An acceleration-sensitive device was used to measure motor activity continuously through the day in 18 hyperactive boys in a day hospital program. The children received methylphenidate, dextroamphetamine, or placebo daily after breakfast and lunch in an 11-week double-blind crossover trial. Differential effectiveness of the two drugs in lowering motor activity was found. Methylphenidate significantly lowered activity measurements in a morning structured classroom and in less structured activities in the afternoon. Dextroamphetamine effects on activity were similar, although they did not differ significantly from placebo effects between 11:00 AM and noon in our classroom setting. Methylphenidate produced a greater decrement in motor activity than did dextroamphetamine between 11:00 AM and 1:00 PM. There were no significant differences in activity level between drug doses within each drug phase across the dose ranges used (for methylphenidate 0.45 to 1.25 mg/kg given twice daily, and for dextroamphetamine 0.2 to 0.6 mg/kg given twice daily). Plasma drug concentrations did not correlate with decrements in activity for either drug.