Specific results on the surgical resection of a large number of pulmonary metastases (PM) are currently unavailable, and the risk-benefit ratio of this aggressive approach may appear questionable. A systematic review of the records of 456 adult patients who underwent thoracic surgery for PM between 1979 and 1990 led to the identification of 44 patients who underwent at least one resection of eight or more PM (range eight to 110), of whom 33 (75%) had PM from osteogenic or soft tissue sarcoma. These 44 patients underwent a total of 77 operations, of which 47 (61%) were bilateral and nine (12%) incomplete resections. The 3- and 5-yr probabilities of survival after the first resection of eight or more PM were 36 and 28%, respectively, and were not significantly different from those of the 412 other patients who underwent surgery for PM over the same period. In this small group of patients, only the quality of resection (complete or incomplete) was found to be a highly significant prognostic factor (p < 0.01). A critical analysis of the reported data supports the view that, at least in patients with osteogenic or soft tissue sarcoma, the prognostic value of the number of PM seems to be more dependent on associated resectability than on the number per se and that, after careful preoperative patient selection, PM that can be resected should be resected, whatever their number.