STUDIES ON PNEUMOCOCCUS GROWTH INHIBITION : V. THE RELATION OF VIRULENCE TO THE PNEUMOCOCCIDAL ACTIVITY OF NORMAL RABBIT SERUM-LEUCOCYTE MIXTURES.

Employing a method, described in an earlier publication, for testing the pneumococcidal activity of serum-leucocyte mixtures, a study has been made of the pneumococcus-destroying properties of the blood of a relatively susceptible animal, the rabbit, for pneumococci of low virulence for the species. It was found that rabbit serum-leucocyte mixtures possessed the power to kill avirulent pneumococci in relatively large numbers but failed to inhibit the growth of virulent organisms even in minute quantities. The results of numerous experiments in which all three types of pneumococci were employed indicated that the ability of a strain of pneumococcus to grow in rabbit blood is dependent on its virulence for the rabbit. The extreme susceptibility of the very young rabbit to strains of pneumococcus of low virulence for the full grown animal, was found to be associated with an absence of pneumococcidal properties in the blood of the young rabbit. These findings suggest that the relatively susceptible animals possess the same type of defense mechanism against pneumococcus infection as do the highly pneumococcus-resistant species.     

STUDIES ON THE ENZYMES OF PNEUMOCOCCUS : II. LIPOLYTIC ENZYMES: ESTERASE.

1. Pneumococci contain an intracellular enzyme of marked lipolytic activity as measured by the acid liberated by its action on tributyrin. 2. Enzyme-containing solutions may be prepared by dissolving pneumococci in bile, or by extraction by other means. 3. The optimum reaction for maximum activity of the endolipase is about pH 7.8, which coincides with the optimum hydrogen ion concentration for growth of pneumococci. 4. Heating the enzyme for 10 minutes at 70°C. destroys its activity. 5. Attenuation of virulence of pneumococcus had no measureable effect on enzyme activity. 6. The possible relation of the endolipase to the mechanism of bile solubility is discussed.     

STUDIES ON NATURAL IMMUNITY TO PNEUMOCOCCUS TYPE III : III. CORRELATION OF THE BEHAVIOR IN VIVO OF PNEUMOCOCCI TYPE III VARYING IN THEIR VIRULENCE FOR RABBITS WITH CERTAIN DIFFERENCES OBSERVED IN VITRO

Among the experimental findings reported in this paper to which we wish to give particular emphasis are the following: 1. The results which follow the intravenous injection into rabbits of two strains of Pneumococcus Type III of different degrees of virulence vary with the state of the capsule. Thus when this structure is completely developed both remain in the blood. A culture of either strain begins to become susceptible to the blood-clearing mechanism contemporaneously with the onset of capsular degeneration and the initiation of other concomitant changes at the surface of the organism (cf Paper II), which occur much earlier with the less virulent strain. 2. When, in either case, removal from the blood stream occurs, this is effected by the phagocytic cells of the body. There is no suggestion that a new or unknown mechanism is involved. The greatest share of the burden is borne by the fixed phagocytic cells of the liver and spleen, and to a less extent by those of the lung and bone marrow. Nevertheless, it has been demonstrated that the polymorphonuclear leucocytes may also participate. 3. Phagocytosis by the leucocytes of the normal animal either in intro or in vivo has been observed only at such a time as the capsule has become impaired. Ingestion of the organisms by the fixed tissue cells appears also to be effective only under the same condition and is accordingly observed with much younger cultures of the less virulent strain. 4. Following their removal from the blood and their accumulation within the fixed phagocytes of the organs, destruction of most of the cocci proceeds within 2 to 4 hours. Both strains are destroyed provided they are in the state favorable to phagocytic attack. 5. Evidence has been presented which indicates that just as in vitro, so in a local area of inflammation within the body, aging with attendant capsular loss and increasing susceptibility to phagocytosis may take place. 6. With organisms from either strain a variable period of lag follows their injection into the blood stream, even when they are introduced in a state of active multiplication and complete encapsulation. 7. Differences in virulence for rabbits of two strains of Pneumococcus Type III do not imply that this animal possesses a defensive mechanism which is absent in other species, since it has been possible to demonstrate similar differences when the organisms are injected intravenously into mice. This fact indicates that the factors determining the degree of virulence of these strains are to be sought in the organisms themselves, rather than in the kind of host.     

STUDIES ON OXIDATION AND REDUCTION BY PNEUMOCOCCUS : II. THE PRODUCTION OF PEROXIDE BY STERILE EXTRACTS OF PNEUMOCOCCUS.

In the present paper methods have been described for the preparation of sterile extracts of pneumococci. These extracts may be obtained by dissolving the bacteria in broth cultures by means of bile, or by extraction of the cellular substances by repeated freezing and thawing of broth or saline suspensions of unwashed cells. Under special precautions these extracts may be passed through Berkefeld filters without loss of potency. In this procedure, as in all other manipulations incident to their preparation, the extracts should be protected as far as possible from contact with air. All extracts were proved sterile by cultural and animal tests. Sterile extracts of unwashed pneumococcus cells promptly form peroxide on exposure to air. Peroxide formation is almost as active in extracts aerated at 2°C. as in those exposed to the air at room temperature. Detectable amounts of peroxide may be produced by these cell extracts within the reaction range of pH 5 to 9, the optimal zone lying at reactions less acid than pH 6. The peroxide-forming activity of the extracts is gradually diminished by prolonged exposure to 55°C., and is completely destroyed by heating at 65°C. for 5 minutes. Cell extracts of pneumococci which have been thoroughly washed prior to extraction in salt or phosphate solutions exhibit no peroxide-forming activity. These extracts of washed cells may be activated by the addition of the cell washings, yeast extract, or muscle infusion.     

STUDIES ON OXIDATION AND REDUCTION BY PNEUMOCOCCUS : VI. THE OXIDATION OF ENZYMES IN STERILE EXTRACTS OF PNEUMOCOCCUS.

1. Certain enzymes of Pneumococcus are destroyed by oxidizing agents formed when sterile extracts of the cellular substances are exposed to air. The carbohydrate-hydrolyzing enzymes (sucrase, raffinase, inulase, and amylase) are the most easily inactivated under these conditions, although the bacteriolytic enzyme is also reduced in activity. Similar treatment is without effect upon the active concentration of pneumococcus lipase and peptonase. 2. The enzymes which are destroyed during the oxidation of unwashed cell extracts are themselves non-reactive with molecular oxygen. The reactions by which they are destroyed seem to represent oxidations of a type similar to those proposed in previous papers for the oxidation of hemotoxin and of hemoglobin. 3. A study has been made of the relative resistance of different pneumococcus enzymes to heat and to the action of hydrogen peroxide. 4. The various enzymes may be arranged in the same order of relative resistance whether the rating be made from the standpoint of resistance to heat or of resistance to oxidation. Nevertheless, it appears that by a proper regulation of conditions of oxidation, certain labile constituents of a mixture of cellular enzymes may be inactivated with less effect upon the activity of other constituents of the mixture than when inactivation is brought about by heat.     

STUDIES ON PNEUMOCOCCUS GROWTH INHIBITION : VI. THE SPECIFIC EFFECT OF PNEUMOCOCCUS SOLUBLE SUBSTANCE ON THE GROWTH OF PNEUMOCOCCI IN NORMAL SERUM-LEUCOCYTE MIXTURES.

By means of a method previously described for demonstrating the growth-inhibitory and pneumococcidal power of normal serum-leucocyte mixtures, a study was made to determine the action of pneumococcus soluble substance on the growth of strains of pneumococci avirulent for rabbits and cats. It was found that the presence of a very small amount of the purified soluble substance of the homologous type markedly altered the conditions in the mixture so that even a small number of avirulent pneumococci were enabled to grow in the serum and leucocytes of animals which possess the power to destroy ordinarily such pneumococci in relatively large numbers. The action of the soluble substance was shown to be highly specific to type. A Type II substance assisted the growth of only Pneumococcus Type II, likewise a Type III substance, the growth of Pneumococcus Type III only. Experiments involving broth filtrates of young pneumococcus cultures in place of the purified soluble substances gave similar results, thereby establishing further the identity of the purified substance with the precipitable substance originally described in the filtrates of pneumococcus cultures. The effect of the soluble specific substance of pneumococcus in the serum-leucocyte mixtures is, therefore, to confer on the avirulent pneumococcus of the homologous type the growth capacity of a virulent organism. The mechanism of the reaction is briefly discussed.     

STUDIES ON HOST FACTORS IN PNEUMOCOCCUS INFECTIONS : II. THE PROTECTIVE ACTION OF TYPE I ANTIPNEUMOCOCCUS SERUM IN RABBITS

The power of specific antipneumococcus serum to protect rabbits against infection with Type I Pneumococcus has been studied with reference to the capacity of the animal to utilize the specific antibodies. Under conditions ensuring relatively controlled genetic factors it was found that heavier animals and those with high white blood cell counts are much better able to utilize the passively conferred immune principles. The interrelationships of the extrinsic and intrinsic factors responsible for immunity have been discussed.     

STUDIES ON THE ENZYMES OF PNEUMOCOCCUS : I. PROTEOLYTIC ENZYMES.

1. Pneumococci contain an intracellular enzyme or enzymes which (a) hydrolyze to some extent intact protein and (b) hydrolyze with striking avidity peptones. The optimum reaction for hydrolysis is pH 7 to 7.8, which also represents the optimum for the growth of pneumococcus. For convenience the terms "protease" and "peptonase" have been used, but no assumption is made as to whether the two actions, proteolysis and peptolysis, are due to two separate enzymes or are two activities of the same enzyme. 2. Solutions of intracellular substance of comparable enzymic activity may be prepared by dissolving the bacteria in bile, in sodium choleate, or by mechanical and autolytic disintegration of the cell. 3. The rapidity with which peptone is hydrolyzed is proportional to the concentration of the enzyme. 4. Heating the enzyme for 10 minutes at 100°C. destroys its activity. 5. Increasing the acidity to pH 5, the acid death-point of pneumococcus, suspends activity but does not destroy the enzyme, for activity is restored by readjustment to pH 7.8. 6. Attenuation of virulence to 1/1,000,000 of the original virulence had no measurable quantitative effect on the enzyme activity.     

STUDIES ON PNEUMOCOCCUS INFECTION IN ANIMALS : SECOND PAPER: ACTION OF IMMUNE SERA ON PNEUMOCOCCUS INFECTION.

From the results of this study of the action of immune sera on pneumococcus infection it is evident that immune sera vary greatly in their curative value. Immune sera possess protective action, but protective action is not necessarily indicative of curative action. Treatment with the serum of normal rabbits may prolong the course of pneumococcus infection in the rabbit. This action, however, is slight and not always manifest. Sera from animals immunized with dead pneumococcus cells which had been washed free from their products, failed to exert materially greater curative action than normal sera. Sera from animals immunized with culture filtrates free from pneumococcus cells possessed, in some instances, a slight curative value, but often this curative action was not apparent. In animals actively immunized, however, the presence of an immunity to culture filtrates was readily demonstrated. In the immunity produced by injections of dead culture material the strength was not sufficiently exalted for the sera to possess a practical curative value. It was only after immunization with virulent living cultures that the blood serum acquired marked curative action. After pneumococcus infection in the rabbit had become established, treatment with this serum induced crisis and cured the animals. From the results of the study of the mechanism of recovery it is evident that, despite the fact that virulent pneumococci are singularly insusceptible to the action of immune sera in the test-tube, pneumococcus infection nevertheless conforms to the general law of infection. Diphtheria and tetanus organisms give rise to powerful toxins, but the parasitism of these organisms is slight and their development is localized. Diseases produced by these organisms are toxemias and neutralization of their toxins by antitoxin puts an end to the disease. The pneumococcus gives rise to toxic substances which are less active or are active only in the body tissues, but the parasitism of this organism is marked and its development is rarely localized. Nevertheless, the manifestations of the disease arise from the action of the bacterial poisons on the tissues. The neutralization of the pneumococcus poisons by immune serum puts an end to the symptoms of the disease, but the pneumococci survive as harmless parasites until destroyed by lysis or phagocytosis. The neutralization of the pneumococcus poison may take place suddenly and completely as in crises; or, it may be incomplete with exacerbations of infection, as in lysis. Crisis, as it occurs in the lobar pneumonia of man and in the bacteriemia of the rabbit, is simply one phase of recovery, and recovery does not differ fundamentally, whether it is sudden and complete as in crisis, or incomplete and prolonged as in lysis, or whether the pneumococci are destroyed by lysis extracellularly as in the rabbit, or intracellularly as in the phagocytosis of the dog and man. Since the recovery of animals from pneumococcus infection differs in no essential from that of man, since the unaided protective mechanism of man as compared with that of susceptible animals is exceptionally efficient, and since it is possible by treatment with sera from animals highly immunized with living cultures of virulent pneumococci to cure pneumococcus infection in the most susceptible animals, it is difficult to conceive of the infection in man failing to yield similarly to the administration of such sera.     

FURTHER OBSERVATIONS ON THE USE OF PNEUMOCOCCUS EXTRACTS IN EFFECTING TRANSFORMATION OF TYPE IN VITRO

Pneumococcus extracts highly active in inducing the in vitro transformation of the specific types of Pneumococcus have been prepared by dissolving S cells with sodium desoxycholate, precipitating the dissolved material in alcohol in which the bile salt remains soluble, and extracting the precipitate in salt solution. Further purification of these active extracts has been attained by the removal of considerable inactive material by charcoal adsorption and by reprecipitation of the adsorbed extract in alcohol or acetone. The importance of using young cultures for extraction, and of preventing autolysis during the preparation of the extracts, is emphasized. Extracts prepared by the method described have been filtered through Berkefeld Candles (V, N, and W) without appreciable loss in activity, provided the reaction of the extract was slightly alkaline at the time of filtration. The purified and filtered extracts are water-clear, and sterile by rigid cultural and animal tests. They have been heated to temperatures of 60°C. for 30 minutes without appreciable loss in their capacity to induce specific changes in type. And although they have generally shown definite decrease in potency after heating to temperatures above 80° C., some extracts have been found active even after an exposure of 10 minutes to a temperature of 90°C. They have been completely inactivated by boiling. Relatively small amounts of extract have been effective when added to a broth medium containing normal serum or serous fluid. In this medium, R pneumococci, irrespective of their type derivation, have developed and thereafter retained all the type-specific characteristics of the encapsulated S cells from which the extract was prepared. The specific action of the extracts is discussed with reference to their transforming and antigenic properties.     

STUDIES ON OXIDATION AND REDUCTION BY PNEUMOCOCCUS : VIII. NATURE OF THE OXIDATION-REDUCTION SYSTEMS IN STERILE PNEUMOCOCCUS EXTRACTS.

1. The systems responsible for oxidation and reduction in the pneumococcus cell have been shown to consist of two components, the nature of which has been discussed. 2. The oxidizing and reducing power of active extracts of pneumococci is inactivated by exposure to air. This loss in activity is due to an inactivation of the thermolabile cellular component consequent on secondary reactions analogous to those which under similar conditions destroy the hemotoxin and enzymes in pneumococcus extracts.     

STUDIES UPON THE EFFECT OF LIGHT ON BLOOD AND TISSUE CELLS : II. THE ACTION OF LIGHT ON ERYTHROCYTES IN VITRO.

1. It was found that when hanging drops of whole blood, drawn from a rabbit, were subjected to irradiation from certain light sources, a striking degeneration of the white and of the red cells occurred. In this degeneration of the red cells there was (a) a preliminary period of 15–30 minutes during which no effect was noted; (b) following this, there was a period during which the cells swelled and became first almost biplanar, then biconvex, and finally spherical. After a spherical shape had been assumed, the liberation of blood pigment from the cells began. This process progressed with varied rapidity, sometimes being very slow, and being incomplete; sometimes being of almost explosive rapidity. This liberation was not accomplished in either case through a gross rupture of the cells, but through the agency of some submicroscopic change produced either in the cell contents, the cell wall, or both. Following this liberation of pigment, in some cultures the red cells showed signs of coagulation, as did the white cells. In other cultures, however, the red cells were reduced to achromatic shadows. 2. When whole blood was diluted with unbuffered Locke solution and then irradiated in hanging drop preparations, the erythrocytes swelled, as in whole blood, but then, instead of an almost instantaneous hemolysis of every cell present, or a slow liberation of hemoglobin from each individual cell, all the red cells hemolyzed, one or two at a time, the hemolyzed cells being left as achromatic shadows. 3. This hemolysis of the red cells, as in the case of the degeneration of the white cells, occurred upon irradiation of the culture with white light, or with light lying in each of the three spectral zones of the visual spectrum, defined by Wratten filters Nos. 45, 58 and 29 respectively, as follows: (a) 430µµ–550µµ; infra-red; (b) 475µµ–630µµ; 690µµ-infra-red; (c) 600µµ-infra-red. 4. Within the range of intensities of light employed, there was little or no difference in the rate at which light of these three regions acted on the red cells. Further, this rate was the same as was the rate when the red cells were irradiated with no colored filter interposed in the optical path, or when a 5 per cent total transmission neutral filter was interposed. 5. It was shown that this degeneration of the red cells under the action of light was not dependent on the presence or absence of serum, with the possible exception of that trace which might have been adherent to the red cells even after repeated washing. 6. The white cells in the culture generally showed no changes until after traces of red cell pigment could be seen free in the surrounding medium. 7. On the other hand, no neutrophils were observed moving around for longer than a few minutes after the liberation of red cell pigment had occurred. 8. The liberation of substances from the red blood cells, as a result of irradiation of such preparations, may play some major rôle in the degeneration of irradiated white cells.     

STUDIES ON THE SOMATIC C POLYSACCHARIDE OF PNEUMOCOCCUS : II. THE PRECIPITATION REACTION IN ANIMALS WITH EXPERIMENTALLY INDUCED PNEUMOCOCCIC INFECTION

The capacity of the serum of rabbits following intradermal pneumococcus infections to precipitate in the presence of pneumococcus C polysaccharide has been studied during the resultant periods of active infection and during recovery. In rabbits infected with Type I, III, or VIII pneumococci, large hemorrhagic lesions are produced which frequently bring about death of the animals after a febrile illness of 3 to 4 days. Repeated precipitation tests with the sera of these animals have been uniformly and consistently negative, not only during the acute illness but in the recovery period as well. On the other hand, the sera of monkeys of the Macacus cynomolgos species actively ill with experimental Type III pneumonia have been shown to react in precipitation tests with the C substance. The serum reaction appears within the first 24 hours after infection, remains positive in high titer for 2 to 3 days during the acute illness, and disappears with the onset of recovery. The precipitation reaction with C also occurs with the sera of monkeys following intradermal and intraperitoneal infection with pneumococci. The results of precipitation tests of the serum of monkeys during experimental pneumonia are similar to those obtained with the sera of patients suffering from pneumococcus lobar pneumonia. From the results of these studies it would appear improbable that the demonstration of the serum precipitation phenomenon with C polysaccharide in monkeys, and possibly also in man, is conditioned by previous exposure to pneumococcus antigen.     

STUDIES ON THE BACTERIOPHAGE OF D'HERELLE : VI. ON THE VIRULENCE OF THE OVERGROWTH IN THE LYSED CULTURES OF BACILLUS PESTIS CAVI? (M. T. II).

Resistants isolated from the overgrowth of cultures of B. pestis caviæ (M. T. II) lysed by various strains of specific bacteriophage proved to be avirulent when administered to mice by feeding, or by intraperitoneal injection. These cultures remained resistant to the action of bacteriophage so long as they were carried on agar. When transferred to broth, however, one group of resistants, namely, those isolated by means of "weak" phages, became susceptible to lysis after five to seven daily passages. The other group of resistants, isolated from the cultures lysed by one of the "strong" phages, failed to become susceptible to lysis even after nearly 200 passages in broth. Simultaneously with the recovery of susceptibility, the cultures of the first group regained a degree of virulence comparable to that of the parent culture of B. pestis caviæ. The cultures of the second group of resistants have failed thus far to recover virulence (10 months after isolation). The latter cultures, apart from lack of both virulence and susceptibility to lysis, are identical with the parent culture of B. pestis caviæ, as indicated by biochemical and antigenic properties. Our findings offer evidence in favor of the view that resistant strains result from selection among variants already existing in the parent culture and do not arise through the inheritance of specific immunity properties produced by the action of phage.     

STUDIES ON IMMUNITY TO PNEUMOCOCCUS MUCOSUS (TYPE III) : II. THE INFECTIVITY OF TYPE III PNEUMOCOCCUS FOR RABBITS.

The observations recorded in this paper on the infectivity of Type III pneumococci for rabbits may be summarized as follows: 1. Of eleven strains of Type III isolated from human sources, ten were found to possess low virulence for rabbits. This was true despite the fact that all the strains tested possessed large capsules and a high degree of virulence for mice. 2. One strain of Type III pneumococcus was rendered highly virulent for rabbits. Since it possessed no other biological property demonstrably different from the other strains, its virulence must reside in some additional property. 3. An initial decrease in the number of circulating organisms following the injection of virulent bacteria is a well known occurrence, and it was observed in rabbits injected with the rabbit virulent strain of Type III. However, the extent of the reduction was in inverse proportion to the degree of virulence of the strain; a fact which makes mechanical explanations of the phenomenon insufficient. 4. The bacteremia produced in rabbits by Type III pneumococci, avirulent for this species, runs a characteristic course. It differs from that produced by non-encapsulated R forms of pneumococci although in both instances survival of the infected animal ensues. This is evidence that the mechanism of resistance against encapsulated and non-encapsulated pneumococci is not identical. 5. Phagocytosis of Type III pneumococci by circulating rabbit leucocytes was not demonstrable by a vital stain technique, whereas under the same conditions the ingestion of non-encapsulated R forms occurred. This is further evidence that the process whereby non-encapsulated pneumococci are disposed of, is insufficient to explain the natural resistance of rabbits to infection with encapsulated Type III pneumococci.     

STUDIES ON VIRULENCE : III. INFLUENCE OF HYDROGEN ION CONCENTRATION AND OF INGREDIENTS OF PLAIN BROTH ON THE VIRULENCE OF PNEUMOCOCCI.

1. A highly virulent strain of pneumococcus, Type I, Neufeld, when grown in an automatic transferring device at 2 hour intervals on different lots of plain broth adjusted respectively to pH = 6.5, 7, 7.5, or 7.8, lost its virulence at a rate in direct ratio to the H ion concentration of the media—the more acid the media, the more rapid the decrease. 2. Growth of a virulent pneumococcus upon plain broth of an H ion concentration changing gradually either to the acid or the alkaline side of neutrality was accompanied by an initial rise in virulence with a subsequent fall, the change being more rapid in the alkaline than in the acid medium. The organism under these conditions underwent an alteration in its behavior toward agglutinating sera; although still specific, its agglutinability became much greater than that of the original organism. This change was more pronounced in the alkaline than in the acid medium. 3. The virulence of a relatively virulent culture of a pneumococcus grown on plain broth underwent different alterations when the organism was transferred at intervals of 1, 2, 4, or 8 hours. With the 1 hour interval, there occurred seemingly an immediate decrease in virulence, while at intervals of 2, 4, and 8 hours, there occurred first a rise and then a fall, the rise being greatest with the 8 hour interval of transfer and least with the 2 hour. 4. Meat infusion adjusted to various H ion concentrations—pH = 7, 7.3, and 7.7—and made with different amounts of meat furnished conditions which caused a decrease in virulence, the unfavorable action varying in inverse proportion to the concentration of the infusion. Dextrose added to both plain broth and to meat infusion neutralized to a marked degree the unfavorable action of these media. 5. The growth of pneumococci on the different ingredients of plain broth, with plain broth made from the same supply as a control, revealed the fact that peptone in 2 per cent solution maintained and even increased the virulence of the strain of pneumococcus studied, while meat infusion caused the usual decrease. The rate of decrease with plain broth was slower than with meat infusion, showing that the 2 per cent peptone neutralized to some extent the unfavorable action of the infusion.     

STUDIES ON OXIDATION AND REDUCTION BY PNEUMOCOCCUS : IV. OXIDATION OF HEMOTOXIN IN STERILE EXTRACTS OF PNEUMOCOCCUS.

In the present work on oxidation and reduction by sterile extracts of pneumococcus, the preparations employed contain among other constituents, a hemolytic substance the properties of which have been described by Cole (1, 2) in his studies on pneumococcus hemotoxin. Pneumococcus extracts prepared by the methods described are actively hemolytic, 0.005 cc. of extract causing complete lysis of 2.5 cc. of a 1 per cent suspension of red cells from rabbit blood. This hemolytic property of pneumococcus extracts is destroyed by 10 minutes exposure to 55°C. When pneumotoxin-containing extracts are protected from the action of molecular oxygen, their hemolytic activity remains unimpaired for considerable periods of time. In the presence of air, on the other hand, the stability of the hemolytic substance depends upon whether the particular type of extract contains a "complete" or "incomplete" oxidation-reduction system. Sterile broth extracts of unwashed pneumococci are reactive with molecular oxygen, and as a result of this union peroxide is formed whenever these extracts are exposed to air. The hemolytic activity of "complete" extracts of this type is rapidly decreased and finally destroyed in the presence of molecular oxygen. On the other hand, the "incomplete" type of extract prepared by saline extraction of washed pneumococci may be exposed to air with little or no loss of hemolytic power. This "incomplete" washed cell extract, unless reactivated, does not undergo autoxidation in the presence of air; under these circumstances peroxide is not formed and the hemolytic activity of this type of extract is not impaired by exposure to air. The stability of the hemolytic agent in the "incomplete" type of extract is evidence that this substance is itself not reactive with or affected by molecular oxygen, even in the presence of the cell enzymes. The destruction of the same hemolytic substance in extracts capable of undergoing autoxidation may be ascribed to the action of some peroxide formed by the union of molecular oxygen with easily oxidized or autoxidizable substances of the extract. It is now known that a peroxide, having the reactions of hydrogen peroxide, accumulates in sterile pneumococcus extracts during oxidation. It has been shown in the present study that the addition of preformed hydrogen peroxide destroys the hemolytic activity of pneumococcus extracts, although higher concentrations were required than were detected in oxidized extracts themselves. These facts and the known action of superoxides in analogous types of reaction make it seem not unlikely that the active agent in the destruction of pneumotoxin in oxidized cell extracts may be a peroxide; either hydrogen peroxide or some higher organic peroxide formed during autoxidation of the extract.