To support his view, sound in principle, Wassermann made an experiment the interpretation of which presents certain difficulties. He injected guinea-pigs with protective antityphoid serum, in a dose insufficient to protect them against a fatal infection. By introducing along with this serum a certain quantity of normal ox serum which, by itself, is also incapable of averting a fatal issue, Wassermann obtains an absolute immunity of his animals. This immunity is due, according to Wassermann, to the cytase of the ox serum acting along with the fixative of the specific serum. The united action of the two ferments causes the death of the micro-organisms. Besredka^[486] has justly observed that normal ox serum contains, in addition to cytases, a substance which exerts a distinct agglutinative action on the typhoid cocco-bacillus and another which stimulates the phagocytic action. These two substances resist a temperature of 55°–60° C., and Besredka shows that with normal ox serum, deprived of its cytases by heating as above, we can obtain the same protective effect as with the same serum unheated.

As the result of another series of experiments, Wassermann^[487] recognises the immunising action of normal serum heated to 60° C. and so entirely deprived of its cytases. Into the peritoneal cavity of guinea-pigs he injects, mixed with heated normal rabbit’s serum, a dose of typhoid cocco-bacilli several times greater than the lethal dose. The guinea-pigs resist this completely. Analysing the mechanism of this immunity, Besredka (l.c. p. 229) attributes it to the combined action of the agglutinin and of the substance which stimulates the phagocytes. We have here another proof that the stimulins which play such an important part in immunity conferred by serums, are found not only in the specific serums, but also in normal serums, whether unheated or heated to 55°–60° C.

The protective property of the normal serums of man and animals against the cholera vibrio has already been referred to. We may now go a little more deeply into the mechanism by which these serums act. This task is an easy one thanks to the important work by Issaeff^[488] carried out in R. Pfeiffer’s laboratory. Having confirmed the observation, made by other investigators, that blood serum from the human subject, whether in health or affected by any disease, is capable of protecting the guinea-pig against the cholera vibrio provided that it is injected 24 hours before the micro-organisms, Issaeff studied the phenomena observed in the peritoneal cavity of the animals experimented upon. By means of small capillary pipettes he drew off at intervals a small quantity of fluid from the peritoneal cavity and examined it in hanging drop or in stained preparations. Some time after the injection this fluid became more and more rich in leucocytes which seized the vibrios, ingested and destroyed them. To obtain this protective effect it was necessary to inject from 0·1 to 5 c.c. of human blood serum. With these doses he could prevent, not only infection of the guinea-pigs by the cholera vibrio, but also the lethal effects of other vibrios. The protective action of normal human serum is general, therefore, and not specific, such as is the immunity conferred by the serums of vaccinated animals or of the human subject who has suffered from an attack of cholera.

Shortly afterwards Funck^[489] confirmed this result in the case of the typhoid cocco-bacillus. He observed that normal horse’s serum, injected as a protective agent in the dose of half a c.c. into the peritoneal cavity of the guinea-pig, preserved this animal from a fatal infection. Pfeiffer and Kolle and Chantemesse and Widal obtained the same results with human serum. The former observers lay special stress on the non-specific character of this protective action of normal serums. As to its mechanism, Funck sums it up as follows: “the specific serum brings about a rapid lysis of the bacilli, normal serum acts in a much more limited fashion; if the dose is very large and if the animal resists infection, the phenomena of extracellular degeneration are rarely appreciable, and it seems that here the specially important factor is the intracellular destruction of the bacteria, in the phagocytes” (p. 70).

Wassermann has shown the protective action of normal serum against the experimental disease produced by the staphylococcus. This action, although not absolutely general, is nevertheless widely distributed. Wassermann^[490], from comparative investigations on this subject, came to the conclusion that “the serum of a different species of animal acts by greatly increasing the resistance, whilst the serum of the same species produces an effect which is not nearly so marked.” As in these normal serums a stimulating influence on the phagocytes is specially marked, it may readily be understood that the serum of the same animal or of the same species does not produce so energetic an effect as the serum of a different species. As these normal serums possess, not only the property of exciting phagocytosis, but often also that of rendering motionless and of agglutinating certain micro-organisms, there might be some difficulty in interpreting the part played by these serums. It may be useful, therefore, to pass in review the protective action of fluids less complicated than blood serums.

Issaeff, in the work already cited, demonstrated that not only normal serums but a whole series of fluids, such as urine, broth, etc., exert a protective effect against microbial infections. These fluids must be injected about 24 hours before the introduction of the bacteria. The best method consists in injecting them directly into the peritoneal cavity, after which the animals acquire an immunity against absolutely fatal doses of cholera vibrios. Funck verified this observation for the infection caused by the typhoid cocco-bacillus, and Bordet confirmed it for the streptococcus. The injection of peptonised broth into the peritoneal cavity of the normal guinea-pig, made 24 hours before an inoculation of double the fatal dose of the streptococcus, exerts a distinct protective action; the infection does not kill the animal. This broth is neither bactericidal, attenuating, nor agglutinative; it forms a good culture medium for the streptococcus and possesses no fixative power. Consequently it does not act directly on the vitality or virulence of the micro-organism; nevertheless, it is distinctly protective.

According to Issaeff’s researches, the protective substances used by him must be arranged in the following order as regards their action against the cholera vibrio. Tuberculin is the most effective; then comes a 2% solution of nuclein, followed by normal human serum, broth, and urine, whilst physiological saline solution is the least active. All prevent infection by the vibrios, but the protection is effective for some days only; this protective action is exerted against various kinds of bacteria, being in no sense specific.

Pfeiffer lays so much stress on the great difference between the protective power of normal serums, as well as of the other fluids mentioned, and that of the anti-infective specific serums, that he even proposes to classify the first group as giving rise to pseudo-immunity or resistance. This view is certainly an exaggerated one, because it is difficult to draw a very distinct line between the two groups of phenomena. There are normal serums, of which 0·1 c.c. is quite sufficient to confer the protective effect, just as there are specific serums of which it is necessary to make use of a much greater dose to attain the same result.

Protective fluids, other than the serums, only manifest their influence by exciting a great phagocytic “superactivity.” As the result of their injection into the peritoneal cavity of normal guinea-pigs, first a transitory phagolysis is induced, this being soon replaced by a very considerable afflux of leucocytes, which is maintained for 24 hours or longer, and then gives place to the normal condition. It is during the period of the greatest leucocytosis of the peritoneal fluid that the animal exhibits the most marked resistance against infective micro-organisms. The vibrios are rapidly ingested by the phagocytes, without having previously been acted upon by the “humours.” Bordet shows that the same thing happens in the case of the streptococcus inoculated into guinea-pigs after a protective injection of peptonised broth.

We have observed the same phenomenon in guinea-pigs and white rats inoculated with the cocco-bacillus of plague. Treated with freshly prepared peptonised broth the day previous to inoculation, these animals oppose to the micro-organism a much more marked resistance than do the control animals. The injection of the cocco-bacillus of plague sets up a marked phagocytosis on the part of the macrophages. These cells ingest large numbers of micro-organisms which, after a time, have all passed into the phagocytes. If a drop of the peritoneal exudation is now withdrawn, we find only intracellular cocco-bacilli (fig. 43). If the drop be kept for some time outside the animal and at a suitable temperature the macrophages may be seen to perish and the micro-organisms to develop in their contents. We thus obtain abundant cultures which pass from the interior of the macrophages into the fluid of the exudation (figs. 42, 44, 45). When the animals are not sufficiently protected the same phenomenon is observed in the peritoneal cavity of the living animal. The macrophages, crammed with cocco-bacilli, burst, allowing the micro-organisms to escape. These multiply in the peritoneal fluid and spread through the animal, which soon dies.

Wassermann affirms that “the artificially increased resistance is nothing but an active and reinforced afflux of the complements (cytases) towards one point in the animal, for the purpose of digestion.” (Ztschr. f. Hyg., Leipzig, 1901, Bd. XXXVII, S. 199.) Wassermann does not explain how this afflux of cytases is produced. The entirely concordant researches on this point by Issaeff, Funck, Bordet, and ourselves, prove that this afflux takes place not through the mediation of the fluids, but solely through the phagocytes, the carriers of the cytases. Consequently it is beyond dispute that in the immunity conferred by physiological saline solution, broth, and several other fluids, we have to do solely with an augmentation of the phagocytic reaction. In the immunity conferred by normal or specific serums, this same stimulating factor still plays the more important part. Along with it, however, there is an intervention more or less pronounced, according to circumstance, and more or less frequent, of cytases, brought by the serums prepared outside the body or that have escaped during phagolysis, as well as of substances truly humoral, such as the fixatives or the agglutinins.

Amongst the non-specific substances which are capable of conferring an immunity more or less stable, must be placed the products of micro-organisms other than those against which we wish to protect the animal. Pasteur^[491] noted that when the anthrax bacillus, mixed with other micro-organisms, in themselves inoffensive, is inoculated into animals, anthrax does not develop and the animals remain well. Later, Emmerich^[492] showed that the streptococcus of erysipelas exerts an antagonistic influence against the anthrax bacillus. He succeeded in immunising and even in curing rabbits inoculated with anthrax, by submitting them to the action of this streptococcus.

These experiments served as the starting-point for several works on the vaccination of animals against anthrax by means of various micro-organisms, as well as by their products. Pawlowsky^[493], Watson-Cheyne^[494], and Bouchard^[495] have proved that bacteria not very pathogenic and even saprophytes, such as the Coccobacillus prodigiosus, Friedländer’s bacillus, and the Bacillus pyocyaneus, were also capable of preventing infection by the anthrax bacillus. Freudenreich^[496] showed that not only did the bacillus of blue pus exert an antagonistic action but that the same effect could be obtained with sterilised cultures of this organism. Woodhead and Cartwright Wood^[497] studied the vaccinating action of these products on rabbits inoculated with virulent anthrax bacilli. The animals resisted completely or survived for some time. Analysing the phenomena produced under such conditions, these two authors came to the conclusion that the action of sterilised cultures of Bacillus pyocyaneus is “indirect and as taking place either by opposing itself to the action of the poison upon the tissues, or by stimulating certain tissues and increasing their functional activity.” With the object of obtaining an exact interpretation of this antagonistic influence I suggested to M. Blagovestchensky^[498] that he should investigate in detail the phenomena which take place in the organism of rabbits inoculated with the anthrax bacillus and submitted to the action of sterilised cultures of the Bacillus pyocyaneus. At the very outset this observer was met by the fact that these cultures act directly upon the vitality of the anthrax bacillus. Thus the association of the former with the anthrax bacillus in vitro was sufficient to interfere with the development of the latter. Under these conditions he had to renounce the investigation of the part played by the cellular elements of the rabbit in the antagonism of the two bacteria.

Friedländer’s bacillus has been found to be much more suitable for this line of research as is shown by work carried out by Freiherr von Dungern^[499] in my laboratory. This observer convinced himself that “anthrax bacilli are weakened neither by the encapsuled bacilli nor by the substances which they contain.” These micro-organisms do not interfere in the slightest with the anthrax bacilli either outside or within the animal, and “when the anthrax infection does not become generalised it is due to the fact that the anthrax bacilli are ingested by the phagocytes at the seat of inoculation and destroyed within these cells” (p. 183).

In this action of foreign micro-organisms upon micro-organisms against which we wish to protect the animal we have to deal with something analogous to the condition we obtain when immunising with normal serums or with any other kind of fluid. In both cases immunity is rapidly established, but it is very transient and is confined to a stimulation of the phagocytic resistance. Direct action may also intervene, as in the case of Bacillus pyocyaneus, but it is not indispensable. The animal whose phagocytes are in a condition of superactivity can do without this direct action, its own resources being sufficient to arrest anthrax.

Following the same lines of investigation as those on the antagonism between the anthrax bacillus and several other micro-organisms, Klein^[500] has demonstrated that, in order to prevent a guinea-pig from contracting experimental cholera peritonitis, it is only necessary to inject into it, the day before infection, a culture of Finkler and Prior’s vibrio or of certain other bacteria. These experiments by Klein served as the point of departure for Issaeff’s work which led to the discovery of the stimulating influence of all kinds of fluids injected into the peritoneal cavity of guinea-pigs.

In this transient immunity obtained with products foreign to the micro-organism against which one is vaccinating, the most constant and consequently most important part is again played by the phagocytes. But there is associated with it an influence, greater or less in degree, of substances present in the serums, such as the microcytases and fixatives, which are able to exercise a direct action on the pathogenic micro-organisms. In all cases known and analysed up to the present, the intervention of the living organism of the animal is indispensable, consequently this form of acquired immunity against micro-organisms cannot be regarded as being really passive.