the epiphyses and radial tuberosity analysis explain 33.3% and 8.53%, respectively, of the variation (total 41.8%). Subsequent PCs explain minimal amounts of the variation and are not considered further (with the exception of PC6 e see below). PC1 reflects the direction of the head and the distal articular surface in relation to the shaft (Fig. 3a). PC2 relates to the length of the radius between the radial tuberosity and 80% level of the shaft and the orientation of the tip of the styloid process (Fig. 3b). PC6 is related to the position of the radial tuberosity (Fig. 3c). The position of the radial tuberosity has been mentioned as a distinct Neanderthal feature and when scatter-plots of the PCs were observed, PC6

Figure 1. Morphological trends for the medial curve of the radius for Neanderthals, early and recent modern humans. (a) Principal component 1: anterior view. Negative values have a higher degree of curvature than positive values. (b) Principal component 2: anterior view. Positive values show an increased medial extension of the proximal interosseous crest and a medial direction of the distal curve (more medially expanded ulnar notch), whereas negative values show no medial expansion of the proximal interosseous crest and an ulnar notch that is not medially projected. (c) Principal component 3: lateral view. Positive values have a more sinusoidal shape, whereas negative values have no sinusoidal shape. Positive and negative visualizations correspond to the most extreme positive and negative scores for each PC.

(4.71% of variation) showed Neanderthals to have mostly positive values and was therefore included in the following analyses. None of these PCs were correlated with centroid size. There was a significant effect of group membership (Neanderthals, early H. sapiens, recent H. sapiens) in lateral curvature of the radius (Medial Curve PC1 d.f. ¼ 2; F ¼ 16.042; P < 0.001 and Lateral Curve PC1 d.f. ¼ 2; F ¼ 5.738; P < 0.005). The degree of mediolateral curvature is the most important PC for both the medial and lateral surface (mcurveAllPC1 and lcurveAllPC1). This is reflected in the significant correlation (r ¼ 0.640, N ¼ 55, P < 0.001) between the two curvature PCs. Post-hoc comparisons show that, despite the overlap in the data range, Neanderthals have the highest degree of lateral curvature and are significantly different from early and recent H. sapiens. The H. sapiens groups are indistinguishable from each other (Fig. 4). There is a significant effect of group membership (Neanderthals, early H. sapiens, recent H. sapiens) on the second principal component of the lateral curvature of the radius (Lateral Curve PC2

400

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Figure 2. Morphological trends for the lateral curve of the radius for Neanderthals, early and recent modern humans. (a) Principal component 1: anterior view. Negative values have a higher degree of curvature whereas positive values have a lower degree of lateral curvature. (b) Principal component 2: anterior view. Positive values have a more proximal apex of curvature and a more laterally projecting styloid process, whereas negative values have their apex of curvature at mid-shaft and lack the lateral projection of the styloid process. (c) Principal component 3: lateral view. Positive values are more sinusoidal. Negative values are not sinusoidal. Positive and negative visualizations correspond to the most extreme positive and negative scores for each PC.

d.f. ¼ 2; F ¼ 6.489; P < 0.005). Post-hoc comparisons indicate that recent H. sapiens have a more proximal apex of lateral curvature and a more projecting styloid process compared with the Paleolithic populations (Neanderthals and early H. sapiens). Despite the initial observation that Neanderthals scored on the positive side of the variation for radial tuberosity orientation (PC6 for the epiphyses), the ANOVA failed to find significant differences between any of the PCs for radial epiphyseal shape (Table 4). The groups (Neanderthals, early H. sapiens, and recent H. sapiens) (Table 4) are significantly different for radial head shape (d.f. ¼ 2; F ¼ 6.891; P < 0.001), relative neck length (d.f. ¼ 2; F ¼ 5.039; P < 0.010), neck-shaft angle (d.f. ¼ 2; F ¼ 3.059; P < 0.050), and maximum length (d.f. ¼ 2; F ¼ 3.034; P < 0.050). The post-hoc comparisons showed that Neanderthals had a more antero-posteriorly ovoid radial head and a relatively longer radial neck, but none of the post-hoc comparisons found significant patterns in the neck-shaft angles and maximum length. A discriminant function with cross-validation using the linear measurements and the most important PCs was used to separate

Figure 3. Morphological trends for the epiphyses of the radius for Neanderthals, early and recent modern humans. All medial view.(a) Principal component 1. Individuals with negative values have a more anteriorly oriented head, whereas those with positive values are more posteriorly oriented. (b) Principal component 2. Negative values indicate a shorter distance between the radial tubercle and the 80% level of the shaft and a more posteriorly located styloid process and positive values have a longer distance and a more anteriorly located styloid process. (c) Principal component 6. Individuals with negative values have a more antero-medially located radial tuberosity compared to those with positive values who have a more medio-posteriorly located tuberosity. Positive and negative visualizations correspond to the most extreme positive and negative scores for each PC.

the three populations in Table 3. The recent H. sapiens sample was summarized into population means. Function 1 explains 60.8% of the variance and best separates the Neanderthals from both groups of H. sapiens. Function 1 (in order of greatest absolute loading of the variable with the function, Table 5) shows the Neanderthal morphology to be a combination of a higher degree of lateral curvature of the radial shaft (McurvePC1), a more antero-posteriorly wide radial head, a relative large shaft, a more antero-posteriorly sinusoidal shape of the shaft (LcurvePC3), a more proximal apex of lateral curvature (LcurvePC2) and a more projecting styloid process (LcurvePC1), a more medially located radial tuberosity (EpiPC6), and a more developed proximal interosseous crest (McurvePC2). Function 2 (39.2%) separated the recent H. sapiens from the early H. sapiens sample. Neanderthals fall in between. The early H. sapiens sample has a larger radial head relative to shaft breadth and to maximum length, is longer overall with a low neck-shaft angle and a rounder mid-shaft shape, reflecting the poor development of the interosseous crest. In the original classification results, 85.7% were correctly classified. Of the fossils, Shanidar 1, Kostienki 14 and Dolni Vestonice 13 were classified as recent H. sapiens, and Skhul 4 was classified as a Neanderthal. The Kazach Medieval pastoralist sample mean was classified as Neanderthal. Cross-validation results correctly classified three out of nine Neanderthals, and seven out of 12 early H. sapiens. This moderate cross-validation result reflects the

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

401

Figure 4. The lateral curve of the radius for Neanderthals, early and recent Homo sapiens. (Horizontal line ¼ mean, Box ¼ 2 S.D., whiskers: range). The lower values for Neanderthals indicate that they are more laterally curved than the modern humans.

variability in the fossil groups and the importance of single individuals for the calculation of the functions. Overall, the crossvalidation results give correct classification in 73.5% of cases. The correlations are presented in Table 6. There are significant correlations between absolute latitude and radius morphology. Individuals from higher latitudes have relatively larger head proportions, increased lateral curvature (LcurvePC1), a more developed proximal interosseous crest (McurvePC2) and a more antero-posteriorly sinusoidal diaphysis (McurPC3). Other correlations reflect the relationship between the size and shape of the radial head and the shaft (Fig. 5). Ulna The first four PCs of the posterior curve of the ulna analysis explain 33.7%, 23.3%, 13.4% and 6.31%, respectively, of the variation (total 76.71%). Subsequent PCs explain minimal amounts of the variation and are not considered further. PC1 reflects differences in medio-lateral curvature (Fig. 6a). PC2 is the sinusoidal shape of the shaft in the medio-lateral plane (Fig. 6b). PC3 relates to the sinusoidal shape of the shaft in the antero-posterior plane and is most related to posterior subtense (Fig. 6c). PC4 is the deflection of the proximal shaft (Fig. 6d). None of the PCs were correlated with centroid size. The first three PCs of the analysis of the proximal articulation of the ulna explain 20.4%, 16.6% and 7.89%, respectively, of the variation (total 44.9%). Subsequent PCs explain minimal amounts of the variation and are not considered further. PC1 reflects differences in the orientation of the proximal ulna in relation to the shaft (Fig. 7a). PC2 relates to the length of the neck (the distance between the 80% level of the shaft and the coronoid process) (Fig. 7b). PC3 shows the orientation of the trochlear notch (Fig. 7c). None of the PCs were correlated with centroid size. The groups (Neanderthals, early H. sapiens, recent H. sapiens) are significantly different for PC2 of the posterior curve (d.f. ¼ 2; F ¼ 4.884; P < 0.02). Neanderthals have a significantly less

sinusoidal shape in the medio-lateral plane of the diaphysis of the ulna than recent H. sapiens. Neanderthals also show a more anterior facing trochlear notch compared with all H. sapiens (ProxAllPC3: d.f. ¼ 2; F ¼ 27.941; P < 0.001). The groups are significantly different for length (d.f. ¼ 2; F ¼ 4.946; P < 0.02), M. brachialis insertion ratio (position of the M. brachialis insertion relative to bone length) (d.f. ¼ 2; F ¼ 19.279; P < 0.001), coronoid olecranon height ratio (relative height of the coronoid compared to the olecranon) (d.f. ¼ 2; F ¼ 3.424; P < 0.040), proximal articulation-shaft size ratio (d.f. ¼ 2; F ¼ 5.545; P < 0.01), and mid-shaft size relative to length (d.f. ¼ 2; F ¼ 5.796; P < 0.005). The post-hoc comparisons between the three groups showed early H. sapiens had significantly longer radii than the recent H. sapiens sample and the Neanderthals. Neanderthals have a higher M. brachialis insertion ratio, indicating the lower M. brachialis insertion relative to bone length compared with both early and recent H. sapiens. The coronoid and olecranon height are more equal in Neanderthals, whereas in early H. sapiens the olecranon is deeper. Neanderthals have a relatively larger proximal articulation compared with early H. sapiens. Neanderthals are larger at the mid-shaft relative to length than recent H. sapiens (Table 4). The correlations between these variable for the recent H. sapiens sample are presented in Table 8 and indicate that the relative size of the articulations is correlated with absolute latitude and overall length of the ulna. The morphology of the proximal articulation is related to the relative insertion of the M. brachialis muscle and the relative size of the articulations and shaft. A discriminant function analysis with cross-validation, using the PCs and the linear measurements included in the analyses above, was used to separate the three samples based on ulna morphology. The recent H. sapiens sample was summarized into population means. Function 1 separates the Neanderthals from both groups of H. sapiens best. Function 2 separates the early from recent H. sapiens groups. Neanderthals are more similar to recent H. sapiens for Function 2 (Fig. 8). Function 1 explains 62.1% of variation in the sample. This variation reflects (in decreasing order based on

402

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Table 4 ANOVA results and descriptives for Neanderthals, early and recent Homo sapiens and the univariate measurements of the lower arm. F Radius Maximal length mm

2.798

0.062

Neck-shaft angle

3.177

<0.05a

Head shape ratio

6.891

<0.001a

Head size rel length

3.201

<0.05a

Head size rel. mid-shaft

2.152

0.118

Shaft shape mid-shaft

0.073

0.930

Relative shaft size

1.790

0.168

4.946

<0.02a

Olecranon/shaft ratio

5.545

<0.010a

Coronoideolecranon ratio

3.424

<0.050a

16.301

<0.001a

Relative shaft size

5.796

<0.005a

Relative head size

1.380

0.261

Head orientation angle

1.484

0.236

Ulna Maximum length

Brachialis ratio

a

Sig.

Mean

S.D.

Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens

9 11 356 9 11 356 9 11 356 9 11 356 9 11 356 9 11 356 9 11 356

227.76 244.68 235.13 31.66 28.61 36.15 116.02 103.90 105.45 30.93 29.88 28.83 146.47 148.52 141.55 120.60 119.20 121.00 12.66 11.74 12.13

26.63 23.34 19.79 13.18 13.23 13.91 9.59 7.47 8.73 2.54 2.08 3.05 16.55 17.09 15.92 25.81 28.87 20.05 1.29 1.38 1.23

Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens Neanderthal Early Homo sapiens Recent Homo sapiens

8 13 344 8 13 344 8 13 344 8 13 344 8 13 344 8 13 344 8 13 344

247.91 268.45 250.35 9.42 8.32 9.10 105.63 107.35 106.75 26.45 23.62 22.97 11.81 11.26 10.48 15.80 14.62 15.59 20.77 24.32 24.09

28.18 19.38 20.65 1.14 1.05 1.00 1.71 1.87 2.47 2.23 1.56 1.81 1.66 1.14 0.89 1.21 2.55 1.87 7.17 7.13 3.68

Significant at a ¼ 0.05.

loading of the variables and the function, Table 7), the more anterior gape of the olecranon (ProxPC3), the lower insertion of the M. brachialis muscle, larger mid-shaft size relative to bone length, a more medio-lateral sinusoidal shaft (PcurvePC2), and a more antero-posterior sinusoidal shaft (PcurvePC3). Function 2 explains the remaining 36.7% of variation and reflects a relatively smaller proximal articulation on a longer ulna, a greater olecranon to Table 5 Discriminant function coefficients e Radius.

McurPC1 Head shape ratio Relative shaft size LcurPC3 LcurPC2 LcurPC1 epiPC6 McurPC2 Head size relative to mid-shaft Head size relative to length Max length Neck-shaft Shaft shape at mid-shaft a

N

Function 1

Function 2

0.559a 0.475a 0.424a 0.413a 0.355a 0.270a 0.174a 0.155a 0.198 0.088 0.312 0.139 0.117

0.441 0.078 0.332 0.019 0.328 0.061 0.069 0.007 0.452a 0.391a 0.365a 0.191a 0.147a

Largest absolute correlation variable and discriminant function.

coronoid height, a more laterally deflected trochlear notch (ProxPC1), a higher degree of medio-lateral diaphyseal curvature (pcurvePC1), a longer neck (ProxPC2), a relatively smaller head rotated away from the shaft-axis and a more anteriorly projected olecranon (pcurvePC4). The original classification resulted in 84.9% of all cases correctly classified. Of the fossils, Shanidar 1 was classified as an early H. sapiens, and Skhul IV and Sungir were classified as Neanderthals. Ein Gev 1 was classified as a recent H. sapiens. The Natufian, African-Americans and Khoikhoi were classified as early H. sapiens. The DFA with cross-validation correctly classified 71.7% of cases. The analysis correctly classified four out of eight Neanderthals and eight out of thirteen early H. sapiens. Eighty-one percent of recent H. sapiens were classified correctly. The correlations (Table 8) demonstrate that there is a positive correlation between latitude and olecranon shaft ratio and head size, indicating increased articulation size in individuals from high latitudes. Both of these variables are negatively correlated to length, indicating that individuals with shorter ulnae have larger articulations. Relative head size is also correlated with a lower M. brachialis insertion and a relatively larger diaphysis at mid-shaft. The more anterior oriented trochlear notch (ProxPC3) is negatively correlated with the insertion of the M. brachialis and the relative

Table 6 Correlations radius. N ¼ 55

Neck-shaft

Head shape ratio Head size rel length Head size rel mid-shaft Shaft shape mid-shaft Rel shaft size McurPC1 McurPC2 McurAPC3 LcurPC1 LcurPC2 LcurPC3 EpiPC6 Centroid Mcurve a

r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig.

0.092 0.606 0.184 0.299 0.031 0.864 0.458a 0.007 0.253 0.149 0.221 0.209 0.290 0.096 0.167 0.346 0.563a 0.001 0.363a 0.035 0.365a 0.034 0.259 0.138 0.266 0.128 0.115 0.517 0.205 0.245

Neckshaft

Max length

0.275a 0.037 0.089 0.508 0.250 0.059 0.132 0.323 0.063 0.641 0.102 0.448 0.026 0.849 0.093 0.489 0.171 0.199 0.015 0.911 0.036 0.790 0.147 0.271 0.009 0.946 0.292a 0.026

0.134 0.317 0.246 0.063 0.028 0.838 0.198 0.137 0.442a 0.001 0.223 0.093 0.110 0.412 0.178 0.181 0.425a 0.001 0.029 0.827 0.017 0.899 0.161 0.235 0.975a 0.000

Head shape ratio

0.441a 0.001 0.069 0.604 0.004 0.974 0.181 0.173 0.345a 0.008 0.326a 0.012 0.101 0.451 0.210 0.114 0.070 0.600 0.307a 0.019 0.201 0.138 0.139 0.300

Rel Head Size

0.449a 0.000 0.128 0.339 0.115 0.391 0.182 0.172 0.135 0.311 0.326a 0.013 0.053 0.690 0.139 0.297 0.288a 0.029 0.233 0.084 0.184 0.167

Head size rel length

Head size rel mid-shaft

0.148 0.266 0.657a 0.000 0.094 0.483 0.060 0.655 0.026 0.847 0.002 0.987 0.096 0.473 0.030 0.825 0.082 0.548 0.057 0.672

0.393a 0.002 0.014 0.918 0.334a 0.010 0.166 0.214 0.058 0.663 0.058 0.668 0.090 0.503 0.055 0.690 0.221 0.095

Rel shaft size

0.166 0.214 0.186 0.161 0.137 0.304 0.287a 0.029 0.246 0.063 0.134 0.316 0.005 0.971 0.501a 0.000

McurPC1

McurPC2

McurPC3

LcurPC1

LcurPC2

LcurPC3

EpiPC6

0.214 0.106 0.076 0.569 0.640a 0.000 0.492a 0.000 0.265a 0.045 0.147 0.280 0.218 0.100

0.016 0.906 0.120 0.370 0.125 0.349 0.186 0.163 0.073 0.592 0.122 0.361

0.187 0.160 0.219 0.098 0.012 0.931 0.007 0.960 0.205 0.123

0.212 0.110 0.194 0.144 0.018 0.897 0.372a 0.004

0.133 0.318 0.102 0.456 0.057 0.669

0.247 0.066 0.055 0.681

0.181 0.182

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Max length

Absolute latitude N ¼ 34

Significant at a ¼ 0.05 (2-tailed).

403

404

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Figure 5. Canonical Linear Discriminant Function 1 and 2 for Neanderthals (rectangles), early Homo sapiens (circles) and recent Homo sapiens (crosses). Solid squared represent group means.

diaphysis size at mid-shaft, showing that individuals with a more anterior oriented trochlear notch have a lower muscle insertion and have larger diaphyses. Discussion The objective of this paper was to investigate the differences in lower arm morphology among Neanderthals, Late Pleistocene H. sapiens and recent H. sapiens. It specifically addresses the

Figure 7. Morphological trends for the proximal ulna for Neanderthals, early and recent modern humans e wireframes. (a) Principal component 1: anterior view. Positive values have a proximal ulna that is medially projected with a medial facing trochlear notch, whereas negative values have a proximal ulna that is laterally projected and has a more lateral facing trochlear notch. (b) Principal component 2: anterior view. Positive values have a longer distance between the 80% and the coronoid process, whereas negative values have short distances. (c) Principal component 3: lateral view. Positive values have a more proximo-anterior facing trochlear notch and negative values have a more anterior facing trochlear notch. Positive and negative visualizations correspond to the most extreme positive and negative scores for each PC.

Figure 6. Morphological trends for the posterior curve of the ulna for Neanderthals, early and recent modern humans. (a) Principal component 1: anterior view. Negative values have a higher degree of medio-lateral curvature, whereas positive values have a lower degree of curvature. (b) Principal component 2: anterior view. Positive values have a less sinusoidal shape in the medio-lateral plane than negative values. (c) Principal component 3: medial view. Positive values are more posteriorly sinusoidal compared to negative values. d: Principal component 4: medial view. Positive values show a bent proximal shaft indicating a more anteriorly projected head, whereas negative values are relatively straight. Positive and negative visualizations correspond to the most extreme positive and negative scores for each PC.

question of whether the morphology of the radius and ulna can be used to distinguish among these groups and how differences in morphology can be related to forearm function. Therefore, morphology of the radius and ulna was described by a combination of linear measurements and three-dimensional shape variables. The results of the univariate and multivariate analyses demonstrate that the morphology of both the radius and ulna can be used to distinguish among the groups investigated here. In agreement with previous findings (Fischer, 1906; Vandermeersch and Trinkaus, 1995; Carretero et al., 1999; Czarnetzki, 2000), Neanderthals have the highest degree of lateral diaphyseal curvature in the radius. In addition, they have a more antero-posteriorly ovoid radial head and a relatively longer radial neck. Although the univariate analyses did not find significant differences between early and more recent H. sapiens, and only a few features were identified for which Neanderthals were significantly different from modern humans, the discriminant function analysis demonstrated that there are trends in the overall morphology of the radius that allow these groups to be distinguished. Neanderthal radii are characterized by a high degree of lateral diaphyseal curvature with

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Figure 8. Canonical Linear Discriminant Function 1 and 2 Neanderthals (rectangles), early Homo sapiens (circles) and recent H. sapiens (crosses). Solid squared represent group means.

a diaphysis that is more antero-posteriorly sinusoidal and has a more proximal apex. The radius also has a more anteroposteriorly wide radial head, a relatively large mid-shaft, a more projecting styloid process, a more developed proximal interosseous crest and, confirming the observations of Trinkaus and Churchill (1988), a more medially located radial tuberosity. This overall radius morphology indicates that Neanderthals had forearms with better mechanical advantage than those of H. sapiens. Increased lateral curvature of the radius not only results in a greater distance between the ulna and the radius and therefore the insertions between the M. pronator quadratus and M. pronator teres, but also allows for larger muscle bellies while maintaining tendon insertions close to the axis of the bone. This maximizes the degree of pronation and supination by maintaining the size and axis of rotation (Yasutomi et al., 2002). The increased supination strength of the Neanderthals is also confirmed by the relatively large midshaft and more developed proximal interosseous crest, the more medial position of the radial tuberosity and the therefore increased lever advantage of the M. biceps brachii (Trinkaus and Churchill, 1988), resulting in a greater range of action of this muscle as a supinator. The discriminant analysis indicated that, compared with recent H. sapiens, early H. sapiens radii are long, which could reflect their more warm adapted body proportions and therefore Table 7 Discriminant function coefficients - Ulna.

ProxPC3 Brachialis Ratio Relative mid-shaft size pcurvePC2 pcurvePC3 Olecranon shaft ratio Max length Coronoid-olecranon ratio ProxPC1 pcurvePC1 ProxPC2 Relative head size pcurvePC4 Head orient angle a

Function 1

Function 2

0.683a 0.561a 0.316a 0.291a 0.130a 0.002 0.073 0.057 0.018 0.144 0.110 0.002 0.053 0.124

0.169 0.177 0.031 0.010 0.066 0.396a 0.363a 0.303a 0.248a 0.228a 0.216a 0.198a 0.179a 0.130a

Largest absolute correlation variable and discriminant function.

405

a thermoregulatory adaptation (Trinkaus, 1981; Ruff, 1991; Holliday, 1995, 1997a,b, 1999; Holliday and Falsetti, 1995; Holliday and Ruff, 1997; Churchill, 1998; Steegmann et al., 2002; Steegmann, 2005; Weaver, 2009). Early H. sapiens are furthermore distinguished by having a larger radial head relative to bone length and width, a lower neck-shaft angle and a weakly developed interosseous crest when compared with recent H. sapiens. The analyses of the ulna confirm the overall strength of the Neanderthal forearm and identified significant morphological differences among the three groups. Characteristic of the Neanderthals is the more distally positioned M. brachialis insertion, increasing lever advantage during flexion, and a large mid-shaft relative to bone length. Also relative to bone length, Neanderthals have larger proximal epiphyses than early and recent modern H. sapiens, reflecting large joint reaction forces. Early H. sapiens ulnae are longer and are also larger at mid-shaft compared with recent H. sapiens. Despite earlier reports on the increased posterior subtense of the ulna (Fischer, 1906) and the more anterior facing trochlear notch (Trinkaus and Churchill, 1988; Churchill et al., 1996; Churchill, 2006; Churchill and Rhodes, 2009) in Neanderthals, the shape analysis and subsequent univariate analyses failed to identify any significant differences. Nevertheless, when the overall morphological variables were used in a discriminant function analysis, distinct groups were identified. Neanderthals are distinguished from early and recent H. sapiens primarily by the more anterior facing trochlear notch and the lower M. brachialis insertion. They also have a larger diaphysis at mid-shaft relative to bone length and a more mediolateral and antero-posterior sinusoidal shaft. This again reflects the relative strength of the Neanderthal forearm and the possibility of packing larger muscle bellies while maintaining the tendons close to the articulations. Relative to recent H. sapiens and Neanderthals, early H. sapiens have a longer, more medio-laterally curved ulnae with relatively smaller epiphyses. Their trochlear notch is projected proximo-anteriorly and is laterally deflected. The long length of their ulnae could be a sign of their warm adapted body proportions and more recent African descent (Trinkaus, 1981; Ruff, 1991; Holliday, 1995, 1997a,b, 1999; Holliday and Falsetti, 1995; Churchill, 1998; Steegmann et al., 2002; Steegmann, 2005; Weaver, 2009). The similarity of the early H. sapiens to the modern Africans is reflected also in the misclassification of the Natufian, African-Americans and KhoieKhoi as early H. sapiens. Despite the risk of colinearity between these variables, the results are consistent with what has been reported in the literature before and are consistent with predictions made upon qualitative observations. The significant correlation between some of the radius and ulna variables with latitude confirms the empirical relationship of climate and body proportions and their effect on post-cranial morphology (Trinkaus, 1981; Ruff, 1991; Holliday, 1997a, 1999; Churchill, 1998; Steegmann et al., 2002; Steegmann, 2005; Weaver, 2009). Therefore, one could argue that the differences observed between Neanderthals and H. sapiens reflect their different climatic environments and are therefore a consequence of cold-adapted forearms. In this case, it appears that it is possible to discriminate between the selective agents of the adaptive changes in this morphology in the view of the concept of equifinality described by Churchill (2006). The forearms of the Neanderthals are on the one hand an adaptation to the cold environment but on the other adapted to maintain optimal functionality. The shorter forearms of the Neanderthals compared with those of early H. sapiens have retained large articulations and muscle attachments despite shortening of the diaphysis. The large articulations indicate large joint reaction forces. Increased curvature not only lengthens the diaphysis but also maintains lever advantage of the muscles and

406

Table 8 Correlations ulna. N ¼ 54

Max length

Head orient angle pcurvePC1 pcurvePC2 pcurvePC3 pcurvePC4 ProxPC1 ProxPC2 ProxPC3 a

Max length

Olecranon shaft ratio

r Sig. r Sig. r Sig.

0.072 0.700 0.495a 0.005 0.076 0.686

0.446a 0.001 0.185 0.181

0.263 0.054

r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig. r Sig.

0.309 0.090 0.274 0.136 0.361a 0.046 0.244 0.186 0.030 0.875 0.087 0.642 0.011 0.954 0.255 0.167 0.299 0.102 0.290 0.114 0.233 0.207

0.098 0.481 0.135 0.332 0.298a 0.029 0.019 0.890 0.213 0.122 0.184 0.183 0.213 0.122 0.047 0.737 0.122 0.378 0.036 0.798 0.120 0.388

0.423a 0.001 0.139 0.317 0.672a 0.000 0.196 0.155 0.111 0.425 0.244 0.075 0.386a 0.004 0.165 0.235 0.287a 0.035 0.529a 0.000 0.015 0.914

Significant at a ¼ 0.05.

Coronoid olecranon ratio

0.117 0.401 0.116 0.405 0.253 0.065 0.158 0.253 0.047 0.734 0.517a 0.000 0.011 0.936 0.023 0.870 0.177 0.200 0.140 0.313 0.064 0.644

Brachialis ratio

0.480a 0.000 0.379a 0.005 0.192 0.164 0.124 0.373 0.393a 0.003 0.032 0.817 0.125 0.369 0.107 0.440 0.697a 0.000 0.571a 0.000

Rel mid-shaft size

0.304a 0.025 0.045 0.748 0.331a 0.014 0.118 0.394 0.017 0.902 0.099 0.475 0.027 0.849 0.283a 0.038 0.417a 0.002

Rel head size

Head orient angle

0.134 0.334 0.043 0.757 0.265 0.053 0.210 0.127 0.029 0.834 0.301a 0.027 0.499a 0.000 0.016 0.906

0.175 0.204 0.071 0.612 0.267 0.051 0.068 0.623 0.006 0.965 0.454a 0.001 0.267 0.051

PcurvePC1

PcurvePC2

PcurvePC3

PcurvePC4

ProxPC1

ProxPC2

0.022 0.873 0.422a 0.001 0.022 0.875 0.008 0.953 0.035 0.799 0.221 0.108

0.067 0.630 0.029 0.838 0.030 0.829 0.307a 0.024 0.276a 0.043

0.064 0.648 0.117 0.400 0.100 0.470 0.243 0.076

0.010 0.943 0.011 0.934 0.007 0.960

0.192 0.165 0.085 0.539

0.120 0.389

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Olecranon shaft ratio Coronoid Olecranon Ratio Brachialis ratio Rel mid-shaft size Rel head size

Absolute latitude N ¼ 31

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

therefore strength. Most of the phenotypic variation that is observed in the Neanderthals, however, may be due to the retention of primitive characters. Apes are also characterized by a more medial radial tuberosity (Trinkaus and Churchill, 1988; Aiello and Dean, 1990; Pearson and Grine, 1997), increased lateral curvature of the radial diaphysis (Martin and Saller, 1959; Knussman, 1967 in Swartz, 1990) and a more anteriorly facing trochlear notch (Drapeau, 2004, 2008). It remains to be investigated whether habitual behavior and habitual positions used by these groups are the source of differences in morphology (Stock and Pfeiffer, 2001, 2004; Stock, 2002, 2006; Pearson et al., 2006; Stock and Shaw, 2007) or whether the morphology is relatively insensitive to locomotor and habitual behavior as has been suggested for the hominoid proximal radius by Patel (2005). In order to investigate this, it is necessary to look at the ontogeny of lower arm morphology and better understand the effect of habitual behavior on the lower arm.

Conclusions This study is the first to consider the overall morphology of the radius and ulna in a comprehensive comparison of the morphological similarities and differences in Neanderthals, Late Pleistocene and recent H. sapiens. This was made possible by methodological advances in 3D geometric morphometrics allowing for certain features, such as longitudinal shaft shape, to be quantified and analyzed as part of the rest of the morphology. The analyses identified both similarities and differences among the three groups. Neanderthals have a radius that is characterized by a more laterally curved diaphysis, a more medially placed radial tuberosity, a longer radial neck, a more ovoid radial head and a well developed proximal interosseous crest. The Neanderthal ulna discriminant function classification distinguished Neanderthals from early and recent H. sapiens with some success. The results presented here confirm many of the differences observed between Neanderthals and H. sapiens that had, to date, not been quantified or compared with both Late Pleistocene and a varied range of recent H. sapiens. The misclassifications and the lack of significant differences between our groups for individual variables indicate that Neanderthals, early and recent H. sapiens are similar in their forearm morphology but that a combination of

407

certain features makes it possible to distinguish among the groups. As is the case for so many Neanderthal features, many of the ranges of variation for the variables presented here overlap with those of H. sapiens, and when considered individually do not distinguish Neanderthals from H. sapiens. It is the suite of the expression of a range of characteristics that defines the derived Neanderthal morphology. This morphology reflects primarily a strong and coldadapted short forearm compared with all H. sapiens. The forearms of H. sapiens are less adapted to extensive use in pronation and supination. The early H. sapiens forearm is longer and reflects a more warm adapted morphology. From a taxonomic and phylogenetic perspective, it is necessary to understand the origin of these differences. Many of the differences between Neanderthals can be explained as a secondary consequence of the differences in body proportions due to the hyper-polar body shape of the Neanderthals (Boule and Vallois, 1952; Trinkaus, 1981; Churchill, 1998; Pearson, 2000a,b; Aiello and Wheeler, 2003; Weaver, 2003; De Groote et al., 2006; Krause et al., 2007; Shackelford, 2007) but also as retentions of the primitive condition of all hominoids. In order to understand this better, more work is necessary on the ontogeny and adaptive pressures on forearm functional morphology in H. sapiens and non-human primates and on an increased number of hominin fossils intermediate to the Neanderthals and Australopithecines.

Acknowledgements I would like to thank Philipp Mitteroecker and Philipp Gunz for the use of their MathematicaÒ functions, and all of the curators at the museums and institutions mentioned in Table 2 for permission to measure specimens. I am indebted to the University College London Graduate School, the Ruggles-Gates Fund, the Leakey Trust, the Wenner Gren Foundation and the European Union Synthesys programme for grant support.

Appendix. Landmark definitions

Radius Landmarks used to calculate the linear measurements Maximal length (Martin no 1) RADL1 RADL2 Neck-shaft angle (Martin no7) HDII4a NSAG2 A 80% Superior head diameter (Martin no 4) HDIS1m HDIS2l HDIS3p HDIS4a Mid-shaft diameters A 50% P 50% M 50% L 50%

Maximum length measured from the most superior point on the articular surface on the head to the most distal point on the styloid process. The most superior point on the articular surface on the head The most distal point on the styloid process Also collo-diaphyseal angle. Martin no7. The angle described by the shaft-axis (going through the middle of the shaft) and the neck-axis (going through the middle of the neck) Most anterior point on a line describing the maximum diameter on the most inferior edge of the head Point where the most narrow diameter of the neck intersects with the anterior neck-axis through the middle of the shaft Most anterior point at 80% level. The 0% shaft level is defined as the most inferior edge of the styloid process; the 100% is the most superior point on the articular surface on the head. Maximum diameter of the radial head on the edge of the articular surface. Used to calculate Head shape ratio and Head size relative to length. Most medial point on a line describing the maximum medio-lateral diameter on the most superior edge of the head Most lateral point on a line describing the maximum medio-lateral diameter on the most superior edge of the head Most posterior point on a line describing the maximum antero-posterior diameter on the most superior edge of the head Most anterior point on a line describing the antero-posterior maximum diameter on the most superior edge of the head Medio-lateral and antero-posterior diameters taken at the 50% level. The 0% shaft level is defined as the most inferior edge of the styloid process; the 100% is the most superior point on the articular surface on the head. Used for shaft shape mid-shaft and relative shaft size. Most anterior point at 50% level. Most posterior point at 50% level. Most medial point at 50% level. Most lateral point at 50% level.

408

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Appendix (continued) Landmarks used to calculate the linear measurements Landmarks used in the epiphysis configuration 80% diameter M 80% L 80% A 80% P 80% The radial tuberosity RADT Distal radial articulation ARTSp ULNT ARTSa RADL2

Medio-lateral diameter taken at the 80% level. The 0% shaft level is defined as the most inferior edge of the styloid process; the 100% is the most superior point on the articular surface on the head. Most medial point at 80% level. Most lateral point at 80% level. Most anterior point at 80% level. Most posterior point at 80% level. The most projecting point on the radial tuberosity The tip where the radial tuberosity projects maximally The middle of the distal radial articular surface edge on the posterior side. The middle of a curved surface Middle of distal radial articular surface edge on the posterior side. Middle of curved surface Middle of medial articular surface on the ulnar notch Middle of distal radial articular surface edge on the anterior side. Middle of curved surface Most distal point on the styloid process

Curvature Medial and lateral curve MCURV LCURV

Curvature of the radius along two sides. Medial curvature from 80% down to the middle of the ulnar notch. Lateral curvature from 80% level down to the tip of the styloid process. Semi-landmarks taken every 5 mm along the medial curve of the radius Semi-landmarks taken every 5 mm along the lateral curve of the radius

Ulna Landmarks used to calculate the linear measurements Maximum length (Martin no1) ULNL1 ULNL2 Olecranon shaft ratio OLTP OLMXp Coronoid-Olecranon ratio (Martin 7a and 8a) TRWD1 TRWD2 CORPR Brachialis ratio (Solan, 1992) ULNL1 BRACH Mid-shaft diameters A 50% P 50% M 50% L 50% Distal articulation diameter HDIAp HDIAa HDIAm HDIAl Head orientation angle A 80%

Maximum length measured from the most superior point on the olecranon process to the most distal point on the articular surface (not styloid process because of preservation issues in archaeological samples) Most superior point on the olecranon process Most distal point on the radial articulation surface Size of the proximal articulation: olecranon size/length * 100. Olecranon size is the depth from the tip of the olecranon to the posterior surface of the trochlear notch. Tip of the olecranon process Most posterior point on the olecranon process Height olecranon/height coronoid * 100. Height olecranon (Martin 7a) ¼ Distance from OLTP to midpoint between TRWD1 and TRWD2. Height coronoid (Martin 8a) ¼ Distance from CORPR to midpoint between TRWD1 and TRWD1 Most medial point on the trochlear notch along the minimum width line perpendicular to the shaft-axis Most lateral point on the trochlear notch along the minimum width line perpendicular to the shaft-axis Tip of the coronoid process The position of the brachialis tuberosity: Distance from the proximal extremity (ULNL1) to the most distal point of the brachialis tuberosity (BRACH2i). Most superior point on the olecranon process The most distal point of the brachialis insertion Medio-lateral and antero-posterior diameters taken at the 50% level. The 0% shaft level is defined as the most distal point on the articular surface; 100% is the most superior point on olecranon process. Used to calculate relative shaft size. Most anterior point at 50% level. Most posterior point at 50% level. Most medial point at 50% level. Most lateral point at 50% level. Antero-posterior and medio-lateral diameter of the superior edge of the distal articulation. Used to calculate Relative head size. Most posterior point on the superior edge of the distal articulation Most anterior point on the superior edge of the distal articulation Most medial point on the superior edge of the distal articulation Most lateral point on the superior edge of the distal articulation The angle at the olecranon when a triangle is formed between the anterior point at 80% shaft length (A80%), the tip of the olecranon (OLTP) and the coronoid (CORPR). Most anterior point at 80% level. The 0% shaft level is defined as the most distal point on the articular surface; the 100% is the most superior point on olecranon process.

Landmarks used in the proximal condyle configuration 80% diameter M 80% L 80% A 80% P 80% Proximal articulation dimension OLTP OLMXm OLMXl

Medio-lateral diameter taken at the 80% level. The 0% shaft level is defined as the most distal point on the articular surface; the 100% is the most superior point on olecranon process. Most medial point at 80% level. Most lateral point at 80% level. Most anterior point at 80% level. Most posterior point at 80% level. The dimensions of the olecranon and coronoid process. Tip of the olecranon process Most medial point on the olecranon process Most lateral point on the olecranon process

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

409

(continued ) Landmarks used to calculate the linear measurements OLMXp TRWD1 TRWD2 CORPR RADNm RADNa RADNp

Most posterior point on the olecranon process Most medial point on the trochlear notch along the minimum width line perpendicular to the shaft-axis Most lateral point on the trochlear notch along the minimum width line perpendicular to the shaft-axis Tip of the coronoid process Most medial point on the radial notch Most anterior notch on the radial notch Most inferior notch on the radial notch

Curvature Posterior curve PCURV

Curvature of the ulna along the posterior surface from 80% down to the tip of the styloid process. Semi-landmarks taken every 5 mm along the posterior curve of the ulna

References Adams, D.C., Rohlf, F.J., Slice, D.E., 2004. Geometric morphometrics: ten years of progress following the ‘revolution’. Int. J. Zool. 71, 5e16. Aiello, L., Dean, C., 1990. Human Evolutionary Anatomy. Elsevier Academic Press, London. Aiello, L., Wheeler, P., 2003. Neanderthal thermoregulation and the glacial climate. In: van Andel, T.H., Davies, W. (Eds.), Neanderthals and Modern Humans in the European Landscape during the Last Glaciation. McDonald Institute Monographs, Oxford, pp. 147e166. Bookstein, F., 1991. Morphometric Tools for Landmark Data. Cambridge University Press, Cambridge. Boule, M., Vallois, H.V., 1952. Fossil Men: a Textbook of Human Palaeontology. Thames and Hudson, London. Carretero, J.M., Lorenzo, C., Arsuaga, J.L., 1999. Axial and appendicular skeleton of Homo antecessor. J. Hum. Evol. 37, 459e499. Churchill, S.E., Rhodes, J.A., 2009. The evolution of the human capacity for “killing at a distance”: the human fossil evidence for the evolution of projectile weaponry. In: Hublin, J.-J., Richards, M.P. (Eds.), The Evolution of Hominin Diets. Springer Netherlands, Dordrecht, pp. 201e210. Churchill, S.E., Pearson, O.M., Grine, F.E., Trinkaus, E., Holliday, T.W., 1996. Morphological affinities of the proximal ulna from Klasies River main site: archaic or modern? J. Hum. Evol. 31, 213e237. Churchill, S.E., 1998. Cold adaptation, heterochrony and Neandertals. Evol. Anthropol. 7, 46e60. Churchill, S.E., 2005. Particulate versus integrated evolution of the upper body in Late Pleistocene humans: a test of two models. Am. J. Phys. Anthropol. 100, 559e583. Churchill, S.E., 2006. Bioenergetic perspectives on Neanderthal thermoregulatory and activity budgets. In: Harvati, K., Harrison, T. (Eds.), Neanderthals Revisited: New Approaches and Perspectives. Springer Netherlands, Dordrecht, pp. 113e133. Czarnetzki, A., 2000. The significance of pathological changes for judging the morphology of classical Neanderthals. In: Orschiedt, J., Weniger, G.-C. (Eds.), Neanderthals and Modern Humans e Discussing the Transition: Central and Eastern Europe from 50.000e30.000 B.P. Neanderthal Museum, Mettmann, pp. 296e302. De Groote, I., Lockwood, C.A., Aiello, L.C., 2006. Long bone curvature in Neanderthals and modern humans: a new method for measuring long bone curvature using 3D landmarks and semilandmarks. Am. J. Phys. Anthropol. Suppl. 42, 82. De Groote, I., Lockwood, C.A., Aiello, L.C., 2010. A New method for measuring long bone curvature using 3D landmarks and semilandmarks. Am. J. Phys. Anthropol. 141, 658e664. De Groote, I., 2011. Femoral curvature in Neanderthals and modern humans: a 3D geometric morphometric analysis. J. Hum. Evol. 60, 540e548. Drapeau, M.S.M., Ward, C.V., Kimbel, W.H., Johanson, D.C., Rak, Y., 2005. Associated cranial and forelimb remains attributed to Australopithecus afarensis from Hadar, Ethiopia. J. Hum. Evol. 48, 593e642. Drapeau, M.S.M., 2004. Functional anatomy of the olecranon process in hominoids and Plio-pleistocene hominins. Am. J. Phys. Anthropol. 124, 297e314. Drapeau, M.S.M., 2008. Articular morphology of the proximal ulna in extant and fossil hominoids and hominins. J. Hum. Evol. 55, 86e102. Dytham, C., 1999. Choosing and Using Statistics: a Biologist’s Guide. Blackwell Science Ltd., Malden. Field, A., 2000. Discovering Statistics using SPSS. Sage Publications Ltd., London. Fischer, E., 1906. Die Variationen an Radius und Ulna des Menschen. Z. Morph. Anthropol. 9, 147e247. Gunz, P., Mitteroecker, P., Bookstein, F., 2005. Semi-landmarks in three dimensions. In: Slice, D.E. (Ed.), Modern Morphometrics in Physical Anthropology. Kluwer Academic, New York, pp. 73e98. Holliday, T.W., Falsetti, A.B., 1995. Lower limb length of European early modern humans in relation to mobility and climate. J. Hum. Evol. 29, 141e153. Holliday, T.W., Ruff, C.B., 1997. Ecogeographical patterning and stature prediction in fossil hominids: comment. Am. J. Phys. Anthropol. 103, 137e140. Holliday, T.W., Ruff, C., 2001. Relative variation in human proximal and distal limb segment lengths. Am. J. Phys. Anthropol. 116, 26e33.

Holliday, T.W., Trinkaus, E., 1991. Limb/trunk proportions in Neandertals and early anatomical modern humans. Am. J. Phys. Anthropol. Suppl. 12, 93e94. Holliday, T.W., 1995. Body size and proportions in the Late Pleistocene Western Old World and the origins of modern humans. Ph.D. Dissertation, University of New Mexico. Holliday, T.W., 1997a. Postcranial evidence of cold adaptation in European Neandertals. Am. J. Phys. Anthropol. 104, 245e258. Holliday, T.W., 1997b. Body proportions in Late Pleistocene Europe and modern human origins. J. Hum. Evol. 32, 423e447. Holliday, T.W., 1999. Brachial and crural indices of European Late Upper Paleolithic and Mesolithic humans. J. Hum. Evol. 36, 549e566. Howell, F.C., 1957. The Evolutionary significance of variation and varieties of “Neanderthal” man. Quart. Rev. Biol. 32, 330e347. Hublin, J.J., Spoor, F., Braun, M., Zonneveld, F., Condemi, S., 1996. A late Neanderthal associated with Upper Palaeolithic artefacts. Nature 381, 224e226. Hublin, J.J., 1989. Climatic changes, paleogeography, and the evolution of the Neandertals. In: Akazawa, T., Aoki, K., Bar-Yosef, O. (Eds.), Neandertals and Modern Humans in Western Asia. Plenum Press, New York, pp. 295e310. Krause, J., Orlando, L., Serre, D., Viola, B., Prufer, K., Richards, M.P., Hublin, J.-J., Hanni, C., Derevianko, A.P., Paabo, S., 2007. Neanderthals in central Asia and Siberia. Nature 449, 902e904. Lanyon, L.E., 1980. The Influence of function on the development of bone curvature an experimental study on the rat tibia. J. Zool. 192, 457e466. Majó, T., Tillier, A.M., Thompson, J.L., Krovitz, G.E., Nelson, A.J., 2003. Postcranial growth in Neandertals and modern humans. In: Foley, R.A., Jablonski, N.G. (Eds.), Patterns of Growth and Development in the Genus Homo. Cambridge University Press, Cambridge, pp. 361e385. Martin, R.D., Saller, K., 1959. Lehrbuch der Anthropologie. Gustav Fischer, Stuttgart. O’Higgins, P., Jones, N., 1998. Facial growth in Cercocebus torquatus: an application of three dimensional geometric morphometric techniques to the study of morphological variation. J. Anat. 193, 251e272. O’Higgins, P., 2000. The study of morphological variation in the hominid fossil record: biology, landmarks and geometry. J. Anat. 197, 103e120. Patel, B.A., 2005. The hominoid proximal radius: re-interpreting locomotor behaviors in early hominins. J. Hum. Evol. 48, 415e432. Patte, E., 1955. Les Néanderthaliens: Anatomie, Physiologie, Comparaisons. Masson et Co., Paris. Pearson, O.M., Grine, F.E., 1997. Re-analysis of the hominid radii from Cave of Hearths and Klasies River Mouth, South Africa. J. Hum. Evol. 32, 577e592. Pearson, O.M., Lieberman, D.E., 2004. The aging of Wolff’s law: ontogeny and responses to mechanical loading in cortical bone. Yearb. Phys. Anthropol. 47, 63e99. Pearson, O.M., Cordero, R., Busby, A., 2006. How different were Neanderthals’ habitual activities? A comparative analysis with diverse groups of recent humans. In: Harvati, K., Harrison, T. (Eds.), Neanderthals Revisited: New Approaches and Perspectives. Springer Netherlands, Dordrecht, pp. 135e156. Pearson, O.M., 2000a. Activity, climate and postcranial robusticity: implications for modern human origins and scenarios of adaptive change. Curr. Anthropol. 41, 569e607. Pearson, O.M., 2000b. Postcranial remains and the origin of modern humans. Evol. Anthropol. 9, 229e247. Rak, Y., Arensburg, B., 1987. Kebara 2 Neanderthal pelvis: first look at a complete inlet. Am. J. Phys. Anthropol. 73, 227e231. Ruff, C.B., Walker, A., 1993. Body size and body shape. In: Walker, A., Leakey, R. (Eds.), The Nariokotome Homo erectus Skeleton. Harvard University Press, Cambridge, pp. 235e263. Ruff, C.B., Trinkaus, E., Walker, A., Larsen, C.S., 1993. Postcranial robusticity in Homo I: temporal trends and mechanical interpretation. Am. J. Phys. Anthropol. 91, 21e53. Ruff, C.B.,1991. Climate and body shape in hominid evolution. J. Hum. Evol. 21, 81e105. Ruff, C.B., 1994. Morphological adaptation to climate in modern and fossil hominids. Am. J. Phys. Anthropol. 37, 65e107. Shackelford, L.L., Trinkaus, E., 2002. Late Pleistocene human femoral diaphyseal curvature. Am. J. Phys. Anthropol. 118, 359e370. Shackelford, L.L., 2007. Regional variation in the postcranial robusticity of late Upper Paleolithic humans. Am. J. Phys. Anthropol. 133, 655e668.

410

I. De Groote / Journal of Human Evolution 61 (2011) 396e410

Slice, D.E., 2005. Modern Morphometrics. Modern Morphometrics in Physical Anthropology. Kluwer Academic Publishers, New York. Solan, M., Day, M.H., 1992. The Baringo (Kapthurin) ulna. J. Hum. Evol. 22, 307e314. Steegmann Jr., A.T., Cerny, F.J., Holliday, T.W., 2002. Neandertal cold adaptation: physiologic and energetic factors. Am. J. Hum. Biol. 14, 566e583. Steegmann Jr., A.T., 2005. Climate, racial category, and body proportions in the U.S. Am. J. Hum. Biol. 17, 393e402. Stock, J.T., Pfeiffer, S.K., 2001. Linking structural variability in long bone diaphyses to habitual behaviors: foragers from the southern African Later Stone Age and the Andaman Islands. Am. J. Phys. Anthropol. 115, 337e348. Stock, J.T., Pfeiffer, S.K., 2004. Long bone robusticity and subsistence behaviour among Later Stone Age foragers of the forest and fynbos biomes of South Africa. J. Archaeol. Sci. 31, 999e1013. Stock, J.T., Shaw, C.N., 2007. Which measures of diaphyseal robusticity are robust? A comparison of external methods of quantifying the strength of long bone diaphyses to cross-sectional geometric properties. Am. J. Phys. Anthropol. 134, 412e423. Stock, J.T., 2002. Climatic and behavioural influences on postcranial robusticity among Holocene foragers. Ph.D. Dissertation, University of Toronto. Stock, J.T., 2006. Hunter-gatherer postcranial robusticity relative to patterns of mobility, climatic adaptation, and selection for tissue economy. Am. J. Phys. Anthropol. 131, 194e204. Stringer, C.B., 1992. Reconstructing recent human evolution. Phil. Trans. R. Soc. B. 337, 217e224. Swartz, S.M., 1990. Curvature of the forelimb bones of anthropoid primates - overall allometric patterns and specializations in suspensory species. Am. J. Phys. Anthropol. 83, 477e498. Thompson, J.L., Nelson, A.J., 2005. The postcranial skeleton of Le Moustier 1. In: Ullrich, H. (Ed.), The Neandertal Adolescent Le Moustier 1 New Aspects, New Results. Die Deutsche Bibliothek, Berlin, pp. 265e281. Tompkins, R.L., Trinkaus, E., 1987. La Ferrassie 6 and the development of Neanderthal pubic morphology. Am. J. Phys. Anthropol. 73, 233e239.

Trinkaus, E., Churchill, S.E., 1988. Neandertal radial tuberosity orientation. Am. J. Phys. Anthropol. 75, 15e21. Trinkaus, E., Ruff, C.B., 1999. Diaphyseal cross-sectional geometry of Near Eastern Middle Palaeolithic humans: the tibia. J. Archaeol. Sci. 26, 1289e1300. Trinkaus, E., Churchill, S.E., Ruff, C.B., 1994. Postcranial robusticity in Homo II: humeral bilateral asymmetry and bone plasticity. Am. J. Phys. Anthropol. 93, 1e34. Trinkaus, E., Ruff, C.B., Churchill, S.E., Vandermeersch, B., 1998. Locomotion and body proportions of the Saint-Césaire 1 Chatelperronian Neandertal. Proc. Natl. Acad. Sci. 95, 5836e5840. Trinkaus, E., 1981. Neanderthal limb proportions and cold adaptation. In: Stringer, C. (Ed.), Aspects of Human Evolution. Taylor and Francis, London, pp. 21e53. Trinkaus, E., 1983. The Shanidar Neandertals. Academic Press, London. Trinkaus, E., 2006. Modern human versus Neandertal evolutionary distinctiveness. Curr. Anthropol. 47, 597e620. Vandermeersch, B., Trinkaus, E., 1995. The post-cranial remains of the Régourdou 1 Neandertal: the shoulder and arm remains. J. Hum. Evol. 28, 439e476. Vl
cek, E., 1961. Die in Pavlov aufgefundenen Menschenreste aus de Jungpleistozen. Pamatky Archeol. II, 46e56. Walker, A., Leakey, R., 1993. The postcranial bones. In: Walker, A., Leakey, R. (Eds.), The Nariokotome Homo erectus Skeleton. Harvard University Press, Cambridge, pp. 95e160. Weaver, T.D., 2002. A multi-causal functional analysis of hominid hip-morphology. Ph.D. Dissertation, Stanford University. Weaver, T.D., 2003. The shape of the Neandertal femur is primarily the consequence of a hyperpolar body form. Proc. Natl. Acad. Sci. 100, 6926e6929. Weaver, T.D., 2009. The meaning of Neandertal skeletal morphology. Proc. Natl. Acad. Sci. 106, 16028e16033. Yasutomi, T., Nakatsuchi, Y., Kioke, H., Uchiyama, S., 2002. Mechanisms of limitation of pronation/supination of the forearm in geometric models of deformities of the forearm bones. Clin. Biomech. 17, 456e463.