The validity of the Durie and Salmon's clinical staging system for multiple myeloma has been tested in 81 consecutive patients studied at the Istituto Nazionale Tumori of Milan from January 1970 to June 1982. Median survival from diagnosis was 48 months for stage I, 41 months for stage II and 23 months for stage III (P = 0.02). Median survival of patients with normal renal function (A) was 35 months and of those with abnormal kidney function (B) 7 months. Almost all early deaths were observed in patients with stage III disease associated with renal failure. No statistically significant difference was found in the median survival in patients with kappa and those with lambda light chains. The analysis of survival according to the three main combinations of chemotherapy used in this study (melphalan-prednisone vs melphalan-procarbazine-prednisone vs adriamycin-prednisone) could not disclose any significant difference. Prognosis was, however, closely related to the response to combination chemotherapy: median survival was 72 months in responders, 36 months in patients with improvement and 25 months in non-responders (P less than 0.01). A lower incidence of response was obtained in patients with stage III myeloma compared to patients with stage I-II. The myeloma staging system used in this study is simple to employ and allows identification of truly comparable patient groups in the evaluation of therapeutic results. Our therapeutic results confirm the effectiveness of melphalan plus prednisone and fail to demonstrate any advantage in the administration of adriamycin as first-line therapy.