Objective: To study the effect of consumption of dark green leafy vegetables (DGLV) and sweet pumpkin on serum b-carotene and retinol concentrations in children treated for Ascaris lumbricoides. Design: Experimental study with a randomised design. Subjects: A total of 110 primary school children aged 8 ± 12 y in northwestern Bangladesh. Interventions: All children were de-wormed and 2 weeks later randomly assigned to one of three groups to receive for 6 days per week, for 6 weeks, one complete meal containing either: (1) 4.4 mg beta-carotene from DGLV (n 37, after 18 dropouts); (2) 1.5 mg beta-carotene from sweet pumpkin (n 36, 18 dropouts); or (3) vegetables containing virtually no beta-carotene (control) (n 37, 18 dropouts). Results: Signi®cant increases (P`0.001) in mean serum b-carotene concentrations were seen in all three study groups, with a statistically higher increase (mmolal) in the DGLV group (0.44; 95% con®dence interval (CI) 0.32, 0.55) compared to the control group (0.20; 95% CI 0.14, 0.26; P 0.002). The increase in serum retinol (mmolal) was statistically signi®cant (P 0.04) only in the DGLV group (mean 0.066; 95% CI 0.002, 0.13), but this increase was not different from the increase in the control group. Conclusion: In children successfully treated for Ascaris lumbricoides, a substantial increase in serum b-carotene was seen after feeding with a moderately high cumulative dose of DGLV for 6 weeks.