AMERICAN JOURNAL OF HUMAN BIOLOGY 19:218227 (2007)

Pearl Memorial Lecture

Human Cold Adaptation: An Unnished Agenda

A. THEODORE STEEGMANN, JR.* Department of Anthropology, University at Buffalo, Buffalo, New York 14261

ABSTRACT 1975 marked the end of a 20-year period of human biology research on physical environment. The focus then shifted from climatic adaptation to problems of nutrition, disease, and stress. However, many questions about human environmental patterns, especially in reference to their evolution, were abandoned rather than resolved. Assumptions about cold protective functions of low surface area/body mass ratio are entrenched in physical anthropology, despite lack of experimental validation. Since heat loss is controlled by vasoregulation and tissue insulation, a simple physics model of SA:mass may not apply. The issue merits investigation, as do the assumed thermal advantages of foreshortened extremities. Physiological assessment remains our primary research tool. In cold climate natives, elevated basal metabolic rates now appear to be genetically induced. During cold exposure, the body manages heat conservation through well known channels but also by specialized thermogenic functions such as metabolism in brown adipose tissue (BAT). The powerful protective capacity of BAT is largely unexplored either within or between populations of cold exposed human adults. An irony of our profession is that many biological variables seem to have minor effects when compared to behavioral cold protections. This is partly because biological anthropologists may have made incorrect assumptions about what most threatens the well being of cold climate people. Contrasts in environmental behaviors when comparing northern cultures such as Inuit, Athabaskan, and Norse are particularly instructive. Adaptations to life in the cold may ultimately reveal their secrets through biocultural research design modeling of environmental research. With both practical and theoretical gains still wide open, the eld needs renewed attention from human biology. Am. J. Hum. Biol. 19:218227, 2007. ' 2007 Wiley-Liss, Inc.

The purpose of this paper is to consider human cold adaptation as a research area within human biology. I will particularly evaluate the progress we have made, outline some of our operating hypotheses, and note challenges that have been abandoned in mid pursuit. Let me open with the most obvious question. Why would we want to study cold adaptation, now, on the frontal edge of global warming? There are at least two reasons. First, it is always cold somewhere within the human domain. Even in areas that are currently tropical, cold can periodically be a real stressor. High latitude areas are going to experience winter cold regardless of higher global temperatures. But there is also a deeper concern. Global warming may disrupt the circulation of warm and cold water in the North Atlantic Ocean. If that happens, the gulf stream that warms the UK and northwestern Europe may also fail and the results would be a climatic disaster. Some indicators suggest that this has already begun (Kerr, 2005). As human biologists we should

reexamine our historic interest in adaptation to physical environment, both hot and cold, and view this as an emerging research arena of great practical importance. One original research agenda within human biology sought answers to a fundamental question. Are some of the biogeographic variations within our species environmentally adaptive? We made some progress with that issue but then largely abandoned the question. Beginning with the early work of the 1950s, anthropologists developed unique understandings of biobehavioral adaptation. These involved thermal, metabolic, and cultural defenses, one example

2006 Raymond Pearl Memorial Lecture. *Correspondence to: A. Theodore Steegmann, Jr., Department of Anthropology, 380 MFAC, University at Buffalo, Buffalo, NY 14261, USA. E-mail: atsjr@buffalo.edu Received 13 November 2006; Accepted 13 November 2006 Published online in Wiley InterScience (www.interscience. wiley.com). DOI 10.1002/ajhb.20614

C V 2007

Wiley-Liss, Inc.

HUMAN COLD ADAPTATION

219

of which I offer from my own eld experience. Cree/Ojibwa shermen use gill nets for winter lake shing. Getting sh out of the nets is quite difcult when the wind and cold numb ones hands almost instantly. Their traditional solution is to grip the sh snout with the front teeth and use both hands to strip the sh clear. The process is fast, and my interpretation is that it increases efciency and reduces hand exposure to the cold. This is a classic biobehavioral action of self-evident value. To the benet of anthropologists, environmental physiologists paralleled our efforts with extensive investigations of thermal response, many in relation to morphological variation. For instance, Wade and Veghte (1977) demonstrated the difference in heat loss patterns in swimmers, comparing thermographic images of volunteers before and after exercise in cool water. Surface heat loss depended in part on the activity level and mass of underlying muscle. Clearly, behavior, bodily insulation, and type of activity will have to be accounted for in understanding human adaptive response to climate. WHY DID HUMAN BIOLOGY ABANDON ENVIRONMENTAL RESEARCH? Early thermal physiology laboratory studies were effective in dening physiological mechanisms, but operated pretty much in the one causeone effect mode. This is expected and necessary when the objective is to run carefully controlled experiments. By the early 1960s, eld studies showed us that multiple intrinsic and extrinsic factors operate to determine environmental response, and they do not all t within physics style models. While our capacity to measure these complex independent variables in the eld was and is still limited, the eld work stimulated construction of complex, biobehavioral, causal models. We grew better at predicting what to include in research designs (Thomas, 1998). This has been a crowning achievement of our eld and continues to inuence the way we think. We had moved from simplistic to more realistic research structure, but our own basic assumptions continued to present an obstacle. My personal epiphany came during a winter spent with subarctic Algonkian trappers. Despite everyday activities going on at extremely low temperatures, cold simply didnt seem to be much of a problem to them, based on what they said and how they acted. I was not then sure whether cold was really a minor issue,

and had been inated by our own romanticism, or that adaptations were running so smoothly that they were mostly invisible to an outsider. Perhaps it was some of each. Before that got sorted out, human biology underwent an agenda shift and old challenges were largely laid aside. Partly due to internal dynamics and also as a result of national trends in biomedical sciences, human biologists began to turn their attention to health, nutrition, and molecular biology. My memory puts this at about the mid 1970s. As an example of national inuences on our eld, the Framingham study was beginning to raise awareness in both the sciences and in ordinary life. Medicine had awakened to the impact of behavioral or lifestyle factors on health. Diet, smoking, and alcohol were increasingly seen as causes of early morbidity and mortality (Dawber, 1980). I have identied ve questions for commentary in this paper, mostly focused on cold defense. These are the abandoned topics. 1. Does low surface area per unit of weight increase heat conservation in the cold? 2. How important is muscle tissue as an insulator? 3. Have we accurately portrayed the bodys thermal core? 4. Do body proportions affect thermoregulation? 5. Do thermally protective behaviors make special biological adaptations unnecessary? CONCEPTUAL ISSUES Our general theoretical premise is that biological variation produces signicant functional outcomes. This is a core concept in human biology and in general biology as well. In the present context we are examining evidence for traits that confer cold protection to humans. Is a signicant functional outcome an adaptation? There is no shortage of argument about concepts of adaptation, but for present purposes, a signicant functional outcome is the denition of adaptation. Signicance needs to be demonstrated for either evolutionary or social structures. Cast in terms of natural selection, functional outcomes, such as manual dexterity or core heat conservation, would not be signicant unless they could be shown to induce differences in fertility or mortality. To my knowledge this has never been studied directly. However, some of the great

American Journal of Human Biology DOI 10.1002/ajhb

220

A.T. STEEGMANN TABLE 1. Correlations to core temperature

speciation events in the hominid lineage are now being linked to periods of cold and dryness (Bobe et al., 2002). Cold related fertility and mortality are problems awaiting investigation within human biology. A second approach, not evolutionary in structure, asks whether independent variables produce differences in productivity or life quality. These outcomes may be caused by proximate factors, whether environmental, biological, or behavioral. The advantage to human biology in models like this is that many elements are directly measurable. A further virtue of this view is that outcomes are usually easy to understand. They are the stuff of daily life. Our research assumption that variables relate to signicant outcomes must be tested. Here are some cautionary tales from my historical work in Northern Ontario. The rst case describes an infant born with nonfunctioning legs. Nevertheless, he grew into a man who was a successful trapper and fathered a large family. A second story relates how one of the best hunters in the community lost his arm due to a gunshot injury. After arm amputation and recovery, he continued to be one of the top hunters in the band. It is hard to imagine biological variables that outrank lack of arm or leg function in such a challenging environment, and yet the functional consequences were negligible. Consequently, we must be mindful of two longstanding tests for adaptiveness. It is not sufcient to show that a response occurs, or that individuals or populations differ in their responses, but rather that some benet is conferred. Necessity is an even more compelling issue. Regardless of demonstrated benet, does one really need it to function adequately? (Mazess, 1975). One of the reasons for decline in environmental studies within human biology was a failure to pick up Mazess challenge.

loss per minute in cold water

r Subscapular skinfold SA/mass Mass Beta Subscapular skinfold Surface area 0.640 0.296 0.809 0.753 0.701 Sig. 0.000 0.031 R 0.809 0.845 r2 0.654 0.576 0.491 R2 0.654 0.741

Data presented by Sloan and Keating (1973) are reanalyzed to address questions in this paper. Both data sets examine relationships between core temperature loss per minute and morphological variables. The rst set shows product moment correlations with morphological variables, and the second gives values from a backward stepwise regression using the same variables.

NEGLECTED PROBLEMS Several interesting biological problems within human environmental adaptation remain to be resolved. Bergmanns rule and cold The geographic relationship between greater body mass and cold within a species range has generally received an adaptive interpretation. It is assumed that reduction of body surface

area per unit of mass acts to conserve body heat. We should question this assumption, partly because the body does not respond to cold stress in a simple, unitary fashion. One of the most interesting experiments to test surface area/mass effects on cold response was reported by Sloan and Keatinge (1973). Their sample of 28 boys and girls swam in 238C (748F) water for up to 40 min. This is a useful design for several reasons. First, cold doubtless acted selectively on children in the past, and they are consequently important research subjects. I suspect that most cold deaths result from falls into cold water. Further, exercise in cold water rapidly drains heat from the skin, compared to air, and swimming is a somewhat natural activity. Finally, the sample is marginally large enough for statistical analysis, in contrast to most experimental samples within environmental physiology; the authors also published individual data, including an important measure of cold resistanceloss of core temperature per minute. From a reanalysis of their data, the following morphological effects are evident. The correlation between surface area and weight is 0.987, between SA/mass ratio and weight is 0.928, and between SA/mass and subscapular skinfold is 0.713. In other words, anthropometric variables are so highly correlated that it is difcult to partition their effects on cold response. Table 1 shows correlations to core temperature loss per minute. Using the data in multiple regression analysis, skinfold explained 65.4% of the temperature loss variance, and surface area added an additional 6%. However, SA/mass did not enter the equa-

American Journal of Human Biology DOI 10.1002/ajhb

HUMAN COLD ADAPTATION

221

Fig. 1. Colors on the two gures approximate tissue temperatures when exposed to high and low ambient temperature conditions, red being hot and blue cool. Note particularly that in cold, the warm area withdraws to the body core, protected by a cooled shell. Consequently, estimation of the thermal surface area is inexact and will vary widely depending on vasoconstriction and body composition.

tion; in a situation like this it should have, had it been of real importance. Several studies of morphological cold defense have made assumptions about the SA/mass ratio but without directly measuring its inuence. McArdle et al. (1984) reported a comparison of men and women of contrasting body fat levels, immersed for an hour in 208C (688F) water. Lean women had twice the amount of body fat as lean men yet showed equivalent core temperature loss. Body fat advantages in

women may have been offset by their higher SA/mass and lower muscle mass, but the effect could not be measured directly because of small sample sizes. Toner et al. (1986) compared cooling responses in two groups of men of contrasting mass, but matched for percent body fat. Despite differences in SA/mass, large and small men showed equivalent core temperature and metabolic patterns. They concluded that surface area/mass . . . has a minimal effect on heat transfer in water, within a similar subject

American Journal of Human Biology DOI 10.1002/ajhb

222

A.T. STEEGMANN TABLE 2. Factors affecting body temperature in heat Body mass (%) 1-h exercise relative to capacity Hot dry 10 Warm humid 18 1-h exercise set rate Hot dry 40 Warm humid 40 VO2 max/kg (%) Residual (%)

population. Quoting results of another study (McArdle et al., 1984) they concluded that the theoretical possibility of [SA/mass] as an important factor in thermal responses was therefore abandoned. These experiments were repeated, but in colder water, and the results were essentially the same (Glickman-Weiss et al., 1993). A summary of our understanding of the role of SA/mass in cold defense was presented in a comprehensive review of morphology and thermal response. Because of confounding effects of increased mass, in individuals with increased surface area, it is difcult to nd conclusive evidence for the exact nature of the relationship between body surface and the degree of cold strain and rate of core cooling in humans. In fact, there is a growing body of research that suggests that, during periods of cold water immersion, differences in SA/m have a minimal effect on heat transfer in similar populations (Anderson, 1999). The bottom line is that cold-protective effects of low SA/mass are not clearly evident physiologically. Not only is surface area closely related to mass, but the body core surface area is smaller than the total body surface area during cold exposure, and confounds functional estimation of SA/mass. Cold causes the body shell (supercial tissues) to vasoconstrict and cool. Further body core protection is provided by subcutaneous adipose tissue and muscle. The extremities are cooled by vasoconstriction as well as by counter current heat exchange. As shown in the cold model of Figure 1, the core body mass that loses heat from its surface is actually smaller than the full body shell surface area. Consequently, estimates of surface area/mass during cold exposure are of questionable value because of individual variation in insulation. Bergmanns rule and heat In contrast to cold, elevated ambient temperatures cause the entire body surface to become hot to rid itself of heat load. One would then assume that the physics model of SA/mass heat loss would be veried through physiological experimentation. That is, one should be at heat advantage with high surface area for mass. It turns out that unloading heat at high ambient temperature is physiologically complex. Havenith et al. (1998) tested 2430 men and women, exercising for 30 min each in two environments. Warm humid conditions were 358C (958F) and 80% relative humidity, and hot dry

33 11 19

90 49 49 41

Values are percentages of explained variance in core body temperatures. The minus sign indicates an inverse relationship. Independent (intrinsic) variables are VO2 max/kg body weight and body weight (mass). Residual is the variance not explained by the independent variables. Data are from Havenith et al. (1998).

was 458C (1138F), 20% RH. One pair of tests was at a set pace of 60 W work and the other was based on a rate relative to the individuals VO2 max. Some of their results are shown in Table 2. Since mass and aerobic capacity are related, the authors used a relative measuremaximum oxygen uptake per kilogram of body weight. In both hot dry and warm humid conditions (absolute exercise rate), core temperature was lower as body mass was higher, and less so, as VO2 max increases. A larger person with higher aerobic capacity has the advantage, probably due to greater sweat production. Where exercise is adjusted as a percentage of VO2 max (equivalent to natural self-pacing), lower core temperature is associated with higher body weight, as before, but higher tness increases core temperature in humid conditions where more sweat is less helpful. Under all of these conditions, mass is more important than SA/mass. Though smaller subjects had higher SA/mass, they also had higher core temperatures, possibly due to less muscle related sweat production. Havenith et al. (1998) drew these conclusions: A. At a xed exercise rate in the heat, high body mass and high aerobic capacity relate to lower core temperature. B. At an exercise rate adjusted as a percentage of aerobic capacity, high body mass and low aerobic capacity go with lower core temperature. C. In all conditions, high body mass was an advantage, and SA/mass behaved almost identically to mass as a determiner of deep body temperature. Smaller subjects had higher core temperatures. If we use these conclusions to interpret thermal adaptation in human evolution, the increase in body size in Homo ergaster and

American Journal of Human Biology DOI 10.1002/ajhb

HUMAN COLD ADAPTATION

223

Homo erectus may have helped control core temperatures during more active foraging on the open plains. By that logic, the large, linear bodies of the African Tutsi peoples may have had little to do with increasing SA/mass, but were instead an adaptation for large body size/ cooler core temperatures in the heat. Relative surface area may have little adaptive role. Body core/body shell While anthropologists have generally accepted the thermal importance of body surface area, that may be in part because the role of the adipose-muscle shell has slipped from view. My old colleague at Buffalo, Don Rennie, studied that problem for most of his career and left us with these interesting insights (1988). A. The skin/adipose tissue/muscle shell is the primary site of cold resistance. B. Thermal conduction across the shell may increase sevenfold with muscle perfusion produced by shivering or exercise, so that insulation declines. C. Fat insulates better than muscle per unit of thickness. However, in a t person, muscle layers are usually much thicker than subcutaneous fat and consequently have higher absolute insulative value. In Rennies experiments, 7074% of total insulation was from nonfat tissues. D. Real world complexity derives from insulation disruption due to vasoconstriction/dilation and exercise metabolism. The body is in constant thermal ux brought on by changes in activity. The message for human biologists is to account for the importance of both muscle and fat as insulators. Foragers today are pretty lean, and that was probably true in most settings in the human past. Rode and Shephard (1971) and Shephard et al. (1973) offer data on recent circumpolar populations. The Inuit, for example, traditionally had high muscle mass and high work capacity, but low body fat. The design of realistic eld experiments must include measures of body composition, activity, and metabolic level, and whether pacing is controlled by the researcher or by the subject. Allens rule and cold Human populations with long histories in cold climates have foreshortened extremities, especially the lower leg (Ruff, 1994). This expression of Allens rule has been interpreted

as a mechanism by which body heat is conserved, as was proposed for low SA/mass. Were that true, the distal extremities would have to be sites of substantial heat loss. To assess this radiator model, we must estimate how much heat is actually lost from those sites. Generally, heat ow is blocked by muscle, subcutaneous fat, and vasoconstriction. However, circulation through the head and neck is fairly unrestricted in the cold, producing heat loss, and the opposite is true of the hands and feet due to vasoconstriction and lack of major muscle and fat masses. Wade et al. (1979) estimated heat loss from different body areas during cold water immersion. Each body section was measured for surface area and its heat ow gauged empirically, to gain estimates of heat loss by area. During 30 min in 25.28C (77.28F) water, average heat loss from the upper torso was 24 W, from the head 17, and from the calf 17. In contrast, the hand and foot each lost 1.5 W. In other words, the distal extremities were so vasoconstricted that they were close to water temperature and were a minor source of body heat loss. One aspect of the complexity of this issue is growth plasticity of the lower leg. Bogin et al. (2002) report on growth of children of Mayan immigrants to the U.S. found children growing up in the new country were larger than their Central American peers, but that leg length increased relatively more than stature. Stinsons review (2000) stated that this is generally seen in secular increases. Based on secular trends in U.S. skeletal populations, Jantz and Jantz (1999) noted that change is greatest in males, lower limbs change more than upper and the greatest proportional change is in distal segments. While past growth conditions should be assessed for those participating in environmental experiments, even under equivalent growth environments population contrasts in body proportions persist. For example, Steegmann (2005) found relative leg length was greater in American soldiers of African ancestry than in those of European origins, even though the samples were of nearly identical stature. Could this reect ancestral genetic adaptation to environment? I am suggesting that reduction of extremity length relative to trunk height makes little difference in heat conservation. It is unlikely that we will ever be in a position to directly examine the action of natural selection by cold on human extremity proportion. However, by morphological and physiological measurement, we could determine experimentally

American Journal of Human Biology DOI 10.1002/ajhb

224

A.T. STEEGMANN

whether those with shorter extremities maintain warmer toe temperatures (indirectly determining potential for cold injury) and whether deep body temperature is inuenced by extremity length. With global warming, heat mortality is on the increase. As part of a larger study on intrinsic factors in heat defense, body proportions and composition should be evaluated along with age, poverty, developmental and current nutrition, and the immediate human made environment. Summary of body morphology and thermal response A. During cold exposure, vasoconstriction and counter-current heat exchange combine to keep the distal extremities so cold that little heat escapes, regardless of length or surface area differences. B. I suspect that length and diameter of the trunk are the body areas most involved in cold defense. For instance, Ruff (1994) found a correlation of 0.886 between pelvic diameter and latitude and that needs to be explained. C. A thermal adaptation interpretation of Bergmanns and Allens rules for humans has been accepted within biological anthropology and is now part of our folklore. D. Richard Mazess suggested that adaptations pass muster only when they have been proven to be both benecial and necessary for survival and function. E. To my knowledge, we have done almost no empirical testing of the advantages of these morphological patterns, in either laboratory or eld. It has seemed sufcient to simply demonstrate their presence. F. While my skepticism may seem a contrary to general positions, it is really a call for investigation. In truth, we still know little about body size and proportion as factors in thermal defense. But as the astronomer Carl Sagan noted, Absence of evidence is not evidence of absence. This problem needs work.

tioned in different geographic zones. We may assume that these were strictly induced by local conditions. Intake was about 3,400 kcal/d in the tropics, 3,800 in temperate climates, and about 4,500 in subarctic/arctic settings. Basal metabolic rate Leonard and his colleagues have conrmed long standing recognition that cold climate natives show elevated basal metabolic rates. Data on native Siberians are in agreement with those of several other high latitude groups. Further, the researchers were able to discount effects of testing anxiety, diet, and lifestyle as causative factors. BMR is elevated 1619%. Morphologically, fat free mass was a better predictor than SA/mass (Leonard et al., 2002, 2005; Snodgrass et al., 2005). In the end, we may assume that genetic causation is the simplest explanation. This is a persuasive case of morphological and physiological cold adaptation. Functional advantages of these patterns, whether in economic status, health, or reproduction, await clarication. BAT The role of brown adipose tissue (BAT) in environmental adaptation is more speculative and virtually unexplored anthropologically. BAT, as the only human tissue dedicated solely to heat production, is a compelling problem in human adaptation, particularly in areas where people live in extreme cold. Infants, with their immature thermoregulation and small muscle mass, have key body functions thermally protected by BAT. Small deposits are distributed over the trunk and neck somewhat like a vest (Lean and James, 1986). BAT produces much more heat than other tissues, per volume, so the total amount may be small and still be very effective. It normally disappears after infancy. Less well recognized is the capacity of adults to either maintain or regenerate BAT if sufciently exposed to cold. As a metabolic adaptation, it is energetically expensive. In other words, this modest tissue may have big benets as well as costs, and could be more important functionally than many other traits that have received so much attention. Age and decline in thermal resistance Sometimes change throws light on an adaptation that may otherwise be hard to understand. As an example, consider aging as a cause of the decline in human capacities to

PROGRESS IN ENVIRONMENTAL HUMAN BIOLOGY Not all aspects of human biology and the physical environment have been neglected. There has been real progress in both nutrition and energetics. These issues are not new. Johnson and Kark (1947) published data on voluntary caloric intake of American soldiers sta-

American Journal of Human Biology DOI 10.1002/ajhb

HUMAN COLD ADAPTATION

225

resist cold. Using cool intravenous uid to lower the body temperature, Frank et al. (2000) tested samples of older and younger men for physiological response. The younger men showed more nger vasoconstrictiona protective reex intended to reduce general body heat loss. As vasoconstriction increased, the younger sample also produced more body heat through a steeper rise in metabolic rate. Consequently, the older men suffered a relatively greater decline in deep body temperaturea classic sign of thermoregulatory inadequacy. This pattern is of particular importance to anthropologists as we turn toward more applied work in our own societies. Demographic trends, especially in Europe, indicate a general increase in the percentages of populations over 65 years old. Cold is hardest on those without the economic resources to buy high quality food, adequate protective clothing, and sufcient home heating. These are problems that respond well to biocultural research design and ultimately, lead to policy advice on intervention. COLD MORBIDITY AND MORTALITY Let us now return to the Mazess injunction: Biological variation is not adaptive unless we can show that it makes a difference in the real world. There is little argument against using premature death as a marker of ultimate adaptive failure. Morbiditya common topic in human biology researchis a pretty close second. We assume that differential mortality was, in the past, responsible for the evolution of mammalian cold adaptation. Human cold adaptation is largely an expression of those longstanding systems, though there may have been periods when our lineage was subjected to new levels of cold exposure. Paleoanthropologists have implicated shifts to colder climate as factors in Pliocene-Pleistocene hominid speciation (Bobe et al., 2002). Populations of Homo ergaster and Homo erectus were living in open habitats with their often low nighttime temperatures and mid-day heat. We may assume these environments were thermally selective, but direct evidence is lacking. However, an oblique approach is offered by modern military morbidity data reported by Rav-Acha et al. (2004). Israel lies between 31 and 33 degrees north latitude, with Mediterranean climate but cool winters. These are probably the conditions experienced by many early hominids. Rav-Acha deter-

TABLE 3. The excess mortality index for the countries listed indicates the percentage by which deaths in cold months exceed those in other monthsa Portugal Ireland Spain Greece UK Italy Austria France Belgium Luxembourg Denmark Holland Germany Finland
a

28 21 21 18 18 16 14 13 13 12 12 11 11 10

Data are from Healey (2003).

mined that Israeli military cold injuries were mainly due to forced inactivity at low temperaturesa common military risk. Hypothermia occurred as a training accident among recruits who experienced wet clothing, no protective gear, and 24-h exposure to wind chill above 08C. In contrast, veteran soldiers sustained peripheral cold injuries, mostly to the feet, during combat operations below 08C windchill and with ongoing exposure for over 48 h. The point is that even in relatively warm climates, cold can cause disabling injuries and potentially, even hypothermia deaths. Regardless of historical analogy, cold continues to be a threat today. There were no deaths in the Israeli sample but the same was not true in Europe. Healey (2003) has shown that European cold can be lethal but the reasons are complex. Values on Table 3 represent the percentage by which mortality during the four coldest months exceeds that of other months. The 28% excess in Portugal, compared to 10% in Finland is counter-intuitive. However, empirically higher cold season death levels related to warmer climate, inadequate home heating, low per capita gross national product, and a shortage of hospital services. Other reports indicate that those in warmer climates dress less warmly and allow dwelling temperatures to fall in colder weather. In other words, contemporary cold injury has sufciently strong behavioral precursors to merit separate treatment. BEHAVIORAL COLD RESISTANCE I will begin this section with a model of contrasting strategies for cold defensethe Eskimo and the Kutchin Athabaskans from Alaska. Interpretations were drawn by anthropologist

American Journal of Human Biology DOI 10.1002/ajhb

226

A.T. STEEGMANN

Nelson (1973) who conducted extensive eldwork with both groups, and I include here only two examples out of many he describes. If a party gets lost during very cold weather, Kutchin hunters keep moving, fearing that a rest stop would lead to sleep and death by freezing. In stark contrast, the Eskimo keep moving just enough to stay warm and avoid exhaustion by resting as needed. A second perception by Nelson was that the Eskimo studied weather prediction more carefully than the Kutchin, and used a body of shared and accurate knowledge to better avoid dangerous situations. In both cases, Eskimos seem to practice higher survival skills and both behaviors are strongly directed by cultural traditions. A second casethe extinction of the Greenlandic Norseis a spectacular example of culturally based failure, driven by climatic cooling and adverse changes in access to the European economy. The Norse colony in Greenland was founded in 986 A.D. during the Medieval Warming Period and reached its peak of about 2,000 people by 1300 A.D. The Little Ice Age began about that time, Norse Inuit conicts started, and the West Settlement was lost by 1350. After 1410 there is no record of contact with the Norse colonists. Most scholars agree that extinction came by about 1450 (Norlund, 1936). There is disagreement on causes of the Norse extinction. It was doubtless due to multiple factors acting in concert (Barlow et al., 1997). The Greenland Norse lived by animal husbandry. Both cooling and overpopulation led to overgrazing and then a decline in domestic herds. At the same time, climate change brought sea ice southward along the coasts. Since the Norse partly depended on imported technology and trade, blockage of the sea lanes and a decline in demand for hides and ivory damaged their adaptive stance. With cooling and ice came Inuit from the northbetter adapted to collecting wild resources and willing to ght competitors. The Norse church resisted adoption of Inuit clothing and equipment as heathen while continuing to drain diminishing resources itself. In the end, the Norse turned to seal hunting but lacked the skills and equipment to hunt on the ice. There is no historical evidence they returned to Iceland or Norway (McGovern, 2000). Rather than being a failure to resist cold, this represents a shortfall in handling a new environment created by the cold. One subtext to this drama does have to do with thermal behavior. The Norse apparently continued to wear fashionable European clothing to the

very end (Arneborg, 2000). Perhaps it was as adaptive to this environment as what the Inuit wore, since Europeans had dealt with cold for a long time. But this is an empirical question, representing one of many opportunities available for anthropological research on the cold. SUMMARY AND CONCLUSION This paper has reviewed two areas within environmental human biology. The rst constitutes a set of intellectual conclusions that we have not pursued fully yet persist in accepting. A second set of points covers progress in the area and future research opportunities. The current status of abandoned problems A. If humans resist cold by having low surface area per unit of body weight, that has not been demonstrated physiologically. Mass alone is probably the key factor in both hot and cold adaptation. Bergmann actually described climatic regularities in body mass, not SA/mass (Mayr, 1963). B. The role of muscle as an insulator was just emerging when human biology largely dropped out of thermal research. Muscle is an excellent insulator but only when inactive. This presents a classic biobehavioral research design challenge involving technology, morphology, physiology, and behavior. Since our mid-Pleistocene ancestors were pretty muscular and in most circumstances males burned up too much energy to maintain subcutaneous fat thickness of real insulative value, this is an issue of both physiological and historical interest. C. Third, we have accepted the heat conservation advantage of foreshortened extremities uncritically. Because of peripheral vasoconstriction as a basic cold response, little heat is lost from the distal extremities, regardless of body proportion differences within our species. The problem goes virtually unexplored. Arenas of progress A. Convincing evidence supports the presence of genetically induced elevation of the basal metabolic rate in many circumpolar populations. This may be a real adaptation if it is actually necessary. B. Secondly and possibly related to BMR, there appear to be cold resistance advantages to BAT, in both infants and adults. This has scarcely been touched anthropologically.

American Journal of Human Biology DOI 10.1002/ajhb

HUMAN COLD ADAPTATION

227

C. Elders, at least in industrial societies, suffer decline in physiological cold resistance. Both mortality and life quality are affected. D. Cold mortality is still an active threat, but can be buffered behaviorally. It is highest in warmer areas of Europe and possibly in other subtropical areas. E. Finally, not all cultures are equally adept at cold protective behaviors or technology. In the Greenland case, Norse failures caused their entire population to vanish. This paper has addressed only a few key questions. Answers elude us, only partly because of lack of investigation. We do not expect this to be easy, but that is nothing new. In short, the relation between organism and environment is everywhere and always mutually reciprocal, and as man is the most highly differentiated and manifoldly diverse in his capabilities of all organisms, so also is his effective environment the most complicated (Pearl, 1939). LITERATURE CITED
Anderson GS. 1999. Human morphology and temperature regulation. Int J Biometeorol 43:99109. Arneborg J. 2000. Greenland and Europe. In: Fitzhugh WW, Ward EI, editors. Vikings: the Norse Atlantic saga. Washington: Smithsonian. p 304318. Barlow LK, Sadler JP, Ogilvie AEJ, Buckland PC, Amorosi T, Ingimundarson JH, Skidmore P, Dugmore AJ, McGovern TH. 1997. Interdisciplinary investigations of the end of Norse Western settlement in Greenland. Holocene 7:489499. Bobe R, Behrensmeyer AK, Chapman RE. 2002. Faunal change, environmental variability and late Pliocene hominin evolution. J Hum Evol 42:475497. Bogin B, Smith P, Orden AB, Varela Silva MI, Loucky J. 2002. Rapid change in height and body proportions of Mayan American children. Am J Hum Biol 14:753761. Dawber TR. 1980. The Framingham study: the epidemiology of atherosclerotic disease. Cambridge: Harvard University Press. Frank SM, Raja SN, Bulcao C, Goldstein DS. 2000. Agerelated thermoregulatory differences during core cooling in humans. Am J Physiol 279:R349R354. Glickman-Weiss EL, Nelson AG, Hearon CM, Goss FL, Robertson RJ, Cassinelli DA. 1993. Effects of body morphology and mass on thermal responses to cold water: revisited. Eur J Appl Physiol 66:299303. Havenith G, Coenen JML, Kistemaker L, Kenney WL. 1998. Relevance of individual characteristics for human heat response is dependent on exercise intensity and climate type. Eur J Appl Physiol 77:231241. Healey JD. 2003. Excess winter mortality in Europe: a cross country analysis identifying key risk factors. J Epidemiol Community Health 57:784789. Jantz LM, Jantz RL. 1999. Secular change in long bone length and proportion in the United States, 18001970. Am J Phys Anthropol 110:5767.

Johnson RE, Kark RM. 1947. Environment and food intake in men. Science 105:378379. Kerr RA. 2005. The Atlantic conveyor may have slowed, but dont panic yet. Science 310:14031404. Lean MEJ, James WPT. 1986. Brown adipose tissue in man. In: Trayhurn P, Nicholls DG, editors. Brown adipose tissue. London: Edward Arnold. p 339365. Leonard WR, Snodgrass JJ, Sorensen MV. 2005. Metabolic adaptation in indigenous Siberian populations. Ann Rev Anthropol 34:451471. Leonard WR, Sorensen MV, Galloway VA, Spencer GJ, Osipova L, Spitsyn VA. 2002. Climatic inuences on basal metabolic rate among circumpolar populations. Am J Hum Biol 14:609620. Mayr E. 1963. Animal species and evolution. Cambridge: Belknap. Mazess RB. 1975. Human adaptation to high altitude. In: Damon A, editor. Physiological anthropology. New York: Oxford University Press. p 167209. McArdle WD, Magel JR, Gergley TJ, Spina RJ, Toner MM. 1984. Thermal adjustment to cold-water exposure in resting men and women. J Appl Physiol 56:15651571. McGovern TH. 2000. The demise of Norse Greenland. In: Fitzhugh WW, Ward EI, editors. Vikings: the Norse Atlantic saga. Washington, DC: Smithsonian. p 327339. Nelson RK. 1973. Hunters of the Northern Forest. Chicago: University of Chicago Press. Norlund P. 1936. Viking settlers in Greenland and their descendants during ve hundred years. London: Cambridge University Press. Reprinted 1971, Kraus Reprint, New York. Pearl R. 1939. The natural history of population. London: Oxford University Press. Rav-Acha M, Heled Y, Moran DS. 2004. Cold injuries among Israeli soldiers operating and training in a semiarid zone: A 10-year review. Military Med 169:702706. Rennie DW. 1988. Tissue heat transfer in water: Lessons from the Korean divers. Med Sci Sports Exerc 20 (Suppl): S177S184. Rode A, Shephard RJ. 1971 Cardiorespiratory tness in an arctic community. J Appl Physiol 31:519526. Ruff CB. 1994 Morphological adaptations to climate in modern and fossil hominids. Yrbk Phys Anthropol 37:65107. Shephard RJ, Hatcher J, Rode A. 1973. On body composition of the Eskimo. Eur J Appl Physiol 30:113. Sloan REG, Keatinge WR. 1973. Cooling rates of young people swimming in cold water. J Appl Physiol 35:371375. Snodgrass JJ, Leonard WR, Tarskaia LA, Alekseev VP, Kirvoshapkin VG. 2005. Basal metabolic rate in the Yakut (Sakha) of Siberia. Am J Hum Biol 17:155172. Steegmann AT Jr. 2005. Climate, racial category and body proportions in the U.S. Am J Hum Biol 7:393402. Stinson S. 2000. Growth variation: Biological and cultural factors. In: Stinson S, Bogin B, Huss-Ashmore R, ORourke D, editors. Human biology: an evolutionary and biocultural perspective. New York: Wiley-Liss. p 425463. Thomas RB. 1998. The evolution of human adaptability paradigms: Toward a biology of poverty. In: Goodman AH, Leatherman TL, editors. Building a new biocultural synthesis. Ann Arbor: Michigan University Press. p 4373. Toner MM, Sawka MN, Foley ME, Pandolf KB. 1986. Effects of body mass and morphology on thermal responses in water. J Appl Physiol 60:221225. Wade CE, Dacanay S, Smith RM. 1979. Regional heat loss in resting man during immersion in 25.2oC water. Aviat Space Environ Med 50:590593. Wade CE, Veghte JH. 1977. Thermographic evaluation of the relative heat loss by area in man after swimming. Aviat Space Environ Med 48:1618.

American Journal of Human Biology DOI 10.1002/ajhb