The biochemist and the anthropologist look at native food problems from very different points of view, but it seems clear that research on African diets will be of scientific value only if there is co-operation between the two. Without chemical analysis it is impossible to know the composition of any particular native diet. Without a knowledge of the customs and language of any tribal area the necessary facts cannot be collected. The present article describes an attempt at co-operation along these lines, based on a study of the diet of a Northern Rhodesian tribe.