Objectives: In The Netherlands, part of the population experienced food restriction and severe famine during World War II. The purpose of this study was to study the e ects of severe undernutrition during adolescence on the risk of breast cancer later in life. Methods: We examined the hypothesis in the Netherlands Cohort Study on diet and cancer (NLCS), among 62,573 women aged 55±69 years. Baseline information on diet and other risk factors was collected with a questionnaire in 1986. Information was collected on residence in the Hunger winter (1944±1945) and War years (1940±1944) and fathers' employment status in 1932±1940 as indicators of exposure. After 6.3 years of follow-up, 1009 incident breast cases were available for analysis. Results: In multivariate case-cohort analysis, residents of the western part of the country in 1944±1945 had an increased breast cancer risk (western city RR = 1.1, 95% CI: 0.9±1.4, western rural area RR = 1.5, 95% CI: 1.1± 1.9). For the War years (1940±1944) we found no association between breast cancer risk and urban vs. rural residence. Women whose fathers were unemployed during the Depression years (1932±1940) had a non-signi®cant decrease in breast cancer risk (RR = 0.9, 95% CI: 0.7±1.2). Exposure to energy restriction during the adolescent growth spurt or during the period between menarche and birth of the ®rst child did not change the RRs substantially. Conclusions: We found no clear evidence in this study for the hypothesis that energy restriction in adolescence leads to a decreased breast cancer risk.