1. The Cornell Medical Index Health Questionnaire was administered to a randomly chosen sample of 47 adolescent Indian boarding school students in an attempt to study the validity of this instrument in detecting psychiatric disturbances in this selected adolescent population.2. Statistically significant differences in responses to CMI were found in the "normal" and "abnormal" groups of subjects; thus confirming previous statements of the value of CMI in detecting psychiatric disturbances, and extending its applicability to Indian beneficiaries.3. A question of the validity of the CMI in detecting psychiatric disturbances in Alaskan adolescents (Eskimos) is raised and deserves further study.Statistical data are presented for results of administration to 365 boarding school subjects of the Cornell Medical Index Health Questionnaire. The subjects were Navajo Indians and Alaska Natives (mostly Eskimos) attending a Federal boarding school. It was found that age, sex, academic level, recent boarding school experience or chronic illness did not affect the response to the CMI.The mean number of responses for the entire group of "Indian" adolescents making up this study was 29.6. On the basis of frequency distribution it might be stated that less than 62 "yes" responses in the Alaskan, and less than 85 "yes" responses in the Navajo would fall statistically within the "normal" range (mean plus 2 S.D.).