The effects of oral administration of 3 doses of ethanol (0.75, 1.0 and 1.5 g/kg) and single doses of morphine (2.5 mg/kg) and diazepam (2.5 mg/kg) on spectral components of the EEG were investigated in the squirrel monkey Saimiri sciureus. Comparisons of mean spectral power in the 4 frequency bands (2-4 Hz, 4-8 Hz, 8-16 Hz, 16-32 Hz) revealed that the lowest dose of ethanol (0.75 g/kg) produced some increase in power in all 4 frequency bands with significant increases in the 8-16 Hz band. Morphine also produced some increase in power in all bands; however, significant increases were prominent in the 16-32 Hz range. Moderate doses of ethanol (1.0, 1.5 g/kg) produced dose-dependent decreases in the lower frequencies which were significant in the 2-4 Hz band. Diazepam also produced significant decreases in the lower frequencies (2-4, 4-8 Hz) but also produced significant decreases in 8-16 Hz and increases in the 16-32 Hz band. The distribution of these power changes as a function of time (power spectral band time series) was also studied. Evaluation of fluctuations in the 8-16 Hz band which is a measure of the periodic shifts in alertness demonstrated that ethanol and diazepam slowed the dominant frequency of the time series; whereas, morphine caused an increase in the frequency. This study suggests that two EEG measures, mean EEG spectral profiles and power spectral band time series analysis, can successfully discriminate the electrophysiological actions of ethanol, and that these markers may provide a means for uncovering ethanol's mechanisms of action.