a Soil radionuclide concentration log-probability plots were con~ structed ff.4 es ft ih a for 99sr, 137¢¢, 239+240p,, | and 24lam at profile depths 0-15 cn, 15-25 cm, 25-40 cm, and 40-60 cm using concentrations in pCi/gm. (Kate), Kidrinen (Lucy), Bokenelap (Mary), Elle (Nancy), Aej (Olive), Aomon (Sally), Bijire (Tilda) Lojwa (Ursula), Alembel (Vera), and Billae We evaluated 70sr and 13’cs concentrations in coconut meat and fluid, and papaya meat from Eneu Island (Bikini Atoll) using the same soil computational algorithms. Our analyses showed the lognormal probability density assumption to be correct for data sets having a majority of concentration above the MDA. The arithretic mean is an adequate estimator compared to the mini- Mum variance estimator, particularly when the coefficient of variance is less than two--this includes 96 percent of the analyzed deta sets. importantly, More this method evaluates the proportion of measurements less than the mean, x, and 3x (Tables 45-47). The analyzed soil and food data 3X values includes an average of 95 + 3.5% of the sample (range 68-100 percent) measurements. estimate doses that These ™ anc 3% céncentrations are then used to include a known fraction of possible measurements; for xX, more than 64 percent of the measurements are included; and for 3™, 95 percent arc included. Calculations for Alternate Dietary and Time Variations There is always an interest in developing dose estimates for living patterns, and option within living patterns, which are not developed in the paper. An enormous number of options could be synthesized and it is - 5o0yt 12 42 - RK, he uZ. oy: yt) & We examined the islands of Enjebi (Janet) and its four quadrants, Mijikadrek (Wilma). FI