Craniometric Variation in the Americas

Craniometric Variation in the Americas Ann H. Ross, Douglas H. Ubelaker, Anthony B. Falsetti Human Biology, Volume 74, Number 6, December 2002, pp. 807-818 (Article) Published by Wayne State University Press DOI: https://doi.org/10.1353/hub.2003.0010 For additional information about this article https://muse.jhu.edu/article/39176 Access provided at 21 May 2019 11:49 GMT from UCLA Library Craniometric Variation in the Americas ANN H. ROSS,1 DOUGLAS H. UBELAKER,2 AND ANTHONY B. FALSETTI1 Abstract Craniofacial variation is investigated in Latin America and the Caribbean. The samples included in this study are two historic and one prehistoric sample from Ecuador; prehistoric and modern Cuban samples; a prehistoric Peruvian sample; two prehistoric Mexican samples and one contemporary admixed Mexican sample; a 16th/17th-century Spanish sample; and Terry blacks. Biological distance is investigated using traditional craniometrics by computing size and shape variables according to Mosimann and colleagues. This study shows that there is much biological variation within the Americas. Two key objectives in physical anthropology are to document the vast range of human variation of past and present populations and to investigate the evolutionary and environmental forces responsible for phenotypic variation, which allows one to address large-scale issues such as migration and expansion. The patterns of human variation among geographic populations have been examined using genetic markers, linguistics, and anthropometrics. In particular, much attention has been directed at investigating genotypic and phenotypic variation among past and present New World populations. Historically, the Americas have been viewed as relatively homogenous, with significant biological variability not occurring until initial European contact and the subsequent influx of European and African populations, a view profoundly influenced by Morton and Hrdlička (Powell and Neves 1999). This view of minimal population variation in the Americas and therefore lack of great antiquity can be found at the core of Americanist studies until recently. Population origins traced to early demic expansion represent one preferred explanation for Native American homogeneity (Hrdlička 1920; Turner 1985, 1994). Linguistic evidence is another area of exploration used in the study of early American migration models. Greenberg (1987) and others (Greenberg and Ruhlen 1992; Greenberg et al. 1986) suggest that all living Amerindians are direct descendants of a single founding population (with the assumption that a common 1 C.A. Pound Human Identification Laboratory, University of Florida, Gainesville, FL 32611. Department of Anthropology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560. 2 Human Biology, December 2002, v. 74, no. 6, pp. 807–818. Copyright © 2003 Wayne State University Press, Detroit, Michigan 48201-1309 KEY WORDS: NEW WORLD, CRANIOMETRICS, MULTIVARIATE STATISTICS, HUMAN VARIATION 808 / ross et al. language indicates a common origin), which acts as a barrier to gene flow and influences the formation of population structures, operating much like geographic isolation (Barbujani and Sokal 1990; Sokal et al. 1993). Alternatively, multiple migration models have been presented by linguistic, genetic, dental, and craniometric studies of late Pleistocene and early Holocene Paleo-Indians (Haydenblit 1996; Lahr 1995; Powell and Neves 1999; Neves and Pucciarelli 1991; Steele and Powell 1993, 1994). Each of the models mentioned above requires that certain assumptions be made about the amount of phenotypic variation present in the Americas prior to European contact. However, a comprehensive and systematic examination of precontact and early-contact Latin American and Caribbean human skeletal remains for the study of microevolutionary processes is still lacking. Most researchers have focused on localized and regionalized studies, specifically South American and South Andean variation (Cocilovo and Rothhammer 1990, 1996, 1999; Rothhammer et al. 1982, 1984, 1985; Varela and Cocilovo 1999, 2000). Rather most researchers have used a single typological group such as W.W. Howells’ Peruvian sample as the morphological representative for all of South America (e.g., Steele and Powell 1994). The purpose of this paper is to present a preliminary study of the craniometric variation in Latin America and the Caribbean using a broad spatial distribution of American groups. If precontact populations in the Americas were relatively heterogeneous, this would be consistent with new findings suggesting high variability among early American fossil crania and would further support new evidence for a multiple population model for peopling of the New World (Jantz and Owsley 2001). Materials and Methods Materials. Ten groups totaling 388 individuals are used in this study. Group names, sample sizes, and provenience are listed in Table 1. The Terry black sample is included for comparison and to provide a comparable sample to represent the imported African population. Only undeformed crania were selected and males and females were pooled. Seven standard craniometric variables were utilized in the analyses and were selected according to measurement availability across samples (Table 2). Statistics. In order to examine craniofacial variation in the Americas while directly accounting for size effects, size and shape variables were computed according to Mosimann and colleagues (Mosimann and James 1979; Darroch and Mosimann 1985) using the raw measurements. Size is defined as the geometric mean (GM) of all seven cranial variables. The GM of n variables is calculated as 冪莦莦 n GMY = n 冲 Yi, i=1 (1) Craniometric Variation in the Americas / 809 Table 1. Sample Composition Group N Euro-Ecuador 18 Ubelaker (1994) 7 Ubelaker (1994) 11 Ubelaker (1981) 5 9 27 9 110 Present study Comas (1943) Comas (1943) Guardia (1942) Howells (1973) Admix-Ecuador Ayalán Tezontepec Tlatelolco Admix-Mexico Precontact Cuba Peru Reference Terry Blacks 87 Galera et al. (1998) Spain 83 Present study Modern Cuba 22 Present study Table 2. Provenience Historic, church Convento de San Francisco, Quito (a.d.1540–a.d.1940) Historic, church Convento de San Francisco, Quito (a.d.1535–a.d.1858) Coastal precontact Urn burials from Ecuador (a.d.730–a.d.1500) Prehistoric Mexico (Chicago Field Museum) Precontact Mexico Contemporary Precontact Ciboney Precontact (mountainous central part of the old province of Yauyos, 50 to 100 km southeast of Lima) Contemporary, housed at the Smithsonian Institution and provided by Richard Jantz Collection from Wamba, near Villanubla and Valladolid in northwestern Spain, 16th or 17th century, Departamento de Biología Animal, Universidad Complutense Madrid 19th-century cemetery collection from Museo de Montane, Havana, Cuba Measurements Measurement Maximum cranial length (g-op) Maximum cranial breadth (eu-eu) Basion-Bregma height (ba-br) Nasal height (na-ns) Nasal breadth (al-al) Orbit breadth (d-ec) Orbit height Description Martin and Saller (1957) Martin and Saller (1957) Martin and Saller (1957) Martin and Saller (1957) Martin and Saller (1957) Howells (1973):176 Martin and Saller (1957) and each variable is then divided by the GM to create shape variables, which are simple ratios of the geometric mean and are scale-free or dimensionless (Falsetti et al. 1993). Such ratios may or may not be correlated with size. While these new variables do not remove absolute size from the analysis, they provide a better insight into the “geometric similarity” among the populations. A one-way analysis of variance (ANOVA) was performed on the size variable to test the null hypothesis that the mean size is not significantly different among the groups. Next, canonical variates, linear combinations of predictor variables that summarize between-population variation, are derived from the newly 810 / ross et al. transformed shape variables to examine the interrelationships among the populations. In addition, a Pearson correlation analysis to measure the strength of the relation between the GM and the canonical axes was conducted. The degree of differentiation among the groups was measured using Mahalanobis D2 or generalized distance, which is a function of the group means and the pooled variances and covariances (Afifi and Clark 1996). D2 is used to test whether group centroids are significantly different. In addition, the Mahalanobis distances were also corrected for possible sampling bias due to small sample sizes using RMET 4.0, a computer program that performs population genetics analyses for metric data that was downloaded from the Web (Relethford n.d.). FST values were also derived from the craniometric data enabling comparisons to genetic data using an estimate of h2 = 0.55. Minimum FST (phenotypic) was first proposed by WilliamsBlangero and Blangero (1989) and Relethford and Blangero (1990) to distinguish it from the genetic FST. The results after correcting for sampling bias and scaling did not differ significantly from the original uncorrected and unscaled results, and thus the original generalized distances are utilized for the subsequent analyses. The UPGMA (unweighted pair group with arithmetic mean) clustering analysis was performed from the generalized distance matrix using the unweighted pair-group method with arithmetic averages (Sneath and Sokal 1973). Clustering methods are useful because they provide a unidimensional representation of the distance matrix and permit one to assess the relationships among the groups as those being more or less similar. These analyses were performed using the SAS system for windows Version 8 (2001). Results The ANOVA for the size variable yielded a significant difference among the groups (R2 = 0.279, PR > F 0.0001). The Pearson correlation revealed a weak relationship between CAN3 and the GM (r = –0.237; p < 0.0001), suggesting that some of variation observed on CAN3 using the transformed shape variables is influenced by size as well as shape. Neither the first, second, nor fourth axes are correlated to size, indicating that the difference observed is shape related and not influenced by size. The generalized distances are presented in Table 3. Table 4 presents the four significant canonical axes, which shows that roughly 46% of the among-group shape variation is accounted for on CAN1, 26% on CAN2, 13% on CAN3, and 9% on CAN4. The total canonical structure, the correlation between the original variables and the canonical variates, for CAN1, CAN2, CAN3, and CAN4 is given in Table 5. The total canonical structure indicates that the variation on the first canonical axis separates the groups with respect to cranial length, while the second axis isolates the groups on orbit breadth. The third canonical axis isolates the groups on cranial breadth and orbit height and the fourth axis is related to vault height. Figure 1 is a graphical representation of the class means on canonical variables illustrating the variation among the Table 3. Mahalanobis D2 Values Ayalán AdmixMex Precontact Cuba EuroEcuador Peru Terry Black Spain Tezon Tlatelolco Modern Cuba 0 5.907 2.900* 9.271 1.453 2.788* 3.324* 3.011* 6.400* 5.237* 3.245* 0 7.581* 9.224 3.595 6.593 13.442 8.854 11.382 3.213 10.095 0 7.640 2.284 1.256 2.841 2.308 3.125 3.490 1.861 0 4.713 11.233 13.698 9.503 12.431 10.009 6.570 0 2.746 4.801 3.536 7.169 4.301 2.404 0 2.069 1.956 4.382 3.412 3.978 0 1.794 7.556 9.536 2.967 0 7.051 6.249 2.593 0 3.104 8.188 0 8.232 0 Note: All groups significant at p < 0.0001 except those marked *, which are significant at p < 0.05. Figures in boldface indicate no significance. Craniometric Variation in the Americas / 811 Admix-Ecuador Ayalán Admix-Mex Precontact Cuba Euro-Ecuador Peru Terry Black Spain Tezon Tlatelolco Modern Cuba AdmixEcuador 812 / ross et al. Table 4. No. 1 2 3 4 Table 5. Significant Canonical Axes (Shape Variables) Eigenvalue Proportion Canonical Correlation Approximate F df PR > F 0.680 0.379 0.191 0.126 0.46 0.26 0.13 0.09 0.636 0.524 0.400 0.335 7.30 5.30 3.84 3.01 70 54 40 28 0.0001 0.0001 0.0001 0.0001 Total Canonical Structure Variable CAN1 CAN2 CAN3 CAN4 GOL XCB BBH NLH NLB OBB OBH –0.726 0.511 0.349 –0.019 –0.264 0.254 0.047 0.411 –0.111 0.278 0.050 –0.408 0.608 –0.305 0.409 0.661 0.198 –0.016 0.182 –0.564 –0.628 0.115 –0.251 0.765 –0.191 0.186 –0.387 –0.191 groups, revealing that all Central and South American populations have relatively short cranial vaults in varying degrees with Ayalán, precontact Cuba, and Tlatelolco having the shortest skulls. Terry blacks have the longest vaults followed by the Spanish series. Peru, Tezontepec, Tlatelolco, and Ayalán have narrow orbits, while the remainder of the groups exhibit relatively wide orbits with Cuba on the extreme. The exception is the Terry black sample and admixed Mexican sample, which fall in between. Figure 2 illustrates that Spain, admixed Mexicans, Ayalán, and Tlatelolco have the broadest vaults and low orbit heights, while the rest of the groups have relatively narrow vaults and high orbit heights in varying grades. Tezontepec exhibit the highest vaults followed Tlatelolco and precontact Cubans. The unbiased minimum FST, which is used to measure the degree of differentiation among subpopulations, is 0.242 (S.E. = 0.0162) assuming average heritability (0.55) and 0.138 (S.E. = 0.0157) assuming complete heritability for craniometric traits suggest that the groups are highly differentiated. Interestingly, these FST values are considerably larger than those obtained by Relethford (1994) comparing different geographic regions. The UPGMA clustering analysis shows that the admixed and European Ecuadorians and modern Cubans cluster together (Figure 3). Contemporary Mexicans are closest to precontact Peruvians, suggesting a strong indigenous component. Given the ethnohistory of Latin America, the proximity of the Terry black Class means on canonical variables (shape variables). Craniometric Variation in the Americas / 813 Figure 1. 814 / ross et al. Figure 2. Class means on canonical variables (shape variables). Craniometric Variation in the Americas / 815 and Spanish crania to the admixed Ecuadorian, modern Cuban, and contemporary Mexican samples is not surprising. Interesting, however, is the similarity of Ayalán to the precontact Mexican groups rather than to the South American groups. Precontact Cubans branch away notably from the rest of the American series, indicating that Cuban crania are very dissimilar from the rest of the Americans, reflecting a dissimilar ancestry. Figure 3. UPGMA phenogram of the generalized distance matrix. 816 / ross et al. Discussion To date, there has not been a systematic examination of craniometric variation among a broad spatial distribution of Latin American groups. Most researchers have focused on localized studies and have used a single group such as the Howells Peruvian sample as the morphological representative for all of South America. However, our results, although preliminary and based on some groups with small sample sizes, indicate that there may have been much diversity among Latin American populations and that the Americas were much more heterogeneous than previously thought. Interestingly, the morphological similarities between precontact Mexico and coastal Ecuador from Ayalán and the dissimilarity to the Howells Peruvian sample seem to contradict conclusions by Ruhlen (1994) and others that South America was populated by a single migration from North America. These results also provide further support for the argument that different populations peopled the New World (Schurr et al. 1999; Schurr and Wallace 1999). Since craniofacial morphological similarities to some degree reflect genetic relationships, we can further extrapolate that the morphological similarity between Mexico and Ecuador may have been a result of early demic expansion concurring with the archaeological evidence of contact between Mexico and coastal Ecuador (Ubelaker 1987). In addition, the morphological dissimilarity of the precontact Cuban sample to the other American populations probably suggests a different origin. The Caribbean has been settled by several waves of migration of preceramic- and ceramic-age settlers (Keegan 1995). Several migration routes have been proposed, which include preceramic colonists crossing the Yucatan passage or moving down from Florida into Cuba and dispersing eastward. However, the most recognized Antillean dispersal hypothesis is a direct jump by agriculturalists from Venezuela followed by dispersal into the Lesser Antilles westward (Keegan 1995; Moreira de Lima 1999). The FST results lend further support for the strong craniometric differentiation among Latin American and Caribbean populations. These FST are much greater than those obtained by Varela and Cocilovo (2002) for the Azapa valley and coast of Chile and those obtained by Rothhammer et al. (1990) for living Aymara groups in Chile and Bolivia. This probably coincides with the broad spatial distribution of our groups, which probably remained more isolated from each other in contrast to the comparisons of localized populations that probably experienced higher levels of external gene flow. Historically, the Americas have been viewed as relatively homogeneous, with biological variability not occurring until after initial European contact. However, these results suggest that there may have been much variation among New World populations before the arrival of the Europeans; they also emphasize the need for systematic and comprehensive sampling of the Americas using methods from both traditional and geometric morphometry before conclusions or interpretations can be made with respect to early Paleo-Indian variation and/or migration models. Craniometric Variation in the Americas / 817 Received 25 July 2001; revision received 17 July 2002. Literature Cited Afifi, A.A., and V. Clark. 1996. Computer-Aided Multivariate Analysis. 3rd ed. London, UK: Chapman & Hall. Barbujani, F., and R.R. Sokal. 1990. Zones of sharp genetic change in Europe are also linguistic boundaries. Proc. Natl. Acad. Sci. 87:1816–1819. Cocilovo, J.A., and F. Rothhammer. 1990. Paleopopulation biology of the Southern Andes—Craniofacial chronological and geographical differentiation. Homo 41:16–31. Cocilovo, J.A., and F. Rothhammer. 1996. Methodological approaches for the solution to ethnohistorical problems: Bioassay of kinship in prehistoric populations of Arica, Chile. Homo 47:177– 190. Cocilovo, J.A., and F. Rothhammer. 1999. Morphological microevolution and extinction of kinship in prehistoric human settlements from the Azapata Valley, Chile. Rev. Chilena His. Natural 72:213–218. Comas, J. 1943. El Metopismo: Sus Causas y Frecuencia en los Craneos Mexicanos. Tomo IV de los Anales del Instituto de Etnografia Americana. Universidad de Cuyo, Mendoza, Argentina, 121–159. Darroch, J.N., and J.E. Mosimann. 1985. Canonical and principal components of shape. Biometrika 72:241–252. Falsetti, A.B., W.L. Jungers, and T.M. Cole. 1993. Morphometrics of the Callitrichid Forelimb: A case study in size and shape. International J. Primatol. 14:551–572. Galera, V., D.H. Ubelaker, and L. Hayek. 1998. Comparison of macroscopic cranial methods of age estimation applied to skeletons from the Terry collection. J. Forensic Sci. 43:933–939. Greenberg, J.H. 1987. Language in the Americas. Stanford, CA: Stanford University Press. Greenberg, J.H., and M. Ruhlen. 1992. Linguistic origins of Native Americans. Sci. Am. 267:94–99. Greenberg, J.H., C.G. Turner II, and S.L. Zegura. 1986. The settlement of the Americas: A comparison of the linguistic, dental, and genetic evidence. Curr. Anthropol. 27:477–497. Guardia, F.R. 1942. Ensayo Sobre Cranea Cubana Precolombina. La Habana, Cuba. Haydenblit, R. 1996. Dental variation among four pre-Hispanic Mexican populations. Am. J. Phys. Anthropol. 100:225–246. Howells, W.W. 1973. Cranial Variation in Man. A Study by Multivariate Analysis of Patterns of Difference among Recent Human Populations. Cambridge, MA: Peabody Museum of Archaeology and Ethnology, Harvard University. Hrdlička, A. 1920. Shovel-shaped teeth. Am. J. Phys. Anthropol. 3:429–465. Jantz, R.L., and D.W. Owsley. 2001. Variation among early North American crania. Am. J. Phys. Anthropol. 12:327–338. Keegan, W.F. 1995. Modeling dispersal in the prehistoric West Indies. World Archaeol. 26:400–420. Lahr, M.M. 1995. Patterns of modern human diversification: Implications for Amerindian origins. Yrbk. Phys. Anthropol. 38:163–198. Martin, R., and K. Saller. 1957. Lehrbuch der Anthropologie. Stuttgart, Germany: Gustav Fisher. Moreira de Lima, L.J. 1999. La Sociedad Comunitaria de Cuba. Editorial Felix Varela, La Habana. Mosimann, J.E., and F.C. James. 1979. New statistical methods for allometry with application to Florida red-winged blackbirds. Evolution 33:444–459. Neves, W.A., and H.M. Pucciarelli. 1991. Morphological affinities of the first Americans: An exploratory analysis based on early South American human remains. J. Hum. Evol. 21:261–273. Powell, J.F., and W.A. Neves. 1999. Craniofacial morphology of the first Americans: Pattern and process in the peopling of the New World. Yrbk. Phys. Anthropol. 42:153–188. Relethford, J.H. n.d. RMET 4.0. Computer Software for Windows 95/98 http://konig.la.utk.edu/. Relethford, J.H. 1994. Craniometric variation among modern human populations. Am. J. Phys. Anthropol. 95:53–62. 818 / ross et al. Relethford, J.H., and J. Blangero. 1990. Detection of differential gene flow patterns of quantitative variation. Hum. Biol. 62:5–25. Rothhammer, F., R. Chakraborty, and R.E. Farrel. 1990. Intratribal genetic differentiation as assessed through electrophoresis. In The Aymara: Strategies in Human Adaptation to a Rigorous Environment, W.J. Schull and F. Rothhammer, eds., 193–201. Rothhammer, F., J.A. Cocilovo, S. Quevedo et al. 1982. Microevolution in Prehistoric Andean populations—Chronologic craniometric variation. Am. J. Phys. Anthropol. 58:391–396. Rothhammer, F., J.A. Cocilovo, S. Quevedo et al. 1985. The settlement of South America. Arch. Biol. Med. Exp. 18:R73. Rothhammer, F., S. Quevedo, J.A. Cocilovo et al. 1984. Microevolution in prehistoric Andean populations—Chronologic nonmetrical cranial variation in Northern Chile. Am. J. Phys. Anthropol. 65:157–162. Ruhlen, M. 1994. Linguistic evidence for the peopling of the Americas. In Method and Theory for Investigating the Peopling of the Americas, R. Bonnichen and D.G. Steele, eds. Peopling of the Americas Publications. Center for the Study of the First Americans, Department of Anthropology, Oregon State University, 177–188. SAS. 2001. System for Windows Version 8, Computer Program. Cary, NC. Schurr, T.G., R.I. Sukernik, Y.B. Starikovsdaya et al. 1999. Mitochondrial DNA variation in the Okhotsk Sea–Bearing region during the Neolithic. Am. J. Phys. Anthropol. 110:271–284. Schurr, T.G., and D.C. Wallace. 1999. MtDNA variation in Native American and Siberians and its implications for the peopling of the New World. In Who Were the First Americans? R. Bonnichsen, ed. Center for the Study of the First Americans, Department of Anthropology, Oregon State University, 41–77. Sneath, P.H.A., and R.R. Sokal. 1973. Numerical Taxonomy. San Francisco, CA: W.H. Freeman. Sokal, R.R, G.M. Jacquez, N.L. Oden et al. 1993. Genetic relationships of European populations reflect their ethnohistorical affinities. Am. J. Phys. Anthropol. 91:55–70. Steele, D.G., and J.F. Powell. 1993. Paleobiology of the first Americans. Evol. Anthropol. 2:138–146. Steele, D.G., and J.F. Powell. 1994. Paleobiological evidence of the peopling of the Americas: A morphometric view. In Method and Theory for Investigating the Peopling of the Americas, R. Bonnichen and D.G. Steele, eds. Peopling of the Americas Publications. Center for the Study of the First Americans, Department of Anthropology, Oregon State University, 141–164. Turner, C.G. 1985. The dental search for Native American origins. In Out of Asia. Peopling the Americas and the Pacific, R. Kirk and E. Szathmary, eds. The Journal of Pacific History, Inc., Canberra, Australia, 31–78. Turner, C.G. 1994. Relating Eurasian and Native American populations through dental morphology. In Method and Theory for Investigating the Peopling of the Americas, R. Bonnichen and D.G. Steele, eds. Peopling of the Americas Publications. Center for the Study of the First Americans, Department of Anthropology, Oregon State University, 131–140. Ubelaker, D.H. 1981. The Ayalan Cemetery. A Late Integration Period Burial Site on the South Coast of Ecuador. Washington, D.C.: Smithsonian Institution Press. Ubelaker, D.H. 1994. Biologia de los Restos Humanos Hallados en el Convento de San Francisco. Quito-Ecuador: Instituto Nacional de Patrimonio Cultural del Ecuador Agencia Espanola de Cooperacion Internacional Smithsonian Institution. Ubelaker, D.H. 1987. Dental alteration in Prehistoric Ecuador: A new example from Jama-coaque. J. Washington Acad. Sciences 77:76–80. Varela, H.H., and J.A. Cocilovo. 1999. Evaluation of the environmental component of the phenotypic variance in prehistoric populations. Homo 50:46–53. Varela, H.H., and J.A. Cocilovo. 2000. Structure of the prehistoric population of San Pedro de Atacama. Curr. Anthropol. 41:125–132. Varela, H.H., and J.A. Cocilovo. 2002. Genetic drift and gene flow in a prehistoric population of the Azapa valley and coast, Chile. Am. J. Phys. Anthropol. 118:259–267. Williams-Blangero, S., and J. Blangero. 1989. Anthropometric variation and the genetic structure of the Jirels of Nepal. Hum. Biol. 61:1–12.