SEARCH

SEARCH BY CITATION

Keywords:

  • biological distance;
  • dental morphology;
  • Mexico;
  • Postclassic period;
  • Southwest US

Abstract

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

The Southwest United States (US) and Mesoamerica are often thought of as disparate regional networks separated by Northern Mexico. Chaco Canyon in the Southwest US, Tlatelolco in Central Mexico and Casas Grandes in Northern Mexico, all had large inter-regional trade centres that economically connected these networks. This study investigated how factors such as geographic distance, shared migration history, trade and political interaction affected biological relationships and population affinities among sites in Mexico and in Southwest US during the Postclassic period (ad 900 ~ 1520). Distances based on cultural and geographic variables derived from archaeological and ethnohistoric data were compared with phenetic distances obtained from dental morphological traits. The results of Mantel tests show trade (corr = 0.441, p = 0.005), shared migration history (corr = 0.496, p = 0.004) and geographic distance (corr = 0.304, p = 0.02) are significantly correlated with phenetic distances, whereas political interaction (corr = 0.157, p = 0.133) is not. Partial Mantel tests show trade (corr = 0.223, p = 0.049) and shared migration history (corr = 0.493, p = 0.003) remain significant when controlling for similarities with geographic distance, although the correlation for trade and phenetic distance is lowered. Geographic distance is not significant when similarities with trade (corr = 0.067, p = 0.681) and shared migration history (corr = 0.148, p = 0.127) are controlled. These results highlight the importance of economic relationships and shared migration history across geographic regions in interpreting biological relationships among contemporaneous populations in prehistoric Mexico and the Southwest US. Copyright © 2014 John Wiley & Sons, Ltd.

Patterns of human biological variation in pre-contact Mexico and the Southwest United States (US) were affected by migration and genetic exchange associated with widespread trade networks, endemic warfare, imperial expansion and rapid population growth. During the Postclassic period (ad 900–1520), these processes intensified with political and economic networks connecting virtually every group in Mexico (Smith, 2001; Smith and Berdan, 2003). In the Southwest US, large-scale migrations occurred throughout the region, likely in response to a series of droughts occurring around ad 1130–1180 (Schlanger and Wilshusen, 1993; Douglass, 2006). Large, inter-regional trade centres connected groups throughout the Southwest US and Mexico. Although relationships among groups in these regions are well documented in the archaeological and ethnohistoric records, biological relationships among these groups are understudied. Here, we evaluate how population structure is shaped by the processes examined by cultural ecology, the study of how humans adapt to their cultural and physical environments (Steward, 1990). We use phenetic distances obtained from dental morphological observations as a proxy for genetic relationships for comparison with geographic distance, shared migration history, trade and political interaction among groups from the Southwest US and North, West and Central Mexico. Previous studies of some of the samples used in this analysis relied upon craniometric and nonmetric data for phenetic comparisons (Corruccini, 1972; Mackey, 1977; El-Najjar, 1978; Schillaci and Stojanowski, 2002a, 2002b; Gonzales-Jose et al., 2007). This project expands on these studies to include Northern Mexico and considers groups within the Southwest US and Mexico as a single inter-regional network to investigate how cultural and ecological factors affected phenetic distances among Postclassic period Mexican and Southwest US populations.

The archaeological and ethnohistoric records

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

Extensive archaeological and ethnohistorical research offers evidence of large-scale migrations, shared ideology, trade and warfare among the various regions of Mexico and the Southwest US. Periods of drought and economic collapse led to several migrations, particularly, throughout the Postclassic period. Cultural similarities such as ballcourts, burial patterns, irrigation techniques and architecture demonstrate possible links between these geographically separate networks (Ross, 1967; McGuire, 1980, 2012; Whalen and Minnis, 1996). Trade items such as cacao, macaws and turquoise also suggest connections between Mesoamerica and the Southwest US through inter-regional trade networks (Ericson and Baugh, 1993; Weigand, 2008; Crown and Hurst, 2009; Melgar Tisoc and Solis Ciriaco, 2009; Melgar Tisoc, 2010). Fortified settlements located near resource production and mining sites in the Southwest US and throughout Mexico suggest a relationship between continuous warfare and trade. Although many of these groups were connected through cultural and ecological relationships, these geographic regions are typically grouped into two separate core cultural regions: the Southwest US, including parts of Northern Mexico; and Mesoamerica, including West and Central Mexico. These geographic regions are shown in Figure 1.

image

Figure 1. Map of sites represented.

Download figure to PowerPoint

Southwest US and Northern Mexico

The Postclassic period of Mexico overlaps the chronological periods of Pueblo II to Pueblo IV in the Southwest US (ad 900–1600). Prior to the Postclassic period, ancestral Puebloan, Hohokam and Mogollon populations primarily occupied the Southwest US. From around ad 850–1130, Chaco Canyon was one of the largest Pueblo sites in the region. Chaco was likely an economic centre for distributing agricultural and luxury goods for the entire San Juan Basin region of Northwest Mexico (Washburn et al., 2011; Crown, 2013). Cacao from the Gulf Coast or Southern Mexico was among the items used and likely traded at Chaco Canyon (Crown and Hurst, 2009). Between ad 1130 and 1180, coinciding with a period of disastrous drought, Chaco Canyon was abandoned, and migrants dispersed south into the Rio Grande Valley and Northern Mexico (Benson et al., 2007). These migrations likely led to population growth and the development of a large inter-regional trade centre at Casas Grandes in Northern Mexico (Kelley, 1993; Lekson, 1999). In the Southwest US, the frequency of hostile interactions increased with time (LeBlanc, 1999; VanPool and O'Brien, 2013). Economic and political relationships shifted often in response to access to resources, population growth and natural phenomena. During the Postclassic period, sites such as Hawiku in the Cibola area and Puye on the Pajarito Plateau were primarily local distribution centres. Some exotic trade items found in these areas were obtained from as far as Northern Mexico (Mills, 1995; Walsh, 2000), and Hawiku likely was involved in the trade of items from West Mexico (Riley, 1975). Hawiku, also participated in inter-group warfare, especially with Hopi and Acoma groups around the region (Hammond and Rey, 1940; LeBlanc, 1999); other hostile interactions within the Rio Grande Valley are not as well understood. Warfare on the Pajarito Plateau area declined as the pueblos grew after about ad 1400 (LeBlanc, 1999, 2000), and it is unlikely that these groups had hostile interactions with the Zuni or Hopi groups to the west. Most of the Pajarito Plateau was abandoned around the time of European contact, and occupation of the Zuni site of Hawiku continued until shortly after the Pueblo Revolt in ad 1680.

Much of Northern Mexico is often referred to as the ‘Greater Southwest’. Among the populations that occupied this region were the semi-nomadic Chichimecs, who appear in the archaeological record during the Late Postclassic period (around ad 1200). The term Chichimec does not only imply a particular ethnic, linguistic or technological identity but also refers to a large number of nomadic groups that inhabited Northern Mexico, the Gulf Coast and parts of Central Mexico (Lopéz Austin and Lopéz Luján, 2008). The Coahuilan groups in the northeast occupied caves and open areas that had access to water and desert plains (Turpin, 1997). To the northwest, the Casas Grandes culture, centred at Paquimé, controlled a large trade centre between the Southwest US and Mexico (DiPeso, 1974; Woosley and Olinger, 1993). Casas Grandes grew from a small agricultural community to a large trade centre shortly after the abandonment of Chaco Canyon, leading some archaeologists to attribute this demographic growth to ancestral Puebloan migrants (Lekson, 1999). Others argue that this growth was caused by migrants from and trade with Central Mexico (DiPeso, 1974; Nelson, 1986; Kelley, 1993). The Casas Grandes culture and its remnants throughout the region actively participated in trade and continuous inter-group warfare. Areas such as Copper Canyon in the Sierra Madre Occidental Mountains were inhabited by nomadic groups that lived in temporary cave dwellings. Many burials found in caves in this area show signs of trauma to the skeleton (Walker, 2006), and historical accounts from Spanish invaders in the 17th century describe hostile relationships between indigenous groups throughout the region. These groups were located near major trade routes along the northwestern mountains and coast of Mexico (Weigand, 2008). In general, groups from Northern Mexico were highly involved in economic and hostile interactions between the Southwest US and Mexico.

Central and West Mexico

Central Mexico is often considered the centre for widespread trade and political expansion during the Postclassic period (Smith, 2001; Lopez Austin and Lopez Lujan, 2005; Melgar Tisoc and Solis Ciriaco, 2009). During the Early Postclassic period, the dominant Toltec populations inhabited the Valley of Mexico and established political dominance and long-distance exchange routes (Cowgill, 2000). After the decline of the Toltecs in the mid-12th century ad, these populations continued to settle in Central Mexico, as well as in the Gulf Coast, and throughout the Maya region (Smith and Schreiber, 2006). Some recent biological distance studies using dental morphological traits have provided evidence of these migrations (Aubry, 2009; Willermet et al., 2013). During the Late Postclassic period, the Aztecs grew in power and achieved dominance over much of Mexico (Hassig, 1995; Lopez Austin and Lopez Lujan, 2005). The Aztecs also maintained large inter-regional markets and trade routes throughout Mexico and beyond, with their goods being found as far north as the Southwest US and as far south as Guatemala and Honduras (Smith and Berdan, 2003). Large markets such as Tlatelolco served as many as 20 000–25 000 per day (Anonymous Conqueror, 1971; Berdan, 2005), attracting people from various regions around Mexico. Until Spanish contact, other cities throughout the Aztec realm participated in various forms of market exchange, war campaigns and paid tribute to the Aztecs.

Throughout the Postclassic period, West Mexico was an important region for trade and for the mining of minerals such as copper. In the Early Postclassic period, several small city-states were widely distributed in defensible positions (Foster and Gorenstein, 2000). During the Late Postclassic period, much of the region was consolidated and controlled by the Tarascan Empire. Like the Aztecs, the Tarascans maintained a relatively large empire, extracting tribute of minerals and luxury items from cities within their range of control (Pollard and Vogel, 1994; Beekman, 2010). Unlike the Aztecs, the Tarascan elite maintained firm political and economic control within their domain. Additionally, the Tarascan elite controlled specific raw materials and land, remaining unchallenged in their control until the arrival of the Spanish in ad 1525 (Pollard, 2003). Many groups from the northwest part of the region participated in inter-regional exchange, independent of the Tarascan Empire. These groups settled in defensive city-states primarily settled along riverine valleys, with a large number of imported items found in houses and burial pits (Gorenstein, 2000). The Tarascans were unable to extend their political control into the north, although they likely maintained trade relationships until European contact in the 16th century.

Previous biological distance studies

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

In Mexico, biological distance studies have focused on migration patterns among the Classic and Postclassic period populations from Northern, West and Central Mexico. Kemp et al. (2005) used mtDNA from samples from Central Mexico to find that samples such as Tlatelolco were more similar to other samples from Central and Southern Mexico and were different from other Uto-Aztecan speaking groups in Northern Mexico and in Southwest US. Using craniofacial measurements, Gonzales-Jose et al. (2007) found Central Mexican samples to be more similar to West Mexican samples and different than earlier Central or Northern Mexican samples. Finally, Gómez-Valdéz (2008) used dental morphological traits to investigate relationships among West Mexico populations, mostly prior to the Postclassic period. Gómez-Valdéz found that Tlatelolco had small biological distances with West Mexico samples, as well as with other Central Mexico samples from the Classic period (ad 300–900). These distances were especially low between Tlatelolco and the West Mexico samples from the Aztatlan region. These results support a shared migration history between West and Central Mexico.

Craniometric and cranial morphological traits from some of the samples studied here have also been used in biological distance analyses to investigate the origins of sites in the Southwest US and Central Mexico. Mackey (1977) used cranial morphological traits in samples from around the Rio Grande Valley to find that the Pajarito Plateau (Puye and Sapawe), Jemez River Valley and Hawiku were very similar to each other but different from a sample from Pueblo Bonito. Studies using craniometric data from these samples found similar results (El-Najjar, 1978; Schillaci and Stojanowski, 2002a, 2002b). However, Corruccini (1972) found contradictory results when he compared Hawiku and Puye with a different sample from Pueblo Bonito, suggesting that Puye and Pueblo Bonito were similar when compared with Hawiku. The difference in these findings may be explained by Schillaci and Stojanowski (2002a, 2002b), who found a distinction between the west and north samples of Pueblo Bonito, with the west being more similar to Puye and Hawiku. These results suggest high within group variation at Pueblo Bonito that may be a result of migration and some population replacement, and support a shared migration history between Chaco Canyon and later Rio Grande Valley sites.

Other previous biological distance studies using dental morphological traits from these regions have focused on testing hypotheses about migration, language group distribution and shared culture (Turner, 1993; LeBlanc et al., 2008). Turner's work (1993) indicated that phenetic distances were related to geographic proximity and cultural group among prehistoric and modern Southwest US and Mexican populations. However, phenetic distances were not related to linguistic classification. LeBlanc et al. (2008) also compared Southwest US and Mexican populations with test cultural and linguistic correlations with phenetic distances using dental morphological traits. The results of this study indicated small phenetic distances among the Southwest US samples of Hawiku, Puye and Pueblo Bonito, as well as between these samples and a combined Central Mexico and Coahuilan (northeast Mexico) sample. The authors combined their Mexican samples to facilitate evaluating the spread of the migrations of Uto-Aztecan speakers. Our study expands this work by looking at geographic, economic and political relationships among samples to examine cultural relationships among site-specific samples across regions.

Materials

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

We observed dental morphological traits in 311 individuals representing 11 archaeological sites from the Southwest US and Mexico. As is true in most bioarchaeological dental studies, observability was limited by missing teeth, heavy dental attrition and caries. Because of these reasons, many individuals used in this study are young or middle adults. Figure 1 shows the geographic location of all the sites from which the samples are derived. Samples were chosen to represent various forms of economic and political structure. Three samples are derived from sites in the Southwest US (n = 80), Pueblo Bonito (ad 828–1130), Puye (ad 1150–1580) and Hawiku (ad 1400–1680). Pueblo Bonito is the largest Great House of Chaco Canyon, in northwestern New Mexico. It predates all the other samples (ad 828–1130) but was included in this study because of its importance in Northern Mexico and Southwest US migration histories. There are two major skeletal samples for Pueblo Bonito: the west and north cemeteries. Here, we use the sample from the north cemetery housed at the Smithsonian National Museum of Natural History. The Puye sample comes from the Pajarito Plateau near the Jemez Mountains in central New Mexico. Finally, Hawiku was a large Zuni site on the New Mexico and Arizona border.

Eight samples are derived from Mexico: three from Northern Mexico (n = 60), two from West Mexico (n = 57) and three from Central Mexico (n = 114). The Northern Mexico sites of Copper Canyon (ad 1300–1700) and Nararachic (ad 1200–1700) represent possible Casas Grandes remnant populations from the Sierra Madre Occidental Mountains in the Mexican state of Chihuahua. Copper Canyon is an archaeological and historic site associated with the historic Tarahumara mines. Nararachic is a small cave burial site in Chihuahua, near a small semi-nomadic site possibly under the control of Paquimé, the centre for the Casas Grandes culture. The third Northern Mexico site is a burial cave in Cuatro Ciénegas (ad 1200–1500), located in the Coahuilan Valley of Northeast Mexico.

The two West Mexico samples are from Zacapú (ad 1200–1540) and Guasave (ad 900–1400). Zacapú, located within the Tarascan Empire, was the first Tarascan settlement in the present day Mexican state of Michoacan. The Guasave sample comes from the archaeological site in the Mexican state of Sinaloa, from the Aztlan region on the northern frontier of West Mexico, best known for participation in long-distance coastal trade between Northern and West Mexico.

The three Central Mexican sites included are Tlatelolco (ad 1335–1520), Tenayuca (ad 1200–1520) and Texcaltitlán (ad 1100–1520). All come from the Valley of Mexico in Central Mexico, once the site of Lake Texcoco and the centre of the Aztec Empire. Tlatelolco represents a sample of the population from the largest inter-regional market in Mexico during the Postclassic period and is also the ‘sister-city’ of the Mexica capital Tenochtitlan. Here, we use Tlatelolco to represent a combined Tlatelolco-Tenochtitlan Aztec sample, in accordance with the history provided by the Codex Cozcatzin. Texcaltitlán and Tenayuca are samples from cities occupied and controlled by the Aztecs. Tenayuca was an early Chichimec settled site that was conquered early by the Aztecs and made a tributary province. Texcaltitlán was a major city for the Matlatzinca culture and was conquered and controlled by the Aztecs as a strategic province on the western frontier of the empire (Berdan et al., 1996). Further information about these sites is provided in Table 1.

Table 1. Samples used in this study
SiteOccupationCultural groupMarketPolitical affiliationN
Hawikuad 1400 –1680ZuniRegionalIndependent20
Puyead 1150 –1580TewaLocalIndependent30
Pueblo Bonitoad 828 –1130AnasaziRegionalIndependent30
Cuatro Ciénegasad 1200 –1500ChichimecLocalIndependent27
Nararachicad 1200 –1700Raramuri?LocalIndependent18
Copper Canyonad 1300 –1700RaramuriLocalIndependent15
Texcaltitlánad 1100 –1520MatlatzincaLocalTributary (Aztec)27
Tlatelolcoad 1335 –1520Aztec-MexicaInter-regionalTributary (Aztec)70
Tenayucaad 1200 –1520ChichimecLocalTributary (Aztec)17
Guasavead 900 –1400AztatlanInter-regionalIndependent17
Zacapúad 1200 –1540TarascanRegionalTributary (Tarascan)40

Methods

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

We used phenetic distances calculated from dental morphological observations as a proxy for genetic relationships among samples. Phenetic distances are biological distances based on morphological (phenotypic) observations. We created model matrices on the basis of expected relationships from geographic distance, shared migration history, trade and political interaction using archaeological and ethnohistoric data (detailed model descriptions in the following text). It is important to note that these relationships are generalisations about intergroup dynamics during the time period of interest. These generalisations are necessary as political and economic relationships sometimes shifted throughout the temporal periods. We compared these model matrices with the matrix of phenetic distances to test for correlations among phenetic, ecological and cultural relationships.

Edgar (HE) collected the samples from Tlatelolco and Guasave, and Ragsdale (CR) collected all other samples. To estimate inter-observer and intra-observer error, we scored a sample of 50 dentitions of prehistoric Native American dentitions housed at the Maxwell Museum of Anthropology Laboratory of Human Osteology, once by HE and twice by CR. Traits with no variation (100% or 0% present) were removed prior to calculation, as these traits were not useful for error testing. We calculated Cohen's kappa, standard error and percent agreement for intra-observer and inter-observer error tests. Kappa values for inter-observer and intra-observer error range from 0.7 to 1.0, and percent agreement range from 86% to 100%, demonstrating high concordance between observations and observers. Intra-observer error for HE is available in the study of Edgar (2002).

Phenetic distances

Dental morphological traits are small variants of the teeth, mostly on the occlusal (chewing) surface, used to track changes in population phenotypes over space and time, as the gene frequencies change because of factors such as genetic drift and gene flow (Turner et al., 1991). These traits are scored on a graded scale in accordance with the Arizona State University Dental Anthropology System (Turner et al., 1991). Phenetic distances using dental morphological variation are useful as a proxy for genetic relationships among archaeological populations, because they are strongly under genetic control (Mizoguchi, 1977; Nichol, 1989, 1990; Townsend et al., 2009). Previous studies using modern Pima (Nichol, 1989) and Euro-Australian (Townsend et al., 2009; Bockmann et al., 2010) sibling–parent relationships show moderate to high levels of genetic inheritance of some of the dental morphological traits used here (heritability has not yet been estimated for the remaining traits). Using 600 individuals from 83 modern Pima families, Nichol (1989) found an average transmissibility estimate of 0.5. However, Nichol (1989) predates the standardisation of scoring procedures for dental traits used here (Turner et al., 1991). More recent studies by Townsend et al. (2009) and Bockmann et al. (2010) examined more of the standardised traits used here, finding heritability estimates averaging 0.7. These studies support a high level of inheritance for the traits used in this study.

We scored traits on any observable adult teeth and scored both left and right sides; the higher score represent the maximum expression of the trait in each individual to account for asymmetry. Raw data are dichotomized for use in biological distance statistics. Presence/absence breakpoints for traits were first drawn from Scott and Turner (1997). We then reviewed raw observations among all samples to explore possible breakpoint adjustments. We found that upper and lower incisor shovelling and upper incisor double shovelling have high frequencies in every sample when the standard breakpoints are used. We adjusted the breakpoints for these traits to better detect variation between groups. These adjustments are highlighted in Table 2.

Table 2. Dental morphological traits used to calculate the pseudo-Mahalanobis D2 phenetic distance matrix
TraitHUPYPUCUNVCCTXTOTNZUGU
Tooth (−, +)(N)%(N)%(N)%(N)%(N)%(N)%(N)%(N)%(N)%(N)%(N)%
  1. Traits are provided with standardised breakpoints, samples size (N) and percent present (%) for each sample. Non-standard breakpoints are italics.

  2. Italicised breakpoints were adjusted according to variation within the data set.

Shovelling
UI1 (0–3, 4–7)(19)70(20)80(20)80(6)90(4)90(5)90(8)70(48)70(15)75(10)80(9)75
Double shovelling
UI2 (0, 1–6)(20)25(17)18(20)40(9)55(5)50(6)83(9)55(44)93(15)33(6)60(11)43
Interruption groove
UI2 (0, 1–4)(21)24(18)44(20)30(8)10(6)10(6)50(9)37(44)41(16)33(15)14(8)20
Tuberculum dentale
UI1 (0–1, 2–6)(20)65(19)63(20)50(8)43(5)90(5)67(8)75(44)36(15)86(10)50(8)83
Tuberculum dentale
UI2 (0–1, 2–6)(19)47(18)28(19)53(8)62(5)90(6)33(9)25(40)42(15)43(14)15(12)86
Tuberculum dentale
UC (0–1, 2–6)(18)72(16)85(21)90(9)57(7)86(8)90(18)89(41)58(15)36(18)70(12)81
Metacone
UM1 (0–4, 5–6)(19)74(27)57(24)62(17)29(16)69(11)40(29)59(54)66(16)66(35)85(11)36
Metacone
UM2 (0–4, 5–6)(19)32(28)92(24)17(17)06(15)25(11)10(28)18(53)17(17)19(35)40(12)01
Hypocone
UM1 (0–4, 5–6)(20)78(28)64(24)83(17)11(16)69(10)89(28)57(54)66(16)80(36)92(9)01
Hypocone
UM2 (0–1, 2–6)(20)60(26)20(24)71(17)69(15)83(10)89(27)73(52)81(17)37(36)83(8)90
Cusp 5
UM1 (0, 1–5)(20)05(28)25(24)13(17)31(11)33(9)12(28)07(49)49(16)06(36)25(9)71
Carabelli's cusp
UM1 (0, 1–7)(20)01(28)04(23)33(19)64(16)11(9)71(23)64(48)56(11)40(36)57(9)80
Parastyle
UM1 (0, 1–6)(18)02(25)64(23)17(16)10(9)12(9)12(28)07(54)13(15)14(36)27(9)01
Shovelling
LI2 (0–2, 3–7)(16)26(14)67(15)71(11)49(5)90(6)80(7)66(45)98(16)50(19)83(8)80
Anterior fovea
LM1 (0–1, 2–4)(15)33(22)14(22)40(11)08(8)20(7)40(18)38(49)49(12)18(26)65(8)40
Protostylid
LM1 (0, 1–7)(15)33(22)23(22)50(13)75(8)43(7)83(21)75(54)62(16)46(27)77(10)71
Protostylid
LM2 (0, 1–7)(18)29(23)61(22)54(16)62(8)60(8)40(18)61(54)49(15)28(24)69(9)77
Cusp 5
LM2 (0, 1–5)(18)22(25)23(24)29(18)44(8)33(8)57(21)40(52)86(15)28(24)33(9)50
Cusp 6
LM1 (0, 1–5)(18)06(20)20(25)04(18)31(7)01(7)33(17)11(45)42(15)01(25)12(9)12

In many dental morphological studies, an observation of absence for the protostylid actually includes the lowest expression of the trait, 1 = pit, or foramen caecum molaris. Our review of the raw data for this study indicated nearly all of the variation exists among these low expressions. Only two individuals from Guasave, one from Zacapú and four from Tlatelolco had protostylids scorable above grade 2. For this reason, we chose to keep the standard presence/absence breakpoint (0/1) for this analysis.

We used pseudo-Mahalanobis' D2 to determine the matrix of phenetic distances among groups (Konigsburg, 1990; Irish, 2010). Pseudo-Mahalanobis D2 extends the squared Mahalanobis' distance for use with dichotomized data by using z-scores and a tetrachoric correlation matrix in the calculation (Konigsburg, 1990). The tetrachoric correlation matrix is a matrix of correlation coefficients computed for two normally distributed dichotomized variables, and it serves to account for inter-trait correlations. We chose this statistic, as opposed to Smith's mean measure of divergence, because it allows for the use of correlated traits. Recent studies comparing pseudo-Mahalanobis' D2 and mean measure of divergence have shown that the two statistics are highly correlated, so both are useful in biological distance studies (Edgar, 2004; Irish, 2010). Tetrachoric correlation matrices and pseudo-Mahalanobis' D2 values were calculated using sas statistical software (SAS Institute, 2007).

We used Ward's method of hierarchical clustering to illustrate phenetic distances between samples using the pseudo-Mahalanobis' D2 matrix and created a three-dimensional principal component (PC) scatterplot to show the relationships between sites on three axes. Ward's cluster method produces tree diagrams by forming clusters that minimise intra-cluster variation while maximising inter-cluster variation (Ward, 1963). The PC analysis provides eigenvalues and the percent of the variance accounted for by each axis provided using the phenetic distance matrix and provides the percentage of variance among samples accounted for on each axis. These analyses were conducted using past (Øyvind Hammer, University of Oslo, Norway) (Hammer et al., 2001).

Geographic and cultural relationships

We used Mantel and partial Mantel tests (1967) to investigate correlations between geographic, cultural and phenetic distances. Mantel tests compare matrices in thousands of permutations, eliminating assumptions about the distribution of the data being used (Manly, 2005). To test for correlations among cultural variables, we calculated a correlation matrix using Pearson's correlation coefficient. To account for correlations among the variables, we conducted partial Mantel tests, which allow two matrices to be compared while controlling for similarities with a third matrix (Legendre and Legendre, 1998). We created a two-dimensional PC plot of the Mantel correlations and correlations between geographic and cultural variables. These analyses were conducted using past (Hammer et al., 2001). The cultural variables tested were geographic distance, shared migration history, economic exchange and political interaction.

Geographic distance

To determine the extent to which phenetic distances were correlated with geographic distances, this study tested isolation by distance model (Wright, 1943). According to this model, as geographic distances increase, biological distances should also increase. Geographic distances were calculated between all sites, following the shortest possible route given geographic barriers such as mountains or large bodies of water, using global positioning system coordinates provided by Google Earth (Figure 1). Archaeological and historical data provide useful information on routes used for trade and military campaigns to give an estimate of a most likely travelled route. These distances were used to form a model matrix, which, like the other three model matrices (described in the following text) was compared with the matrix of phenetic distances.

Shared migration history

We created a model matrix of expected group relationships using archaeological, ethnohistoric and linguistic data for the Postclassic period. Migration histories among populations used here were drawn from archaeological data (Matos Moctezuma, 1989; Townsend, 1992; Lekson and Cameron, 1995; Benson et al., 2007; Beekman, 2010) and from ethnohistoric migration documents such as the Boturini Codex, Codex Xolotl (Douglas, 2010), Codex Cozcatzin and Codex Chimalpopoca (Bierhorst, 1992). Distances used for this matrix were based on branches representing time since a migratory split between populations. We scored a ‘1’ for diverging branches from parent populations to account for migration, as well as for 300-year temporal intervals to account for drift. These scores were added up to give a shared migration distance between any two samples in the study. This model is represented in Figure 2(A).

image

Figure 2. (A) Model for shared migration history: Pueblo Bonito (PUE), Puye (PUY), Hawiku (HAW), Copper Canyon (COP), Nararachic (NAR), Cuatro Cienegas (CUA), Guasave (GUA), Zacapu (ZAC), Texcaltitlan (TEX), Tenayuca (TEN) and Tlatelolco (TCO). (B) Model for economic exchange. (C) Model for political interaction. Distances represent an average interaction between sites.

Download figure to PowerPoint

Economic exchange

During the Postclassic period, trade existed on a variety of scales among different regions throughout Mexico and the Southwest US. Trade relationships were drawn from the archaeological record (Riley, 1975; Smith and Berdan, 2003; Weigand, 2008; Crown and Hurst, 2009; Beekman, 2010; Washburn et al., 2011) and ethnohistoric sources such as the Codex Mendoza (Berdan and Anawalt, 1997) and the Florentine Codex (Sahagun and Anderson, 1982). We coded each sample as a local market, regional market or inter-regional market. Local markets were periodic events where local food and utility items were exchanged. They acted as venues for intra-population trade. Regional markets were more frequent and often perennial events that incorporated luxury items as well as utility items and were venues for intra-regional interaction among populations. Lastly, inter-regional markets were year-round, large markets that exchanged a wide variety of items from various regions around Mexico and the Southwest US. Trade distances were measured in branch lengths between samples using various market exchange forms. We scored a ‘1’ for interactions between a local market and their respective regional/inter-regional market, a ‘2’ for interactions between local markets within the same regional trade networks and for interactions between inter-regional/regional markets, a ‘3’ for interactions between local markets and regional/inter-regional markets from another regional trade network and a ‘4’ for interactions between local markets from different regional trade networks. We also combined scores for samples not in direct contact through trade but for which goods were likely exchanged through another site. In the statistical analysis, a ‘99’ represented cases where there was no economic interaction between groups. Pueblo Bonito was not included in this analysis as the sample predates all other samples. A representation of this model is provided in Figure 2(B).

Political interaction

During the Postclassic period, political expansion and endemic warfare existed throughout Mexico and the Southwest US. Political alliances, military conquest and the control of resources led to powerful empires such as the Aztec and Tarascan Empires. We coded political interaction on the basis of political structure and relationships among populations (Berdan et al., 1996; LeBlanc, 1999; Lopez Austin and Lopez Lujan, 2005; VanPool and O'Brien, 2013). We recorded each sample as a city-state or a capital city-state. City-states refer to all populations in the study that are not a political capital but are independent cities within a larger polity. These city-states may also be important trade or religious centres. City-state capitals refer to capital cities within a city-state cultural region or polity; generally, these capitals control other city-states and towns with the same political affiliation. Political distances were measured for different political relationships for each city-state in the study. We scored a ‘1’ for political allies, a ‘2’ for strict dominant–subordinate relationships between samples such as tributary provinces in the Aztec Empire, a ‘3’ for loose dominant–subordinate relationships between samples such as client-state or frontier provinces around the Aztec Empire and a ‘4’ for hostile or enemy relationships between samples. For samples with no political interactions, or for political interactions that have not been documented, a ‘99’ was used in the statistical analysis. Pueblo Bonito was not included in this analysis as the sample predates all other samples. Figure 2(C) shows this model.

Results

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

Pseudo-Mahalanobis' D2 distances among samples were calculated using data from 21 dental morphological traits. These traits are provided in Table 2. Pseudo-Mahalanobis' D2, cultural and geographic model distances are provided in Table 3. A Ward's cluster dendrogram (Figure 3) and three-dimensional PC scatterplot (Figure 4) show the phenetic distances graphically. Consistent with other biological distance studies of the Southwest US samples (Mackey, 1977; El-Najjar, 1978; Turner, 1993; Schillaci and Stojanowski, 2002a, 2002b; LeBlanc et al., 2008), Hawiku and Puye are relatively close to each other and distant from Pueblo Bonito. Interestingly, Pueblo Bonito is closer to the Northern Mexico samples from Copper Canyon and Nararachic than others in the Southwest US. Two of the Central Mexico samples, Tlatelolco and Texcaltitlán, and the two samples from West Mexico, are closer to each other, compared with samples from Northern Mexico and the Southwest US. One of the Central Mexico sites, Tenayuca, is closer to the Northern Mexico samples and Pueblo Bonito than to the Central Mexico and West Mexico samples. This result is surprising, especially Tenayuca's relatively close relationships with the Coahuilan sample from Cuatro Ciénegas, as shown in the Ward's cluster dendrogram. The result of the three-dimensional PC scatterplot shows the greatest distance between Cuatro Ciénegas and all the other samples, as could be expected given the site's geographic location and economic structure.

Table 3. Distances for pseudo-Mahalanobis D2 and other variables used in Mantel and partial Mantel tests
  HUPYPUCUNRCCTXTOTNZUGU
  1. ‘--’ represents no relationship, coded as ‘99’ in statistical analyses.

HawikuD20          
 Geog0          
 Migration0          
 Trade0          
 Political0          
PuyeD28.9410         
 Geog2680         
 Migration30         
 Trade10         
 Political--0         
PuebloD213.61317.9160        
BonitoGeog1541600        
 Migration210        
 Trade----0        
 Political----0        
CuatroD214.12726.21313.7030       
CiénegasGeog1104107211500       
 Migration9870       
 Trade------0       
 Political------0       
Narar-D215.42414.1769.26611.1620      
achicGeog8098989065010      
 Migration43290      
 Trade3------0      
 Political--------0      
CopperD216.17434.6779.39023.63923.7400     
CanyonGeog8689849815771170     
 Migration432920     
 Trade3------20     
 Political--------40     
Texcal- D236.48725.35216.48420.86721.93715.8930     
titlánGeog200419822100916124312310    
 Migration9878870    
 Trade------------0    
 Political------------0    
TlatelocoD230.73024.94819.03913.80018.89112.49216.3810   
 Geog19771960203288812361242920   
 Migration98728820   
 Trade------3----10   
 Political------4----30   
TenayucaD222.55825.68922.71522.33516.18620.94424.00324.0750  
 Geog1965194820208761224122996120  
 Migration987288330  
 Trade------4----210  
 Political------4------20  
ZacapúD231.67237.47527.09338.26128.26731.16610.66715.42728.4120 
 Geog182418411898798104210292232742820 
 Migration9878994480 
 Trade--------5--4340 
 Political------------44--0 
GuasaveD234.69925.33319.27213.3787.24530.76712.2578.50219.9399.9760
 Geog1073116611876572941831147117611359250
 Migration87678833730
 Trade--------3--32310
 Political------------------40
image

Figure 3. Ward's cluster dendrogram of phenetic distances using the pseudo-Mahalanobis' D2 matrix.

Download figure to PowerPoint

image

Figure 4. Three-dimensional principle component analysis of phenetic distances using the pseudo-Mahalanobis' D2 matrix. Site name abbreviated as in Figure 2(A).

Download figure to PowerPoint

The results of the Mantel tests show shared migration history (corr = 0.496, p = 0.004), trade (corr = 0.441, p = 0.005) and geographic distance (corr = 0.304, p = 0.02) are significantly correlated with pseudo-Mahalanobis' D2 phenetic distances. Political interaction (corr = 0.157, p = 0.133) is not significantly correlated with phenetic distances. Partial Mantel tests indicate that geographic distance is not significantly correlated with phenetic distances when similarities with shared migration history (corr = 0.148, p = 0.127) or trade (corr = 0.067, p = 0.681) are controlled. However, shared migration history (corr = 0.493, p = 0.003) and trade (corr = 0.223, p = 0.049) remain correlated when similarities with geographic distance are controlled. The results of the Mantel tests also show all geographic and cultural variables are correlated with the exception of shared migration history and trade (corr = 0.06, p = 0.593), emphasising the usefulness of the partial Mantel tests. Results of the Mantel and partial Mantel test are listed in Table 4. A two-dimensional PC plot of the correlations and Mantel tests (Figure 5) shows all variables are relatively close, as well as the relative proximity of each variable to phenetic distances. Shorter distances correspond to higher correlations.

Table 4. Mantel test results for correlations with phenetic distances
Mantel and partial Mantel tests (α = 0.05)
VariableCorrelationp
  • *

    p < 0.05.

  • **

    p < 0.005.

Mantel tests for geographic/cultural variables  
Geographic distance * shared migration0.5180.004**
Geographic distance * trade0.5140.003**
Geographic distance * political interaction0.4670.003**
Shared migration * trade0.060.593
Shared migration * political interaction0.440.032*
Trade * political interaction0.6070.002**
Mantel tests for phenetic distances  
Geographic distance0.3040.02*
Shared migration0.4960.004**
Trade0.4410.005**
Political interaction0.1570.133
Partial Mantel tests for phenetic distances  
Geographic distance + shared migration0.1480.127
Geographic distance + trade0.0670.681
Shared migration + geographic distance0.4930.003**
Trade + geographic distance0.2230.049*
image

Figure 5. Two-dimensional principle component analysis of correlations among cultural and geographic variables and phenetic distances.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

Although most samples clustered within regional networks or with adjacent regional networks, many samples were not as phenetically similar as the nearest geographically located groups. Pueblo Bonito and Tenayuca are grouped with the Northern Mexico samples, which are located much further away than other samples from within their respective geographic regions. Mantel and partial Mantel tests show that geographic distance is correlated with phenetic distances but not when trade and shared migration history are controlled.

Northern Mexico and Southwest US samples group together in the Ward's cluster and PC analyses. Pueblo Bonito is less distant from the Northern Mexico samples than to the other Rio Grande Valley samples of Hawiku and Puye, probably because of shared migration history between the Northwest Mexico samples and Pueblo Bonito. The West Mexico and Central Mexico samples also have a shared migration history and are grouped together, with the exception of Tenayuca in Central Mexico. Tenayuca was likely established by Chichimec populations from Northern Mexico (Lopez Austin and Lopez Lujan, 2005), which may explain the close phenetic similarity between Tenayuca and the Northern Mexico samples. Tlatelolco and Texcaltitlán are most similar to the West Mexico samples of Guasave and Zacapú. These similarities are concordant with migration histories of Postclassic Central Mexican groups from an area called Aztlán, believed to be located in West Mexico, from which the Aztec origin myths are derived (Berdan et al., 1996; Berdan, 2005).

The Mantel and partial Mantel tests show that economic exchange is correlated with phenetic distance, even when controlling for geographic distance. Generally, most of the samples are grouped together on the basis of their respective regional trade networks or with groups connected through major trade routes. This suggests that although inter-regional exchange may have had some effect on population movement and genetic exchange, the relationships seen through phenetic distances among the samples are mostly consistent with regional trade relationships. This result is expected, because many of the samples are from the Southwest US and Northern Mexico, where regional markets were often all that was accessible. Some evidence of inter-regional exchange facilitating population movement is seen in the close phenetic relationships between the West and Central Mexico samples, where trade existed across political boundaries. For example, Zacapú is located within the Tarascan Empire, where trade was regulated by the capital Tzintuntzan (Pollard, 2003), although the sample is phenetically similar to samples within the neighbouring Aztec Empire. This relationship is confounded by migration history, however, and more data from samples within both regions are needed to further test these relationships.

Finally, the Mantel and partial Mantel tests show no correlation between political interaction and phenetic distance. The only capital city-state among the samples is Tlatelolco, representing the dominant Tlatelolco-Tenochtitlan group in the Valley of Mexico. Texcaltitlán and Tenayuca maintained different political relationships with Tlatelolco and Tenochtitlan. Tenayuca was located in a tributary province, whereas Texcaltitlán was located in a strategic province. No other dominant–subordinate relationship occurred among any of the samples. With semi-nomadic populations in Northern Mexico, geographically isolated sites in the Southwest US, and specific imperial borders in West and Central Mexico, it is not surprising that political relationships had little to no effect on phenetic distances. Future research should include data from more political capitals and subordinate towns or cities.

Conclusion

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

Data for shared migration histories and trade relationships provided by archaeological and ethnohistoric records are plentiful for Postclassic period Mexican populations, making these groups an ideal choice for testing correlations between biological and cultural relationships. The results presented here suggest cultural processes affect phenetic distances among Postclassic Mexican populations. Here, biological affinities are shaped primarily by shared migration history and economic relationships. Phenetic distance relationships among the samples chosen here are congruent with what is known about migration patterns in the Southwest US and Mexico. It is interesting that interaction through trade has an effect on population interaction across large geographic areas and separate political boundaries. This is particularly true for the groups from West and Central Mexico, which were connected only through migration history and trade.

Although Northern Mexico is often considered a region peripheral to the Southwest US or Mesoamerica, we suggest that groups throughout this region played an important role in shaping population affinity in Mexico prior to European contact. These are preliminary results for testing population interaction in prehistoric Mexico and may be improved upon with more data from other geographic regions and temporal periods. Further, we conclude the model used here may be usefully applied to more samples from varying political and economic relationships in Mexico and the Southwest US. This study demonstrates a new approach using modelling to test cultural relationships using dental morphological observations as a proxy for genetic relationships. The model presented here can be extended to other geographic regions or temporal periods.

Acknowledgements

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References

The authors are grateful to the individuals at the institutions from which data were collected for this study for their support and assistance: José Antonio Pompa y Padilla and David Volcanes, Instituto Nacional de Antropologia e Historia; David Hunt, National Museum of Natural History; Ian Tattersall and Gisselle Garcia, American Museum of Natural History and the staff of the Maxwell Museum Laboratory of Human Osteology, Albuquerque, New Mexico. We are grateful for the financial support of this study provided by the Research Collections Grant, American Museum of Natural History; the Field Site Development Grant, Department of Anthropology University of New Mexico and the Latin American and Iberian Institute, University of New Mexico. For valuable insights, we thank Frances Berdan and reviewers of a previous version of this paper.

References

  1. Top of page
  2. Abstract
  3. The archaeological and ethnohistoric records
  4. Previous biological distance studies
  5. Materials
  6. Methods
  7. Results
  8. Discussion
  9. Conclusion
  10. Acknowledgements
  11. References
  • Anonymous Conqueror. 1971. Relacion de algunas cosas de la Nueva Espana y de la Gran Ciudad de Temestitan Mexico: Escrita por un companero de Hernan Cortes. In Coleccion de Documentos para la Historia de Mexico, JC Icazbalceta (ed.). Antigua Libraria: Mexico City.
  • Aubry BS. 2009. Population structure and interregional interaction in Prehispanic Mesoamerica: A biodistance study. Ph.D. dissertation, The Ohio State University, Columbus.
  • Beekman C. 2010. Recent research in Western Mexican Archaeology. Journal of Archaeological Research 18: 41109.
  • Benson L, Peterson K, Stein J. 2007. Anasazi (Pre-Columbian Native-American) migrations during the middle-12th and late-13th centuries – were they drought induced? Climatic Change 83: 187213.
  • Berdan FF. 2005. The Aztecs of Central Mexico: An Imperial Society. Thomson and Wadsworth: Belmont, USA.
  • Berdan FF, Anawalt PR. 1997. The Essential Codex Mendoza. University of California Press: Berkeley and Los Angeles, CA.
  • Berdan FF, Blanton RE, Boone, EH, Hodge MG, Smith ME, Umberger EE. 1996. Aztec Imperial Strategies. Dumbarton Oaks Research Library and Collection: Washington, DC.
  • Bierhorst J. 1992. History and Mythology of the Aztecs: The Codex Chimalpopoca. University of Arizona Press: Tucson, AZ.
  • Bockmann MR, Hughes TE, Townsend GC. 2010. Genetic modeling of primary tooth emergence: A study of Australian twins. Twin Research and Human Genetics 6: 573581.
  • Corruccini RS. 1972. The biological relationships of some prehistoric and historic Pueblo populations. American Journal of Physical Anthropology 37: 37388.
  • Cowgill GL. 2000. The central Mexican highlands from the rise of Teotihuacan to the decline of Tula. The Cambridge History of the Native Peoples of the Americas, Volume 2, Mesoamerica, Part 1, RE Adams, MJ MacLeod (eds.). Cambridge University Press: Cambridge, UK.
  • Crown PL. 2013. Pre-Hispanic use of cocoa. Chocolate in Health and Nutrition. Humana Press: New York, NY; 39.
  • Crown PL, Hurst JW. 2009. Evidence of cacao use in the Prehispanic American Southwest. PNAS 106: 7.
  • DiPeso CC. 1974. Casas Grandes, A Fallen Trade Center of the Gran Chichimeca. American Anthropologist 78: 13.
  • Douglas EDJ. 2010. In the Palace of Nezahualcoyotl: Painting Manuscripts, Writing the Pre-Hispanic Past in Early Colonial Period Tetzcoco, Mexico. University of Texas Press: Austin, TX.
  • Douglass AE. 2006. The Secret of the Southwest Solved by Talkative Tree Rings. Kessinger Publishing: Whitefish, MT.
  • Edgar HJH. 2002. Biological Distance and the African American Dentition. Ph.D. dissertation, The Ohio State University, Columbus.
  • Edgar HJH. 2004. Dentitions, distance, and difficulty: A comparison of two statistical techniques for dental morphological data. Dental Anthropology 17(2): 5562.
  • El-Najjar MY. 1978. Southwestern physical anthropology: do the cultural and biological parameters correspond? American Journal of Physical Anthropology 48: 151158.
  • Ericson JE, Baugh TG. 1993. The American Southwest and Mesoamerica: Systems of prehistoric exchange. Interdisciplinary Contributions to Archaeology.
  • Foster MS, Gorenstein S. 2000. Greater Mesoamerica: The Archaeology of West and Northwest Mexico. University of Utah Press: Salt Lake City, UT.
  • Gómez-Valdéz JA. 2008. Antropologia dental en poblaciones del Occidente de Mesoamérica. Instituto Nacional de Antropologia e Historia: Mexico City, MX.
  • Gonzales-Jose R, Martinez-Abadias N, Gonzales-Martin A, Bautista-Martinez J, Gomez-Valdez J, Quinto M, Hernandez M. 2007. Detection of a population replacement at the classic-postclassic transition in Mexico. Proceedings of the Royal Society 274: 681688.
  • Gorenstein S. 2000. Western and northwestern Mexico. In Cambridge History of the Native Peoples of the Americas, Volume 2, Part 1, BG Trigger, WE Washburn, RE Adams (eds.). Cambridge University Press: Cambridge, UK.
  • Hammer Ө, Harper DA, Ryan PD. 2001. past: paleontological statistics software for education and data analysis. Palaeontological Association.
  • Hammond GP, Rey A. 1940. Narratives of the Coronado Expedition 1540–1542. University of New Mexico Press: Albuquerque, NM.
  • Hassig R. 1995. Aztec Warfare: Imperial Expansion and Political Control. University of Oklahoma Press: Norman, OK.
  • Irish JD. 2010. The mean measure of divergence: Its utility in model-free and model bound analyses relative to the Mahalanobis D2 distance for nonmetric traits. American Journal of Human Biology 22: 378395.
  • Kelley JC. 1993. Zenith passage: The view from Chalchihuites. In Culture and contact: Charles C DiPeso's Gran Chichimeca, AI Woosley, JC Ravesloot (eds.). American Foundation and University of New Mexico Press: Albuquerque, NM.
  • Kemp BM, Resendez A, Roman Berrelleza JA, Malhi RS, Smith DG. 2005. An analysis of ancient Aztec mtDNA from Tlatelolco: Pre-Columbian relations and the spread of Uto-Aztecan. In Biomolecular Archaeology, Genetic Approaches to the Past, DM Reed (ed.). Southern Illinois University Press: Carbondale, IL.
  • Konigsburg LW. 1990. Analysis of prehistoric biological variation under a model of isolation by geographic and temporal distance. Human Biology 62: 4970.
  • LeBlanc SA. 1999. Prehistoric Warfare in the American Southwest. University of Utah Press: Salt Lake City, UT.
  • LeBlanc SA. 2000. Regional interaction and warfare in the late prehistoric Southwest. In The Archaeology of Regional Interaction: Proceedings of the 1996 Southwest Symposium, M Hegmom (ed.). University Press of Colorado: Boulder, CO.
  • LeBlanc SA, Turner CG, Morgan ME. 2008. Genetic relationships based on discrete dental traits: Basketmaker II and Mimbres. International Journal of Osteoarchaeology 18: 109130.
  • Legendre P, Legendre L. 1998. Numerical Ecology. Elsevier: New York, NY.
  • Lekson SH. 1999. The Chaco Meridian: Centers of Political Power in the Ancient Southwest. Alta Mira Press: Lanham, MD.
  • Lekson SH, Cameron CM. 1995. The abandonment of Chaco Canyon, the Mesa Verde migrations, and the reorganization of the Pueblo world. Journal of Anthropological Archaeology 14: 184202.
  • Lopéz Austin A, Lopéz Luján L. 2005 Mexico's Indigenous Past. University of Oklahoma Press: Norman, OK.
  • Lopéz Austin A, Lopéz Luján L. 2008. Aztec human sacrifice. In The Aztec World, ME Brumfiel, GM Feinman (eds.) Abrams Press: Mexico City, DF.
  • Mackey J. 1977. A multivariate, osteological approach to Towa culture history. American Journal of Physical Anthropology 46: 477482.
  • Manly BF. 2005. Multivariate statistical methods: A primer. Chapman and Hall. Mantel N. 1967. The detection of disease clustering and a generalized regression approach. Cancer Research 27: 209220.
  • Matos Moctezuma E. 1989. The Aztecs. Rizzoli: New York, NY.
  • McGuire RH. 1980. The Mesoamerican connection in the Southwest. Kiva 46: 12.
  • McGuire RH. 2012. Mesoamerica and the Southwest/Northwest. In The Oxford Handbook of Mesoamerican Archaeology, DL Nichols, CA Pool (eds.). Oxford University Press: Oxford, UK.
  • Melgar Tisoc ER. 2010. Una Relectura del Commercio de la Turquesa: Entre Yacimientos, Talleres y Consumidores. Caminos y Mercados de México. Universidad Autonoma de Mexico: Distrito Federal, México.
  • Melgar Tisoc ER, Solis Ciriaco RB. 2009. Manufacturing techniques of the Turquoise Mosaics from the Great Temple of Tenochtitlan, Mexico. 2nd Latin-American Symposium on Physical and Chemical Methods in Archaeology, Art and Cultural Heritage Conservation & Archaeological and Arts Issues in Materials Science. Mexico City, Mexico.
  • Mills BJ. 1995. The organization of the protohistoric Zuni ceramic production. In Ceramic Production in the American Southwest, BJ Mills, PL Crown (eds.). University of Arizona Press: Tucson, AZ.
  • Mizoguchi Y. 1977. Genetic variability of tooth crowns as ascertained from twin data. Journal of the Anthropological Society of Japan 85: 301309.
  • Nelson RS. 1986. Pochtecas and Prestige: Mesoamerican Artifacts in Hohokam Sites. In Ripples in the Chichimec Sea: New Considerations of Southwestern-Mesoamerican Interactions. FJ Mathien, RH McGuire (eds.). Southern Illinois University Press: Carbondale and Edwardsville, IL.
  • Nichol CR. 1989. Complex segregation analysis of dental morphological variants. American Journal of Physical Anthropology 78: 3759.
  • Nichol CR. 1990. Dental genetics and biological relationships of the Pima Indians of Arizona. Ph.D. Thesis, Arizona State University, Tempe.
  • Pollard HP. 2003. West Mexico beyond the Tarascan Frontier. In The Postclassic Mesoamerican World, ME Smith, FF Berdan (eds.). University of Utah Press: Salt Lake City, UT.
  • Pollard HP, Vogel TA. 1994. Late Postclassic imperial expansion and economic exchange within the Tarascan Domain. In Economies and Polities in the Aztec Realm, MG Hodge, ME Smith (eds.). Institute for Mesoamerican Studies: Albany, NY.
  • Riley CL. 1975. The Road to Hawikuh: Trade and trade routes to Cibola-Zuni during Late Prehistoric and Early Historic Times. The Kiva 41: 2.
  • Ross SH. 1967. Prehistoric trade routes between Mesoamerica and the American Southwest: A tentative assessment. The California Geographer 8: 919.
  • Sahagun B, Anderson AJ. 1982. Florentine Codex: General History of the Things of New Spain. University of Utah Press: Salt Lake City, UT.
  • SAS Institute, Incorporated. 2007. sas Software v.9.1.3. Cary, NC, USA.
  • Schillaci MA, Stojanowski, CM. 2002a. Investigating social organization at Aztec Ruins using determinant ratio analysis. In Forward into the Past: Papers in Honor of Teddy Lou and Francis Stickney, RN Wiseman, TC O'Laughlin, CT Snow (eds.). Albuquerque: Archaeological Society of New Mexico.
  • Schillaci MA, Stojanowski CM. 2002b. A reassessment of Matrilocality in Chacoan Culture. American Antiquity 60(2): 343356.
  • Schlanger SH, Wilshusen RH. 1993. Local abandonments and regional conditions in the North American Southwest. Abandonment of settlements and regions: Ethnoarchaeological and archaeological approaches 1: 8598.
  • Scott GR, Turner II CG. 1997. The Anthropology of Modern Human Teeth: Dental Morphology and its Variation in Recent Human Populations. Cambridge University Press: Cambridge, UK.
  • Smith ME. 2001. The Aztec Empire and the Mesoamerican World System. In Empires: Comparative Perspectives from Archaeology and History, SE Alcock, TN D'Altroy, KD Morrison, CM Sinopoli (eds.). Cambridge University Press: Cambridge, UK.
  • Smith ME, Berdan FF. 2003. Postclassic Mesoamerica. In The Postclassic Mesoamerican World, ME Smith, FF Berdan (eds.). University of Utah Press: Salt Lake City, UT.
  • Smith ME, Schreiber KJ. 2006. New world states and empires: Politics, religion, and urbanism. Journal of Archaeological Research 14: 152.
  • Steward JH. 1990. Theory of Culture Change: The Methodology of Multilinear Evolution. University of Illinois Press: Champaign, IL.
  • Townsend RF. 1992. The Aztecs. Thames and Hudson Ltd.: London, UK.
  • Townsend GC, Hughes TE, Luciano M, Bockmann MR, Brook A. 2009. Genetic and environmental influences on human dental variation: A critical evaluation of studies involving twins. Archives of Oral Biology 54: 4551.
  • Turner CG. 1993. Southwest Indians: Prehistory through dentition. National Geographic Research and Exploration 9(1): 3253.
  • Turner II CG, Nichol CR, Scott GR. 1991. Scoring procedures for key morphological traits of the permanent dentition: The Arizona State University dental anthropology system. In Advances in Dental Anthropology, MA Kelley, CS Larson (eds.). Wiley Liss.
  • Turpin SA. 1997. Cradles, cribs, and mattresses: Prehistoric sleeping accommodation in the Chihuahuan Desert. Journal of Big Bend Studies 9: 118.
  • VanPool TL, O'Brien MJ. 2013. Sociopolitical complexity and the bow and arrow in the American Southwest. Evolutionary Anthropology 22: 111117.
  • Walker CM. 2006. The bioarchaeology of newly discovered burial caves in the Sierra Tarahumara. Ph.D. dissertation, University of Oregon, Eugene.
  • Walsh MR. 2000. Material evidence for social boundaries on the Pajarito Plateau, New Mexico. Kiva 65: 197213.
  • Ward JH. 1963. Hierarchical grouping to optimize an objective function. Journal of the American Statistics Association 58.
  • Washburn DK, Washburn WN, Shipkova PA. 2011. The prehistoric drug trade: Widespread consumption of cacao in ancestral Pueblo and Hohokam communities in the American Southwest. Journal of Archaeological Science 38: 16341640.
  • Weigand PC. 2008. Turquoise: formal economic interrelationships between Mesoamerica and the North American Southwest. In Archaeology Without Borders: Contact, Commerce, and Change in the U.S. Southwest and Northwestern Mexico, LD Webster, ME McBrinn (eds.). University of Colorado Press: Boulder, CO.
  • Whalen ME, Minnis PE. 1996. Ballcourts and political centralization in the Casas Grandes region. American Antiquity 61(4).
  • Willermet C, Edgar HJH, Ragsdale C, Aubry BS. 2013. Biodistances among Mexica, Maya, Toltec, and Totonac groups of Central and Coastal Mexico. Chungará 45(3): 447459.
  • Woosley AI, Olinger B. 1993. The Casas Grandes ceramic tradition: production and interregional exchange of Ramos Polychrome. In Culture and Contact: Charles C. DiPeso's Gran Chichimeca, AC Woosley, JC Ravesloot (eds.) Amerind Foundation: Albuquerque, NM.
  • Wright S. 1943. Isolation by distance. Genetics 28: 11413.