Both insulin resistance and decreased insulin secretion have been hypothesized as precursors of noninsulin-dependent diabetes mellitus (NIDDM). However, there are few data on insulin resistance and secretion in relationship to parental history of diabetes in ethnic groups at different risks for NIDDM. We examined the relationship of fasting insulin (as a marker of insulin resistance) and the ratio of the 30-min change in insulin to the 30-min change in glucose (delta I30/delta G30) during an oral glucose tolerance test (as a marker of insulin secretion) in relation to parental history of diabetes in 1672 nondiabetic Mexican Americans and 894 nondiabetic non-Hispanic whites. A parental history of diabetes was associated with increased fasting insulin concentrations. The association between decreased insulin secretion and parental history of diabetes was not significant. However, when insulin resistance and insulin secretion were included in the same regression model, both fasting insulin (odds ratio = 1.80; 95% confidence interval = 1.17, 2.76) and decreased insulin secretion (odds ratio = 0.70, 95% confidence interval = 0.52, 0.95) were significantly associated with a parental history of diabetes. These results were little changed after adjustment for obesity, body fat distribution, and glucose tolerance. These results were similar in both Mexican Americans and non-Hispanic whites and in both subjects with impaired and those with normal glucose tolerance. A parental history of diabetes in both Mexican Americans and non-Hispanic whites is associated with both increased fasting insulin and decreased delta I30/delta G30. These data suggest that both increased insulin resistance and early decreased insulin secretion in response to an oral glucose challenge may be important factors in the pathogenesis of NIDDM in populations at high and low risk of NIDDM. Our results also emphasize the importance of adjusting for insulin resistance in evaluating insulin secretion.