"European Contact and Indian Depopulation in the Northeast: The Timing of the First Epidemics" by Dean R. Snow and Kim M. Lanphear, University at Albany, SUNY in "Ethnohistory" (Vol. 35:1, Winter 1988, pp. 15-33) Abstract In order to estimate prehistoric Indian population sizes in the New World, it is first necessary to gain a better understanding of the demographic effects of European-introduced diseases. To that end we have reexamined the timing of the first introduction of these diseases into one region, the Northeast. A careful reexamination of the ethnohistoric record combined with a study of the history and process of smallpox has lead us to conclude that this and other such diseases did not enter the Northeast until the seventeenth century, long after the well-documented initial epidemics of the Caribbean and Mexico. Reasons for the lag are suggested. Most archaeological investigations of prehistoric American Indian demography have suffered from inadequately developed theory, and all have been denied the benefit of estimates based on preepidemic population levels around A.D. 1500. The latter problem stems from inadequate assessment of the impacts of epidemics and a consequent uncertainty about what fractions of original populations the known postepidemic remnants represent. The principal difficulty is that mortality rates were so extreme that even when accurate data on remnant populations exist, small differences in assumed mortality rates yield very different results (Snow 1980: 35). The efforts of Dobyns (1966, 1983) notwithstanding, there is still little certain knowledge about pre-1500 population levels. Without them we continue to have difficulty comparing early historic populations with prehistoric ones known only archaeologically (e.g., Ramenofsky 1982). By focusing on clarification of events in the sixteenth century, this paper attempts to narrow the gap between demographic archaeology and demographic studies based on documentary sources, in order to bridge the domains of archaeology and ethnohistory. It seems clear that a better understanding of the one known major demographic shift in North America, the abrupt Indian population decline that resulted from the introduction of European diseases, is necessary for an understanding of preepidemic population levels. The latter in turn is prerequisite to a clear understanding of earlier population levels and hypothesized shifts during the course of prehistory. Paleodemography is systematically linked to other cultural variables in key ways, and it is important to bring theory, empirical archaeology, and ethnohistory to bear on it. Many sources indicate that lethal Old World diseases were introduced to the New World by 1518, when smallpox first appeared in Santo Domingo (Crosby 1986: 199-208; Dobyns 1983). American Indian societies had not been large enough or dense enough for long enough to have spawned their own array of acute community infections (Cockburn 1971). Measles, for example, requires a minimum population of 200,000 people to persist (Black 1966). The ability to isolate and snuff out smallpox in recent years illustrates how vulnerable the disease is to extinction in small populations that are naturally or artificially isolated from other human groups. It also provides emphasis for the observation that smallpox must be a relatively recent feature of human history. The latter point also supports the view that the comparatively large array of domesticated animal species in Eurasia as compared to the Americas facilitated the differential rise of lethal crown infections there (VanBlerkom 1984). In 1518, European populations were in some sense adapted to measles and smallpox as endemic crown infections, but the same cannot be said for American Indians. Once introduced, the diseases spread rapidly, aided by nearly universal susceptibility and by native treatment techniques, such as sweat baths, that often actually increased the severity of the diseases and promoted their rapid spread (Lanphear 1983). Smallpox was carried to Mexico by someone in the company of Cortes, where it quickly devastated the population. This and subsequent epidemics in Mexico are well documented. Gibson (1964: 448-51) summarizes the evidence, and Cook and Borah (1971, 1974, 1979) have dealt with it at length. These sources claim that by the end of a single century the population of central Mexico had declined to only 3 percent of its level in 1520. Documentary evidence of epidemics north of Mexico is sparse for the sixteenth century, and our understanding of epidemics there necessarily depends upon assumptions about how they spread (if they occurred at all). The product of our assumptions can only occasionally be tested by documentary or archaeological evidence, and the evidence is so scanty and ambiguous that it can be made to support almost any reasonable hypothesis. Dobyns (1983) has dealt with the problem at the theoretical level by assuming that all epidemics were pandemics, at least within the Eastern Woodlands. According to this assumption, evidence of an epidemic anywhere in the Eastern Woodlands during the sixteenth century is evidence for an epidemic everywhere in the region in the year for which it is documented. The problem with Dobyns's assumption is that North American population densities were generally lower than those in Mexico at the time, and there were often buffer zones between population concentrations or isolates that would have impeded the spread of epidemics. Milner (1980: 47) has stated that "pandemics ... might have occurred, but the usual pattern was perhaps one of exogenous disease introduction and restricted geographical spread closely corresponding to a socially heterogeneous landscape." On a historiographical level, Dobyns has been accused of misusing the few scraps of documentary evidence we have in an effort to sustain his argument for widespread sixteenth-century epidemics (Henige 1986). Many of the firsthand descriptions of the effects of epidemics on seventeenth-century North American Indian populations either implicitly or explicitly argue that the affected people had little or no previous experience with severe epidemics (Crosby 1976). Some documentary accounts indicate that locally severe epidemics were quite selective in their spread. In short, it has not been demonstrated that most or all sixteenth-century epidemics can be assumed to have been pandemics. A more complex model and new efforts to test it using documentary, archaeological, and medical data are required. The Ethnohistoric Record Current evidence suggests that epidemics did not spread northward from Mexico. Further, where epidemic diseases were introduced to North America, the likelihood of their spread beyond the point of introduction depended on the size of the infected population, the sizes of surrounding populations, the length of common territorial boundaries, the depths of buffer zones, levels of preexisting hostilities, and simple chance. Documentary sources for epidemics north of Mexico are not numerous, and a review of them will serve to make our point. First, there is no evidence that the Mexican smallpox epidemic of 1520-24 spread northward past Texas. Cabeza de Vaca (1961: 60) mentioned what might have been a local typhoid epidemic on the Texas coast in 1528. Later, in 1535, he came across a village in which a majority of the people were blind in one or both eyes, possible evidence for an earlier smallpox epidemic (Cabeza de Vaca 1961: 103). But Cabeza de Vaca did not describe other villages in a way that would suggest that they had experienced epidemics. Indeed, his astonishment in 1535 suggests that this case was very unusual, and the evidence is at best consistent with what one would have seen after a smallpox epidemic burned itself out in the relatively sparse population of Texas. Henige (1986: 296) has also pointed out that this passage is consistent with locally endemic congenital cataracts. Indeed, Cabeza de Vaca did not mention pock marks at all, and we are left with almost no evidence that smallpox was even sporadically present in the region. Second and coincidentally, Cartier noted scurvy among his men on the St. Lawrence in 1535, following some sort of epidemic among the St. Lawrence Iroquois of Stadacona (Burrage 1906: 72-76). He incorrectly inferred that the Indians had the same disease. We now know the inference to be incorrect because Cartier later made it clear that the Indians knew how to effectively treat scurvy (Trigger 1976: 194). Cartier also did not observe any sick Indians firsthand. It is possible that the Indians were suffering from influenza or some rhinovirus brought by Cartier and his men. Over fifty Indians died, but this was not an epidemic on a scale anything like what was to come later. Third, according to Garcilaso de la Vega (1962: 298), de Soto visited a village in Florida named Cofachiqui (Cofitachiqui) in 1540 and was told that there had been an epidemic in 1538, which Cofachiqui had escaped but which had devastated other villages. The Gentleman from Elvas tells a similar story (Smith 1968: 63). While this might be evidence for localized epidemic, it is clear evidence against the idea that it would have been a pandemic, since not all villages in the area were attacked. There were renewed epidemics in Mexico during the following twenty-five years (Gibson 1964: 448-51), but little to indicate that any of them made significant impact in the Eastern Woodlands. Dobyns (1983: 30) cites McNeill (1976: 209, 339) and Gibson (1964: 449) in support of the hypothesis that influenza was epidemic in the Southeast in 1559. McNeill (not a primary source) says only that there was an influenza epidemic about the same time in Europe. He raises the possibility that this epidemic was global by citing some dubious evidence from Japan but admits that there was no epidemic at all in China at this time. Gibson indicates that there was an epidemic of unknown type (but like that of 1545) in Mexico at this same time, but neither he nor McNeill mentions any evidence from the remainder of North America. We recognize that the lack of sources to document real epidemics could be just that. However, given the known behavior of epidemics and the contents of the few documents we do have, we conclude that the argument for widespread but unrecorded epidemics is fatally weak, and that the evidence against them is relatively substantial. Lewis and Loomie (1953: 85, 89) provide a transcription and translation of a 1570 letter by Luis de Quiros and Juan Baptista de Segura to Juan de Hinistrosa, which describes a severe epidemic in the area of Virginia around what was later to be Jamestown. The epidemic began six years prior to the writing of the letter, that is, in 1564. The disease cannot be identified from what they wrote, but there were epidemics of both smallpox and measles in Mexico at this time (Gibson 1964: 449). Curiously, there was another epidemic just a few years later around Roanoke Island, which was described by the English colonist Thomas Harriot. Harriot (1590: 28) said that "the disease also so strange, that they neither knew what it was, nor how to cure it; the like by report of the oldest men in the country never happened before, time out of mind." This apparent contradiction can be accounted for in several possible ways. The 1564 epidemic was accompanied by drought and crop failure and might have been due as much to famine as to disease. Second, the 1586 epidemic might have been measles (Quinn 1955, 1: 378), different in character from the earlier disease (if perhaps only a more virulent form) and therefore unlike anything experienced previously. Third, Harriot's old informants may have been dissembling for reasons unknown to him or to us. Fourth, the disease experience of the Indians around Jamestown might have been different from that of those around Roanoke Island, over one hundred miles away. There are still other possibilities, but the fourth mentioned here is the most economical hypothesis in light of what we know about both how diseases spread and the specifics of historical sources from the Southeast. Although the Southeastern populations were locally dense, there apparently were large buffer zones between population concentrations that could have impeded the spread of disease. There is considerable evidence for an epidemic in Mexico in 1576, which Gibson (1964: 449) equates in type with those of 1545 and 1559. The symptoms suggest yellow fever, but it has been generally accepted that yellow fever was not introduced into the New World until 1647. These same symptoms could also indicate typhoid, but Dobyns (1983: 30) is almost certainly incorrect in describing it as bubonic plague. The plague-carrying rat fleas, especially Xenopsylla cheopis and Nosopsyllus (Cevatophyllus) fasciatus (Shresbury 1970: 2), are carried by the black rat (Rattus rattus) which is today restricted to ships and coastal areas. It was apparently driven out of Europe by the more aggressive brown rat (Rattus norvegicus), and its retreat was accompanied by an end to bubonic plague there (Banfield 1974: 221-23). The black rat was similarly unable to compete with the now widespread brown rat when both were introduced to the Americas, and it is unlikely that plague rat fleas could have established themselves quickly enough on alternative hosts (e.g., squirrels) to raise bubonic plague to the status of epidemic disease this early in the interior of America. Quinn (1955, 1: 378) cites the epidemic mentioned by Harriot (1590: 28) and speculates that it was probably measles or smallpox, apparently on the basis of his reading of Drake's account of a Florida epidemic in European origin. The episode seems not to have slowed the Algonquins in their usual activities, and whatever its origin, the fever apparently did not take on epidemic proportions. The situation was much different by the end of the decade. Only five years after Smith, Thomas Dermer described a much different scene on coastal New England: I passed alongst the Coast where I found some ancient plantations, not long since populous now utterly void; in other places remnent remaines, but not free of sicknesse. Their disease the Plague for wee might perceiue the sores of some that escaped, who described the spots of such as usually die. (Winship 1905: 251) Josselyn, writing after 1663 but referencing the 1630s says Not long before the English came to this Countrey, happened a great mortality amongst them, especially where the English afterwards planted, the East and Northern parts were sore smitten with the Contagion; first the plague, afterwards when the English came by the small pox, the three Kingdoms or Sagamoreships of the Massachusets were very populous, ... but by the plague were brought from 30000 to 300. There are not many now to the Eastward, the Pequots were destroyed by the English: the Mohacks [probably Mohegans] are about five hundreds. (1865: 96-97) Thomas Morton also uses the word "plague" to describe the epidemic, adding that "they died on heapes, as they lay in their house" (Adams 1883: 132). However, Morton might be using "plague" as a synonym for "epidemic," not as a specific identification of the disease, and Dermer might have been similarly general. We have already argued that it is unlikely that bubonic plague was part of the set of epidemic diseases that reduced interior America. Still, it might have been the culprit in coastal New England in 1616. Other possibilities, including yellow fever and trichinosis, have been suggested by various scholars (see Spiess and Spiess 1987 with references) but many primary sources refer to the epidemic as "plague," "the plague," and "Plague" (Winslow 1622; Purchas 1625; Anonymous 1802: 226), so it seems that Williams's (1909) original diagnosis of bubonic plague might still stand. Gookin (1970: 9) appears to err in dating the epidemic to 1612-13, but usefully identifies a principal symptom as yellow skin. This has led Spiess and Spiess (1987) to suspect hepatic failure and to suggest a hepatitis virus as the probably cause. According to the Jesuit missionary Biard, the epidemic hit the Maine coast in 1616, and the Indians had experienced nothing like it previously. "They are astonished and often complain that since the French mingle and carry on trade with them, they are dying fast, and the population thinning out" (Thwaites 1959, 3: 105). Bourque and Whitehead (1985: 337) see some evidence that localized epidemics my have struck the Indian communities around the Gulf of St. Lawrence. But even here there is no evidence of disease prior to Champlain's 1604 exploration, and clear indications that the epidemics did not reach the interior of New England. Richard Vines and his men spent a winter on the Saco River at the height of the epidemic, and although they inhabited the same dwellings as sick and dying Indians, they never contracted the illness themselves. Gorges (1890: 1:100), who had sent Vines to Saco in the first place, reports that although the English seem to have been immune, many Indian settlements were so reduced that there was no one left to bury the dead. The testimony of the Indians themselves and the effects the epidemic had on the population combine to indicate that this was the first major epidemic in the Northeast. While the 1616 epidemic was the first to appear in the Northeast, the sources clearly indicate that it did not spread far into the interior. The first epidemic to reach the interior was probably the 1633 smallpox epidemic. This epidemic was so prevalent and so destructive that nearly every contemporary writer mentions it and all Northeastern Indian groups appear to have suffered from it (Bradford 1908: 302-3; Jameson 1909: 141; Thwaites 1959, 7:87, 8:43, 12:263-65; Winthrop 1908: 1:111-14). The sources include reports from English, Dutch, and French observers in various locations in the Northeast, and they clearly report a widespread epidemic affecting the Algonquians of New England and the Iroquois (specifically the Mohawk). They imply that other groups were involved as well. An epidemic, possibly smallpox, hit the Huron in the following year (Trigger 1985: 230). Trigger (1976: 499) concludes that this was the first serious epidemic in Huronia. Its mildness suggests that it was either a strain of smallpox (Jackes 1983), measles (Dobyns 1983: 17, Duffy 1953: 165), or influenza (Talbot 1956: 131). We know that virgin soil epidemics have their strongest effect when they are just that, experienced for the first time (Crosby 1976; Neel 1977). Furthermore, the Indians consistently stated that they had no such diseases before the Europeans came. Thus the 1633 epidemic appears to have been the first time that the Indians of the Northeast had encountered smallpox. The documentary evidence for the Northeast is internally consistent. Moreover, it is consistent with archaeological evidence currently emerging from the Mohawk Valley Project (Snow and Starna, in press). There were probably about eighty-one hundred Mohawks living in four communities in 1633. Iroquois villages having populations much over two thousand were apparently unstable, for they were rare if they existed at all. Had earlier epidemics reduced the Mohawks to this level from an earlier larger population we would expect to find evidence of more villages and/or larger villages than we know to have existed in the sixteenth century. In other words, the archaeological evidence barely accounts for eighty-one hundred Mohawks in four comtemporaneous communities during the sixteenth century and will not support an argument for any hypothesized larger Mohawk population. Table 1. Native populations in the Northeast Preepidemic Postepidemic Group population population Mortality ---------------------------------------------------------------- Maliseet-Passamaquoddy 7,600 2,500 67% Eastern Abenaki 13,800 3,000 78% Western Abenaki 12,000 250 98% Massachusett 44,000 6,400 86% Mohegan-Pequot 16,000 3,000 81% Pocumtuck 18,400 920 95% Quiripi-Unquachog 29,900 1,500 95% Mahican 6,400 500 92% Mohawk 8,100 2,000 75% ---------------------------------------------------------------- Note: Preepidemic population estimates apply to the first decade of the seventeenth century, whereas postepidemic levels are those reached by the middle of that century. ---------------------------------------------------------------- The Mohawk data are consistent with other data from the Northeast. Table 1 is adapted from Snow (1980: 34), with some additional refinements based on more recent work. Revisions in figures for groups other than the Mohawk result from the new assumption that adult men made up 25 percent rather than the previously assumed 30 percent of the population. The revised assumption produces slightly higher figures when populations are derived from warrior counts. Figures for the Munsee, for whom the evidence is poorest, are omitted from this presentation. The figures show the relationships between observed postepidemic population levels, high seventeenth-century mortality rates, and probable preepidemic levels. All estimates are subject to unknown error resulting from migration during the period covered. Although still approximate, the data are completely consistent with the hypothesis that population collapse in the Northeast did not begin until the second decade of the seventeenth century. The Medical Evidence If we accept the evidence that indicates that there were no major epidemics in the Northeast prior to 1616 and no smallpox until 1633, question remains as to why there was a delay between the first contact with Europeans in the area and the advent of epidemic diseases. Scattered population and large buffer zones prevented pandemics spreading from the South, but why were there virtually no epidemics at all when we know that the Indians in the Northeast had fairly frequent direct contacts with Europeans? The reason for this time lag was possibly that small fishing boat crews and the long voyage (six weeks) did not allow the smallpox virus to survive the transatlantic crossing. Only after about A.D. 1600, when the ocean passage was shortened to one month, could the smallpox virus be maintained by a small crew and thereby be transmitted to the Indians (Snow 1980: 32). This hypothesis is reasonable. Given a noninfectious incubation period of twelve to fourteen days and a disease period of fourteen to sixteen days (Horsfall and Tamm 1965: 938-39), it seems very unlikely but nevertheless possible that a crew of five to six susceptible men could maintain smallpox for a month and a half if it was contracted just prior to departure. One infected person would not become contagious until two weeks into the trip. Even if the rest of the crew contracted the disease as early as it appeared in the first member, they would be followed by a two-week period of illness, giving a shortest possible disease life of thirty-eight days, or almost six weeks. But given the ubiquity of smallpox in Europe at the time, it is unlikely that small fishing and trading crews made up of adult males would contain any members who would be susceptible to smallpox. Even if one member of the crew was susceptible and contracted the disease before departure, the remainder would be immune. Another problem rests in the viability of the smallpox virus itself. MacCallum and McDonald (1957) have shown that the smallpox virus can remain active and infectious for up to seventeen months if wrapped in raw cotton and kept in an environment in which temperature and humidity are controlled. Laboratory conditions are certainly not equivalent to those of a sixteenth-century fishing boat, and it may be impossible to perform the kinds of experiments that would simulate those conditions now that smallpox has been exterminated. Nevertheless, the raw cotton experiment showed that the virus remained viable for many weeks. Thus a fisherman that died of smallpox might have been thrown over the side, but his blanket could have remained on board, later to be traded to the Indians. Smallpox was introduced to South Africa in this way. In 1713 a group of settlers came to the area from India. There were several smallpox deaths in the course of the voyage, but no sick people or corpses left the boat. However, the clothing of the ill was brought ashore to be washed, and smallpox was contracted by those who did the laundry (Dixon 1962: 207-8). Because of the relatively long disease period for smallpox and the lengthy viability of the virus, it now seems that this disease could well have survived the month-and-a-half-long transatlantic crossing, even if the crew was small and composed of few susceptibles. Therefore, the lengthy sea voyage and small crew sizes alone are not a satisfactory set of reasons for the lack of the disease prior to 1633. A more complete explanation for the gap between the first contact and the first smallpox epidemics lies in a closer look at the history of the virus and in an understanding of the disease process itself. Smallpox was endemic in Europe in the sixteenth and seventeenth centuries (Duffy 1953: 16). It was considered a disease of children because surviving to adulthood in this population was usually accomplished by contracting and surviving the disease as a child and thereby acquiring lifelong immunity. Only rarely did adults contract smallpox, and they did so only if they had somehow been isolated from it as children (Razzell 1977: 113-15). It is unlikely, therefore, that small boat crews of adults contained infected men. It seems much more plausible that any disease would run its course well within the limits of a six- week-long sea voyage and not be transmitted to Indians. The one question that remains concerns why smallpox spread so rapidly from Spain to Mesoamerica. We suggest that the answer lies in the differential approach to colonization in the Northeast and New Spain. The goal in New Spain was to conquer the Indians. Permanent settlements of large numbers of colonists with children were established early and used as bases for further conquest. For example, Cortes sailed in 1519 from such a settlement in Cuba to Mexico with some five hundred men (Diaz del Castillo 1568: 60; Aiton 1936: 151). These large crews and large settlement sizes greatly increased the chances of bringing smallpox victims first across the ocean and then into Mexico. Another factor in the introduction of epidemic diseases to New Spain may have been the early and massive importation of African slaves. The first epidemic in New Spain was apparently brought to the area by an infected African slave in the company of Cortes (Gibson 1964: 448). By 1576 there were forty thousand Africans in Mesoamerica (Aiton 1936). Smallpox was quickly introduced to New Spain by large crews of soldiers and slaves moving out from early permanent settlements. Importation of slaves did not begin until 1619 in the United States. The lag between the first contact in the Northeast and the first epidemics was therefore probably based on the fact that the first explorers and traders there were adults, who came to the New World as members of small crews. While the catalyst for the epidemics in New Spain were larger crews, early settlements, and African slaves, the catalyst in the Northeast appears to have been the introduction of children. It is probably more than coincidence that the first record of smallpox in the Northeast does not appear until after colonists with children enter the area, most notably during the mass migrations into New England in the 1630s that were led by John Winthrop (Duffy 1953: 43). Prior to Winthrop's movement of vast numbers of people into the Northeast, very few children entered the New World north of Mexico. Of those children who did enter the area, almost all were either born in passage or were more than seven years old. They were, therefore, not exposed or well beyond the age when they would be likely to be susceptible to smallpox. The cutoff age of seven comes from an examination of variolation records. Variolation, the precursor to vaccination, was a process whereby pus from a smallpox victim is injected into an uninfected child. This would bring on a case of smallpox that was milder and less likely to be fatal than smallpox contracted conventionally (Duffy 1953: 24). The variolation process became fashionable in Europe in the early eighteenth century, but it was never used on children over seven years of age (Duffy 1953: 27). This was so because by age seven children were thought to have already been exposed to the disease and it was consequently unreasonable to risk variolation. This is consistent with the endemicity of smallpox in Europe during the seventeenth and into the eighteenth centuries, and indeed by age seven all children were likely to have been exposed to the disease (Duffy 1953: 27). Children under the age of seven rarely traveled with the early settlers (see Hutchinson 1665; Tappan 1908). Most often wives and children were left behind until a colony was settled, and then they were sent for (Quinn 1985: 265). Although there were a few exceptions--there may have been an infant on the Roanoke voyage of 1587 (Quinn 1985: 286)--children were rare, and because of what we have already discussed about the spread and maintenance of smallpox, these early infants probably did not bring the disease to the New World. The picture changed dramatically after 1630. There had been a trading post at Fort Nassau near Albany as early as 1614, and settlers were clustered around its successor, Fort Orange, by 1630. There is a persistent story that there had been an earlier French post at modern Albany in 1540. Even some very recent authors have followed older sources such as Weise (1884: 344-63) in repeating this claim. However, there is no evidence for Europeans on the upper Hudson until the 1609 voyage of Hudson himself (Snow 1987). This gives clear meaning to the 1656 statement that "the Indians [of the upper Hudson River] also affirm, that before the arrival of the Christians, and before the small pox broke out amongst them, they were ten times as numerous as they now are, and that their population had been melted down by this disease, whereof nine- tenths of them have died" (van der Donck 1968: 64). The Massachusetts Bay Colony had established several successful towns beginning with Plymouth in 1620, Jamestown was growing, and many other areas were ready for wives, children, and immigrants. Between 1630 and 1640, over twenty thousand Europeans entered New England alone (Adams 1921), among them many children. It is during this period that deaths due to smallpox on the transatlantic crossing begin to appear in the documents (see Josselyn 1865: 218). We conclude that it was the introduction of susceptible and possibly infected children along with the shortened transatlantic crossing that determined the timing of the first smallpox epidemics in the Northeast. Conclusion The two epidemics discussed here (an unknown disease in 1616 and smallpox in 1633) and subsequent epidemics in the Northeast reduced some Indian groups by up to 95 percent during the seventeenth century (Snow 1980: 34). The epidemics may have been delayed for several decades in the Northeast, compared with Mexico, but their effects were similar. It is unnecessary to assume a series of pandemics in the Eastern Woodlands during the sixteenth century. Indeed, given the known effects of the epidemics in the Northeast, the assumption of earlier equally severe epidemics would necessarily entail the projection of unrealistically high population levels for 1520. The demographic record of New England during the seventeenth century has been examined (Snow 1980: 31-42). There is enough documentation for that region to indicate that the overall mortality from the 1616-19 and 1633-39 epidemics was no less than 86 percent of the Indian population. In some localities it was much higher, reaching 95 percent or more for this twenty-three-year period. Subsequent epidemics and wars reduced the Northeastern population even more, such that Dobyns's (1966) estimate that American Indian populations typically declined by 95 percent overall does not seem at all unreasonable in this case. However, to reach this level of mortality we do not have to resort to the assumption that there were several unrecorded epidemic episodes during the sixteenth century in the Northeast. In other words, what we know of epidemics in the seventeenth century is consistent with both Dobyns's general views regarding the effects of epidemics and with documentary sources indicating that there were no sixteenth-century epidemics of consequence in the Northeast. Notes Portions of this paper drafted by Dean Snow were originally part of his contribution to a joint paper with William Starna, delivered at the 1984 meeting of the Society of American Archaeology in Portland, Oregon. The remainder of that paper has been expanded by Snow and Starna (in press) for separate publication. Starna's helpful comments on the present paper are acknowledged with our thanks. Finally, we thank four anonymous readers for many specific comments that led to a much-improved final version of this paper. References Adams, C.F. 1883 New English Canaan of Thomas Morton. Boston: The Prince Society. Adams, J.T. 1921 The Founding of New England. Boston: Atlantic Monthly Press. Aiton, A.S. 1936 The Spanish Conquest and Settlement of the Indies. in Colonial Hispanic America. A.C. Wilgus, ed., pp. 148-66. Washington: George Washington University Press. Anonymous 1802 A Description and History of Eastham, in the County of Barnstable. Massachusetts Historical Society Collections, ser. 1, 8: 154-86. Arber, E., and A.G. Bradley, eds. 1910 Travels and Works of John Smith. 2 vols. Edinburgh: J. Grant. Banfield, A.W.F. 1974 The Mammals of Canada. Toronto: University of Toronto Press. Black, F.L. 1966 Measles Endemicity in Insular populations: Critical Community Size and Its Evolutionary Implication. Journal of Theoretical Biology II: 207-11. Bourque, B.J., and R.H. Whitehead 1985 Tarrentines and the Introduction of European Trade Goods in the Gulf of Main. Ethnohistory 32: 327-41. Bradford, W. 1908 Bradford's History of Plymouth Plantation. W.T. Davis, ed. New York: Barnes and Noble. Burrage, H.S., ed. 1906 Early English and French Voyages, Chiefly from Hakluyt, 1534-1608. New York: Barnes and Noble. Cabeza de Vaca, A.N. 1961 Adventures in the Unknown Interior of America. C. Covey, trans. and ed. Albuquerque: University of New Mexico Press. Champlain, S. de 1907 Voyages of Samuel de Champlain, 1604-1618. New York: Barnes and Noble. Chupack, H. 1969 Roger Williams. New York: Twayne Publishers. Cockburn, T.A. 1971 Infectious Disease in Ancient Populations. Current Anthropology 12: 45-62. Cook, S.F., and W. Borah 1971 Essays in Population History: Mexico and the Caribbean. Vol. 1. Berkeley: University of California Press. 1974 Essays in Population History: Mexico and the Caribbean. Vol. 2. Berkeley: University of California Press. 1979 Essays in Population History: Mexico and the Caribbean. Vol. 3. Berkeley: University of California Press. Crosby, A.W. 1976 Virgin Soil Epidemics as a Factor in the Aboriginal Depopulation of America. William and Mary Quarterly, 3rd ser., 33: 289-99. 1986 Ecological Imperialism: The Biological Expansion of Europe, 900-1900. New York: Cambridge University Press. Diaz del Castillo, B. 1568 The True History of the Conquest of Mexico. Reprinted 1927. New York: R.M. McBride. Dixon, C.W. 1962 Smallpox. London: J. and A. Churchill. Dobyns, H.F. 1966 Estimating Aboriginal American Population: An Appraisal of Techniques with a New Hemispheric Estimate. Current Anthropology 7: 395-416 1983 Their Number Become Thinned. Knoxville: University of Tennessee Press. Duffy, J. 1953 Epidemics in Colonial America. Baton Rouge: Louisiana State University Press. Fox, J.P., C.E. Hall, and L.A. Elveback 1970 Epidemiology: Man and Disease. London: Macmillan. Gibson, C. 1964 The Aztecs under Spanish Rule: A History of the Indians of the Valley of Mexico, 1519-1810. Stanford: Stanford University Press. Gookin, D. 1970 Historical Collections of the Indians in New England. Boston: Towtaid. Gorges, F. 1890 Sir Ferdinando Gorges and His Province of Maine. 2 vols. Publications of the Prince Society, vol. 2, no. 19. New York: Burt Franklin. Harriot, T. 1590 A Briefe and True Report of the New Found Land of Virginia. Reprinted 1972. New York: Dover. Henige, David 1986 Primary Source by Primary Source? On the Role of Epidemics in New World Depopulation. Ethnohistory 33: 293-312. Horsfall, F.L., and I. Tamm 1965 Viral and Rickettsial Infections of Man. Philadelphia: J.B. Lippincott Company. Hutchinson 1665 The History of the Colony of Massachusetts. London: M. Richardson. Jackes, M.K. 1983 Osteological Evidence for Smallpox: A Possible Case from Seventeenth Century Ontario. American Journal of Physical Anthropology 60: 75-81. Jameson, J.F., ed. 1909 Narrative of a Journey into the Mohawk and Oneida Country. In Narratives of New Netherland. pp. 135-62. New York: Barnes and Noble. Josselyn, J. 1865 An Account of Two Voyages to New England, Made during the Years 1638, 1633. Boston: William Veazie. Lanphear, K.M. 1983 Biocultural Interactions: Smallpox and the Mohawk Iriquois. M.A. thesis. Department of Anthropology, State University of New York at Albany. Lewis, C.M., and A.J. Loomie 1953 The Spanish Jesuit Mission in Virginia, 1570-1572. Chapel Hill: Virginia Historical Society. MacCallum, F.O., and J.R. McDonald 1957 Survival of the Variola Virus in Raw Cotton. Bulletin of the World Health Organization 16: 247-54. McNeill, W.H. 1976 Plagues and Peoples. Garden City: Doubleday/Anchor. Milner, G.R. 1980 Epidemic Disease in the Postcontact Southeast: A Reappraisal. Mid-Continental Journal of Archaeology 5: 39-56. Neel, J.V. 1977 Infectious Disease among Amerindians. Medical Anthropology 5: 39-56. Purchas, S. 1625 The Description of the Countrey of Mawooshen, Discovered by the English in the Yeere 1602.3.5.7.8. and 9. In Hakluytus Posthumus; or, Purchas His Pilgrims. Vol. 4. pp. 1873-75. London: Henry Fetherston. Quinn, D.B. 1955 The Roanoke Voyages, 1584-1590. 2 vols. London: Hakluyt Society. 1985 Set Fair for Roanoke: Voyages and Colonies, 1584-1606. Chapel Hill: University of North Carolina Press. Ramenofsky, A.F. 1982 The Archaeology of Population Collapse: Native American Response to the Introduction of Infectious Disease. Ph.D. diss. University of Washington, Seattle. Razzell, P. 1977 The Conquest of Smallpox. Sussex: Caliban Books. Sauer, C.O. 1935 Aboriginal Populations of Northwestern Mexico. Ibero- Americana 10. Berkeley: University of California Press. Shrewsbury, J.F.D. 1970 A History of Bubonic Plague in the British Isles. Cambridge: Cambridge University Press. Smith, B., ed. 1968 Narratives of de Soto in the Conquest of Florida as Told by a Gentleman of Elvas and in a Relation by Luys Hernandez de Biedma. Gainesville, FL: Palmetto. Snow, D.R. 1980 The Archaeology of New England. New York: Academic Press. 1987 Indians and Europeans: The Confrontation of Cultures in the Contact Period. Selected Rensselaerswick Seminar Papers. New Netherlands Project. In press. Snow, D.R., and W.A. Starna in press Sixteenth Century Depopulation: A View from the Mohawk Valley. American Anthropologist. Spiess, A.E., and B.D. Spiess 1987 New England Pandemic of 1616-1622: Cause and Archaeological Implication. Man in the Northeast 34: 71-83. Starna, W.A., G.R. Hamell, and W.L. Butts 1984 Northern Iroquoian Horticulture and Insect Infestation: A Cause for Village Removal. Ethnohistory 31: 197-208. Talbot, F. 1956 Saint among the Hurons: The Life of Jean de Brebeuf. Garden City: Image Books. Tappan, E.M. 1908 Letters from Colonial Children. Boston: Houghton Mifflin. Thwaites, R.G., ed. 1959 The Jesuit Relations and Allied Documents, 1610-1791. 73 vols. New York: Pageant. Trigger, B.G. 1976 The Children of Aataentsic: A History of the Huron People to 1660. 2 vols. Montreal: McGill-Queen's University Press. 1985 Natives and Newcomers: Canada's "Heroic Age" Reconsidered. Montreal: McGill-Queen's University Press. van der Donck, Adriaen 1968 A Description of the New Netherlands. Syracuse: Syracuse University Press. VanBlerkon, L.M. 1984 Animals and the Origins of Human Infectious Diseases. Paper presented at the 83rd annual meeting of the American Anthropological Association, Denver. Vega, G. de la 1962 The Florida of the Inca. J.G. Varner and J.J. Varner, trans, and ed. Austin: University of Texas Press. Weise, A.J. 1884 The Discoveries of America to the Year 1525. New York: G.P. Putnam's Sons. Williams, H.U. 1909 The Epidemic of the Indians of New England, 1616-1620. John Hopkins Hospital Bulletin 20: 340-49. Williams, R. 1863 Letter to John Winthrop. Massachusetts Historical Society Collections, ser. 1, 6: 229-30. Winship, G.P., ed. 1905 Sailors Narratives of Voyages along the New England Coast, 1524-1624. New York: Burt Franklin. Winslow, E. 1622 Good News from New England. Massachusetts Historical Society Collections, ser. 1, 8: 239-76. Winslow, C.-E.A. 1943 The Conquest of Epidemic Disease. Princeton: Princeton University Press. Winslow, C.-E.A., W.G. Smillie, J.A. Doull, and J.E. Gordon 1952 The History of American Epidemiology. St. Louis: C.V. Mosby. Winthrop, J. 1908 Winthrop's Journal "History of New England." 2 vols. J.K. Hosmer, ed. New York: Barnes and Noble. Wray, C.F. 1973 Manual for Seneca Iroquois Archaeology. Honeoye Falls, NY: Cultures Primitive. Wray, C.F., and H.L. Schoff 1953 A Preliminary Report on the Seneca Sequence in Western New York. Pennsylvania Archaeologist 23: 53-63 Submitted 3 October 1986 Accepted 25 March 1987 Final revisions received 9 December 1987