The effect of therapy with immune serum has been studied in thirty-two cases of Plasmodium circumflexum infection, all of them produced by blood inoculation. Eighteen of these cases never showed parasites, and seven others developed infections which were definitely milder than those of the controls. The therapeutic serum was in all cases obtained from chronic cases which had previously been superinfected to raise the immune titre. It seems justifiable to conclude that: 1. Passive immunity can be conferred in avian malaria, at least when caused by Plasmodium circumflexum just as it can be in certain types of monkey malaria, and perhaps in human malaria as well. 2. Whatever the nature of the protective substances present in the serum of chronic cases may be, they are present in very low concentration. Their concentration can be raised by superinfection, however. These substances may be strain-specific or species-specific, but the results of these experiments do not give any clear-cut answer to this question. 3. Serum therapy previous to infection seems to be more effective than when given afterward. 4. The administration of normal serum or even of physiological saline in a dosage comparable to that employed with the immune serum used in these experiments produced similar macroscopic changes in the size of the spleen. 5. Agglutination of cells parasitized by Plasmodium circumflexurn when mixed with immune serum was observed.