Metastasectomy in the treatment of solitary metastases has been recommended almost unanimously. The basis for this recommendation has been that 5-year survival after metastasectomy is around 30%, which is just as good as after operation for bronchial carcinoma. It has been assumed, implied, or claimed that the 5-year survival without operation is nil. Control material is, however, lacking. Seventy surgically treated patients were compared with a small, historical control group of 12 patients. There was no difference in 5-year survival. Because of these findings and after a study of the literature, we postulate that patients with lung metastases fulfilling the criteria for operation constitute a selected group with a favorable natural history. Five-year survival, therefore, is an insufficient way of describing the effect of metastasectomy. However, patients with 10-year survival are rare in the literature. In some patients with a favorable tumor-host relationship or with possibilities for effective chemotherapy, cure or prolongation of life has been achieved. Such patients should undergo operation. Randomized studies are needed in all groups for which we do not have sufficiently strong evidence that metastasectomy contributes to the longevity of the patient.