In a Latin-square double-crossover with random assignment to sequence, 18 boys, aged 6-12 years, with attention-deficit hyperactivity disorder received 1 month each of placebo, D-amphetamine, and Efamol (evening primrose oil containing gamma-linolenic acid, with vitamin E as preservative). Parents' ratings were noncontributory. Teachers' ratings showed a trend of Efamol effect between placebo and D-amphetamine. The trend reached significance (p less than 0.05) only on Conners Hyperactivity Factor. Dosage may be crucial; 8 Efamol capsules per day were used in this study. Heuristic data scrutiny suggested possible interaction (sequence effect). Further study with a different design and dose is suggested. This study does not establish Efamol as an effective treatment.