Highly dilute solutions of the alkaline oleates, which do not suffice to alter appreciably the morphology or reproductive power of the pneumococci, nevertheless produce profound changes in their structure. Pneumococci treated with sodium oleate become more subject to autolysis, as is indicated both by the rapidity and the perfection of the process of self-digestion, and at the same time they become subject to serum-lysis. The serum-lysis of the soaped pneumococci tends to be incomplete with normal serum and to be perfect with an immune antipneumococcus serum. When normal serum is employed, the surviving pneumococci subsequently multiply either in the test tube or in the animal body, in the latter case producing fatal infection. When, on the other hand, an immune serum is employed, lysis is complete, no multiplication occurs, the test tube mixture is sterilized, and the inoculated animal is protected from infection. The inhibition of their activity which the soaps ordinarily suffer in the presence of protein, can be prevented by the addition of an appropriate quantity of boric acid, so that suitable mixtures of serum, soap, and boric acid can continue to exert a deleterious and solvent influence on the pneumococci, and the effect is greater when immune serum is employed in the mixtures. Infection can not only be prevented when the mixture of immune serum, soap, and boric acid is added to the pneumococci before injection into the peritoneal cavity of small animals, but the infection can likewise be prevented when a therapeutic injection of a mixture of the three substances mentioned is made to follow the inoculation of normal, highly virulent diplococci. The limits of the activity of the therapeutic mixture are determined, in part by the amount of protein to be overcome, and in part by the peculiarities of the infection occurring in highly susceptible animal species. The virulence of the pneumococci is somewhat diminished by the soap treatment, but the treated organisms are not rendered more subject to phagocytosis. It would appear that the action of the soap is exerted upon the lipoidal moiety of the bacterial cells, through which they are rendered more pervious to serum constituents and brought under their deleterious and dissolving influence. The changes in the pneuocci here described probably have a prototype in the resolving exudate of a pneumonic process; so that it may be considered that they occur in the animal body in the course of spontaneous infection and constitute one of the conditions of the conquest of the organism by the body's forces. A certain conformity exists between the manner of destruction of the pneumococci in a pneumonic exudate and that in the artificially prepared soap-serum mixtures. In order to imitate outside of the body the conditions of the removal of pneumococci within the body, it does not suffice merely to study the reactions to leucocytes and serum and to conclude from these reactions the means which the body employs for the disposal of the diplococci; but it is necessary to invoke still other factors, among which are the effects of chemical substances present in exudates, of which the soaps represent one class. The failure hitherto to unify the reactions in test tubes with those occurring in the body in connection with the pneumococcus may be due to the fact that account was not taken of this class of chemical bodies. Whether the principle here presented can be made applicable to the treatment of local pneumococcus infections in human beings is a pressing question. Its application to the treatment of local infections, to the seats of which the serum, soap, and boric acid mixtures can be directly applied, particularly after evacuation of an inflammatory exudate, seems to offer promise.