Objective The aim was to test the hypothesis that the difference in body mass index (BMI) between twins and singletons may be attributed to a difference in birth weight. Methods Data from the Medical Birth Registry of Norway were linked with data from the Norwegian Conscript Service. This study used data on 411,186 males who were born in single or twin births in Norway during 1967 to 1984, and who were examined at the mandatory military conscription (98% were examined the year they turned 18 or 19 years of age). Results The difference in BMI between singletons and twins was 0.49 kg/m2 (95% confidence interval [CI]: 0.41, 0.56) after adjustment for background factors and gestational age at birth. When birth weight was added to the model, the difference was reduced to 0.17 kg/m2 (95% CI: 0.09, 0.25). The corresponding figures for sibships that included both singletons and twins were 0.63 kg/m2 (95% CI: 0.39, 0.86) and 0.38 kg/m2 (95% CI: 0.06, 0.70) when twins were compared with their singleton brothers. In both the analysis of the total study population and the analysis of the sibships, birth weight was significantly associated with the BMI after adjustment for twin status, gestational age at birth, and background factors. Conclusions Male twins born in Norway during 1967 to 1984 had lower BMI at the military conscription than had singletons, and a part of this difference may probably be attributed to lower intrauterine growth rate among twins. Am. J. Hum. Biol. 28:566–573, 2016. © 2016 Wiley Periodicals, Inc.