Regular physical activity may prevent or postpone type 2 diabetes, and is thought to be related to an increase of insulin sensitivity. We studied whether physically active, glucose-tolerant first-degree relatives of type 2 diabetes patients differ in glucose tolerance (oral glucose tolerance test [OGTT]) and insulin secretion (hyperglycemic glucose clamp) from less active first-degree relatives. A group of 37 relatives was split into 2 subgroups according to the sex-specific median of the sports index, assessed by a questionnaire, as the cutoff point. Blood glucose levels during the OGTT were lower in the highly active subgroup versus the less active counterparts (multivariate ANOVA [MANOVA], P = .011), but the plasma insulin levels were similar. First-phase secretion was not different in the highly active group versus the less active group, but second-phase secretion (average plasma insulin in the third hour) was significantly lower (P = .016). As expected, the insulin sensitivity index (ISI) was higher in the highly active subgroup (P= .011). Subdivision into subgroups with high or low maximal O2 consumption (VO2max) resulted in similar differences, but these were not significant. In a group of 21 controls, the results resembled the values in the relatives but were less often statistically significant. In conclusion, regular physical activity not only is associated with increased insulin sensitivity but also downregulates the pancreatic beta cell. This downregulation may provide an extra mechanism by which physical activity diminishes the development of type 2 diabetes.