It has been shown by Keynes (5) that survival rates equivalent to those obtainable by radical surgical methods can be obtained by radiological methods in the treatment of primary, operable cancer of the breast. This plies, however, only to interstitial radiation and not to external radiation technics. The result is that the present methods of external radiation of cancer of the breast are based largely upon the limitations imposed by the tolerance of normal tissues, especially the skin, rather than upon a more rational basis. This concept of normal tissue tolerance itself cannot, at present, be defined in terms of physical or other constants. An analysis of Keynes' interstitial radiation technic in terms of the roentgen may serve, in lieu of statistical survival data, to place the technic of x-ray treatment of cancer of the breast upon a more rational basis. Such an analysis is now valid for the roentgen has been adopted provisionally as the unit of quantity not only for x-rays but also for gamma rays (10). The intensity of radiation in terms of this unit has been found by numerous observers (2, 3, 4, 7, II, 12) to be in the region of 8.5 roentgens the milligram element hour at a distance of 1 cm. from a point source filtered by 0.5 mm. of platinum. This value, 8.5 r/mg.-hr.-cm., will be used in this analysis. The radium-bearing needles employed by Keynes were constructed of platinum with a wall thickness of 0.5 mm. Pt and no correction for this factor is necessary. The theory that doses of 200 kv. roentgen and gamma rays, which are equivalent in terms of the physical constant, the roentgen, are equivalent in terms of biological reactions is not accepted by all. It should be emphasized, however, that results published by those who do not accept this theory are not consistent. This applies especially to the value in roentgens of the threshold erythema dose. This dose, as determined by Quimby (9), is 525 r (measured in air) for 200 kv. x-rays filtered by 0.5 mm. Cu and 1.0 mm. Al delivered at the rate of 60 r /rnin. to a 10 × 10 cm. field, 50 cm. distant from the tube target. This dose as estimated by Failla (1) for gamma rays is 1,010 r which is considerably higher. If millicuries in radon seeds required to deliver specified doses to masses of various diameters, then the results are again different. For example, the millicuries required to deliver 1 T.E.D. to masses of various diameters is: But the T.E.D. in terms of the roentgen, using the value 8.5 mg.-hr.-cm., will be, for the values listed above: These values are consistent neither with Failla's value of 1,010 r nor with the inverse square law.