[en] The long-term risks induced by radiation are of much concern to patients and clinicians alike. As an example, perceived radiation risks are frequently cited in a woman's decision to choose a radical mastectomy over lumpectomy + radiation. In consequence, the actual radiation risks are often considerably overstated, or unreasonably downplayed. In this lecture we will discuss just what is known about the long term risks following radiotherapy, both from the human experience and from the laboratory. We will discuss risks both to the patient and to radiotherapy personnel. A good deal is known about the carcinogenic effects of high and low doses of radiation, in large part thanks to the careful study of the survivors of the atomic bombing in Japan, as well as studies of individuals exposed to medical x rays. It is possible to make an estimate, which is probably good to within a factor of, perhaps, three to five, of the cancer risks faced by a patient of a particular age and sex who is going to undergo a particular radiotherapeutic regimen. It is also possible to make an estimate of the risks faced by radiotherapy and nursing staff exposed to low doses. Brachytherapy related risk estimates are likely to be somewhat more uncertain, due to the poorly known sparing effects of the low dose rates used; for the radiotherapy personnel in brachytherapy, because of the doses which can be received, the risks can be quite significant. A recent complication in external-beam radiotherapy is the advent of high-energy linacs, which can produce a significant fast neutron dose which, dose for dose, may be ten to fifty times more carcinogenic than gamma rays. Data relating to the risks of hereditary effects of radiation come almost entirely from laboratory experiments in animals. Studies involving several million mice form the basis of most of our current understanding of hereditary effects. The results of these studies indicate that radiation is a relatively inefficient mutagen. The first generation of offspring of survivors of Hiroshima and Nagasaki show no statistically significant increase in hereditary effects, but the numbers are small, and not inconsistent with risk estimates based on animal data. These data will be discussed, and we will try to conclude just what can currently be reasonably said about hereditary risks to a young patient facing the radiotherapy decision. Finally, risks from prenatal radiation exposure are now of little relevance to radiotherapy patients, but are certainly of considerable relevance to radiotherapy personnel, particularly those involved in brachytherapy. The developing embryo and fetus are extremely radiosensitive and, at certain stages of gestation, risk estimates far exceed those for carcinogenic or hereditary effects