A relationship between sudden infant death syndrome and the use of phenothiazine-containing medications is suspected. In order to investigate the influence of phenothiazines upon cardiorespiratory and sleep characteristics, four normal infants (mean age 15.8 weeks) had polygraphic recordings made for one night. The recordings were repeated three nights later, after the daily administration of a usual therapeutic dose of a standard phenothiazine syrup (promethazine, 1 mg/kg of body weight per day). The infants were monitored for six days after the first treatment. Findings from the first polygraphic recordings were normal for each infant. The second recordings disclosed an increase in sleep time, a reduction in the number and duration of awakenings, an increase in non-rapid eye movement (NREM) III sleep state, and a reduction in body movements. During the second night, each infant showed an increase in the number of central apneas, and each infant also had several episodes of obstructive apneas (median duration four seconds). It is concluded that promethazine depresses the arousal and respiratory mechanisms in normal infants during sleep. This observation reinforces the opinion that CNS depressants should be avoided in infants less than 1 year of age; CNS depressants could result in sudden death in apnea-prone infants.