It is commonly supposed that radioactive isotopes decay in a strictly exponential way, so that the process can be characterized by a half-life; and that the half-life depends only on the isotope, not being influenced at all by surroundings. Now both of these assumptions are challenged: it is questionable whether the decay is always strictly exponential, and there is evidence to show that in some cases at least the decay may be influenced by the surroundings, or by something else external to the nuclei. The importance of this possibility in trying to establish ages with the use of carbon 14 is obvious; and the question is of first-rate importance to physics generally.