8. The more recent report of Walburg [W1] is essentially a critique of the concept of non-specific life-shortening, at least at the low doses of practical interest. Walburg came to the conclusion that life-shortening effects after irradiation may principally or perhaps exclusively be explained by the induction or acceleration of neoplastic diseases. supported by Storer {S6!. This conclusion was also All authors recognized that at higher doses other mechanisms of death would be prevailing. B. 9. METHODOLOGY Life-shortening can only be assessed on the basis of death, an end-point that can be defined rather precisely in time. However, it is usually more informative to know the reasons why an animal dies than to identify with great prevision the time at which the event occurs. To ascertain the cause of death is often a difficult exercise and, in some cases, an impossible one since death is often the result of a variety of causes all acting jointly. This is particularly true as animals grow older [D1, A2, D2] because aging animals of all species (but particularly of the long-lived ones) die with multiple lesions, contrary to what happens early in life, where the cause of death is more often related to a single pathological entity. In old animals the number of possible causes of death increases and the primary or precipitating cause is difficult to diagnose. Also, most irradiated animals die of diseases which are unrelated to radiation exposure and this complicates the identification of the terminal pathological syndromes. Thus, multiple disease conditions, and interactions between diseases in the same animals should be correlated with parameters such as age and dose to provide a meaningful interpretation of the pathology at death. 10. The first difficulty with much of the work reviewed, particularly with the oldest contributions, is the lack of careful pathological observations on the animals at death, let alone the refined multifactorial analysis mentioned above. Many experimental series are therefore difficult to interpret, since the representation - as accurate as it maz be in time - of the life-span shortening, marks the complexity of the biological end-points. Another difficulty lies with the fact that even when good pathology is available, this is usually collected at death. Under these conditions it is impossible