A total of 107 children who had been hospitalized following a febrile con vulsion were enrolled into the trial. By random allocation, 55 children were treated with pyridoxine hydrochloride (20 mgs twice daily) and the remain ing 52 children were treated with a placebo until there had been either a fur ther convulsion or a year had passed without recurrence. Eighty children were adequately followed up and of these, 17 had a recurrent febrile convulsion while receiving medication. Recurrences occurrences occurred in 7 of the 38 children receiving pyriddxine and in 10 of the 42 childreh receiving placebo (χ2 = .346, p > 0.5). Initial tryptophan load tests had been abnormal in 34 children, and of these, recurrences occurred in 3 of the 17 who received pyridoxine and in 3 of the 17 who received placebo. It has yet to be shown that pyridoxine supplementation protects children from recurrent febrile convulsions.