The relationships between the Fletcher and Manchester methods in dosimetry of cancer of the uterine cervix were explored in order to compare dose prescriptions between the systems, to optimize the best features of both systems, and to establish the necessity for computerized dosimetry. A total of 91 Fletcher-Suit radium applications was analyzed by a linear least-square regression analysis to compare Point A and Point B doses of the Manchester system will milligram-hours of the fletcher system. Although moderately high correlations were found between milligram-hours of radium and doses at Point A and Point B, it was concluded that direct comparisons, particularly between individual patients, are fraught with dose uncertainties of clinical significance. In addition, the correlation between milligram-hours of radium and Point A dose was markedly affected by the position of the colpostats and tandem, thus making it difficult to formulate a simple conversion factor between the two systems.