Using spatial analysis to estimate depop

Journal of Anthropological Archaeology 31 (2012) 83–92 Contents lists available at SciVerse ScienceDirect Journal of Anthropological Archaeology journal homepage: www.elsevier.com/locate/jaa Using spatial analysis to estimate depopulation for Native American populations in northeastern North America, AD 1616–1645 Eric E. Jones a,⇑, Sharon N. DeWitte b a Department of Anthropology, Wake Forest University, P.O. Box 7807, Winston-Salem, NC 27109, United States b Departments of Anthropology and Biology, 1512 Pendleton Street, Hamilton College, Room 317 University of South Carolina, Columbia, SC 29208, United States a r t i c l e i n f o a b s t r a c t Article history: Eight years ago, Ramenofsky et al. (2003) characterized the discussion of the impact of Old World dis- Received 28 February 2011 eases on Native American populations as almost exclusively historical in nature. They specifically argued Revision received 9 August 2011 for the application of more evolutionary, genetic, and epidemiological theory to research into this topic. Available online 15 November 2011 We agree with their assessment and further suggest that such research would greatly benefit from spatial analyses of disease spread as well. Using trend surface analysis of existing ethnohistorical and archaeo- Keywords: logical data pertaining to population sizes and disease events, we examine the spatiotemporal dimen- Native American depopulation sions of 17th century depopulation in northeastern North America. The subsequent results allow us to Old World diseases Northeast predict possible depopulation rates for populations with very little demographic data. Further, our use Interpolation of biological, historical, and cultural data to interpret the results represents an attempt to provide a more Kriging complex explanation for the variability in cultural survivability across the region and several possible Genetic diversity avenues for productive future research. We believe research like this can significantly improve our under- standing of how Old World diseases affected historic Native American populations and cultures and con- tinue to impact them today. Ó 2011 Elsevier Inc. All rights reserved. Introduction and biological factors that were likely influencing depopulation rates. Thornton (1997) similarly critiqued the lack of demographic Determining the magnitude of depopulation from the introduc- theory being applied to this area of research. We agree with these tion of Old World diseases among Native American societies has assessments and further suggest that this topic would benefit many benefits, from helping us to understand cultural develop- greatly from spatial analyses of existing population data in order ment, to providing information that may be helpful for under- to explore spatial trends in the data and even possibly predict pop- standing current demographic patterns in Native American ulation losses among groups with no current data. communities, and to helping us understand disease transmission Our first goal for this research is to simply map what we cur- and effects among human populations. There currently exists a sig- rently know about the population distributions and depopulation nificant amount of data, albeit highly variable in accuracy and pre- (i.e. percentage of population lost) for 17th-century Native Ameri- cision, on pre- and post-contact Native American population sizes, can societies in northeastern North America. We will use this infor- depopulation rates, and the timing of Old World disease events. mation to examine spatial trends in the depopulation—something However, large geographic, demographic, and temporal gaps exist that has yet to be done. Our second goal is to use spatial interpola- in the data when viewed on regional scales. There is a need to fill in tion to determine depopulation for groups in northeastern North those gaps to better understand the variability of depopulation and America with no current population information. This will fill in the nature of disease spread. With more complete data, we may be the gaps in our current information and help us to better under- able to synthesize and analyze population and disease patterns in stand the nature of depopulation in this region. To accomplish this more detail. goal, we map 17th-century population distributions and existing Ramenofsky et al. (2003) contended that discussions of pre-con- historic and archaeological depopulation data to examine simple tact populations and the impact of Old World diseases on Native spatial patterning in depopulation rates and analyze existing his- American populations have tended to focus on historical questions toric and archaeological data using kriging, a spatial interpolation and have yet to adequately examine the complex set of cultural method, in order to predict depopulation rates. Our final goal is to use our results and existing historic and archaeological popula- ⇑ Corresponding author. tion data to begin discussions about the cultural, biological, and E-mail addresses:

[email protected]

(E.E. Jones),

[email protected]

(S.N. historical factors that may have influenced DeWitte). 0278-4165/$ - see front matter Ó 2011 Elsevier Inc. All doi:10.1016/j.jaa.2011.10.004 84 E.E. Jones, S.N. use current hypotheses about host immune function and genetic variability, cultural information from these groups, and historical information from the region to explore possible causes of observed patterns of depopulation in the Northeast. Background The reduction of Native American populations during the six- teenth through nineteenth centuries was culturally and demo- graphically devastating and has had lasting effects among Native American societies. As a result, numerous researchers, including archaeologists, paleodemographers, and historians, have been interested in gaining a better understanding of this phenomenon. Although warfare, slavery, and other mistreatments were causes, introduced Old World diseases were responsible for the over- whelming majority of deaths (Black, 1991). As such, these diseases have rightfully been the focus of research into the details of Native American depopulation. In this body of work, researchers have used numerous methods for estimating depopulation percentages and rates from disease. These have included the use of historical data (e.g. Dobyns, 1966, 1983; Kroeber, 1939; Palkovich, 1985; Thornton, 1987, 1997; Ubelaker, 1976, 1988), archaeological data (e.g. Jones, 2010a, 2010b; Snow, 1995a, 2001; Warrick, 2008), and combinations of both (Ramenofsky, 1987). To date, there have been almost no attempts to estimate depopulation figures or ex- plore existing data using spatial analysis techniques. Pre- and post-contact Native American populations in the Northeast present perhaps the best cases for examining existing depopulation patterns with new methods. Demographic archaeo- logical research on Iroquoian-speaking populations, such as the Haudenosaunee and Wendat (Jones, 2010a, 2010b; Snow, 1995a, 1995b; Snow and Starna, 1989; Warrick, 2008), has produced an empirically based population dataset for several populations for the periods just prior to and around the time of the arrival of Euro- peans. For other groups in the region, population data in the form of total counts, warrior counts, and impacts of diseases come from historical texts dating to the very earliest 17th century interactions between Europeans and Native Americans through the 18th cen- tury (for examples see Jameson, 1909, p. 141; Thwaites, 1959, vol. 7: p. 87, vol. 8: p. 43, vol. 12: pp. 263–265; Winthrop, 1908, vol 1: pp. 111–114). However, there are a small but significant number of populations in the region that have little or no popula- tion data. These include the Erie, Wenro, Neutral, Susquehannock, Kennebec, Munsee, and Naragansett-Pokanoket. These gaps have made discussing large-scale patterns of disease spread and depop- ulation in the region difficult. However, this type of dataset with several reliable sources of population data and a few gaps is almost ideal for using spatial interpolation methods to estimate depopula- tion rates for those groups with no relevant data. Interpolation methods use trends in existing data to estimate values for holes in the dataset. Datasets with more data will produce more accurate estimated values. Once these gaps are filled, the depopulation data for the Northeast will be one of the more complete population datasets for the continent. Such a dataset would allow us to ex- plore the biological, cultural, and historical factors behind the depopulation trends in much more detail than has been possible in the past. While a great deal of research has gone into producing pre- and post-contact Native American population numbers, there have been far fewer attempts to explain the patterns of the diseases and other causes that gave rise to these numbers. In addition to generating new depopulation data, we attempt to build on the small number of influential studies that sought to explain the depopulation patterns. Black (1991) discussed the role of human genetic diversity in disease virulence, particularly within native E.E. Jones, Fig. 1. The geographic distribution of 17th century Native Abenaki, (3) Western Abenaki, (4) Kennebec, (5) Penobscot, Mahican, (12) Munsee, (13) Minisink, (14) Susquehannock, Wendat, (24) Tíonontaté. Table 1 The 17th-century Native American cultures of the Northeast the source of their geographic information. Those in italics population information but have depopulation percentage below. Culture Sources on geographic distributions Maliseet-Passamaquoddy Snow (1980, p. 341) Mohegan-Pequot Dincauze (1990) Kennebec Piotrowski (2002) Mahican Bradley (2007) Eastern Abenaki Snow (1980, p. 62) Western Abenaki Snow (1980, p. 342) Massachusett Dincauze (1990) and Bragdon (1996) Pocumtuc Dincauze (1990) Quiripi-Unquachog Dincauze (1990) Narragansett-Pokanoket Dincauze (1990) Mohawk Snow (1995a) Oneida Jones (2010a, 2010c) Onondaga Jones (2010a, 2010c) Cayuga Jones (2010c) Seneca Wray et al. (1987) Wendat Engelbrecht (2003) Neutral Snow (1994) and Engelbrecht (2003) Wenro Snow (1994) and Engelbrecht (2003) Erie Snow (1994) and Munsee Dincauze (1990) Susquehannock Snow (1994) and Minisink Kraft (1991) and mapping done by one of the authors (EEJ) during previous set- tlement and population research for the 16th- and 17th-century Haudenosaunee (Jones, 2006, 2010a, 2010b, 2010c). The informa- tion from existing publications comes from regional specialists who published detailed maps, from descriptions of the extent of 86 E.E. Jones, Table 2 Population data for the 17th-century Native American populations populations for the depopulation event during the indicated Population Date Initial Resulting population Maliseet-Passamaquoddy 1633 7600 2500 Mohegan-Pequot 1633 16,000 3000 Mahican 1633 6400 500 Eastern Abenaki 1633 13,800 3000 Western Abenaki 1633 12,000 250 Massachusett 1633 44,000 6400 Pocumtuc 1633 18,400 920 Quiripi-Unquachog 1633 29,900 1500 Mohawk 1633 8000 1750 Seneca 1634 4850 3150 Oneida 1635 1500 400 Wendat 1639 20,000 10,000 Onondaga 1645 2700 1150 as the Haudenosaunee and Wendat-Tionontaté numbers. However, they have sufficient reliability to use them to examine large-scale depopulation trends. We calculated depopulation percentages from the initial and resulting population numbers. To georeference the depopulation data, we recorded each percentage as occurring at a single location and placed it at the center of the early 17th-century territory for the associated society. This created the spatial component neces- sary for accomplishing our first goal, which was to observe spatial patterns in the depopulation data. To achieve our second goal of estimating depopulation rates for populations with no current data, we performed kriging analysis on the depopulation data to estimate the depopulation rates among groups with no population information. Kriging is a type of trend surface analysis, which in turn is an interpolation tech- nique. Interpolation is ‘‘the prediction of exact values at unsampled locations from measurements made at control points within the same area’’ (O’Sullivan and Unwin, 2003, ates a continuous surface of values from a number of points an area defined by the extent of those points. Interpolation monly used to predict climatological phenomena such as rainfall amounts or temperatures. It these cases, as it is with it can be very informative to have a continuous surface to a set of isolated points. It is important to note that only produce a surface that is an approximation of the iation in the dataset (O’Sullivan and Unwin, 2003, p. counted for anomalies will not be predicted. Kriging involves three distinct steps. The first is semivariogram, which is a description of the spatial variation dataset. The second is summarizing this data using a regular ematical function. The final step is determining interpolation weights (O’Sullivan and Unwin, 2003, pp. 266–281). all analyses using the Geostatistical Analyst tool in ArcGISÒ Unlike the Spatial Analyst tool, this tool allows the operator choose from a wider array of mathematical functions, or models, based on the actual semivariogram. We chose ordinary kriging and a spherical model, which provided the best-fit line tial variation of our dataset. We chose kriging over other interpolation methods, verse distance weighting, because other methods tend to deter- mine the weight of interpolated points without consideration of the nature of the dataset being used (O’Sullivan and Unwin, 2003, p. 265). As such, these other methods are somewhat arbi- trary, which can be problematic for irregular datasets. kriging is preferred over inverse distance weighting in cases where positive autocorrelation is a factor and the density and distribution of points is irregular (Connolly and Lake, 2006, pp. 97–98). Positive E.E. Jones, Fig. 2. The percentage of population lost in each disease Fig. and influenza (i.e. several suspected causes of the 17th century epi- demics in the Northeast) among other Native American popula- tions. For example, over 50% of people in several Northern Plains groups were killed during an epidemic of smallpox in 1782 (Deck- 88 E.E. Jones, Table 3 Estimated depopulation rates based on the Kriging results. Culture Wenro 35–44.5 Neutral 35–56.1 Erie 35–56.1 Cayuga Susquehannock Kennebec Munsee 85–92.8 Narragansett-Pokanoket 79.8–85 coast of Washington ranged from 22% to 83% (Boyd, 1990). A mea- sles epidemic in 1819 killed 25–50% of people in several Northern Plains groups (Decker, 1991). In 1900, measles and influenza epi- demics killed an estimated 22% of native Alaskans at St. Lawrence Island (and other, though less reliable, estimates place mortality from these two diseases at nearly 75% in other native Alaskan vil- lages) (Wolfe, 1982). Influenza alone killed 25–50% of Woodland Cree during an epidemic in 1835 Decker (1991), and nearly 80% of people in an Inuit community died during the 1918 epidemic (Markham, 1986). The kriging results, when overlapped with prehistoric culture ranges, produce new population information for several groups in the Northeast. This has the potential to help us understanding historical and cultural developments among these cultures as well as explain differential cultural survivability in the region. We can be fairly certain that every Iroquoian culture in the Northeast underwent significant demographic and cultural changes during the mid-17th century (Bradley, 1987; Fitzgerald, 2001; Jones, 2010a, 2010b; Snow, 1995a, 1995b; Trigger, 1976; Warrick, 2008). However, we have very little information on Iroquoian pop- ulation estimates outside of the Haudenosaunee and Wendat-Tío- nontaté. In addition, we are certain about the reasons why some groups survived while others did not. Demographic stress almost certainly played a significant role in cultural survivability, so our population data provide us with a starting point for exploring this topic. We know that depopulation during the 17th century had signif- icant impacts on cultural and historical developments in the re- gion. The Mohawk suffered the greatest losses from disease during this period compared to other Haudenosaunee cultures. (We must exclude the Cayuga from this discussion because we do not have reliable population data.) Of the Mohawk, Onondaga, Oneida, and Seneca, the Mohawk are the only group that did not retain some portion of their traditional homeland (Snow, 1994, p. 199). Although we also have to consider the high religious conver- sion rates and subsequent migration to Canada and the compara- tively frequent participation in warfare by the Mohawks (Snow, 1994), their severe depopulation likely played a significant role in their dispersal and migration. Given the Mohawk situation, it is productive to examine our depopulation estimates with respect to the cultural survivabilit of groups in the Northeast. The Neutral Iroquoians were dispersed during the 1650s, and while individuals almost certainly survived, the Neutral ceased to exist as a discreet cultural unit thereafter (Fitzgerald, 2001). There is some historic documentation of the timing of diseases among the Neutral (Fitzgerald, 2001, p. 37), but no data exists on the actual numbers lost. Our results here offer a possible explanation for their dispersal. They show that groups like the Neutral and Erie, both of whom did not survive culturally, may have experienced higher depopulation rates, as much as 56%, compared to their closest surviving neighbor, the Seneca, who experienced a 35% depopulation. This would have offered the Sen- eca a distinct advantage over the Neutral Iroquoians, similar to the advantage they had over the Wendat. E.E. Jones, S.N. did live along the Connecticut River and not very far inland. It is not a stretch to assume that they also contracted the unidentified 1616 disease. The Western Abenaki, however, were located much farther up the Connecticut River—farther from the coast than several other groups who experienced much less population loss. It is not surprising that groups that experienced two disease events had higher rates of depopulation than many of those that experienced just one. However, disease dynamics are more com- plex than simple mathematics; other factors likely further in- creased risks of severe depopulation. Researchers studying modern disease dynamics have found that infection with some dis- eases can have deleterious effects on the immune system’s ability to respond to subsequent disease exposure. Infection with numer- ous viruses, such as measles, influenza, chickenpox, human immu- nodeficiency virus (HIV), Epstein-Barr, and respiratory syncytial virus (RSV) can cause immunosuppression through a variety of mechanisms (Beadling and Slifka, 2004; Louie et al., 1995; Man- chester et al., 2002; McChesney and Oldstone, 1987; Mims, 1986; Notkins et al., 1970; Roberts, 1982; Wolfe, 1982). For example, people infected with measles virus have reduced immunity and thus increased susceptibility to secondary infections; measles infection is associated with increased risks of morbidity and mor- tality from such pathogens as Mycobacterium tuberculosis and Staphylococcus aureus (Casali et al., 1984; Manchester et al., 2002; Shaheen et al., 1996; Slifka et al., 2003). Researchers have observed that people who survive acute measles have reduced cell-mediated immunity and are more likely to die during the sub- sequent months from secondary infections than their vaccinated peers, and there is evidence that this suppression of immunity can last for years following the initial measles infection (i.e. after viral clearance) (Shaheen et al., 1996). Infection with influenza and RSV can lead to changes in immunity that increase an individ- ual’s susceptibility to and risk of mortality from bacterial pneumo- nia long after the initial infection has ended (Didierlaurent et al., 2008). Thus, not only were groups like the Massachusett and Qui- ripi-Unquachog inflicted with two separate diseases, but also the diseases were less than a generation apart. Thus, impacts from the first disease could have led to even higher mortality during the second event than the second disease would have normally in- flicted. These two factors combined might partly explain the large disparity in depopulation rates. Although we are not entirely certain of the identity of the dis- eases that impacted these populations during the early 17th cen- tury, several diseases including smallpox, measles, influenza, yellow fever, and leptospirosis have been cited as possible causes (Bratton, 1988; Carter, 1931; Crosby, 1976; Duffy, 1953; Jackes, 1983; Marr and Cathey, 2010; Snow and Lanphear, 1988; Spiess and Spiess, 1987; Talbot, 1956). These diseases could have nega- tively affected immune function as described above. Further, sev- eral studies (Allison et al., 1973; Arriaza et al., 1995; Buikstra and Cook, 1981; Mackowiak et al., 2005; Owsley and Bass, 1979; Powell, 1991; Rieder, 1989; Stone et al., 2009) suggest that tuber- culosis and other bacterial diseases could have been present in North America prior to contact. Thus, the introduction of Old World viruses could have made people more susceptible to bacte- rial infections that were already present in these populations, thereby increasing infectious disease mortality rates. Settlement location and type We are still left with the interesting case of the Western Abe- naki and the fact that they experienced the highest depopulation percentage of any group in the Northeast. Although they differ geo- graphically from the other three groups with the highest depopu- lation percentages, their position along the Connecticut River may explain their high depopulation percentage. As previously 90 E.E. Jones, S.N. groups are the most genetically divergent of all the groups in- cluded in their study with respect to the hypervariable segment I (Malhi et al., 2004). Given that there are modern genetic differ- ences between the two groups, there might have been genetic fac- tors that affected disease susceptibility and mortality in these populations at the time of 17th-century disease events. To date, however, no data have been published on differences between Iro- quoian and Algonquian populations with respect to genes associ- ated with immune function or susceptibility to particular infectious diseases. Furthermore, the patterns of genetic diversity that exist among living populations do not necessarily directly re- flect the genetic diversity of their ancestral populations. The mod- ern genetic diversity of populations in the Northeast has been shaped not only by the diseases themselves but also by the com- bined effects of selection, mutation, admixture, and genetic drift during the intervening years since the 17th-century disease events. Differences in mortality rates between Haudenosaunee and their Algonquin neighbors might also have been the result of pop- ulation differences in levels of heterozygosity. This line of thinking is similar to Black’s (1991) work but examines the role of genetic diversity on a smaller scale. Studies have found that genetic homo- zygosity, at least in vertebrates, is associated with reduced im- mune competence and increased susceptibility to pathogens, whereas heterozygosity is associated with immune advantages (e.g. Acevedo-Whitehouse et al., 2005; Keller and Waller, 2002; Lie et al., 2009; Lyons et al., 2009a; MacDougall-Shackleton et al., 2005). In humans, for example, homozygosity is associated with elevated risks of tuberculosis, hepatitis, and bacterial meningitis (Lyons et al., 2009a, 2009b). The health advantages of genetic het- erozygosity are apparently not just limited to infectious diseases, as several studies have demonstrated an association between homozygosity and increased susceptibility to degenerative dis- eases in humans such as hypertension, high cholesterol, and cancer (Lie et al., 2009; Bener et al., 2007; Rudan et al., 2003). The genes of the major histocompatibility complex (MHC) are among the most highly polymorphic genes in vertebrates, and they have been the focus of extensive study because of their role in immunity (Lenz et al., 2009). MHC genes encode cell-surface mol- ecules that play a vital role in the immune system’s recognition of pathogens (Lenz et al., 2009). MHC-heterozygotes have immune systems that can respond to and thus confer resistance to a much wider array of pathogens than is true of homozygotes (Apanius et al., 1997; Doherty and Zinkernagel, 1975; Grob et al., 1998; Woelfing et al., 2009). Studies indicate that MHC heterozygotes have faster virus clearance rates (Rapin et al., 2010), lower parasite loads and better overall body condition (Lenz et al., 2009) than homozygotes. The immune benefits associated with genetic diver- sity are not unique to the MHC; Lie et al. (2009) found that greater genetic diversity at non-MHC loci was associated with fewer re- ported symptoms of infectious disease in humans. Levels of heterozygosity might have been affected by the cul- tural practices of Northeast populations. As mentioned, the Haude- nosaunee had a long-standing cultural practice of replacing lost community members through capture and adoption of enemies. This practice had spiritual importance as well as economic, social, and kinship functions (Richter, 1992, p. 32). During the 17th cen- tury when large numbers of people were dying from diseases and warfare, adoption increased in scale to the point where entire communities of outside people were adopted into Haudenosaunee cultures (Snow, 1995a, pp. 364, 403; Thwaites, 1959, vol. 52: pp. 52–55). Although these two large-scale cases occurred after the disease events in question, the practice was likely long-standing in Haudenosaunee culture (see Wray et al. (1987) for evidence of pre-contact adoption of outsiders). In addition to bolstering dwin- dling population sizes, adoption among the Haudenosaunee might have also have inadvertently served to increase their genetic E.E. is in need of investigation. Wars and other mistreatments at the hands of Europeans cannot be discredited as causes either. For example, serious conflicts between New England Algonquian groups and settlers occurred throughout the early 1600s, while major conflicts between Iroquoian groups and Europeans did not begin until later in the 1600s. A full understanding of the depopu- lation rates and subsequent effects on these cultures requires a comprehensive analysis combining demographic, archaeological, epidemiological, and historical factors. Conclusions Researchers interested in the health effects of contact between Europeans and Native American populations have, for the most part, been content with reporting estimates of populations losses derived primarily from historic sources. However, other methods for estimating depopulation numbers and rates should be ex- plored. Through spatial analyses of archaeological and historical data, this new research reveals clear differences in mortality rates between populations in the Northeast. The results also allowed us to predict depopulation rates for groups that previously lacked suf- ficient population data with which to do so. Further, only a small number of studies have gone beyond esti- mates to explore the possible genetic, demographic, environmen- tal, cultural, and historical reasons for observed patterns in depopulation. We suggest several possible reasons for the ob- served variation in depopulation: prior disease exposure resulting in compromised immune function, genetic variants that increase or decrease disease susceptibility, and cultural practices (particularly adoption among the Haudenosaunee) that may have increased ge- netic diversity in beneficial ways. This portion of our research is certainly a very early step and somewhat speculative. However, we hope that these questions will encourage future collaborative research between archaeologists and biological anthropologists focusing on demographic archaeology and genetic data. In particu- lar, research into regions of the genome associated with immune function, from modern consenting Native American populations, might reveal which potential factors most likely influenced mortal- ity rates among Northeast populations (and those in other regions) and how the observed patterns of depopulation impacted historic and modern Native American peoples and cultures. Acknowledgments We would like to thank Dean Snow and three anonymous reviewers for their comments and suggestions, which were instru- mental in this research. Part of this research was funded by a grant from the Wake Forest University Office of Sponsored Research and Programs. References Acevedo-Whitehouse, K., Vicente, J., Gortazar, C., Höfle, Amos, W., 2005. Genetic resistance to bovine tuberculosis boar. Molecular Ecology 14 (10), 3209–3217. Allison, M.J., Mendoza, A., Pezzia, A., 1973. Documentation in Pre-Columbian America. The American Review of Respiratory (6), 985–991. Apanius, V., Penn, D., Slev, P.R., Ruff, L.R., Potts, W.K., on the major histocompatibility complex. Critical (2), 179–224. Arriaza, B.T., Salo, W., Aufderheide, A.C., Holcomb, T.A., tuberculosis in northern Chile: molecular and skeletal Journal of Physical Anthropology 98 (1), 37–45. Beadling, C., Slifka, M.K., 2004. How do viral infections bacterial infections? Current Opinion in Infectious Bener, A., Hussain, R., Teebi, A.S., 2007. Consanguineous on common adult diseases: studies from an endogamous Principles and Practice: International Journal of Science Centre 16 (4), 262–267. 92 E.E. Jones, E.E., 2010c. An analysis of the factors influencing century Haudenosaunee (Iroquois) settlement locations. Journal of Anthropological Archaeology 29, 1–14. Keller, L.F., Waller, D.M., 2002. Inbreeding effects in Ecology and Evolution 17 (5), 230–241. Kraft, H.C., 1991. The Minisink Indians. In: Orr, D.G., People of Minisink: Papers from the 1989 Delaware Water Gap Symposium. National Park Service, Philadelphia, pp. 21–46. Kroeber, A.L., 1939. Cultural and Natural Areas of Native of California Press, Berkeley. Langdon, S.P., 1995. Biological relationships among the International Record of Research 67 (3), 355–374. Lenz, T.L., Wells, K., Pfeiffer, M., Sommer, S., 2009. increases parasite resistance and body condition in (Leopoldamys sabanus). BMC Evolutionary Biology 9, 269. Libert, F., Cochaux, P., Beckman, G., Samson, M., Aksenova, Claustres, M., de la Rúa, C., Ferrari, M., Ferrec, Kucinskas, V., Lavinha, J., Mercier, B., Ogur, G., Schwartz, M., Spitsyn, V., Timar, L., Beckman, L., 1998. The deltaccr5 mutation conferring protection populations has a single and recent origin in Northeastern Molecular Genetics 7 (3), 399–406. Lie, H.C., Simmons, L.W., Rhodes, G., 2009. Does genetic humans? PloS One 4 (7), e6391. Louie, M., Yoo, J., Moran, T., Mayer, L., Sperber, K., function after influenza virus infection. Clinical Immunology 2 (4), 426–433. Lyons, E.J., Amos, W., Berkley, J.A., Mwangi, I., Shafi, Peshu, N., Marsh, K., Scott, J.A.G., et al., 2009a. childhood death due to invasive bacterial disease. Lyons, E.J., Frodsham, A.J., Zhang, L., Hill, A.V.S., Amos, susceptibility to infectious diseases in humans. Biology MacDougall-Shackleton, E.A., Derryberry, E.P., Foufopoulos, T.P., 2005. Parasite-mediated heterozygote advantage population. Biology Letters 1 (1), 105–107. Mackowiak, P.A., Blos, V.T., Aguilar, M., Buikstra, J.E., American tuberculosis. Clinical Infectious Diseases: the Infectious Diseases Society of America 41 (4), 515–518. Malhi, R.S., Breece, K.E., Shook, B.A.S., Kaestle, F.A., Smith, D.G., 2004. Patterns of mtDNA diversity in Human Biology; an International Record of Research 76 (1), 33–54. Manchester, M., Smith, K.A., Eto, D.S., Perkin, H.B., Torbett, hematopoietic suppression of human CD34+ cells by Virology 76 (13), 6636–6642. Markham, N., 1986. The north coast of Labrador and the Them Days 11, 4–5. Marr, J.S., Cathey, J.T., 2010. New hypothesis for cause Americans in New England, 1616–1619. Emerging Infectious 281–286. McChesney, M.B., Oldstone, M.B.A., 1987. Viruses perturb selected principles characterizing virus-induced immunosuppression. Review of Immunology 5 (1), 279–304. Milner, G.R., Chaplin, G., 2010. Eastern North American American Antiquity 75 (4), 707–726. Milner, G.R., Anderson, D.G., Smith, M.T., 2001. The distribution of Eastern Woodlands people. In: Brose, D., Cowan, C., Mainfort, Eclipse: Archaeology of the Eastern Woodlands Indians, Smithsonian, Washington, pp. 9–18. Mims, C.A., 1986. Interactions of viruses with the immune Experimental Immunology 66 (1), 1–16. Notkins, A.L., Mergenhagen, S.E., Howard, R.J., 1970. Effect function of the immune system. Annual Review of Microbiology 538. Novembre, J., Galvani, A.P., Slatkin, M., 2005. The geographic Delta32 HIV-resistance allele. PLoS Biology 3 (11), e339. O’Sullivan, D., Unwin, D.J., 2003. Geographical Information Sons, Hoboken. Owsley, D.W., Bass, W.M., 1979. A demographic analysis of skeletons from the Larson site (39WW2) Walworth County, South Dakota: vital statistics. American Journal of Physical Anthropology 51 (2), 145–154. Palkovich, A.M., 1985. Historic population of the Eastern Journal of Anthropological Research 41 (4), 401–426. Piotrowski, T., 2002. Introduction. In: Piotrowski, T. New Hampshire and Northern New England. McFarland and Company, Jefferson, NC, pp. 1–19. Powell, M.L., 1991. Endemic treponematosis and tuberculosis southeastern United States: biological costs of chronic Ortner, D.J., Aufderheide, A.C. (Eds.), Human Paleopathology: and Future Options. Smithsonian Institution Press, Ramenofsky, A.F., 1987. Vectors of Death: The Archaeology University of New Mexico, Albuquerque. Ramenofsky, A.F., 1990. Loss of innocence: explanations in the sixteenth-century southeast. In: Thomas, D.H. (Ed.), Columbian Consequences: Archaeological and Historical Perspectives in the Spanish depopulation rates among Native American groups. To accomplish the final goal, we rights reserved. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 South American populations, which tend to be relatively homoge- nous. Such host homogeneity can increase the virulence of a dis- ease, because a pathogen that has infected and adapted to one host is thereafter preadapted to individuals who are genetically similar to the original host. There have been few attempts to apply Black’s findings to other results. Ramenofsky (1990) examined differential survivability as a re- sult of Old World diseases for groups in the Southeast using a se- lectionist approach (i.e. Darwinian evolution). She developed a model based largely on settlement location and type and how characteristics, such as sedentism and distance to major water- ways, would impact survivability of a group. Her work introduces several possible explanations for differential depopulation rates among groups in the same region. Like Black’s work, little follow- up research has been done to test her hypotheses. Below, we dis- cuss our results in reference to Ramenofsky’s model. Thornton et al. (1991) sought to bring a more complex demo- graphic approach to the discussion of depopulation. They noted that it is an oversimplification to talk solely of mortality. We must take into account, in particular, fertility patterns before and after disease events in order to fully understand the overall depopula- tion rate that is observed. For example, they noted that a popula- tion with a 1% growth rate could completely recover from a 40% loss in population within 35 years (Thornton et al., 1991, p. 37). While we do not specifically address fertility below, it is important to keep their findings in mind when discussing depopulation rates. Specifically in the Northeast where our research focuses (region and cultures shown in Fig. 1), it appears that Old World diseases had drastic negative impacts on Native American populations only after protracted contact with Europeans (Jones, 2010a, 2010b; Snow, 1995a, 1995b; Snow and Starna, 1989; Warrick, 2008). Snow and Lanphear (1988) sought to explain early results using modern knowledge of the virulence and vectors of transmission of small- pox, the most likely disease to have first struck Northeastern groups. They hypothesized that adults on the initial voyages to North America would have begun the journey with immunity to smallpox, as otherwise the disease would have burned itself out during the voyage. Only when a significant number of possible hosts (i.e. children) were present on boats could the disease infect and spread among susceptible individuals and thus be maintained on the long voyage. Although speculative, this is one of the only at- tempts to determine disease vectors and their role in the timing of diseases. Using a different approach, Milner et al. (2001) and Milner and Chaplin (2010) sought to explain differential disease impacts and depopulation through first-level spatial analyses of the geographic ranges of Late Prehistoric cultures in eastern North America. First- level spatial statistics are the initial observations of patterns, and these two studies identified a clustered pattern of settlement that included significant areas of uninhabited space across the eastern portion of the continent. The authors proposed that these uninhab- ited areas acted as barriers to disease spread, preventing pandemics. All of these studies were informative and represented the beginning of a wave of analyses that sought to move beyond num- bers to explain the patterns of depopulation. Unfortunately, many of these studies (aside from Milner and associates’ work) are over 20 years old. Few studies have sought to continue this approach over the last decade. Methods Information about the spatial distribution of early 17th-century Native American populations in the Northeast came from existing publications, which are listed in Table 1, and archaeological survey S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 85 American cultures of the Northeast used in the study. The populations are (1) Maliseet-Passamaquoddy, (2) Eastern (6) Massachusett, (7) Narragansett-Pokanoket, (8) Mohegan-Pequot, (9) Pocumtuc, (10) Quiripi-Unquachog, (11) (15) Mohawk, (16) Oneida, (17) Onondaga, (18) Cayuga, (19) Seneca, (20) Wenro, (21) Erie, (22) Neutral, (23) settlement of a group during the late sixteenth or early 17th cen- used in this study and tury, and from published research that compiled geographic settle- previously had no ment data from unpublished regional sources. Using ESRI™ estimates from our results ArcInfo™ software, we digitized the information from these sources into a geographic information system (GIS). Previous re- search by EEJ produced regional settlement pattern data using pe- destrian surveys and differential geographic positioning systems (GPS) technology. EEJ originally used these data to study the fac- tors influencing settlement patterns (Jones, 2010c) as well as the relationship between settlement size and population size (Jones, 2010a, 2010b); our study here represents a new use of these exist- ing data. Given our study period of 1616–1645, we used the loca- and Piotrowski (2002) tions of sites occupied during this period and within a GIS, simply created a shapefile of the geographic range that contained them. These data were combined with the aforementioned data from existing publications to create the total dataset for this research. Fig. 1 displays the geographic ranges for all of the groups. Pre- and post-contact population estimates came from the same and Jones (2010c) and Warrick (2008) previous research (Jones, 2010a, 2010b) as well as published sources. All population numbers are estimates and carry varying levels of accuracy. The population numbers, depopulation percent- Engelbrecht (2003) ages, and sources are listed in Table 2. The most accurate numbers Engelbrecht (2003) are those of the included Haudenosaunee cultures (i.e. Mohawk, Oneida, Onondaga, and Seneca) and the Wendat-Tionontaté. These data were generated from long-term research projects that gener- ated population sizes and changes over time using several lines of evidence, including archaeological settlement data and historic ac- counts. Population data for the other cultures come primarily from historic warrior counts or population counts by European traders and explorers. The sources from Snow and Lanphear (1988) are re- vised numbers from Snow (1980). Because they are based on a sin- gle line of evidence, the historic estimates are likely not as accurate S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 used in this study, including the year of the depopulation event being studied, beginning and ending year, percentage of total population that died, and sources. Percent References population lost 67 Snow and Lanphear (1988, p. 24) 81 Bradford (1952, p. 260) 92 Snow and Lanphear (1988, p. 24) 78 Snow and Lanphear (1988, p. 24) 98 Snow and Lanphear (1988, p. 24) 86 Snow and Lanphear (1988, p. 24) 95 Snow and Lanphear (1988, p. 24) 95 Snow and Lanphear (1988, p. 24) 78 Snow (1995a, 1995b), Jones (2010a), and Snow and Lanphear (1988, p. 24) 35 Jones (2010b) 73 Jones (2010a) 50 Heidenreich (1976), Trigger (1976, pp. 588–589), and Warrick (2008) 57 Jones (2010a) autocorrelation refers to situations where nearby events are likely to be similar to one another. In this case, disease events in close proximity to one another are likely to have similar depopulation rates and there is a large amount of variability in the depopulation rates at different locations (i.e. between different groups). We used the locations and depopulation percentages as the in- put data for the analysis. Kriging created a surface from our data points that predicted the depopulation percentages for the areas between the points and represented different values with different shades. We then overlaid our map of the geographic ranges of 17th century cultures in the Northeast with the depopulation percent- age surface. Next, we examined these overlaid layers focusing spe- cifically on those groups with no existing depopulation data. The intersection of these layers provided us with a range of depopula- tion percentages for each of these groups. of attributes Results p. 220). It cre- within Fig. 2 displays the overlay of graduated depopulation severity is com- markers and the location of Native American populations around AD 1600. Iroquoian-speaking and Algonquian-speaking popula- disease data, tions are distinguished on the map. The map shows a correlation as opposed between high depopulation rates and Algonquian populations. This kriging can is also geographically correlated with cultures located in the cen- spatial var- tral portion of the region (along the Connecticut and Hudson Riv- 271). Unac- ers) experiencing greater overall depopulation. Fig. 3 represents the results of kriging performed on the per- the creation of a centage of population lost for societies in the Northeast. This in the map represents 29 years of disease history; the earliest record is math- from AD 1616 among several groups in New England and the latest is AD 1645 among the Onondaga of modern-day central New York. We performed Like Fig. 2, the surface shows the largest losses along the Connect- 9.3.1. icut River and in the area between there and the Hudson River. to Mortality lessens as one moves away from this area. The trend sur- face overlaps with several groups, such as the Erie, Wenro, Neutral, Susquehannock, Kennebec, Munsee, and Naragansett-Pokanoket, for the spa- for whom we previously did not have depopulation data. Estimated depopulation rates for these groups derived from the kriging re- such as in- sults can be found in Table 3. It is important to reiterate that kri- ging does not predict anomalies. Discussion For example, The depopulation rates in Tables 2 and 3 range from 35% to 98%. Most of these rates fall within the range of reliably estimated mor- tality rates associated with historic outbreaks of smallpox, measles S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 87 event overlaid on the distribution of Iroquoian-speaking and Algonquian-speaking population distributions. 3. Results of kriging based on existing depopulation data. er, 1991). Haida villages lost about 81% of their populations be- tween 1839 and 1884, and Boyd (1992) attributes this precipitous drop to a smallpox epidemic in 1862–1863. During this same smallpox epidemic, mortality among groups living along the S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 Wendat history offers an analogy for the fate of the Neutral and Erie. The Wendat also did not survive the century culturally intact; Depopulation (% loss) they experienced a 50% decline in population. Historical docu- ments and archaeological data indicate that the Seneca adopted a large number of Wendat people in the mid-17th century after sev- eral decades of sustained warfare (Jones, 2010b; Thwaites, 1959, 44.5–56.1 vol. 52: pp. 52–55). The Wendat people survived biologically, but 56.1–73.4 65.6–73.4 their geographic dispersal and adoption by other groups led to the destruction of their traditional culture. The same may have happened with the Neutral (and the Erie) given the similarities in rates of depopulation among these two groups and the Wendat. Below we further discuss these adoption practices and the genetic implications thereof for Haudenosaunee people. Similarly, the Susquehannocks did not survive the 17th century culturally intact and may have experienced greater depopulation than neighboring groups. Most historic evidence indicates that the migration and eventual extinction of the Susquehannocks as a culture was the direct result of European actions (Jennings, 1968). Their possible comparatively heavy loss of population dur- ing the 17th century may have expedited this unfortunate result. Conversely, the Wenro may not have had a depopulation rate significantly different from that of the Seneca, but they also did not survive the 16th century culturally intact. This may have been more of a factor of their original population size than mortality rates per se (i.e. if their original population was small, they may have been at an initial disadvantage). However, we do not cur- rently have the pre-contact population data to know the size of the Wenro population. The Munsee may have experienced one of the highest depopulation percentages in the region, but they sur- vived biologically and culturally intact. Clearly other factors out- side of demography were at play with these two cultures. Kriging appears to be a viable method for producing demo- graphic data for Native populations in regions where reliable data already exists. More generally, looking at depopulation data from a spatial perspective allows us to explore topics like differential sur- vivability. For several cultures in the Northeast, depopulation per- centages appear to have been indirectly correlated with survivability. However, there are exceptions. Several other factors certainly played a role in determining which cultures survived this tumultuous time, such as alliances with European nations and ori- ginal population sizes, but in many cases depopulation from dis- ease almost certainly was the primary contributor to cultural extinctions. In addition, this dataset, which is now even more ro- bust, is a good place to revive discussions of the complex interac- tion of demography, genetics, cultural practices, and historical events that shaped historic Native American populations in the Northeast. Explaining the patterns of depopulation in the Northeast Multiple diseases and their impact on immune competence y There are several possible explanations for the variable mortal- ity rates observed across the Northeast. The groups living along the New England coast were impacted by two waves of diseases, in 1616 and 1633, compared to one wave over this span of time for the other populations further north and to the west (Snow and Lanphear, 1988, p. 23). By most historic accounts, the 1616 epi- demic did not spread far inland (Snow and Lanphear, 1988). Inter- estingly, of the four groups with the highest depopulation percentages, Western Abenaki, Massachusett, Quiripi-Unquachog, and Pocumtuc, only two, Massachusett and Quiripi-Unquachog, were coastal groups. Groups like the Western Abenaki and Pocum- tuc actually had similar or higher depopulation percentages even though they did not live along the coast. The Pocumtuc, however, DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 89 mentioned, Ramenofsky (1990, p. 42) developed a model that pre- dicted disease impacts based on geographic location and settle- ment. In her model, sedentary groups living along navigable waterways were at the highest risk of being severely affected by disease because they would have had consistent and frequent con- tact with Europeans, who were the primary vectors. The Western Abenaki were semi-sedentary (Carson, 2002; Snow, 1980) and lived along a major waterway in the region, so they meet the qual- ifications described by Ramenofsky. Although most accounts state that the 1616 epidemic did not travel very far inland, it is very pos- sible that it traveled longer distances with Europeans along major waterways like the Connecticut River. The most likely explanation for the high depopulation percentage among the Western Abenaki is that they were one of the few inland groups that experienced both the 1616 and 1633 disease events because of their degree of sedentism and location along a major waterway. In addition to the Western Abenaki, Fig. 2 shows that other groups that settled along major rivers tend to have depopulation percentages in the highest two categories (74–85% and 86–98%). Groups along secondary drainages tend to have lower depopula- tion percentages. These data also fit Ramenofsky’s model. In addi- tion, the kriging results suggest that the Munsee, Susquehannock and Kennebec had the highest depopulation percentages among the groups with no data. The Munsee and Susquehannock resided along major rivers and the Kennebec settled along the coast. Genetic variability As mentioned, the data used in this study also reveal a pattern of depopulation percentages varying by language group (i.e. Iro- quoian or Algonquian). As explained, the highest rates do occur among Algonquian populations in western New England, and Iro- quoian populations have lower mortality figures than these groups. The only exception is the Maliseet-Passamaquoddy. In addition to the geographic factors discussed above, there are several possible genetic explanations that could explain why Algonquian popula- tions suffered higher mortality rates. One possibility is that the Iroquoian people had genetic variants that conferred immunity to certain diseases, and these variants did not exist in Algonquian-speaking peoples. This topic is worth investigating further because numerous genes associated with sus- ceptibility or resistance to disease have been identified in human populations, such as mutations of the G6PD gene, which are asso- ciated with resistance to malaria, and the CCR5-delta32 allele, which confers resistance to HIV in homozygotes and slows pro- gression to AIDS in heterozygotes (Dean et al., 1996; Libert et al., 1998; Novembre et al., 2005; Stephens et al., 1998). Some human major histocompatibility complex (MHC, also often referred to as human leukocyte antigen, or HLA) haplotypes are associated with resistance to specific diseases (e.g. DRB1⁄1302 is associated with reduced risks of malaria) whereas other haplotypes result in in- creased susceptibility to certain diseases (e.g. HLA-DR2 is associ- ated with increased risks of both leprosy and tuberculosis); there are also haplotypes that simultaneously confer resistance to some diseases while increasing susceptibility to others (Apanius et al., 1997). Further, the genetic background of hosts can affect MHC- dependent pathogen resistance, so even if the same MHC haplo- type exists in two populations, immune responses might differ if there are differences between the two populations with respect to other genetic loci. Genetically, Iroquoian-speaking peoples are different from Algonquian-speaking peoples (Malhi et al., 2004). Malhi et al. (2004) examined mitochondrial DNA haplotype diversity among living Iroquoian and Algonquian populations and found significant differences between the haplotype frequency distributions of the two groups. They also found that the Iroquoian and Algonquian DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 diversity, which could have enhanced their immune functions rel- ative to groups that did not practice adoption to the same extent. There is some evidence that rates of adoption may have affected genetic heterozygosity among Northeast native populations. Lang- don’s (1995) analysis of anthropometric data (which are informa- tive about genetic relationships), collected between 1891 and 1893, revealed that among the Haudenosaunee, the Seneca and Onondaga have the highest mean phenotypic variance, which might indicate that they experienced relatively high levels of external gene flow, at least some of which might have resulted from adoption of people from other groups. Historical accounts of the Seneca detail their adoption of an entire Wendat village dur- ing the 17th century (Thwaites, 1959, vol. 52: pp. 52–55), and pre- vious archaeological research suggests that adoption of Wenro and other outside women was common prior to contact (Wray et al., 1987, pp. 242–248; Wray et al., 1991, p. 396). In addition, several studies (Bradley, 1987; Jones, 2010a) provide evidence for the adoption of Hochelagan and Stadaconan Iroquoians into Onondaga society during the late 16th century. Further, as the information in Table 1 displays, the Seneca and Onondaga had the lowest depop- ulation rates of the four Haudenosaunee groups with population estimates. Patterns of genetic diversity offer one possible explana- tion for this pattern. Despite the corroborating historical and archaeological data for these two groups, there are currently few published data available to further test hypotheses about the possible relationships be- tween levels of genetic heterozygosity and disease resistance/sus- ceptibility among historic Northeastern populations. Further, Langdon’s (1995) study uses late 20th century anthropometric data, and like modern haplotype diversity data (Malhi et al., 2004), such relatively modern data may not accurately reflect the phenotypic and genetic diversity that existed in the ancestral pop- ulations at the time of disease events in 17th century. Examination of genetic diversity between Iroquoian and Algonquian popula- tions, particularly in regions of the genome associated with im- mune function, would be helpful in fully examining the variable depopulation rates among Native American populations. Other possible explanations Although we have provided several explanations for the pat- terns of depopulation in the Northeast, we must remain open to the possibility of other factors. First, we must be aware that the largest depopulation rates occurred among the groups for which only historic data are available. It is possible that these data are less reliable because they are based on a single line of evidence. While unlikely, this could account for the higher depopulation per- centages. We believe this is unlikely because historic data were used, at least in part, to estimate population sizes for all of the groups included in our study. Such historic data were simply com- bined with archaeological data when available for many of the Iro- quoian cultures. In addition to possible biases in the data, there are other possi- ble biological and cultural factors that affected variation in depop- ulation rates among Northeast groups. As mentioned, adoption processes among the Haudenosaunee undoubtedly would have masked some of the population loss (Jones, 2010a), and that is a likely explanation for lower depopulation rates among them. The adoptees were also coming from surrounding populations and these emigrants could be counting toward some of the other depopulation rates elsewhere. However, at this time we have no way of confirming such migratory activity. Conversely, the Wendat had a much lower depopulation rate than groups in New England, and they did not adopt large numbers of outsiders during this per- iod. Thus, the differences in depopulation rates between Iroquoian and Algonquian groups is not solely due to adoption practices and Jones, S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 91 Black, F.L., 1991. Why did they die? Science 258, 1739–1740. Boyd, R.T., 1990. Demographic history, 1774–1874. In: Northwest Coast. In: Suttles, W. (Ed.), . Handbook of Native American Indians, vol. 7. Smithsonian Institution Press, Washington, pp. 135–148. Boyd, R.T., 1992. In: Verano, J.W., Ubelaker, D.H. (Eds.), Disease and Demography in the Americas. Smithsonian Institution Press, Washington, pp. 249–255. Bradford, W., 1952. In: Morison, Samuel Eliot (Ed.), Of Plymouth Plantation, 1620– 1647. Alfred A. Knopf, New York. Bradley, J.W., 1987. Evolution of the Onondaga Iroquois. Syracuse University Press, Syracuse. Bradley, J.W., 2007. Before Albany: An Archaeology of Native-Dutch Relations in the Capital Region, 1600–1664. New York State Museum Bulletin 509, Albany. Bragdon, K.J., 1996. Native People of Southern New England, 1500–1650. University of Oklahoma Press, Norman. Bratton, T.L., 1988. The identity of the New England Indian epidemic of 1616–19. Bulletin of the History of Medicine 62, 351–383. Buikstra, J.E., Cook, D.C., 1981. Pre-Columbian tuberculosis in West-Central Illinois: prehistoric disease in biocultural perspective. Northwestern University Archaeological Program Scientific Papers No. 5. In: Buikstra, J.E. (Ed.), Prehistoric Tuberculosis in the Americas. Northwestern University Archaeological Program, Evanston, IL, pp. 115–139. Carson, J.A., 2002. Indians of New Hampshire. In: Piotrowski, T. (Ed.), The Indian Heritage of New Hampshire and Northern New England. McFarland and Company, Jefferson, NC, pp. 108–118. Carter, H.R., 1931. Yellow Fever: An Epidemiological and Historical Study of Its Place of Origin. Williams and Wilkins, Baltimore. Casali, P., Rice, G.P., Oldstone, M.B., 1984. Viruses disrupt functions of human lymphocytes. Effects of measles virus and influenza virus on lymphocyte- mediated killing and antibody production. The Journal of Experimental Medicine 159 (5), 1322–1337. Connolly, J., Lake, M., 2006. Geographical Information Systems in Archaeology. Cambridge, London. Crosby, A.W., 1976. Virgin soil epidemics as a factor in the aboriginal population in America. The William and Mary Quarterly 33, 289–299. Dean, M., Carrington, M., Winkler, C., Huttley, G.A., Smith, M.W., Allikmets, R., Goedert, J.J., Buchbinder, S.P., Vittinghoff, E., Gomperts, E., Donfield, S., Vlahov, D., Kaslow, R., Saah, A., Rinaldo, C., Detels, R., O’Brien, S.J., 1996. Genetic restriction of HIV-1 infection and progression to AIDS by a deletion allele of the CKR5 structural gene. Hemophilia Growth and Development Study, Multicenter AIDS Cohort Study, Multicenter Hemophilia Cohort Study, San Francisco City Cohort, ALIVE Study. Science 273 (5283), 1856–1862. Decker, J.F., 1991. Depopulation of the Northern Plains natives. Social Science and Medicine (1982) 33 (4), 381–393. Didierlaurent, A., Goulding, J., Patel, S., Snelgrove, R., Low, L., Bebien, M., Lawrence, T., van Rijt, L.S., Lambrecht, B.N., Sirard, J.-C., et al., 2008. Sustained desensitization to bacterial Toll-like receptor ligands after resolution of respiratory influenza infection. The Journal of Experimental Medicine 205 (2), 323–329. Dincauze, D.F., 1990. A capsule prehistory of southern New England. In: Hauptman, L.M., Wherry, J.D. (Eds.), The Pequots in Southern New England: The Fall and Rise of an American Indian Nation. University of Oklahoma, Norman, pp. 19–32. Dobyns, H.F., 1966. Estimating aboriginal American population. Current Anthropology 7 (4), 395–416. Dobyns, H.F., 1983. Their Number Become Thinned: Native American Population Dynamics in Eastern North America. University of Tennessee Press, Knoxville. Doherty, P.C., Zinkernagel, R.M., 1975. Enhanced immunological surveillance in mice heterozygous at the H-2 gene complex. Nature 256 (5512), 50–52. Duffy, J., 1953. Epidemics in Colonial North America. Louisiana State University Press, Baton Rouge. Engelbrecht, W., 2003. Iroquoia: The Development of a Native World. Syracuse University Press, Syracuse. Fitzgerald, W.R., 2001. Contact, neutral Iroquoian transformation, and the little ice age. In: Brose, D., Cowan, C., Mainfort, R., Jr. (Eds.), Societies in Eclipse: Archaeology of the Eastern Woodlands Indians, A.D. 1400–1700. Smithsonian, Washington, pp. 37–48. Grob, B., Knapp, L.A., Martin, R.D., Anzenberger, G., 1998. The major histocompatibility complex and mate choice: inbreeding avoidance and selection of good genes. Experimental and Clinical Immunogenetics 15 (3), 119–129. U., Fernández-de-Mera, I.G., Heidenreich, C.E., 1976. Settlements and missionaries, 1615–1650 (Plate 34). In: in the Iberian wild Harris, R.C. (Ed.), Historical Atlas of Canada, vol. 1. University of Toronto, Toronto. of a case of tuberculosis Jackes, M.K., 1983. Osteological evidence for smallpox: a possible case from Disease 107 seventeenth-century Ontario. American Journal of Physical Anthropology 60, 75–81. 1997. The nature of selection Jameson, J.F., 1909. Narrative of a journey into the Mohawk and Oneida country. In: Reviews in Immunology 17 Narratives of New Netherland. Barnes and Noble, New York, pp. 135–162. Jennings, F., 1968. Glory, death, and transfiguration: the Susquehannock Indians in 1995. Pre-Columbian the seventeenth century. Proceedings of the American Philosophical Society 112 evidence. American (1), 15–53. Jones, E.E., 2006. Using viewshed analysis to explore settlement choice: a case study predispose patients to of the onondaga iroquois. American Antiquity 71, 523–538. Diseases 17 (3), 185–191. Jones, E.E., 2010a. Population history of the Onondaga and Oneida Iroquois, AD marriages and their effects 1500–1700. American Antiquity 75 (2), 387–407. population. Medical Jones, E.E., 2010b. Sixteenth and seventeenth century Haudenosaunee (Iroquois) the Kuwait University, Health population trends in Northeastern North America. Journal of Field Archaeology. Jones, S.N. DeWitte / Journal of Anthropological Archaeology 31 (2012) 83–92 sixteenth and seventeenth Borderlands East, vol. 2. Smithsonian Institution Press, Washington, DC, pp. 31–49. Ramenofsky, A.F., Wilbur, A.K., Stone, A.C., 2003. Native American disease history: wild populations. Trends in past, present and future directions. World Archaeology 35 (2), 241–257. Rapin, N., Lund, O., Bernaschi, M., Castiglione, F., 2010. Computational immunology Campana, D.V. (Eds.), The meets bioinformatics: the use of prediction tools for molecular binding in the simulation of the immune system. PloS One 5 (4), e9862. Richter, D.K., 1992. The Ordeal of the Longhouse: The Peoples of the Iroquois League North America. University in the Era of European Colonization. University of North Carolina Press, Chapel Hill. Iroquois. Human Biology; an Rieder, H.L., 1989. Tuberculosis among American Indians of the contiguous United States. Public Health Reports (Washington, D.C.: 1974) 104(6), 653–657. Diverse MHC IIB allele repertoire Roberts, N.J., 1982. Different effects of influenza virus, respiratory syncytial virus, the Long-tailed giant rat and Sendai virus on human lymphocytes and macrophages. Infection and Immunity 35 (3), 1142–1146. M., Cao, A., Czeizel, A., Rudan, I., Rudan, D., Campbell, H., Carothers, A., Wright, A., Smolej-Narancic, N., C., Glover, G., Grinde, B., Güran, S., Janicijevic, B., Jin, L., Chakraborty, R., Deka, R., et al., 2003. Inbreeding and risk of Peltonen, L., Rosatelli, C., late onset complex disease. Journal of Medical Genetics 40 (12), 925–932. Parmentier, M., Vassart, G., Shaheen, S.O., Aaby, P., Hall, A.J., Barker, D.J., Heyes, C.B., Shiell, A.W., Goudiaby, A., against HIV-1 in Caucasian 1996. Cell mediated immunity after measles in Guinea-Bissau: historical cohort Europe. Human study. BMJ (Clinical Research Ed.) 313 (7063), 969–974. Slifka, M.K., Homann, D., Tishon, A., Pagarigan, R., Oldstone, M.B.A., 2003. Measles diversity predict health in virus infection results in suppression of both innate and adaptive immune responses to secondary bacterial infection. The Journal of Clinical Investigation 1995. Impairment of monocytic 111 (6), 805–810. and Diagnostic Laboratory Snow, D.R., 1980. The Archaeology of New England. Academic Press, New York. Snow, D.R., 1994. The Iroquois. Blackwell, Cambridge. M., Williams, T.N., Newton, C.R., Snow, D.R., 1995a. Mohawk Valley Archaeology: The Sites. Pennsylvania State Homozygosity and risk of University Press, University Park. BMC Medical Genetics 10, 55. Snow, D.R., 1995b. Microchronology and demographic evidence relating to the size W., 2009b. Consanguinity and of pre-Columbian North American Indian populations. Science 268, 1601–1604. Letters 5 (4), 574–576. Snow, D.R., 2001. Evolution of the Mohawk Iroquois. In: Brose, D., Cowan, C., J., Dobson, A.P., Hahn, Mainfort, R., Jr. (Eds.), Societies in Eclipse: Archaeology of the Eastern in an outbred songbird Woodlands Indians, A.D. 1400–1700. Smithsonian, Washington, pp. 19–25. Snow, D.R., Lanphear, K.M., 1988. European contact and Indian depopulation in the 2005. On the origin of Northeast: the timing of the first epidemics. Ethnohistory 35 (1), 15–33. An Official Publication of Snow, D.R., Starna, W.A., 1989. Sixteenth-century depopulation: a view from the Mohawk Valley. American Anthropologist 91 (1), 142–149. Chatters, J.C., Hackenberger, S., Spiess, A.E., Spiess, B.D., 1987. New England pandemic of 1616–1622: cause and northwestern North America. archeological implication. Man in the Northeast 34, 71–83. Stephens, J.C., Reich, D.E., Goldstein, D.B., Shin, H.D., Smith, M.W., Carrington, M., B.E., 2002. Targeting and Winkler, C., Huttley, G.A., Allikmets, R., Schriml, L., Gerrard, B., Malasky, M., measles virus. Journal of Ramos, M.D., Morlot, S., Tzetis, M., Oddoux, C., di Giovine, F.S., Nasioulas, G., Chandler, D., Aseev, M., Hanson, M., Kalaydjieva, L., Glavac, D., Gasparini, P., Spanish influenza of 1918. Kanavakis, E., Claustres, M., Kambouris, M., Ostrer, H., Duff, G., Baranov, V., Sibul, H., Metspalu, A., Goldman, D., Martin, N., Duffy, D., Schmidtke, J., Estivill, X., of epidemic among Native O’Brien, S.J., Dean, M., 1998. Dating the origin of the CCR5-Delta32 AIDS- Diseases 16 (2), resistance allele by the coalescence of haplotypes. American Journal of Human Genetics 62 (6), 1507–1515. lymphocyte functions: Stone, A.C., Wilbur, A.K., Buikstra, J.E., Roberts, C.A., 2009. Tuberculosis and leprosy Annual in perspective. American Journal of Physical Anthropology 140 (S49), 66–94. Talbot, F., 1956. Saint Among the Hurons: The Life of Jean de Brebeuf. Image Books, population at ca. A.D. 1500. Garden City. Thornton, R., 1987. American Indian Holocaust and Survival: A Population History Since 1492. University of Oklahoma Press, London. R., Jr. (Eds.), Societies in Thornton, R., 1997. Aboriginal North American Population Decline and Rates of A.D. 1400–1700. Decline, ca. A.D. 1500–1900. Current Anthropology 38, 310–315. Thornton, R., Miller, T., Warren, J., 1991. American Indian population recovery system. Clinical and following smallpox epidemics. American Anthropologist 93 (1), 28–45. Thwaites, R.G., 1959. Jesuit Relations and Related Documents, 73 vols. of virus infections on the Trigger, B.G., 1976. The Children of the Aataentsic: A History of the Huron People to 24 (1), 525– 1660. McGill-Queen’s University Press, Montreal. Ubelaker, D.H., 1976. Prehistoric New World population size: historical review and spread of the CCR5 current appraisal of North American estimates. American Journal of Physical Anthropology 45, 661–666. Analysis. John Wiley & Ubelaker, D.H., 1988. North American population size, A.D. 1500 to 1985. American Journal of Physical Anthropology 77, 289–294. Warrick, G., 2008. A Population History of the Huron-Petun, A.D. 500–1650. Cambridge, New York. Winthrop, J., 1908. In: Hosmer, J.K. (Ed.), Winthrop’s Journal ‘‘History of New Pueblos: 1540–1910. England.’’ 2 vols. Barnes and Noble, New York. Woelfing, B., Traulsen, A., Milinski, M., Boehm, T., 2009. Does intra-individual major (Ed.), The Indian Heritage of histocompatibility complex diversity keep a golden mean? Philosophical Transactions of the Royal Society of London Series B, Biological Sciences 364 (1513), 117–128. in the prehistoric Wolfe, R.J., 1982. Alaska’s great sickness, 1900: an epidemic of measles and endemic disease. In: influenza in a virgin soil population. Proceedings of the American Philosophical Current Syntheses Society 126 (2), 91–121. Washington, pp. 173–180. Wray, C.F., Sempowski, M.L., Saunders, L.P., Cervone, G.C., 1987. The Adams and of European Contact. Culbertson Sites. Rochester Museum and Science Center, Research Records, No. 19. Rochester Museum and Science Center, Rochester. of differential persistence Wray, C.F., Sempowski, M.L., Saunders, L.P., 1991. Tram and Cameron: Two Early Contact Era Seneca Sites. Rochester Museum and Science Center, Research Records 21. Rochester Museum and Science Center, Rochester.