The influence of triazolam on cardiac and respiratory activity of healthy male subjects was examined during nights disturbed by airplane noises and during undisturbed nights. Twenty-four subjects, divided into three groups of eight, slept in the laboratory for 7 nights (N0-N6). Following a double blind design, group A (control group) received a placebo every night. Group B received 0.25 mg triazolam and group C received 0.5 mg on nights N3, N4 and N5. On the other nights, they received a placebo. For all three groups, the nights N0, N3 and N5 were disturbed by 32 semi-randomly distributed airplane noises. Air and wall temperatures (20 degrees C) and air humidity (10 degrees C, 52%) were kept constant. Sleep measures, heart rate and respiratory rate were continuously recorded. Results showed that the largest dose of hypnotic drug produced an increase in tonic heart rate in the first part of each night throughout the treatment period (N3, N4, N5). When compared to baseline disturbed night N0, the phasic cardiac response to the noises was significantly attenuated on only the 1st treatment night (N3). Triazolam had no significant effect on nocturnal respiratory rate. No after-effects of the drug were observed for cardiac and respiratory activity on the withdrawal night (N6). The results suggest that, with regard to the drug action, there was either an increase in arousal threshold or a dissociation between long-lasting and short-lasting modifications of heart rate. Contrary to the single night attenuation of phasic cardiac responses, there was no drug tolerance for the hypnotic-related increase in tonic heart rate.