Introduction. We tested the hypothesis that family history of type 2 diabetes (FHD) is associated with reduced birth weight and reduced insulin secretion later in life. Materials and Methods. Birth weight, body composition by whole-body dual-energy X-ray absorptiometry, and homeostasis model assessment-insulin resistance were compared between Japanese women aged 20 years with positive (n=73) and negative (n=258) FHD. A subsample of 153 women (57 with positive FHD) underwent a 75 g oral glucose tolerance test. Multivariate logistic regression analyses were used to identify the most important determinants of FHD. Results. Women with positive as compared with negative FHD had lower birth weight (3132 +/- 364 vs. 3238 +/- 418 g, p=0.04). However, the current fat mass index and trunk/leg fat ratio, sophisticated measures of general and abdominal fat accumulation, respectively, did not differ. Women with positive FHD had a lower insulinogenic index (2.4 +/- 7.3 vs. 6.2 +/- 16, p=0.007) and higher area under the glucose curve (217 +/- 47 vs. 198 +/- 36 mg/dL/2 h, p=0.006). However, fasting and postload insulinemia, homeostasis model assessment-insulin resistance, and Matsuda index did not differ. In multivariate logistic regression analysis, birth weight was marginally associated with FHD (odds ratio, 0.999; 95% confidential interval, 0.98-1.00000; p=0.0509). Conclusions. FHD was associated not only with reduced birth weight but also with decreased early-phase insulin secretion and increased postload glucose concentrations in Japanese women aged 20 years. These findings may be in keeping with the fetal insulin hypothesis and provide some evidence that FHD can alter size at birth, probably through genetic and shared environmental components, which consequently resulted in decreased early-phase insulin secretion and increased glucose excursion in the early twenties. FHD was not related to sophisticated measures of general and abdominal adiposity and insulin resistance/sensitivity.