a memorandum provided insight into such things as the overall survey procedure, the estimated uncertainties in some reported values, the preferences in preparation and consumption of many food items, and the can conversion data for some food items (grams of food per 12-oz can). a Ujelang for local foods when imports are available and unavailable for adult males (18 to 80 y); adult females (18 to 78 y); and children in the 0- through 3-, 4- through I1-, and 12Results for imported foods (normal conditions only) are summarized in Tables 8 through 10. The maximum diet (adult female) from the MLSC survey was also used to estimate doses for Enewetak and Bikini Atolls. In the summary of a survey conducted during July and August of 1967 at Majuro Atoll, the average coconut use was reported to be approximately 0.5 coconut per day per person.-° This included young drinking coconuts, old nuts used for grated meat and pressed for small volumes of milk, and sprouting nuts used for the sweet, soft core. Recent data from Eneu Island shows that an average drinking coconut contains 325 mL of fluid (standard deviation = 125 mL) so that even if the entire average coconut use of 0.5/d were all drinking nuts, the average intake would be about 160 g/d. This is in agreement with the results from the MLSC survey at Ujelang. The recent BNL report on dietary information on Rongelap, Utirik, and Ailuk was developed by the authors from personal observations while living with the Marshallese and from answers to questionnaires.7° The observations and questionnaires were directed more toward estimating the food prepared for a family rather than the amount of food actually consuméd. Because food is shared and some food prepared is fed to pigs or chickens, these two are not necessarily the same. In the draft report the authors state: This attempt then to seek estimates from the islanders themselves concerning the actual amounts of local foods in the contemporary diet should be used not as an answer to the question of what constitutes the "typical average" but rather as a feasibility study on the possibility of obtaining the desired information in this way. We feel the averages which we obtained from the interview study are for one reason or another consistently overestimated and should be considered maximum estimates or overestimates until such time as further study proves them accurate or (more likely) provides average factors for food sharing and wasting which can be folded ms the study to provide more accurate, reduced estimates* (Ref. 20). The diet patterns are divided into three categories representing three types of communities. * Underlined for emphasis. 3¥ 3 a “S| Tables 3 through 7 summarize the dietary intake results from the MLSC survey at through !7-y age ranges, respectively. i 20 : x3 4| z te A €