less than 37 Becquerels (1 nCi) 137¢5, we can conclude that the detection efficiency of the field system is less than that of the whole body counter at the BNL Medical Department. However, once the activity of an individual significantly exceeds the minimum sensitivity of the field system, the agreement between the results from the two systems is within the 2 sigma counting statistic error. This is seen from the body potassium measurements. Finally, two Marshallese subjects were counted for quality control purposes. The first person was a recount to determine the ex~ pected variability from counting an individual more than once. teplicate count was within 2% of the initial count. Marshallese subject counted was from Rongelap Atoll. The The second This individual's 13705 result (11 kBq or 291 nCi) compared well with his previous 13705 result in April 1977 (14 kBq or 371 nCi) (CO 77). The difference of 22% is close to that which would be predicted from the 12% yearly decrease in the Rongelap population. RESULTS Tables 3 and 4 present a list of adult individuals who were counted in 1974 (CO 75), 1977 (CO 77), 1978 and 1979. There is a gen~ eral increase in the body burdens of adult males from 1974 to 1977 by a factor of 13.3, and from 1977 to 1978 by a factor of 1.8. The gen- eral increase for adult females from 1977 to 1978 was slightly higher than that for males over the same period. In most cases, the 1979 data are significantly lower than the 1978 data with an average reduc- tion in the 137 Cs body burden by a factor of 2.9.