In the United States, milk (usually cow's milk) is widely considered an "essential food" to support bone growth among post-weaning age children, as evident in government-sponsored nutrition policies that mandate milk for children. Milk contains calories, protein, and calcium, among other nutrients, and bioactive components such as insulin-like growth factor-I (IGF-I), all of which may facilitate bone growth. There is a large literature on milk and/or calcium intake and its effects on bone density, but one aspect of bone mass--height--is not well studied in relation to milk consumption. Limited experimental studies show no consistent relationship across populations. To investigate this linkage among American children, analysis of the NHANES 1999-2002 was undertaken. NHANES data allow two hypotheses to be tested: (1) reported frequency of childhood milk consumption will be positively related to adult height and (2) height of children 5-18 years will be predicted by the reported frequency of milk consumption and/or milk intake from a 24-h dietary recall. Results indicate that adult height was positively associated with milk consumption at ages 5-12 and 13-17, after controlling for sex, education, and ethnicity. Among contemporary children, milk consumption had no effect on the height of 5-11 year olds after controlling for age, birthweight, energy intake, and ethnicity. In contrast, milk consumption frequency and milk intake (measured as grams of milk, or protein or calcium from milk) were significant predictors of the height of 12-18 year olds, along with age, sex, household income, and ethnicity. The greatest ethnic contrasts were between Mexican Americans and non-Hispanic whites and blacks, and milk variables remained significant predictors of height in these comparisons. Thus NHANES data show substantial variability in the effects of milk consumption on height.