To test the hypothesis that police would arrest disruptive but nondangerous individuals to expedite their removal from the community, the outcomes of all incidents involving a mentally ill individual during a 5-month period in the 1975-1979 study time frame were reviewed. Data showed that mental-illness-related incidents increased 227.6 percent from 1975 to 1979, whereas felonies increased only 5.6 percent and the total number of incidents excluding these categories decreased by 9 percent. A total of 62 (29 percent) of the 248 incidents involving mentally ill individuals neither met the dangerousness criteria of the State's Mental Health Procedures Act nor fell within the penal category. Another 62 calls did meet the Act's dangerousness criteria and involved individuals for whom emergency commitment proceedings had been initiated, committed patients who had eloped from the State hospital, and self-destructive and assaultive individuals. The police referred all but 1 of the 10 individuals charged with violence against persons to the local mental health center for evaluation. The remaining 92 incidents (42 percent) involved mentally ill individuals whose behavior constituted grounds for arrest on such charges as disorderly conduct, terroristic threats, harassment, defiant trespassing, or indecent exposure. The data do not support the hypothesis that the noncommittable mentally ill are being arrested and jailed as an expedient means of removing them from the community. Interviewed police officers were found to believe that mentally ill individuals should not be held responsible for minor criminal offenses. Tabular data and eight footnotes are provided.