We compared the diagnostic accuracy (sensitivity and specificity) of the Geriatric Depression Scale (GDS) and the Mini-Mult Depression Scale (Mini-D) in a sample of 87 geriatric medical outpatients who were classified as depressed or non-depressed with the DSM-III-based Symptom Checklist for Major Depressive Disorders (SCMDD). In addition, we evaluated the relationship between GDS and Mini-D depression classifications and performances on three tests of the overall cognitive functioning. Although GDS and Mini-D classifications were in moderate agreement with those of the SCMDD (71%), the former measure produced more false-positive errors and the latter produced more false-negative errors. Because neither the GDS nor the Mini-D affords entirely satisfactory diagnostic accuracy, appreciation of these operating characteristics will enable practitioners to select the instrument that yields the most acceptable balance of Type I and Type II errors within their particular clinical settings. Although participants demonstrated signs of a mild cognitive compromise, no relationship was noted between depression classification and overall cognitive functioning. We believe that this finding reflects our dichotomous, rather than continuous, operationalization of “depression”.