Earlier results have shown that some infants born by elective Caesarean section start to sweat in a warm environment while others do not, and that sweating can be inhibited by feeding cold glucose. To determine whether these earlier observations, indicating a difference in postnatal temperature adaptation, could be reproduced in vaginally born infants, we measured the rate of evaporation from the skin surface, body and skin temperatures from several sites, skin blood flow and respiratory rate in newborn infants nursed in a warm environment, before and after feeding cold water. In all infants the body and skin temperatures increased in the warm environment (p < 0.01), with a decreasing difference between oesophageal and leg skin temperature (p < 0.01). Visible sweating occurred in 9/14 infants at a rectal temperature of 37.5 degrees C. In the infants who started to sweat, evaporation rate increased from 5.6 +/- 2.8 (SD) g/m2/h 15 min before sweating to 15.7 +/- 10.6 g/m2/h (p < 0.05) when sweat became visible and the infants were fed cold water. After feeding of cold water the evaporation rate decreased and within 10 min returned to a value not significantly different from the pre-sweating value. Interscapular skin blood flow had increased by 42% (p < 0.01) at the time of sweating and decreased by 22% (p < 0.01) after feeding cold water. In the infants who did not start to sweat, no increase in evaporation rate was noted and the changes in skin blood flow were not statistically significant. The infants who started to sweat did not differ from those who did not regarding maternal medication during delivery. We conclude that some, but not all, newborn infants start to sweat at a body temperature of 37.5 degrees C. In the infants who start to sweat, sweating and an increase in skin blood flow can be inhibited by feeding cold water. There seem to be individual differences in the regulation of body temperature in newborn infants, possibly due to a delayed change in the central temperature set-point in some infants.