Three strains of Plasmodium falciparum, Vietnam Oak Knoll (FVO), Uganda Palo Alto (Hawaii) (FUP-H) and Uganda Palo Alto (Cayenne) (FUP-C), were examined in 154 Aotus nancymai monkeys as suitable models for testing blood-stage vaccines. The Vietnam Oak Knoll strain had the greatest number of animals with maximum parasite counts > 200,000/µl. Uniformity of the parasitemia curve increased from passage 4 to passage 6 with an accompanying decrease in the number of days required to reach maximum parasitemia or required treatment. The Uganda Palo Alto (Hawaii) strain was highly infectious, but many animals had extended prepatent periods and extended days to maximum parasitemia. The FUP-H strain would require a greater number of animals per test group to detect partial protection because of the greater number of low-density maximum parasite counts in control animals. The Uganda Palo Alto (Cayenne) strain was poorly adapted to intact A. nancymai. However, five of six splenectomized monkeys inoculated during passage 6 with 105 parasites had maximum parasite counts > 200,000/µl. For the testing of vaccines against primary parasitemia in the A. nancymai model system, the FVO at passage 4 level would appear preferable to passage 6 parasites following a challenge with 105 parasites. A similar pattern could be obtained using FUP-H if the challenge was 106 parasites. To measure immune memory against recrudescence or rechallenge infection, FUP-C at an early passage in splenectomized A. nancymai would appear to be the appropriate model.