Amitriptyline was prescribed for 100 depressed out-patients, randomly allocated to one of two groups. Patients in Group A were forewarned about side-effects and patients in Group B were not forewarned. None of the patients had, so far as could be ascertained, previously received amitriptyline. The patients were seen again after two weeks and questioned regarding side-effects and continuation of medication. Ninety-three patients were included in the final analysis. The results fail to confirm the hypotheses that forewarning patients of possible side-effects either causes a greater number of patients to complain of such side-effects or causes less frequent discontinuance of therapy. The results call for similar studies with other drugs.