73 posure under extreme conditions, with no corrective measures taken to reduce internal absorption of radioisotopes prior to evacuation of the exposed people from the fallout area. It did not involve the contaminated pasture-cattle-milk cycle, which might be an important pathway ofradioiodine to manin other types of accidents (such as the Windscale accident).!76 Civil Defense planning can provide for several measures that will reduce the hazard of thyroid exposure due to radioiodine absorption and thus largely preclude the degree of thyroid injury sus- tained by the Marshallese. Since the hazard from radioiodine is acute for a period only of days, early protective measures are extremely important. These include avoiding inhalation of radioiodines by shelter protection; consuming food and wateronly from closed containers: feeding cowsin contaminated areas protected fodder; and tem- porarily withholding contaminated milk supplies and diverting them into processed products with a shelf life longer than the life of the isotope. The addition of stable iodine to food or water during the first week woulld provide a relatively inexpensive method of reducing thyroid uptake of radioiodines by isotope dilution and saturation with non-radioactive iodine; this should rarely produce anyseriousside effects and would beof particular value in children and pregnant women. When exposure of the thyroid has already occurred, prophylactic treatment with thyroid hormone, now being used in the Marshallese, may help prevent developmentof thyroid tumors, and even after tumor development, surgical excision may reduce mortality due to malignancy. The thyroid effects in the Marshallese were not anticipated at the timeof the accidentor during the early years afterwards. In retrospectthis is not sur- prising, for several reasons. At that time the thyroid was thought to berelatively radioresistant, particularly with regard to radioiodine exposure (on the basis of animalstudies and diagnostic and therapeutic use of 131] in people), and the calculated thyroid doses in the Marshallese were considered to be below the levels likely to produce tumors. In addition, neither the importanceof the exposureto short-lived iodine isotopes in fallout nor the thyroid dose differential in children due to the smaller size of their gland was fully appreciated. It is quite likely that the final results of thyroid lesions in the Marshallese are incompleteat this time since new lesions are still occurring. The meanlatent period for radiation-induced thyroid tumors may be as long as 30 years. Cases have been seen as late as 40 years after exposure. Furthermore, on the basis of the present data the risk of developing radiation-induced thyroid neoplasia is probably underestimated,since surgical removal of potentially malignanttissues may have occurred and the hormonetreati:.ent may have inhibited the development of some tumors, althoughthelatter is questionable. As has been pointed out, also, the true carcinogenic potential of the exposure, particularly in the children receiving the higher dose, may have been masked byexcessive cellular destruction. The recent finding that subclinical thyroid deficiency is present in some of the exposed people who have not shown any thyroid abnormalities indicates that the thyroid effects in the Marshallese may not yet be completely manifest, and continued careful surveillance of this population is necessary. Acknowledgments Because of the broad scope of the surveys and the complexity of operations it would be impossible to acknowledge all the individuals and organizations that have contributed to the success of the surveys over the past 20 years. We would like, however, to express thanks to some whohavecontributed vitally in more recent years. Our foremost appreciation is to the people of Rongelap and Utirik who haveparticipated in the examinations, including the unexposed people comprising the control population. We treasure the friendship and trust of these people. We are grateful to many Rongelap and Utirik people for assisting the examining teams overthe years,including the Magistrates of both islands - at Rongelap: John Anjain, Billiet Edmond, Jabwe Jorju, and Niktimos Antak; at Utirik: Kabto Campus, Maja Lena, Harold, the school principal, and Joata, the health aide. Thesenior authoris most grateful to the many fine medical and otherspecialists (listed in Appendix 2) who have so willingly given of their time and talents, without remuneration, to participate in the surveys. Withouttheir dedicated efforts the surveys could never have succeeded. Heis also greatly indebted to the many technicalassistants who have contributed much to the success of the surveys, particularly to Mr. W. Scott and Mr. D.