If the spermotoxin were diffused in the plasma and other fluids of the guinea-pig which furnishes it, it ought to render motionless the spermatozoa contained in the genital organs. Experiment demonstrates, however, that this is not the case. If the male organs be removed from a guinea-pig whose serum is very autospermotoxic in vitro, we find, especially in the epididymis, a mass of very virile spermatozoa which for a long time retain their motility in physiological salt solution. The macrocytase, then, has not reached the spermatozoa in the living animal; this is because it is not found in the plasmas. Let us inject into a guinea-pig, whose serum is strongly autospermotoxic, one portion of sperm into the subcutaneous tissue and another portion into the peritoneal cavity. In the first site a soft oedema, filled with transuded fluid, in which the very active spermatozoa retain their motility for a couple of hours, is produced. In the peritoneal fluid the same spermatozoa become motionless in a few minutes. This great difference is explained by the fact that, under the skin, there are no, or almost no pre-existing leucocytes, whilst in the peritoneal fluid they are abundant. The phagocytes injured by the introduction of sperm into the peritoneal cavity, abandon a portion of their macrocytase, sufficient to render the spermatozoa motionless. But when Metalnikoff injected physiological salt solution into the peritoneal cavity of his autospermotoxic guinea-pigs, and then, on the following day, a quantity of sperm, the spermatozoa continued very active for more than an hour. In this case phagolysis is very transitory and insignificant; it is soon followed by a great afflux of leucocytes which bring about a rapid ingestion of the spermatozoa. Many of these elements are devoured in a living state; for even when their body is enclosed in the macrophage, their tail, left outside, continues to move very actively.

All these experiments demonstrate that in the normal state the macrocytase remains within the phagocytes and only escapes during phagolysis, or at the moment when the blood, after it has been withdrawn from the organism, coagulates. Is it the same for the fixative? It is easy to prove that this soluble ferment circulates in the plasmas of the living organism. We have already said that the spermatozoa of a guinea-pig whose serum is very autospermotoxic, remain alive for some time in the physiological salt solution. But if we introduce them, in vitro, into the serum of a normal guinea-pig they remain motile but a short time (some 10–20 minutes), whilst the spermatozoa of a normal guinea-pig will live in the same serum for several hours. This difference is explained by the fact that the spermatozoa of the autospermotoxic guinea-pig, although very active, have absorbed the fixative during the life of the animal. This fixative is, as we have stated, found in the body fluids and has been able to penetrate to the male organs. Here the spermatozoa become charged with the fixative and, once transported into the serum of the normal guinea-pig, rich in macrocytase, they lose their movements very quickly. At the same time the spermatozoa used for control, not having absorbed any fixative, are able to live for a long time in the same serum.

As the macrocytase remains fixed to the phagocytes there can be no doubt as to its origin; it is elaborated by these cells. Whence however comes the fixative which is free in the body fluids and which is precisely the substance that is developed in so large a quantity in the treated animals? The exact solution of this question is not easy; nevertheless there are many facts which indicate that this fixative is also of phagocytic origin. We know already that the serums of normal animals contain only small quantities or sometimes, perhaps, none of the fixative. This fixative only appears abundantly as the result of the resorption of the corresponding elements, red corpuscles or spermatozoa. This resorption, as we have said, is almost exclusively the work of the macrophages. It is just in those cases where the red corpuscles, injected into the peritoneal cavity of an animal of the same species, pass directly into the lymph, without being injured or, save exceptionally, ingested by the phagocytes, that the fixative is not formed. When the red blood corpuscles of the goose, introduced with defibrinated blood below the skin of a guinea-pig, undergo there a partial solution in the fluid of the exudation, and where the phagocytosis is more limited than in the peritoneal cavity, the production of fixative is small. When the injection of the same goose’s blood is made into the peritoneal cavity of a guinea-pig and is followed by complete phagocytosis, the fixative is produced in greater abundance. There exists, then, in all these cases a constant relation between the degree of phagocytosis and the amount of the fixative produced. As this fixative facilitates the access of the cytase to the cells and as the resorption of these elements takes place specially in the macrophages, we are bound to come to the conclusion that the fixative is a second phagocytic ferment which is produced in abundance during the process of intracellular digestion. Only, instead of remaining in the substance of the phagocytes, this fixative is in part thrown out from these elements. It passes into the plasma of the blood and into the other fluids and ends by disappearing from the organism, probably being eliminated by the excretory channels.

In the Invertebrata, where, as we have seen, the alien red blood corpuscles are also digested within the phagocytes, we have never been able to demonstrate any haemolytic property of the blood fluid, even after repeated injections of blood. We must conclude from this that in these animals the quantity of fixative is merely sufficient to bring about the solution of the red corpuscles which are within the phagocytes. In the case of fishes and higher animals (we may recall the example of the red corpuscles of the guinea-pig when resorbed into the organism of the gold-fish) the production of the fixative is much more abundant, and this ferment can be easily demonstrated by its action in vitro.

This over-production of a ferment which acts in the phagocytic resorption, finds its analogue in the passage of certain digestive ferments, such as amylase and pepsin in man and the dog, into the blood and urine, as mentioned in the preceding chapter.

One of the best arguments in favour of the thesis here developed, has been furnished to us by the analysis of the phenomena observed in connection with the autospermotoxic serums of the guinea-pig. This idea of autotoxins was originally put forward by Ehrlich in his memoirs, published in conjunction with Morgenroth and already repeatedly cited. Ehrlich asked himself whether the organism which resorbs, not red corpuscles of an alien species, but red corpuscles of its own species, would also be capable of developing haemolytic substances. With this object he injected blood obtained from goats into these same goats or into other individuals of the same species. He and Morgenroth^[140] were, under these conditions, able to obtain isotoxic serums, that is to say serums which dissolve the red corpuscles of the goat, coming from other individuals than those which had been treated by the blood and which furnished the serum. In order to obtain this result, however, they had to inject, not unaltered blood but blood mixed with water. The red corpuscles of the unaltered blood pass readily into the circulation of the animal of the same species, without being attacked by the phagocytes. Now, we know from the experiments of Bordet that the stromas of the red corpuscles suffice for the production of the fixative, whilst the haemoglobin does not incite to the development of this ferment by the organism. As the stromas, injected with a mixture of blood and water, must be devoured by the macrophages, we can readily understand that these phagocytes may serve for the elaboration of the fixative.

The resorption of the red corpuscles and that of spermatozoa which we have presented as examples, may serve as types for the resorption phenomena of formed elements in general. When other species of cells are introduced into the organism, the resulting process always reveals the same character: inflammatory reaction with preponderant intervention of the macrophages; intraphagocytic digestion of the introduced elements; excessive production and excretion of the fixatives. Whilst the macrocytase is always the same in the same species of animal, the fixatives are different and specific. In addition to the haemofixatives and spermofixatives already described, we may obtain, as the result of the injection of the corresponding cells, leucofixatives, nephrofixatives, hepatofixatives, trichofixatives, etc. It does not enter into our programme to treat the subject here^[141]. We wish simply to insist on those aspects of the resorption of cells which are closely connected with the problem of Immunity. In the next chapter we must, however, recur to certain features of the phenomena of resorption.