Therefore, this attempt to seek estimates from the islanders themselves concerning the actual amounts of local foods in thelr contemporary diet should be used not as a definitive answer to the question of what constitutes the "typical average." Rather it should be regarded as a feasibility study on the In the authors' possibility of obtaining the desired information in this way. judgement, the averages obtained from the interview study represent itil such time as further study They should be so considere overestimates. proves them accurate or (more likely) provig:s representative estimates of food sharing and wastage, which could be folded into the study to provide more accurate consumption estimates. Until such time as the factors involved are more (horoughiy understood, the feasibility of obtaining a "typical average" estimate from the interview method is in question. However, the present study establishes an upper limit, which has been confirmed by (a) an estimate of the calorie intake based on calorie value of foods jl, 2), and (b) the quantity of .ood that is available and is gathered on the islands. Results The data obtained from the interviews and observations made by the authors since 1970 suggests that the diet patterns can be divided into three typical categories or communities. -haracteristics?: These communities have the following Community A: a. Maximum availability of local foods b. Highly depressed local economy - living within income provided by c. Low population d. Little or no ability to purchase imported food selling copra Community B: a. Low availability of local foods ~ except fish (which can form as much as 33% of the total diet as a result of excellent fishing in the area). b. Overpopulated - resulting in low per capita availability of local foods. c. A good supply of imported foods (supply boat comes in every two to three weeks) along with the availability of jobs. Community C: a. Low availability of local foods, even the fishing is poor b. Large government food program