The number of drugs which have been proved to sensitize cells to ionizing radiation is limited. Clinically, differential sensitization must be obtained in which the drug, when combined with irradiation, results in a greater effect on the tumor than on the normal tissues. No such therapeutic effect has been observed with available sensitizers. The halogenated pyrimidines, 5-bromodeoxyuridine (BUDR) (3) and 5-iododeoxyuridine (IUDR) (5), have been shown to sensitize cells grown in tissue culture to ionizing radiation. IUDR was found to markedly increase the effectiveness of ionizing radiation in patients (2, 10). In those subjects in whom relatively low-energy x rays were used, however, a marked effect was also observed on the skin. A suitable measure of the relative reaction of the skin as compared to tumor response was not available, so proof is still lacking that IUDR represents a suitable clinical radiation-sensitizer. Other drugs such as 5-fluorouracil (5-FU) are thought by some (8) to sensitize human tumors to ionizing radiation; other workers believe that the effect is purely additive (6,14). While these attempts to establish a successful sensitizing drug to be used concomitantly with radiation continue, the possibility is explored of improving the clinical result by employing drugs and radiation in suitable sequence as previously suggested (10). This method does not require that a drug be a true sensitizer. Instead, it takes advantage of the information which has become available to us through the laboratory and certain clinical studies. Tissue-culture studies show that a small number of cells require a smaller total x-ray dose for complete destruction than a larger group of cells. Small tumor masses in patients require smaller doses of radiation for complete regression than larger tumors. Two factors are operative. One is based on the work of Elkind and Sutton (4) which demonstrates that a given dose of radiation, the mean lethal dose, will reduce a tissue-culture cell population to 50 per cent of its original value. Each succeeding mean lethal dose reduces the number of cells remaining from the preceding dose by 50 per cent. Till and McCulloch (13) have shown that a similar mean lethal dose exists in mice with respect to the bone marrow. The other factor is that smaller tumors are less apt to have areas protected by relative anoxia (12). Methotrexate reduces the size of a wide variety of head and neck cancers (11). One can expect from 35 to 50 per cent of such tumors to respond, but, following a period of shrinkage, the neoplasms regrow. This drug is toxic to the mucous membranes of the alimentary tract, particularly the oral mucosa on which a pseudodiphtheritic membrane and ulceration develop in severe instances. There is also an effect on the hematopoietic system, and pancytopenia may result with toxic doses.