A retrospective study of patients with osteosarcoma was undertaken to determine whether there was a relationship between biopsy and survival. Fifty-seven patients treated at the Karolinska Hospital, Stockholm, between 1938 and 1959 were included in this study, all of whom were less than thirty years old, had a metaphysial osteosarcoma in a long bone but had no pulmonary metastases at the time of diagnosis; all were treated by amputation. No clinical variants of osteosarcoma were included. Twenty-four of the fifty-seven patients had an amputation without a prior biopsy; the others had biopsies before amputation. These two groups were fairly closely matched in age, sex, site and size of tumour, and in the level of amputation; some patients in each group received radiation before operation. Evaluation of these two groups of patients revealed that the performance of a biopsy, with or without a delay of not more than thirty days between the biopsy and the definitive operation, had no adverse effect on survival.