A double-blind, placebo-controlled, crossover study of methylphenidate (0.4 mg/kgday) and different doses of fenfluramine (1.0, 1.5, or 2.0 mg/g/day) in children with mental retardation or borderline IQ and ADHD was conducted. Parents, teachers, examiners, and physicians rated the children. There were relatively few significant drug effects by condition. When the optimal fenfluramine dose for each child was compared with placebo and methylphenidate, significant improvements occurred for fenfluramine on several parent and teacher subscales; teachers rated the children as somewhat improved with methylphenidate. The highest dose of fenfluramine produced more behavior compliance but apparently at the cost of cognitive efficiency. Most side effects (drowsiness, dizziness, anorexia) occurred with fenfluramine. Both drugs appear to be effective treatments for children with ADHD and mental retardation, although there is a possible neurotoxic action with fenfluramine. We recommend a gradual phase-in of fenfluramine dosage, up to 1.5 mg/kg/day, for most children.