Plasma glucose and insulin concentrations were measured in 45 new-born piglets both before and at regular intervals throughout a 2-h period following an i.v. administration of glucose (1 g/kg body weight), at environmental temperatures of 17, 24 and 33.degree. C. As anticipated, rectal temperature (TR) was dependent upon the environmental temperature. During the glucose tolerance test (GTT), TR remained constant at both 24 and 33.degree. C, 37.0 .+-. 0.2 compared with 38.0 .+-. 0.1.degree. C. However, at 17.degree. C, 12 piglets were able to maintain a constant TR (35.8 .+-. 0.2.degree. C), whereas the remaining 6 piglets became hypothermic (32.8 .+-. 0.4.degree. C), with TR decreasing continuously throughout the GTT. Their body weight was 210 g below that of the other piglets (P < 0.05). Those piglets at 17 and 24.degree. C which maintained homeothermy had a higher glucose disappearance rate (KG; 2.00 and 2.31%/min, respectively), than those kept within thermal neutrality (1.66%/min). The insulin response to glucose was similar. During hypothermia, on the other hand, both KG (0.76 .+-. 0.12%/min) and the insulin response were significantly less (P < 0.05) than in all other piglets. The glucose uptake by the skeletal muscle is increased in cold-exposed, normothermic piglets and results in an increased metabolic rate. Failure to maintain homeothermy diminishes glucose uptake and inhibits insulin release.