One hundred and eighty-two Cocopah Indians, aged five years and over and representing 81 per cent of the study population, were given a 75 gm. glucose equivalent load and determinations were made two hours later of plasma glucose levels. Seventeen per cent of the population at the time were either receiving hypoglycemic therapy or had plasma glucose levels of at least 160 mg. per 100 ml., which was considered indicative of “diabetes.” In those aged thirty-five years and over, the prevalence was 34 per cent, similar to that reported in the Cherokee but lower than that in the Pima Indians. The frequency of obesity in those aged fifteen and over was higher in the Cocopah than in the Pima Indians but, as the prevalence of “diabetes” was lower in the Cocopah, the frequency of obesity did not appear to account for the difference in the prevalences of diabetes between these tribes.