This article investigates the extent to which mentally ill jail detainees are treated while they are in the custody of the criminal justice system. A random sample of 728 subjects were administered the National Institute of Mental Health Diagnostic Interview Schedule (NIMH-DIS) during jail intake and then followed up throughout their stay in jail. Of those subjects who met criteria for severe mental illness (psychosis or major affective disorder), only one third were given treatment within 1 week of intake. Log-linear analysis revealed that treatment decisions were influenced by treatment history (the strongest predictor), the type of mental disorder (schizophrenia vs. depression), type of crime, and whether symptoms were documented by jail intake personnel.