Summary and Conclusions 1. Normal gametogenesis in two species of bird malaria, P. cathemerium and P. relictum var. matutinum was studied. Criteria for distinguishing pre-gametocytes and micro- and macro-gametocytes based on cytological characteristics are offered for both species. 2. Gametocyte production in these two species which have a strict quotidian periodicity of the asexual cycle, was found to have a periodicity also. Pregametocytes are more prevalent at the segmentation time of asexual parasites, the numbers are augmented periodically, and new broods are distinguishable. The process of maturation required more than 24 hours, usually 30 to 36 hours. 3. When the time of segmentation of the asexual forms was shifted by changes in the schedule of light and darkness, so that it occurred 12 hours later, the appearance of new broods of gametocytes coincided with the changed segmentation time. When the synchronism was disturbed by prolonged periods of light and darkness, gametocyte production became irregular. 4. Observations made over a 24 hour period indicated that when mature gametocytes are most plentiful, few gametocytes are found, and vice versa. 5. When gametocytes are mature the ratio of the two sexes is more nearly equal. When gametocytes are prevalent most of them appear to be pre-macrogametocytes but these may be undifferentiated forms. 6. The mean number of merozoites per segmenter in this strain of P. cathemerium was 12.95 ± .17 in 1935, and 13.72 ± .16 in 1936. 7. The initial occurrence of gametocytes in P. cathemerium was studied by injections of whole blood, whole blood that was allowed to stand at room temperature, blood meals from mosquitoes and intravenous injections of sporozoites. Gametocytes often appeared simultaneously with the asexual forms, and increased in numbers as the asexual forms increased. 8. Long continued passage by subinoculation tended to decrease the numbers of gametocytes in P. cathemerium. Injections from birds whose infections had shown a high percentage of gametocytes and had been latent over a month or more always yielded subsequent infections with a large number of gametocytes. 9. Mosquito transmission for three complete generations of the parasite did not increase the percentage of gametocytes. 10. Further observtions were made on a strain of Plasmodium cathemerium which does not produce gametocytes. It differs from the parent strain in that it has a “ragged” appearance, lacks pronounced quotidian synchronism and has a smaller number of merozoites per segmenter. 11. The atypical strain failed to show gametocytes when inoculated into its natural host, the English sparrow. 12. Attempts to infect susceptible mosquitoes with the atypical strain and to transfer by mosquitoes were unsuccessful. 13. The atypical strain showed a complete cross immunity to the typical strain and had no inimical effect on the typical strain in mixed infections. 14. Serum from birds with infections of the atypical strain when mixed with parasites of the typical strain had no effect on gametocyte production. Serum from birds with typical infections did not stimulate gametocyte production when mixed with parasites lacking gametocytes. 15. Infection with the atypical strain was still present but latent in two birds after a period of three and one-half years respectively. Subinoculations from these birds resulted in infections that continued to show atypical characteristics. 16. Two birds used in cross immunity experiments were alive after three years, and were reexamined and found to have undergone no changes in the type of infection which they had.