within the 3 to 30 per cent range of life-shortening the days lost/rad at the various doses varied between 0.56 and 0.02 at progressively higher doses, with differences between male and female animals. With fast neutrons between about 20 and 50 per cent of life-shortening, the days lost/rad were between 0.21 and 0.03, again with oscillations between the two sexes. In these animals death was characteristically associated with neoplastic and degenerative diseases common to the natural aging, except for animals treated with high doses in which death was attributed to necrosis and aplasia of the lymphatic and haemopoietic tissues. The shape of the dose-survival curve (which shows with both neutrons and x rays) could conceivably be explained by difference in the effects responsible for life-shortening, in that not all effects which might contribute to earlier death are identical in dose-response relationships. Leukaemia and other neoplasms could not entirely account for life-shortening in this series of experiments. 69. Data on induction of neoplasia in the above-described experiments were reported in a paper by Upton, Randolph and Conklin et al. [U9]. There is no specific discussion in this paper about the relationships with life-span- shortening but some of the data (G6 irradiation at high dose rates for single doses of 100 and 300 rad) were reanalysed by Walburg !W1], on the basis of rather careful macroscopic examination of the animals at death. There was no significant difference between control and irradiated animals if all causes of death other than neoplasia were considered. But when all causes of death including tumours were analysed together the difference between control and irradiated mice became very significant. It could thus be concluded that there was no significant residual life-shortening when only the non-neoplastic causes of death were considered. This conclusion, which is partly at vari- ance with the conclusions of the authors themselves is to be attributed, in Walburg's view [W1], to the use of a more refined analysis of the lethality data. TO. Darden et al. [D1] also reported data on RF/Un female mice exposed to graded doses of 14 MeV neutrons (dose rate 1 - 2 rad/min). The mean age at death of animals surviving beyond 30 days decreased with increasing dose, with a maximum difference between control and irradiated animals being observed in the 400 rad group and amounting to 151 days of 27 per cent of the control life-span. Life-shortening was an approximately constant or slowly decreasing function of dose up to about 200 rad, but at higher doses the ef- ficiency/rad tended to decrease, as in the series by Upton [U7]. Tumour in-