Local Cold Exposure of Andean Indians during Normal and Simulated Activities R. E. JONES,' M. A. LITTLE,' R. B . THOMAS,Z C . J. HOFF.3 AND D. L. D U F O U R ' 1 Department of Anthropology, Stcctr University of N e w York, Binghcimton, NZW York 13901; 2 Department of Anthropology, Cornell University, Ithnccc, Nt'w York 14850; 3 Department of Anthropology, University of Oregon, Eugene, Oregon 97403

KEY W O R D S Extremity temperatures Cold stress . Andean Indians . Sex differences . Normal activities. ABSTRACT Thermal responses of Andean Indians were measured during several customary tasks associated with cold exposure in the highlands of southern Peru. These included surface temperature measurements of women while they washed clothing in the river and similar measurements of men while they constructed a diversion channel i n the same river. A third test measured the effects of alcohol consumption on body temperatures during light activity. Women maintained slightly warmer hand than foot temperatures. Men maintained nearly equal hand and foot temperatures during the exposure period. Among male subjects the foot rewarmed at a faster rate than the hand. The results from the field studies compared favorably with results from earlier laboratory exposure tests. Comparisons between the river water exposure tests for males and females showed a consistent pattern where females maintained warmer hand and foot temperatures than males. These findings were in accord with previous laboratory studies among Quechua Indians and with the findings reported for other ethnic groups who experience natural cold stress. Alcohol ingestion appeared to have minimal effect in mitigating cold stress response during light activity. This finding was counter to earlier laboratory tests of resting subjects.

Much of the information that we have on variations in whole-body and extremity cold stress responses of highland Quechua Indians is derived from cold tolerance tests conducted in the laboratory (Baker et al., '67; Little, '69; Little et al., '71; Little et al., '73; Mazess and Larsen, '72). The advantages of standardized and controlled laboratory tests are quite clear: for example, different groups may be compared directly through statistical analysis, subjects can begin tests in equivalent states of thermal equilibrium, and extraneous stresses that contribute to unexplained variance can be eliminated. Despite the advantages of laboratory tests, several investigators have questioned the validity of such tests in that they frequently do not simulate conditions of natural cold exposure. It has also been suggested that Andean Indians, in fact, are not cold stressed because of the effectiveness of the Am. J. PHYS. ANTHROP.,44: 305414

insulative properties of their clothing and bedding materials (Hanna, '76). It is important, however, to distinguish between generalized (whole-body) and localized (face and extremities) cold stress in approaching these questions. A number of studies have been conducted to define the microenvironment as a measure of the degree of cultural buffering against cold. Baker ('66) measured body temperatures of men, women, and children at night during normal sleep and concluded that, although children may be slightly chilled during the initial hours, adults do not appear to be significantly cold stressed. These findings were verified in the laboratory by Mazess and Larsen ('72) who found that native bedding substantially dimishes cold stress and permits sleep in what otherwise would be intolerable conditions. Larsen ('73) has come to the same conclusion from 305

306

JONES, LITTLE. THOMAS, HOFF A N D D U F O U R

field studies of the microenvironment. In the only survey of highland Indians' body temperatures during normal daytime activities, Hanna ('70) found that head, trunk, and rectal temperatures were at reasonably comfortable levels but that extremity surface temperatures tended to be lower. There is considerable evidence, then, from both laboratory tests of cold responses and field observations and tests that generalized cold stress is minimal under what might be defined as average conditions. In contrast to the evidence from generalized cold exposure studies, the extremities of highland Indians are routinely exposed to low temperatures during normal activities. Moreover, during both whole-body (Elsner and Bolstad, '63; Baker et al., '67) and extremity (Little, '69; Little et al., '71; '73) exposures, Quechua Indians have consistently maintained high levels of blood flow to the limbs and correspondingly warmer extremity surface temperatures than whites. The physiological mechanisms that enable Indians to maintain elevated extremity temperatures during cold exposures have not been fully defined. It appears, however, that the microcirculation is somehow involved, such that blood flow to the surface of the hands and feet persists under conditions that would ordinarily produce a pronounced vasoconstriction in unacclimatized whites. Under conditions of moderate cold stress to the limbs (cold air), warmer surface temperatures result from reduced peripheral vascoconstriction. During exposure to a more severe stress (cold water), the more specialized circulatory response of cold-induced vasodilation (rapid rewarming) is activated. We still have little information about microenvironmental influences and the actual extent of extremity cold exposure experienced by Indians during day-to-day activities. One aspect of the present study was therefore designed to test the natural acute cold exposure responses of highland Indians during daily activities. These results were compared with those from prior laboratory studies of acute cold exposure. Two field tests were conducted: surface temperature measurements of women while they washed laundry in the river and similar measurements of men while they constructed a diversion channel in the same

river. A third test was designed to measure the effects of alcohol consumption on body temperatures during light activity. This was based upon earlier laboratory results that demonstrated a slight warming effect of alcohol on extremities in resting subjects who were cold stressed (Little, '70). METHODS

Subjects Studies were conducted at or adjacent to the high altitude laboratory in the altiplano town of Nuiioa, Peru (4,000 meters) during the months of July and August, 1972. All subjects were of Quechua Indian derivation and had been lifelong residents at high altitude. Measurements of size and body composition were not included for analysis because of the persistent lack of association found in previous studies between these measures and temperature responses of the hand and foot (Little, '69; Little et al., '71). The physical characteristics of the three groups of subjects did not differ significantly from larger samples of the population used in previous cold tests (Baker et al.,'67; Hanna, '70; Little et al., '71; '73). Many of the subjects had participated in earlier studies conducted at the Nuiioa laboratory and therefore were familiar with test procedures and equipment. Procedures Since tests were to be conducted in the Nufioa River, water temperatures were recorded over a three-day period in late July. Twenty-four-hour variations in river water temperature are illustrated in figure 1. The dry months of May, June, and July tend to be the coldest of the year; hence, river temperatures would be expected to be at min"C

I

"C

J I

2400

~

Olm

OBa)

12m

urn

mn

~

21M

TIME OF DAV. hr

Fig. 1 NuAoa River temperatures in late July during the dry season. Each point signifies one measu reme n t.

LOCAL COLD EXPOSURE OF ANDEAN INDIANS

imal levels when these measurements were made. Clothes-washing test. Nine women between the ages of 20 and 50 years (% = 34 yr) were tested while washing clothing in the river for a period of 60 minutes. Since the test began between 1O:OO A.M. and 1 1 : O O A . M . , subjects were at least three hours post prandial. Three skin temperature sites were monitored at 5-minute intervals throughout the test: the dorsal surfaces of the right and left hands and the left foot. For this and subsequent tests, temperatures were recorded with standardized thermistors read to the nearest 0.1 O C on a battery-operated telethermometer. Subjects were requested to wash their clothing in typical fashion. Clothes washing procedures were judged through questions and observations made on a large sample of women several days earlier. Subjects rested on the river bank for 15 minutes before beginning the test. The washing procedure involved first wading into the water to a depth of 5 to 25 cm, after which the clothes were placed in the water and walked upon to facilitate soaking and preliminary rinsing. Following washing with detergent and beating of the clothing against rocks, the articles were scrubbed with the hands and rinsed thoroughly in deeper water (to midcalf). The entire activity produced constant exposure of the feet to river water at about 10°C with intermittent exposure of the hands. Both hands and feet were continuously wet and subject to evaporative cooling in the dry ambient air. The clothes washing activity was characterized as moderate with periods of work punctuated by brief intervals for conversation. Channel-construction test. Ten men ranging in age from 22 to 48 years ( X = 34 yr) were tested for a period of 30 minutes while constructing a channel in the same river. Since the test began at between 1O:OO A . M . and 11:OO A . M . , subjects were at least three hours post prandial. Subjects also rested on the river bank for 15 minutes before beginning the work, and measurements were continued for a 15-minute recovery period at the end of the test. Skin temperatures were recorded at 5-minute intervals on the dorsal surfaces of the left hand and foot. The work activity was concerned with diverting the main stream of flow closer to

307

the river bank by constructing a wall about 1 m in height of river stones. The task was accomplished by the subjects picking up stones weighing as much as 15 kg to construct the channel wall. This work could be identified as moderate to heavy activity. During the course of the test, the legs and bare feet of the subjects were continuously exposed to cold water (about 10°C) at a depth of 25 to 40 cm, while the hands were again intermittently exposed. Two separate tests were performed with groups of five men each and the results were pooled for analysis. Alcohol-consumption test. The same ten men who participated in the channel-construction work were employed as subjects for the alcohol test. The test consisted of a 30-minute rest period during which subjects were seated, followed by 90 minutes of walking at a constant rate (3.0 to 3.5 I

Washing

I

I

I

I

1

I

I

I

I

I

I

I

I

I

I

-10

-5

0

5

10

15

20

25

30

35

40

45

50

55

M]

TIME, min Fig. 2 Skin temperatures of the dorsal surfaces of the right and left hands and left foot of Quechua Indian women ( n = 9) during clothes washing in the NuAoa River a t a water temperature of 1 0 ° C .

the women were right-handed, which, be- water continuously during the test while cause of the scrubbing pattern, may have the hand was only alternately submersed contributed to the slightly elevated right and withdrawn. hand dorsal temperatures. At no time Alcohol-consumption test during the one-hour test did any of the Figure 4 represents the results of oral women appear to be in a state of discomadministration of alcohol on rectal, chest, fort. hand, and foot temperatures during the Channel-construction test two-hour test period. Mean hand temperaHand and foot skin temperature responses ture was 3.0"C higher than foot temperato cold river water during the channel-build- ture at the onset of the test and dropped to ing activity are presented in figure 3. Men 9.8"C below the initial level during the first maintained nearly equal hand and foot tem- 90 minutes. During the last 30 minutes of peratures during the 30-minute activity the test, hand temperature rose slightly to period. Surface temperatures declined mark- a final value that was about 9.1"C below edly in the f i s t five minutes of the test. the initial resting level. Foot temperature During the 15-minute recovery period, the declined at a more moderate rate, showing left foot showed a greater rate of recovery a rise beginning at minute 80 of the test. at 0.8"C per minute than the left hand at Final foot temperature was only 2.9"C be0.6"C per minute. At the end of this period, low the initial resting level. the foot was about 2.5"C higher than the Activity alone did not appear to arrest hand. Although the difference was not sta- the drop in hand and foot temperatures, tistically significant, it is nevertheless inter- since limb temperatures were still falling esting since the foot remained in the cold after 40 minutes of walking (minute 70 of

LOCAL COLD EXPOSURE OF ANDEAN

NDIANS

309

0

C

35

0

30

0 4

Left Foot Dorsum Left Hand Dorsum River Water

25

20

15

10

5

I I

-_-

T---' I I I I

tR.d 1 -10

+

Construction

1

I

I

I

I

-5

0

5

lo

15

I 20

-- Rest I 25

I

30

I 35

I

I

40

45

TIME, min

Fig. 3 Skin temperatures of the dorsal surfaces of the left hand and foot of Quechua Indian men (n = 10)during channel construction in the Nuiioa River at a water temperature of 10°C.

the test). It is possible that the more gradual decline in foot temperature when compared with the hand was influenced by walking. These temperature declines, however, were halted during the period when alcohol was being consumed, suggesting a very slight effect of the drug. Chest surface temperature declined at steady rates throughout the test and did not appear to be influenced by either the activity of walking or by alcohol ingestion. Rectal temperature showed a slight, but statistically nonsignificant, rise between the eightieth and ninetieth minute.

avoidable. Exposure to cold of the hands and feet is associated with the subsistence tasks of herding and cultivation, house construction, travel and seasonal movements, and the supplying of daily needs. What are especially stressful are contacts of the hands and feet with cold water which acts both as a heat-sink and a medium for evaporative cooling. During the wet season, which extends from September to April, rain, hail and snow storms are commonplace. Indians of all ages are routinely exposed to wet-cold conditions by walking barefoot through light snow, slush and rainpools, and by the DISCUSSION need to ford swollen streams. The planting Natural conditions of wet-cold stress season, which coincides with the onset of Highland Quechua Indians appear to be the wet season, requires the performance of relatively well insulated from the cold tem- an activity that must be carried out under peratures of the altiplano through their use all weather conditions. Throughout the year of homespun wool clothing that covers the other tasks contribute to wet-cold exposure. trunk and much of the limbs. Hands and Women, for example, wash cooking pots and feet, however, are unprotected by cover- prepare a form of dried potato (chuiio) in ings of any kind since walking barefoot or cold water. Another form of potato (morwith rubber-tire sandals is customary. For aya) must be dried for storage by a process this reason and because of the Andean way whereby the potatoes are first left to soak of life, extremity exposure to cold is un- in a stream, an elaborate activity generally

310

JONES, LITTLE, THOMAS, HOFF AND DUFOUR

"C Rectal

36

I

I

I

I

I

I

I

I

I

I

Chest

20

1 l50

0

Left Hand Dorsum

L L

I

I

-I4

0

10

20

Rest

I

I 30

1

Walking 1 40

I

1

alcohol

I

I

1

1

I

I

I

50

60

70

80

90

100

110

1 120

TIME, min Fig, 4 Temperatures of the chest, left hand and foot dorsal, and rectal temperature of Quechua Indian men (n = 1 0 ) during alcohol consumption and light activity at a mean air temperature of 8OC.

performed by men. Men also expose their feet to wetcold during the preparation of mud used for the manufacture of adobe bricks, the dried earth cooking stove, and caulking for stone huts. Other required duties include construction and maintenance of irrigation ditches and channels by men and the laundering of clothing by women. Simulated tests in cold water The results of the two river tests conducted in the present study should be reasonable measures of the cold stress experienced under normal patterns of behavior. Under

each of the conditions of clothes-washing and channelconstruction (figs. 2, 3 ) , hand and foot temperatures dropped rapidly to levels that were comparable to the responses elicited during laboratory exposures. Figure 5 is a comparison between foot temperature responses to water at 1 0 ° C among (1) men tested at rest in the laboratory (Little, 'SQ), ( 2 ) the men who worked at channel-construction, and (3) the women who laundered clothes. The correspondence between the laboratory and field tests of males is close despite the moderate to heavy activity required during channel-construction. There are no equivalent laboratory data for women

LOCAL COLD EXPOSURE OF ANDEAN INDIANS "C 35

e--Channel Construction, 6 0

30

w

River Water : 0 Channel Construction Clothes Washing

25

a 3

t-

U

a 20 W n

5

15

t-

i

I I

----+

10

35

I

I

-

-

,--I 0

h\

I\

I \ I \ I \\

30-

I

I I I

g3 z u a w GrnP n \

c

&

15

&

I

:

\

\

I

II \

\\

\\ \\\

Channel Construction, c? Clothes Washing, River Water: 0 Channel Construction Clothes Washing

0

-

\' recovery

'

exposure

I I

-

I I

-

10 - ; Y - = + z = - i I

I

-5

0

5

10

-

I I I I I

I I

L

.

I

I

-10

-

\

I I

5

4

&

I

I

L

5

---* q___-

1---

I I

15

20

25

30

-

; -

4-

d

35

40

45

50

55

60

T I M E , min Fig. 6 Comparison of htritd temperatures of Indian women ( n = 9) during clothes washing and Indian m e n ( n = 10) during channel construction in the Nulioa River a t a water temperature of 10°C.

312

JONES, LITTLE, THOMAS. HOFF AND DUFOUR

with which to compare the results of the clothes-washing test. The results for men, nevertheless, would tend to validate the use of certain laboratory tests, at least where laboratory conditions approximate natural exposure conditions. Another interesting relationship that arises from an inspection of figure 5 is the marked difference in responses between field tests of men and women (see also fig. 6). Women consistently maintained warmer hand and foot temperatures than men with most differences (by t-test) a t levels of statistical significance (p < 0.05 and p < 0.01). There is some evidence in the literature that supports the present findings. For example, Japanese and Ainu women showed greater tolerance than men during icewater immersion of the fingers (Kondo, '69; Suzuki, '69), and Eskimo women had greater heat output from the hands than did men during cold water immersion of the hands (Lund-Larsen et al., '70; Krog and Wika, '71-'72). Moreover, in hand cooling tests in water of Nunoa Indians in the laboratory, women maintained warmer hand surface temperatures than men during the equilibration and immersion periods, and recovered more rapidly during post-immersion (Little et al., '73). The sex differences in the present study should be interpreted with some degree of caution, however, since the two field tests were not identical. Alcohol consumption and activity The results from the test of alcohol consumption were not in full accord with earlier laboratory studies that demonstrated a distinct short-term thermal advantage of alcohol intake during cold exposure of the foot (Little, '70). The principal objective of the present test was to simulate the patterns of Andean Indian behavior associated with fiestas and evening drinking bouts, where adults walk long distances to visit friends and neighbors, while at the same time, consuming substantial amounts of alcohol. Two factors may have minimized alcohol effects in the present study. First, in the earlier laboratory test (Little, '70), 1.1 g of alcohol per kilogram of body weight was consumed within a ten-minute period, a procedure which elicited marked warming about ten minutes after alcohol ingestion. In the activity test reported here, a n equal amount of alcohol was consumed, yet over

a 40-minute period. This time extension probably reduced the effectiveness of the alcohol. Second, the present test involved light activity whereas the laboratory test was conducted on resting subjects. Hence, the increased activity may have contributed to more rapid metabolizing of the drug. ACKNOWLEDGMENTS

The authors wish to thank Nicole M. Dufour, Louisa and Victor Barreda, and Juliana Hoff, all of whom provided field assistance at various levels of the study. Funds for the study were provided by the Research Foundation of the State University of New York at Binghamton (No. 40-7225-A), Cornell University, the University of Oregon, and the Wenner-Gren Foundation. In addition, The Pennsylvania State University provided logistic support and financial aid through grant No. 5 TO1 GMO 1748-04. This work is a part of the U.S. contribution to the International Biological Program. LITERATURE CITED Baker, P. T. 1966 Micro-environment cold in a high altitude Peruvian population. In: Human Adaptability and its Methodology. H. Yoshimura and J. S. Weiner, eds. Japan Society for the Promotion of Science, Tokyo, pp. 67-77. Baker, P. T., E. R. Buskirk, J. Kollias and R. B. Mazess 1967 Temperature regulation at high altitude: Quechua Indians and U . S. whites during total body cold exposure. Hum. Biol., 39: 155-169. Elmer, R. W., and A. Bolstad 1963 Thermal and metabolic responses to cold of Peruvian Indians native to high altitude. Tech. Report No. AALTDR-62-64, Arctic Aeromed. Lab., Ft. Wainwright, Alaska. Hanna, J. M. 1970 A comparison of laboratory and field studies of cold responses. Am. J. Phys. Anthrop., 32: 227-232. 1976 Natural cold exposure. in: Man i n the Andes: A Multidisciplinary Study of High Altitude Quechua. P. T. Baker and M. A. Little, eds. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania, i n press. Kondo, S. 1969 A study on acclimatization of the Ainu and the Japanese with reference to hunting temperature reaction. J. Fac. Sci. Tokyo Univ., Sec. 5, 3: 253-265. Krog, J . , and M. Wika 1971-72 Studies of the peripheral circulation in the hand of the igloolik Eskimo. In: International Biological Program, Human Adaptability Project (Igloolik, N. W. T.) Annual Report No. 4. Anthropological Series, No. 11, University of Toronto, Ontario, pp. 173181. Larsen, R. M. 1973 The thermal microenvironment of a highland Quechua population; biocult u r d adjustment to the cold. M. A. Thesis, Dept. of Anthropology, University of Wisconsin, Madison.

LOCAL COLD EXPOSURE OF ANDEAN INDIANS Little, M. A. 1969 Temperature regulation a t high altitude: Quechua Indians and U.S. whites during foot exposure to cold water and cold air. Hum. Biol., 41: 519-535. 1970 Effects of alcohol and coca on foot temperature responses of highland Peruvians during localized cold exposure. Am. J. Phys. Anthrop., 32: 233-242. Little, M. A,, R. B. Thomas and J. W. Larrick 1973 Skin temperature and cold pressor responses of Andean Indians during hand immersion i n water at 4°C. Hum. Biol., 45: 643-662. Little, M A , , R. B. Thomas, R. B. Mazess and P. T.

313

Baker 1971 Population differences and developmental changes i n extremity temperature responses to cold among Andean Indians. Hum. Biol., 43: 70-91. Lund-Larsen, K., M. Wika a n d J. Krog 1970 Circulatory responses of the hand of Greenlanders to local cold stimulation. Arctic Anthropol., 7: 21-25. Mazess, R. B.,and R. M. Larsen 1972 Responses of Andean highlanders to night cold. Int. J. Biometeor., 16: 181-192. Suzuki, M. 1969 Peripheral response to cold. J. Anthropol. SOC.Nippon, 77: 213-223.