A population-based case-control study was conducted to test the hypothesis that the risk of the occurrence of neural-tube defects in infants with no other birth defects (isolated neural-tube defects) is associated inversely with the maternal dietary intake of free and/or total folate in early pregnancy. Information was collected from the mothers of 77 case subjects with isolated neural-tube defects who were born in Western Australia from 1982 to 1984, from the mothers of 77 control subjects with birth defects other than neural-tube defects (control group 1) and from the mothers of 154 control subjects with no birth defects (control group 2). The case and control subjects were matched individually by the date of the mother's last menstrual period. Odds ratios were adjusted for a number of potentially-confounding variables, such as the country of birth of the parents, paternal social class, previous pregnancy outcome, interval between index and previous pregnancy and pregnancy order. Crude and adjusted odds ratios showed a protective effect of an increasing intake of free folate in the first six weeks of pregnancy. Adjusted odds ratios, with reference to the lowest quartile of intake, (and their 95% confidence intervals) were 0.72 (0.25-2.08), 0.37 (0.11-1.23) and 0.31 (0.10-0.97) for quartiles 2-4 when control group 1 was used, and 0.44 (0.17-1.13), 0.34 (0.13-0.90) and 0.16 (0.06-0.49) when control group 2 was used. Similar, but weaker, trends were seen when total folate intake was the exposure variable. These findings support the hypothesis that the dietary intake of folate in early pregnancy protects against the occurrence of isolated neural-tube defects in infants. Measures of postpartum dietary folate and of postpartum serum and red-cell folate levels showed no association with the occurrence of neural-tube defects in infants.