Mental hospitalization is an important part of the national picture ofbospitalization. Total inpatient days for mental disorders, and their proportion of total hospital days for all disorders, is a statistic of considerable national import, yet there has never been a complete description oftotal inpatient daysfor mental disorders. The authors present data from the National Center for Health Statistics for nine years (data for only two years bad been published previously), and correct them, and national totals, I or the previous exclusion of data f rom residential treatment centers and community mental health centers. From 1969 through 1 978, inpatient days for mental disorders fell from 168 million to 95 million. This decrease was more than accounted for by decreasedpsycbiatric inpatient days at two sites: state mental hospitals and Veterans Administration psychiatric hospitals. The number of psychiatric inpatient days at all other sites increased during this time penod. The authors also note a decrease in total inpatient days for all disorders over the decade. However, 85 percent of the total national decrease was accounted f or by the decrease in psychiatric inpatient days. The authors stress the implications for public policy and the need for a more adequate national data base.