According to the "fetal origins hypothesis," monozygotic (MZ) twins may be more prone to develop various metabolic abnormalities compared with dizygotic (DZ) twins, and twins all together may be more predisposed to metabolic defects compared with singletons. To determine the impact of twin and zygosity status as well as birth size on in vivo measures of glucose metabolism, we examined 123 young (aged 22-31 years) and 103 elderly (aged 57-66 years) MZ and DZ twins and age-matched singleton control subjects. All participants were born at term with available birth records. Peripheral and hepatic insulin action and intracellular glucose partitioning was determined by a euglycemic-hyperinsulinemic clamp using tritiated glucose combined with indirect calorimetry. In elderly subjects, zygosity status influenced nonoxidative glucose metabolism, while twin status per se was associated with elevated hepatic glucose production during both steady-state periods. Birth weight was associated with nonoxidative glucose metabolism in a nongenetic manner within twins and with a high glucose and low lipid oxidation in singletons. In younger subjects, twin status influenced glucose and lipid oxidation rates. We demonstrate a complex age- or time-dependent relationship between independent markers of fetal environment and glucose homeostasis in twins. The documented differential programming effects associated with either low birth weight and twin or zygosity status all represent known defects of glucose homeostasis in type 2 diabetes.