I The effects of single intravenous doses of diazepam, clonazepam, lorazepam, frusemide, mexiletine and (+)-propranolol on the frequency of electroencephalographic (EEG) interictal spike discharges were assessed in three double-blind placebo-controlled trials in fourteen epileptic patients. 2 Diazepam (5 mg) produced a significant mean decrease of spike discharges to about one third of the control frequency. The effect developed rapidly and persisted for the duration of the study (up to 25 min after injection). A larger dose (10 mg) failed to produce a further reduction in the frequency of the spike discharges. 3 Clonazepam (0.5 mg), frusemide (40 mg) and lorazepam (2 mg) appeared to be as effective as diazepam in reducing spike discharges, although the effect of the latter drug seemed to develop with some delay. 4 Mexiletine (100 mg) and (+)-propranolol (50 mg) did not reduce paroxysmal EEG discharges, their effects being not significantly different from that of saline. 5 EEG spike counting after intravenous administration of single test-doses appeared to be both a simple and promising technique for the rapid preliminary evaluation of the clinical efficacy of new antiepileptic drugs.