The effect of differences in size at birth on growth and glucose homeostasis between female twin lambs during juvenile life was examined. Twenty-six sets of twins were entered into the study, of which ten were used for organ sampling at birth and 16 were studied over the first year of life. Eleven sets were defined as being mismatched for birthweight as the weight difference between twins was >25%, with light lambs weighing 4.1 +/- 0.3 kg and heavy lambs weighing 5.1 +/- 0.1 kg. All remaining twins were matched in bodyweight, weighing 4.6 +/- 0.5 kg. During the rapid period of juvenile growth (i.e. one, three and six months of age) and following stabilization of bodyweight (i.e. 12 months of age) glucose tolerance tests were performed by intravenously injecting 0.8 mg kg(-1) bodyweight glucose. This was followed the next day with an insulin tolerance test, performed by intravenously injecting 0.08 units kg(-1) bodyweight insulin. At birth there were no differences in organ weight as a fraction of total bodyweight between matched and mismatched twins, but the ratio of liver to brain weight was lower in light compared with heavy twins. Light lambs remained lighter than their twins up to six months of age, and crown-rump length was also shorter. At one and three months of age there were no differences in basal plasma glucose concentrations between the groups, but glucose tolerance was greater in light compared with heavy lambs at one and six months of age. Insulin tolerance was greater in light compared with matched lambs at one and six months of age. In conclusion, it has been shown that size at birth of one twin in relation to its co-twin is one factor determining glucose regulation during postnatal life. This not only affects glucose and insulin tolerance but also growth over the first six months of age.