Malaria incidence and prevalence surveys were performed from December, 1967 to February, 1969 among the indigenous Nilotic inhabitants of Gambela, a small administrative centre in the western lowlands of Ethiopia. Entomological data suggested that malaria transmission was seasonal and this was consistent with monthly P. falciparum parasite rates. Monthly P. malariae parasite rates, however, were consistent with an hypothesis of homogeneity. The age-specific incidence of quartan malaria among 26 children zero to 11 years old at the start of study was examined at 28 day intervals over a 15-month period. The resulting data suggested that parasite acquisition was a slow process and an annual P. malariae incidence of 0.17 was derived. This statistic was supported by studies performed five years later: The incidence of P. malariae among 102 infants followed from birth up to 48 months of age was 0.16-0.20. An attempt was then made to account for the prevalence of P. malariae in terms of the entomological conditions observed in Gambela. Macdonald's formula for the sporozoite rate was used to derive hypothetical relative proportions of P. falciparum and P. malariae among the observed sporozoite-positive mosquito populations. About 4% of the sporozoite challenges were estimated to be of P. malariae. An hypothetical annual entomological P. malariae inoculation rate was then made by multiplying the number of observed sporozoite inoculations per person (approximately 10/year) by the proportion of them estimated to be of P. malariae. The annual P. malariae sporozoite challenge was thus estimated at 0.4 per person, in good agreement with the annual incidence estimates from parasite rates in children.