28. Ina paper published in 1956 Sacher (S2) proposed a model wherebz all individuals in a population are assumed to be initiallz identical. However, with time, owing to fluctuations in the phzsiological state of the animals and to external stresses, there would be a progressive change and a disperSion in the phzsiological state of the population. When a fluctuation of large amplitude in the homeostasis of an animal takes place, death will ensue. By a mathematical description of a physiologic fluctuation process, it is possible to derive an approximate relationship between the rate of mortality and the mean physiologic state of the population having the form of a linear function between the logarithm of the mortality rate and the mean physiologic state of the population at any given age. 29. A more analytical presentation of this theory and of the derived func- tions can also be found in other papers [S4, S12]. In a later contribution [S13] attempts were made to interpret lethality in terms of very simple cell population kinetic, without building into the models known parameters of cell kinetics like maturation time or feed-back control of self-renewing systems. Other features of cell Kinetics, with special regard to lengthening of the generation cycle upon continuous irradiation, were also discussed in another paper [S5| as possible causes for a cumulative lesion related to the life- shortening effect. Cytogenetic injury due to rearrangements of chromosomes has also been considered to account for radiation-induced life-span-shortening (14). 30. A critical comparison between the model of Blair and that of Sacher is contained in Sacher and Grahn [S4]. The latter model assumes linearity and additivity functions formally equivalent to those contained in the former, al. though Blair postulated the existence of only one component of recovery from injury, operating soon after exposure and causing injury to fade away expo- nentially. When the equations derived by Blair were fitted to the cumulant lethality function of Sacher the fit failed, owing to significant systematic deviations. In addition, the mean recovery time estimated from Sacher's data would be 4 to 10 times longer than the recovery value of 5 days accepted by Blair on the basis of fractionation experiments. Grahn [S4] For these reasons Sacher and rejected the central assumption of Blair referring to the single linear component of recoverable injury. All the remaining assumptions of Blair are included in the more generalized formulations of Sacher: radiation injury is proportional to dose and to dose-rate; the recovery rate from this injury