different from the incidences in other world populations, which suggests that there is nothing unusual in the environment of the Marshall Islands that might predispose to thyroid abnormalities. The numbers of thyroid abnormalities detected in the various exposed Marshallese groups were roughly related to the calculated thyroid doses in the groups, i.e., greatest in the Rongelap, less in the Ailingnae, and least in the Utirik group, as indicated in Table 1 and Figure 3. However, because of the small numbers of people involved and uncertainties of the doses received, the data do not lend themselves to dose-response analysis. Numerous animal studies have shown the association between radiation exposure and the later development of thyroid neoplasia (143-145,198,199). There are increasing reports of the development of thyroid abnormalities in children many years after therapeutic irradiation of the head and neck (with thyroid exposure) (146-171,200-203). The incidence of thyroid tumors was increased in the survivors of the atomic bombings of Japan (84-86,172). Re- ports of tumorigenic effects of radioiodines are more limited in man (173- 180), and !3ly has considerably less effect than x-radiation and other penetrating radiations in this regard. The few studies of late thyroid effects following diagnostic use of 13ly have indicated generally negative findings (175,176).* Besides the Marshallese exposed in 1954, the only other people that may possibly have had a similar type of exposure were those residing in the Nevada-Utah area during the Nevada testing program. Children living in this area were later examined for thyroid abnormalities. The thyroid dose in these children has been estimated at about 18 rem (maximum 120 rem) (181). The amount of short-lived iodine isotopes involved is not known. No increase in the incidence of thyroid abnormalities was detected in these children compared with unexposed children in Arizona, also examined in that study. Several points should be made regarding the thyroid dose estimates in the Marshallese children. Table 9 (see also Appendix IV, Table 4) shows that radiation risk estimates for the development of thyroid nodularities or carcinoma in the Marshallese approximate those for populations exposed to x- radiation and for Japanese atomic bomb survivors. The internal absorption of radioiodines before evacuation of the Marshallese from their home islands was a most important source of radiation exposure to the thyroid gland. On the basis of known facts regarding risk, the estimated doses of gamma radiation alone would not have been sufficient to induce the observed findings. It is known that 13l_ is considerably less effective in producing thyroid abnormalities than x-radiation (183-185,187), and, since it has been demonstrated that the shorter-lived isotopes of iodine are more destructive to the thyroid than 1317 (188-192) because of greater penetration of their beta rays and faster dose rate, it seems likely that their presence played an important role in thyroid injury and may partly account for the unexpectedly higher risk estimates than would be expected from 13ly alone. However, several points should be repeated. The dose estimates in the Marshallese are only approximate, and *Doses from diagnostic use of 131] vary from about 10 to 200 rads ~77- (194).