inventories of imported foods are expected to be such that the total absence of imported foods from the diet is most unlikely. A few general conclusions can be drawn from evaluating all of the available data on dietary habits in the Marshall Islands. 1. Coconut consumptionis the major source of 137Cs intake in the diet model; the diet model does predict the 157Cs body burden observed in actual whole-body counting of the adult population for two atolls. Consequently, the 137Cs intake in the model is very close to reality —— at least at these atolls. 2. The dietary habits are] to a degree, atoll-specific and should be generplized from one atoll to another only when supporting atoil- specific data are unavailable. 3. There is still some u ainty as to what an average diet really is at any dtoll. 4. Many factors can affect the average diet over any specific year. 5. Further atoll-specific djetary data are needed to improve the precisign of the dose assessmentfor each resettlement fituation. Dose Methodology Although the selection of this particular time distribution is arbifrary, general discussions with Marshallefe people and observations made while we Have been in the To predict the effective dose to a population on Rongelap Island, we calculated both the potential external and internal effective dose from the available data and information. The sources of exposure and methods of calculation are different for external and internal exposure. islands make the selection reasbnable. The external exposure ratPs in 4.R h-! are converted to equivalent. dosq rates in tissue using a factor of 0.0075 Sv peF Roentgen (0.75 rem per Roentgen) and assumirgg a quality factor of 1.0 for gamma radiation (WNSCEAR, 1972; ICRP, 1973; ICRU, 1985). Seferal researchers have evaluated the conversgon of exposure doses in air to absorbed dose specific organs (Kerr, 1980; O'Brien and Sanga, 1976). These conversion factors range from (10049 to 0.0075 Sv per Roentgen (0.49 to 0.75 regh per Roentgen), depending on the organ. Wefhave chosen the External Exposure Estimates of external exposure include both gamma and beta radiation. The method of calculation for each is described below. Gamma Radiation The external exposure calculations for gamma radiation are based on measurements made on Rongelap Island in 1978 and 1988, and decay corrected to 1995. The following arbitrary distribution of time was used to develop the average external exposure for 157Cs for a 1995 resettlement: 1. Nine h d-! are spent in the house where the exposure rate is 0.83 .R h-! (see Table 2). 2. Six h d-! around the house and village area where the exposure rate is assumedto be 2.0 pR h-! (weighted average of outside house and general village sites). 3. Seven h d-! in the interior region of the island where the average exposureis 3.0 uR h-! (Tipton and Meibaum, 1981). 4. Two h d-! on the beach or lagoon where the exposure is 0.089 .R h-l, based on EG&G data (Tipton and Meibaum, 1981). conversion factor for testes, dne of the higher factors, and used it for the whdle body and bone marrow. A result of this choice is that the whole-body doses listed in this report can be used to estimate genetic effect’ based on gonad dose. Based on the conversiqn factor of 0.0075 Sv in tissue per Roentgen exposure in air, the conversion factor to mSv yj! from pR h-! is 0.066. The resultant contributions of 137Cs to the annual average effective dosp in the first year of occupancy of various island areas described in the above scenario are: 1. Inside houses—0.022 vy (2.2 mrem y71). 2. Elsewhere in the houging and village area—0.031 mSv y-! (B.1 mrem y—}), 24