AIMS: To investigate if a second drop of 2.5% povidone-iodine ophthalmic solution placed within the first postnatal day would achieve better prophylaxis against ophthalmia neonatorum than a single drop applied at birth. METHODS: A masked, prospective, controlled trial was conducted over a 2 year period in a Kenyan hospital. Randomisation was achieved by alternating weeks of one or two eye drop application to both eyes. All 719 neonates received one drop of the povidone-iodine solution to both eyes at birth, while 317 received a second drop at hospital discharge or 24 (SD 4) hours after delivery, whichever was first. All infants developing conjunctivitis within a month after birth underwent microbiological analysis using Gram and Giemsa stains, direct fluorescent antibody assay for Chlamydia trachomatis, and culture. RESULTS: Of the neonates receiving the one eye drop application, 18.4% returned with a red eye with discharge, 4.0% had organisms found on the initial smear, and 8.2% had a positive culture. The corresponding proportions for the multidrop group were 24.3%, 4.7%, and 10.4%. Of those returning with an inflamed eye, there were no cases of Neisseria gonorrhoeae, 4.2% in the single dose group and 3.9% in the double dose group were positive for C trachomatis, and 5.4% and 6.5% respectively for Staphylococcus aureus. At discharge, the eyelid oedema score of the double dose group was mildly greater than the single dose group (1.4 (0.67) v 1.2 (0.73), p=0.0002). There was no statistically significant difference between the groups in any other category. CONCLUSION: There is no advantage to administering povidone-iodine prophylaxis against ophthalmia neonatorum twice in the first postnatal day over a single application at birth.