Multiple heterogeneous groups of subjects (both sexes and a wide range of maximal oxygen uptake VO2max, body mass, body surface area (AD),% body fat, and AD/mass coefficient) exercised on a cycle ergometer at a relative (%VO2max, REL) or an absolute (60 W) exercise intensity in a cool (CO 21 degrees C, 50% relative humidity), warm humid (WH 35 degrees C, 80%) and a hot dry (HD 45 degrees C, 20%) environment. Rectal temperature (Tre) responses were analysed for the influence of the individual's characteristics, environment and exercise intensity. Exposures consisted of 30-min rest, followed by 60-min exercise. The Tre was negatively correlated with mass in all conditions. Body mass acted as a passive heat sink in all the conditions tested. While negatively correlated with VO2max and VO2max per kilogram body mass in most climates, Tre was positively correlated with VO2max and VO2max per kilogram body mass in the WH/REL condition. Thus, when evaporative heat loss was limited as in WH, the higher heat production of the fitter subjects in the REL trials determined Tre and not the greater efficiency for heat loss associated with high VO2max. Body fatness significantly affected Tre only in the CO condition, where, with low skin blood flows (measured as increases in forearm blood flow), the insulative effect of fat was pronounced. In the warmer environments, high skin blood flows offset the resistance offered by peripheral adipose tissue. Contrary to other studies, Tre was positively correlated with AD/mass coefficient for all conditions tested. For both exercise types used, being big (a high heat loss area and heat capacity) was apparently more beneficial from a heat strain standpoint than having a favourable AD/mass coefficient (high in small subjects). The total amount of variance in Tre responses which could be attributed to individual characteristics was dependent on the climate and the type of exercise. Though substantial for absolute exercise intensities (52%-58%) the variance explained in Tre differed markedly for relative intensities: 72% for the WH climate with its limited evaporative capacity, and only 10%-26% for the HD and CO climates. The results showed that individual characteristics play a significant role in determining the responses of body core temperature in all conditions tested, but their contribution was low for relative exercise intensities when evaporative heat loss was not restricted. This study demonstrated that effects of individual characteristics on human responses to heat stress cannot be interpreted without taking into consideration both the heat transfer properties of the environment and the metabolic heat production resulting from the exercise type and intensity chosen. Their impact varies substantially among conditions.