THE destructive effects of radiation have been studied for 80 yr. Most techniques involve looking at the surviving cells, which tend to be the more resistant cells of the tissue. On the assumption that the results are representative of all cells in the tissue, many conclusions have been drawn. On the other hand, Cheng and Leblond have used tritiated thymidine (³HTdR) to kill cells synthesising DNA in the crypts of the small intestine¹. Two surprising features of their experiments have provoked little comment. First, very low doses (40–50 µCi per mouse) of ³HTdR caused measurable cell killing and second, the killing (evident from the presence of labelled apoptotic-like² phagosomes¹) was not random throughout the crypt but occurred selectively at the crypt base where relatively few cells are in S (refs 3 and 4) and where the stem cells are presumably located^1,3–5. I report here that the presence of hypersensitive cells at the base of the crypts can be demonstrated after whole-body X or γ irradiation, and to describe the time sequence for the production and loss of these killed cells together with their dose-response relationship.