Objective: To assess the effect on birthweight of antenatal mebendazole plus iron vs. placebo plus iron in a highly hookworm-endemic area. Methods: Double-blind, randomized controlled trial set in rural and peri-urban communities in the Peruvian Amazon region. A total of 1042 second trimester pregnant women between the ages of 18 and 44 years were recruited from April to November 2003, and followed to July 2004. Women were randomly assigned to receive either mebendazole (500 mg single dose) plus iron supplements (60 mg elemental iron daily) or placebo plus iron supplements. The primary outcome was mean infant birthweight and secondary measures included proportion of low birthweight babies and maternal anaemia. Results: The prevalence of hookworm infection was 47.5%. There were no differences between intervention groups in mean birthweight (3104 g vs. 3090 g, P = 0.629), proportion of low birthweight (<2500 g; 8.1%vs. 8.7%, P = 0.755) or maternal anaemia in the third trimester [33.0% (158/479) vs. 32.3% (152/471), P = 0.815]. However, the proportion of very low birthweight (<1500 g) was significantly lower in the mebendazole group [0% (0/479) vs. 1.5% (7/471), P = 0.007]. Conclusions: This trial provides additional evidence for the use of anthelmintics, over and above iron supplementation, within antenatal care programmes in hookworm-endemic areas. Benefits of de-worming may be higher in countries not having an antenatal iron supplementation programme or where intensity of hookworm infections is higher. © 2006 Blackwell Publishing Ltd.