Objective: The authors determined the roles of the physician and the patient in melanoma recurrence detection. Methods: The University of Alabama Melanoma Registry, consisting of 1475 patients surgically treated for cutaneous melanoma from 1958 to 1984, was searched to find 195 evaluable cases of melanoma recurrence. Patients were grouped by the type of return visit Group I returned on a previously determined date, whereas group II returned before the scheduled visit. Results: Symptoms of recurrence were present in 90% of group I patients and 93% of group II and correlated with the site of recurrence in more than two thirds of cases recurrence sites were local, regional, and distant in 35%, 31%, and 29% of group I, respectively, and 42%, 25%, and 29% of group II. The median interval to recurrence was 24.2 months in group I and 37.7 months in group II (p = 0.059). Median overall survival was 57 months in group I and 62 months in Group II (p = 0.210). Conclusions: Symptoms are present in 90% of the patients with recurrent melanoma and accurately predict the site of recurrence. Overall survival is not affected by the type of patient return visit.