In interstitial implantations of linear radioactive sources, such as radium needles, it is usual for the radiotherapist to base the treatment upon an estimate of a single parameter called the effective minimum dose. At best, the dose to a few selected In interstitial implantations of linear radioactive sources, such as radium needles, it is usual for the radiotherapist to base the treatment upon an estimate of a single parameter called the effective minimum dose. At best, the dose to a few selected points may be calculated. By means of an automatic computer it is possible to produce a complete isodose distribution in a cross-sectional plane of an implantation, providing the same quality of information about the dose distribution as is normally available for external beam radiation treatments. The following steps, which can be taken by relatively unskilled personnel, provide a precise calculation of dose in a plane through the implantation. 1. The cross-sectional geometry of the implantation is obtained by transverse tomography (1), as is shown in Figure 1. A co-ordinate scale is overlayed upon the radiograph to obtain locations of the needles. 2. The dose, based upon the type and location of the needles, is calculated to points on a l-mm. grid by a digital computer. 3. Listing of the dose points by the printer is controlled so that the printed doses bear the correct (but enlarged) geometrical relation to each other when the printed sheets are pasted together to produce a map of the implantation. 4. The isodose contours are drawn on the map of the implantation, and a photographic reduction is made for inclusion in the patient's chart. Thus far, steps 1 and 2 have been accomplished as planned, but steps 3 and 4 have been done by more laborious hand methods to produce distributions such as are shown in Figure 2. The IBM RAMAC computer in this institution requires approximately ten hours of computation time for an average calculation. It is planned to use a faster computer, such as the IBM 1620, which can do the same calculation in approximately five minutes. By hand, the computation would require several man-weeks. It is possible that the increased knowledge of dose distribution promised by this method, together with clinical insight, could result in substantial improvement in interstitial treatment. To gain knowledge on this point, the dose distributions in previously treated cases will be calculated and compared to the known effectiveness of the treatment. Particularly, those cases will be investigated in which the clinical results were better or worse than anticipated. Acknowledgments: It is a pleasure to acknowledge the aid given in programming by Miss Mary Edith Horrocks and Mr. David Harbin of the International Business Machines Corporation.