ued lower level of food intake. The marked decrease initially was thought to be due to the initial dehydration; the continued lower level of intake was due to the adaptation to the increased ambient temperature. Body weights in the growing rats dropped initially by as much as 30 grams and then remained constant until the heat stress was removed. These studies have been used as a demonstration of the concept of a body weight set-point lowering effect due to a hot environment (Thompson, 1980).
The "thermostatic" theory of food intake (a decrease in the body weight or fat set-point in response to hot environments) has been proposed as the method by which the body may thermoregulate, in part by decreasing the insulating amount of body fat (Brobeck, 1948). Studies in a number of experimental animals demonstrate a cessation of eating at high temperatures, which indicates that continued eating would probably lead to hyperthermia. The decrease in food intake is thus followed by a decrease in body weight and fat (Jakubczak, 1976).
In investigations of the theory that animals stop eating to prevent hyperthermia, the differences in the resulting thermic effect of food ingested (specific dynamic action) have been implicated. In the series of experiments by Hamilton (1963a), calorie intake of rats fed special diets during mild heat stress was inversely related to the thermogenic effect of the diet selected. It appears that fats may be the preferred energy source in heat stress (Salganik, 1956), and that in severe heat stress, protein is avoided due to the comparatively high amount of heat created (Hamilton, 1963a). With this theory, body temperature should be highly correlated with hunger and satiety. However, there is no consistent observed relationship between the two. Although in 1936 Booth and Strang reported that skin temperature in adults increased 2°C within 10 minutes of eating a high protein meal, a postprandial increase in skin temperature was not found by Stunkard et al. (1962). In dogs, Passmore and Ritchie (1957) found an extremely small rise in skin temperature after a high protein meal while Hamilton (1963a) determined that food consumption and rectal temperature in rats decreased incrementally with temperatures between 7° and 32 °C, but at 35°C, rectal temperature became elevated, while food intake continued to decrease.
Andersson and Larsson (1961) have shown that heating of the preoptic and anterior hypothalamic regions of the brain (areas that are known to be involved in regulating body temperature) inhibits feeding in animals. Opposite results, however, were obtained by Spector et al. (1968) in rats, when heating of the preoptic medialis region caused increased eating when the temperature of the area was raised to 43°C. Decreased eating occurred in their study when the ambient temperature was raised to 35°C. Local temperature in the anterior hypothalamic area has been reported to drop at the onset of feeding, which is opposite of what would be expected (Hamilton, 1963b). Thus it appears that the effect of brain temperature on feeding may