Summary and Conclusions Attention is directed to the difficulties we have experienced in the propagation by either natural or artificial means of various strains of Plasmodium falciparum. Difficulties arising during the course of natural propagation are manifested by the failure of all or a part of the inoculations to take, even though performed by demonstrably infected mosquitoes. In view of the dissection criteria employed to check inoculations, it would not appear that the gametocyte level in the patient from whom the mosquitoes were infected has been a factor. Neither would it appear that refractoriness of the patients inoculated has been a frequent obstacle, since takes have commonly resulted on the reinoculation of such patients. An increasing proportion of failures is associated with increasing age of the sporozoites present in the mosquitoes, a characteristic of which we were already aware. This deterioration is more rapid than is the case with vivax sporozoites. However, since many failures occurred when fresh or relatively fresh sporozoites were used, this cannot be the sole factor involved. A further factor elicited results from the comparatively recent ingestion of quinine by the patient on whom the mosquitoes were infected. It has not prevented completion of the sporogonous cycle, but nevertheless it has impaired the vitality of the resulting sporozoites. When certain of these strains were propagated artificially, gametogeny deteriorated suddenly or gradually. This change has been observed in strains propagated in colored patients. Its occurrence does not appear to be consequent on a heterologous immunity in the patients, since it has more commonly been observed in those presumed to be susceptible. This change does not appear related either to the administration of quinine to the patient or to its amount. When once initiated it appears to persist throughout subsequent passages, and we have not been able to overcome this deterioration. Failure of gametogeny in strains of Plasmodium vivax which have been artificially propagated for extended periods in the application of malaria therapy has been reported. We succeeded in acquiring two strains with such a reputed characteristic, and, while these were being propagated on our service abundant gametocytes were detected, mosquitoes were infected, and infection was transmitted therewith. Our experience with this species of parasite leads us to suspect that reports of failure of gametogeny indicates lack of familiarity with the morphology of the vivax gametocytes on the part of the microscopists concerned. On the other hand, the morphology of the gametocytes of Plasmodium falciparum is sufficiently distinctive to render credible any reports of their presence or absence.