One of the earliest systematic reports of the effect of heat on appetite was Johnson and Kark's (1947) summary of ad lib intakes by soldiers in various geographic areas ranging from the mobile force "Musk Ox," stationed in the Canadian Arctic, to infantry troops on Luzon in the Philippines. Troops stationed in the tropics ate on average 3100 calories, whereas arctic troops ate 4900 calories; in both cases, troops were allegedly offered as much food as they wanted, although the skeptical reader might be able to imagine some interpretive confoundings in these results. Whether the size of the heat effect on appetite is as profound as suggested by Johnson and Kark is debatable, but the direction of the effect has not been disputed. Edholm et al. (1964) confirmed this pattern, observing a 25 percent decrease in food intake by soldiers in Aden compared to the United Kingdom (see also Collins and Weiner, 1968). More parochially, one Toronto restaurateur (unpublished data from consumer survey, 1991) conceded that "sales plunge during a heat wave .... People do not have the appetite for a large heavy meal when it is hot." Beller (1977) neatly summarized the effect of heat on appetite, explaining it on the basis of the thermic effect of food:

Logue (1986) makes the same point more colloquially: "An easy way to quell appetites at a summer dinner party is to turn off the air-conditioning." These generalizations suggest that acute temperature variations have a strong effect on appetite: specifically, heat impairs appetite. (And note that the implication of the turned-off air-conditioning scenario is that had the air-conditioning stayed on, appetite would have remained unquelled, despite it being summer. This suggests that acute temperature variations