SINCE Bradley and his co-workers1,2 reported the effects of racemic amphetamine on school performance and on psychological tests, there have been very few investigations on this subject. In view of the report of the effectiveness of this stimulant on school performance, as well as on behavior problems,3,4 it is surprising that so little systematic effort has been made to evaluate the earlier claims. The amphetamines (both the DL and D forms) have been shown to produce beneficial effects on a wide variety of behavioral tasks in normal adults with very few side effects5; but there is little available work on the effects of amphetamines in children who are not selected specifically for hyperkinetic behavior problems. Neither is there evidence to specify the behavioral locus of the drug effects. (The dextrorotatory [D] isomer is stated to be three to four times more potent than the laevo [L]