To provide quantitative data on the risk of thyroid cancer after exposure to 131I, 34,104 patients administered 131I for diagnostic purposes were followed for up to 40 years. The mean thyroid dose was estimated as 1.1 Gy, and 67 thyroid cancers occurred in contrast to 49.7 expected (standardized incidence ratio = 1.35; 95% confidence interval 1.05-1.71). Excess cancers were apparent only among patients referred because of a suspected thyroid tumor, and no increased risk was seen among those referred for other reasons. Further, risk was not related to radiation dose to the thyroid gland, time since exposure or age at exposure. The slight excess of thyroid cancer thus appeared to be due to the underlying thyroid condition and not radiation exposure. Among those under age 20 years when 131I was administered, a small excess risk (3 cancers compared to 1.8 expected) was about 2-10 times lower than that predicted from data for the A-bomb survivors. These data suggest that protraction of dose may result in a lower risk than an acute X-ray exposure of the same total dose.