sampling in March 1978. Both immediate follow-up periods extended through September 1979, but a "final'’ long-term sampling follow-up is planned for November 1980, possibly as a prelude to institution of long-term parasitesuppressive efforts. The treatment plan for the two atolls, though not theoretically ideal, was based on logistical realities as to what would be attainable in the context of health surveys and physician visits. Following initial stool surveys, treatment with each agent was to be given at quarterly intervals to all inhabi- . tants of the respective atolls except pregnant women and children aged <2 years. Follow-up surveys were to be conducted just before the third quarterly treatment and 9 months thereafter. On Rongelap the schedule was maintained, but ship damage caused schedule alterations during the eradication/suppression attempt on Utirik which resulted in some variation from the treatment/sampling plan. The actual schedules attained are shown in Table l. Sampling (S) and treatment (T) schedules on Rongelap and Utirik Atolls, 1977-1979. PHHiwAt | S - I 4) - 4 - rimi March June September i January wa 1979: — September rretbtArAsA June I January March myn 1978: June September (mht 1977: Utirik l Ronge lap mt Table 1. Sampling. To the extent that they were in any way reachable, all inhabitants of Rongelap Atoll were prevailed upon to submit stool specimens at each sampling period. In spite of strong fecal taboos, samples were obtained from at least 90% .of the population by diligent follow-up of dispensed stool containers. The larger population of Utirik (4300) precluded the possibility of total sampling during the limited survey period. Instead, a cohort of 20+% of the population was followed to the extent possible, and was used as an index of the population as a whole. If cohort members were not present on the island during a survey (away at school, on a visit, etc.), replacements were picked at random to fill out the three categories: adult males, adult fe- males, and children. Again, diligent follow-up of dispensed containers provided the requisite population samples. Processing of Specimens. An aliquot of 0.5 to 1 g of each specimen returned each day was put into an individually identified Il-dram vial we - 37 -