Precentral motor cortex neurons were studied under non-drug and drug conditions in three trained monkeys during the performance of a go-no go visual "attention" task. The two drugs studied, secobarbital and chlorpromazine, produced differing patterns of effect on components of the motor sequence involved in reaction time. The following components were considered: The SF interval or the period from stimulus onset to change in neuronal firing; the FR interval, which is the period from change in firing to the beginning of the task response; and MT or movement time, which is the time necessary to complete the response. Secobarbital produced an increase of 80% in SF and a relatively small average change in FR although there was considerable variability in the latter. MT was decreased in most secobarbital experiments. Anatomical factors relating to the FR variability were considered, and the MT decrease was discussed in terms of possible excitatory effects of the drug. Chlorpromazine produced small increases in SF, FR and MT, alternating with periods of complete abolition of performance. The results were discussed in terms of theories of attention deficit following administration of secobarbital and chlorpromazine.