The objective of this study was to evaluate the association between in utero diagnostic X-rays and childhood acute lymphoblastic leukemia (ALL) and the less well-studied relationship of this malignancy to preconception and postnatal diagnostic X-rays or fetal ultrasound exposures. The Children's Cancer Group conducted a case-control study including interviews with parents of 1842 ALL cases diagnosed under the age of 15 years and 1986 individually matched controls. Associations of self-reported parental preconception, in utero, and postnatal X-ray exposure with risk of childhood ALL were examined using odds ratios (ORs) and corresponding 95% confidence intervals (CIs) obtained from logistic regression models among the overall group of ALL cases as well as immunophenotypic and age-specific subgroups. Overall, in utero pelvimetric diagnostic X-rays were not associated with the risk of pediatric ALL (OR, 1.2; 95% CI, 0.8-1.7). Childhood ALL, all types combined (OR, 1.1; 95% CI, 0.9-1.2) and specific types were also not linked with postnatal diagnostic X-ray exposures. Neither maternal (OR, 0.9; 95% CI, 0.8-1.2) nor paternal (OR, 1.1; 95% CI, 0.8-1.4) lower abdominal preconception diagnostic X-rays were associated with risk of childhood ALL. Among the multiple comparisons for age-, sex-, and subtype-specific subgroups, we observed an elevated risk of total ALL among children ages 11-14 at diagnosis (OR, 2.4; 95% CI, 1.1-5.0) in relation to in utero pelvimetric diagnostic X-ray exposures and a small increase in pre-B ALL for all ages combined (OR, 1.7; 95% CI, 1.1-2.7) in relation to postnatal diagnostic X-rays. In utero diagnostic ultrasound tests were not linked with risk of childhood ALL. We found little consistent evidence that in utero diagnostic ultrasound tests or X-rays were linked with an increased risk of childhood ALL. Small increases in total or pre-B ALL risks for children in selected age groups to very low ionizing radiation exposures from postnatal or preconception diagnostic X-ray exposures may represent chance findings or biases. Future studies of diagnostic X-rays and childhood leukemia in the United States will require extensive additional efforts and resources to quantify risk because of declining in utero exposures in the general population (thus necessitating large numbers of subjects, particularly cases) and the difficulty in validating reported exposures.