A prospective study of 582 pediatric cranial computed tomographic (CT) examinations was made in order to determine the efficacy and safety of two sedation regimens and general anesthesia. Two hundred seventy-nine outpatients were randomly given oral chloral hydrate (80 mg/kg) or an intramuscular preparation composed of atropine, meperidine, promethazine, and secobarbital (AMPS). Three hundred three inpatients were randomly administered chloral hydrate, the AMPS, or endotracheal anesthesia. Intravenous supplementation of the sedation was limited by the protocol to a maximum secobarbital dosage of 2 mg/kg. An additional retrospective analysis was made of 316 scan attempts obtained outside the study during the same period. A failed CT examination was defined as one that could not be completed or one that showed motion artifacts on two or more scan pairs after two scan pair repeats. The failure rate was 15% for the chloral hydrate group and 12% for the AMPS group. No motion artifacts were present on 94% of the completed studies. There was no mortality, but major and minor complications occurred in 3.5% of the randomized group. It was concluded that each of the methods had proved acceptably safe and effective and that measures can be taken to decrease complications and sedation failures further.