The renewed application of fast neutrons in clinical radiotherapy has been stimulated by fundamental radiobiological findings. The biological effects of high LET radiation, including fast neutrons, are different from those obtained with X-rays in at least three respects: the oxygen enhancement ratio, the sensitivity of cells at different phases of the cell cycle, and the contribution of sublethal damage to cell reproductive death. Furthermore, wide variations in relative biological effectiveness (RBE) have been observed for different tumors and normal tissues. Measurements of volume changes in human pulmonary metastases indicate that the RBE for slowly growing tumors which are generally well-differentiated is higher than that for poorly differentiated lesions. Six thousand patients have now been treated with fast neutron beams. The results of the clinical applications vary according to the method of application and to the type of cancer involved: treatment of inoperable malignancies of the salivary gland is very encouraging; the therapeutic gain is rather small for bladder and rectal cancers, soft tissue sarcomas and advanced carcinomas of the cervix; the responses of brain tumors are very disappointing. Most neutron radiotherapy applications have been less than optimal because of inadequate physical and technical conditions. Despite these difficulties, some interesting clinical data have become available. Due to the technical shortcomings, the possible advantages of fast neutrons are probably underestimated for many tumor sites. Well-designed clinical trials, preferably performed with high energy cyclotrons in clinical environments, will provide a decisive answer to the question of the usefulness of the new radiation modality.