This paper has examined the possibilities of applying significant pharmacologic help to a variety of psychiatric problems that may accompany narcotic addiction. It has been shown that many of the patients do have such difficulties, with affective disorders being most common. As far as the various psychotropic drugs are concerned, neuroleptics for schizophrenia and lithium for manic disorders are generally agreed upon. A more extensive trial of lithium in a variety of situations seems indicated. Minor tranquilizers for anxiety and MAO-inhibitors for depression are both seen as problematic in this population--the former because of the possibility of abuse, the latter because of the danger of drug interaction associated with the addict's careless lifestyle. Tricyclic antidepressants may clearly have a role in treating major depression in opiate addicts on or off methadone, but the lability of the syndrome over time with frequent spontaneous remission argues against their routine use until it is clear that depression has persisted 3-6 months into methadone. Disulfiram appears to be a useful adjunct for drug abusers with serious alcohol problems. Psychotropic agents are most helpful to opiate addicts when used to treat coexisting psychopathology. While there is no clear evidence that such agents will reduce or affect the addiction itself, they may help keep patients available for rehabilitation efforts. Failure to intervene may make treatment dropout and recidivism more likely. Given the relative frequency of potentially treatable psychiatric disorders in these patients and the consequences of undiagnosed and untreated conditions, it is important for clinicians to maintain a high index of suspicion for concomitant psychiatric illness and for programs to have a mechanism for routinely diagnosing either all patients or, at a minimum, all patients not doing well. If programs used a standard instrument such as the SADS, it would be possible to compare various programs on this factor; in addition, it would provide a rich source of data for outcome studies.