Increased ambient temperature affects apparent oral glucose tolerance to an extent which may have clinical implications for the diagnosis of impaired glucose tolerance and gestational diabetes. As a first step in order to better define the nature of this effect, we have examined, in a climate chamber, the effects of ambient temperature at four levels (20, 25, 30, and 35 degrees C) on glucose and insulin responses to a standard 75 g oral glucose tolerance test in seven non-diabetic male subjects. Plasma glucose responses to ambient temperature were compared with the responses of core (auditory canal) and skin temperatures. The 2-h plasma glucose was affected in a nonlinear manner by ambient temperature (5.4 +/- 0.2, 5.3 +/- 0.4, 6.5 +/- 0.3, 6.4 +/- 0.4 mmol/l at 20, 25, 30, and 35 degrees C, P = 0.015) with the effect localised between 25 and 30 degrees C (P = 0.012). Core temperature responded in a similar manner (36.6 +/- 0.1, 36.6 +/- 0.1, 36.9 +/- 0.1, 37.0 +/- 0.1, (P = 0.0005) with the effect localised 25 and 30 degrees C (P = 0.011). However skin temperature increased significantly with each 5 degrees C increase in ambient temperature (30.2 +/- 0.5, 33.0 +/- 0.5, 34.2 +/- 0.2, 35.2 +/- 0.2, P < or = 0.0001). We conclude that the acute effect of ambient temperature on apparent glucose tolerance is most likely due to redistribution of blood flow between cutaneous and visceral beds driven by changes in core temperature. The absence of temperature effects between the two lowest, and between the two highest temperatures, provides workable guidelines for the standardisation of conditions during oral glucose tolerance tests in circumstances where temperature may have clinically significant effects.