Genetic Continuity After the Collapse of the Wari Empire: Mitochondrial DNA Profiles from Wari and Post-Wari Populations in the Ancient Andes Brian M. Kemp, 1 * Tiffiny A. Tung, 2 and Marshall L. Summar 3 1 Department of Anthropology and the School of Biological Sciences, Washington State University, Pullman, WA 99164 2 Department of Anthropology, Vanderbilt University, Nashville, TN 37240 3 Center for Human Genetic Research and the Vanderbilt University Medical Center, Nashville, TN 37240 KEY WORDS mtDNA; ancient DNA; bioarchaeology; Huari; Peru ABSTRACT The Wari empire flourished in the central, highland Peruvian Andes from AD 600–1000, and although the events that led to its demise are unknown, archaeologi- cal evidence indicates that Wari control waned at the end of the first millennium. Here, we test the hypothesis that, de- spite the major shift in social and political organization at the fall of the Wari empire, the mitochondrial DNA (mtDNA) composition of populations from the Ayacucho Basin, the for- mer imperial heartland of the empire, remained essentially unchanged. Results show that mtDNA haplogroup fre- quencies among the Wari and post-Wari groups differ, but the difference is not statistically significant (v 2 5 5.886, df 5 3, P 5 0.1172). This is the first study in the Andes to use haplotypic data to evaluate the observed genetic distance between two temporally distinct prehispanic populations (F ST 5 0.029) against modeled expectations of four possible evolutionary scenarios. None of these simulations allowed the rejection of continuity. In total, at both the haplogroup and haplotype levels these data do not allow us to reject the hypothesis that post-Wari individuals sampled in this study are the maternal descendants of those sampled from the Wari era site of Conchopata. However, genetic homogeneity in the mitochondrial gene pool, as seen in the late prehis- panic southern Andes, may also characterize our study region. But, prior to this research, this was unknown. If our new data show mtDNA homogeneity, then this could limit the detection of female migration if, in fact, it occurred. Nonetheless, the novel mtDNA data presented here cur- rently do not support the hypothesis that there was an influx of genetically distinct females into the former Wari heartland after the Wari collapse. Am J Phys Anthropol 140:80–91, 2009. V V C 2009 Wiley-Liss, Inc. Analysis of mitochondrial DNA (mtDNA) has been instrumental in aiding the reconstruction of Native Amer- ican prehistory by documenting genetic relationships between various precontact populations. The study of an- cient DNA (aDNA) combined with archaeological data pro- vides a unique way to reconstruct the past because assumptions about particular aspects of genetic continuity or discontinuity based on cultural proxies can be directly tested. As a complementary view to archaeological and osteological data that shows the collapse of the Wari empire in the Peruvian Andes about AD 1000, we examine if this significant cultural transformation led to altera- tions in the mtDNA composition of those living in the Wari imperial heartland from Wari to post-Wari times. Specifically, we inquire if natal populations living in the imperial heartland were heavily obliterated or pushed out of the region after Wari collapse, perhaps to be replaced by groups that carried substantially different mtDNA pro- files. Alternatively, the subsequent inhabitants in the for- mer Wari imperial heartland could have been descendent populations of those that lived at an earlier Wari settle- ment. To address this, we compare mtDNA variation exhibited by Wari and post-Wari populations. The Wari samples come from the site of Conchopata in the modern city of Ayacucho, and the post-Wari samples are from the Late Intermediate Period (post-Wari) sectors at the site of Huari, located 10 km north of Conchopata. We propose that, if there are distinct mtDNA composi- tions between the Wari (ca. AD 600–800) and post-Wari sample (AD 1100–1400), then this could have resulted from an influx of genetically distinct females into the heartland population after Wari collapse. Gene flow is the explanation for evolutionary change, when one has controlled for the evolutionary forces of mutation and genetic drift, as explained in the simulations in the Methods section. Also, a fluctuation in the gene pool in this short time span is likely not due to natural selec- tion, as mtDNA is under weak selection (Kivisild et al., 2006). In contrast, if the mtDNA compositions of the two sample groups do not significantly differ, then this could suggest that there was no influx of nonlocal females (i.e., female population remains essentially unchanged). How- ever, if mtDNA homogeneity of late prehispanic Andean populations, as detected in Southern Peru and Northern Chile (Lewis, In Press), extends to our study region, then it is possible that long-range population movements and/or replacements might not be detectable based on Grant sponsor: Vanderbilt University Interdisciplinary Discovery Grant. Grant sponsor: Vanderbilt University Center for Americas. *Correspondence to: Brian M. Kemp, Department of Anthropol- ogy, Washington State University, Pullman, WA 99164-4910, USA. E-mail: bmkemp@wsu.edu Received 21 July 2008; accepted 29 December 2008 DOI 10.1002/ajpa.21037 Published online 17 March 2009 in Wiley InterScience (www.interscience.wiley.com). V V C 2009 WILEY-LISS, INC. AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 140:80–91 (2009)