From 1963 to 1993, 890 patients were treated with 3518 cast gold restorations by students and postgraduate dentists. The longevity of these restorations was studied retrospectively using the patient files. Longevity was calculated using the method described by Kaplan and Meier. After the observation period, 111 (3.2%) of the examined restorations were not in place anymore. The most frequent reasons for failure were caries (33.7%), lack of retention (32.7%), endodontic treatment (29.6%), insufficient marginal adaptation (3.1%) and extraction (1%). The cumulative survival rate and a 95% interval of confidence was calculated for all restorations and for each of the locations and surfaces included in the trial. The 10-year survival rate for occlusal inlays was, 76.1% (12.1) for MO inlays 88.3% (4.2), for DO inlays 83.4% (4.6), for MOD inlays 87.5% (2.4), for partial crowns 86.1% (3.3) and 85.7% (1.7) for all restorations. Based on the statistical method used, the cast gold restorations demonstrated satisfactory longevity results.