1 month could be assumed, similar conditions of exposure in man might be expected to cause a reduction of the mean after survival time of 9 to 10 days /rad of gamma-ray exposure. 232. Grahn and Sacher [G1] based their extrapolations on the linearity of the log mean after-survival in days as a function of daily dose in rad or R, for mean after-survival of 25 per cent or more of the control values. This linear trend of the mean after-survival had been shown previously to hold for the mouse [S4] (see Figure XV). The coefficient of this regression is a species constant [G9] and reflects the days of life lost per rad or the fraction lost per day. It may therefore allow, when two points referring to a given species are Known, to define the slope of the curve applying to that species. The ratio of the life-shortening coefficients would be in a direct proportion to the ratio of life expectancies for non-irradiated populations or to the ratio of the agespecific mortality rate slopes. A radio susceptibility scale could thus be con- structed for man, dog, guinea-pig and mouse, on the basis of the slopes being in the ratio of 33:10:3:1 for life expectancies of 16500, 4000, 1400 and 500 days, respectively. It could therefore be deduced that the relative sensitivity of the various species (expressed in per cent life-shortening for exposure of 1 rad) Was approximately the same for all species and in the ratio of: guinea-pig = 1.8; dog = 1.25; and man = 1. mouse = 1; Similar conclusions were also drawn by Sacher [S23, S14]. 233. Grahn [G6] explored further the problem of inter-species comparisons, starting from the notion of the exponential decline of mean after-survival with daily dose, already mentioned repeatedly (see Figure XV). By comparing the vi- tal statistics of two selected populations of male men and mice, he established that the ratio of time scale to equate the two populations is 10 mouse days = 1 man year, or 1 mouse day = 36.5 man days, a factor slightly higher than he previously used (33:1) (20:1) [F4]. [G1] and much higher than that used by Failla and McClement In consideration of these other estimates Grahn selected for his calculations a ratio of 1:30 and established that the daily dose to induce a 20 per cent reduction of life expectancy in man would be 0.65 R/day, to be compared with a 19.4 R/day in the mouse. Calculations for man and for other spe- cies showed that the guinea-pig would be a relatively sensitive species in a framework defined by the mouse, dog and man. It should be pointed out that the life-span and radiosensitivity values used for these calculations are very much at variance with those in the 1968 paper by the same authors [G11 and therefore the relative sensitivity scales in the two papers do not correspond.