But the example of the antispermotoxin of the rabbits in my own experiments is very different. In order that it might manifest its action the serum of these rabbits did not need to be heated to 56° C.; it was not necessary to rid it of its own macrocytase which could have acted under the influence of the fixative, if this latter for want of antifixative had remained free in the added spermotoxin. This antifixative, then, is undoubtedly found in the serum of castrated males which have shown themselves capable of producing not only anticytase, but also antifixative. This result has been further verified by comparative experiments on castrated male rabbits, some of which received spermotoxic guinea-pig’s serum whilst the others received only normal guinea-pig’s serum. It has been demonstrated that the amount of cytases remains almost constant in both normal and vaccinated animals^[172]. If, then, the antispermotoxins contain only anticytase, the injection of specific guinea-pig’s serum and that of normal guinea-pig’s serum should produce the same result, that is to say the serums of castrated rabbits, when treated by these two kinds of guinea-pig’s serum, should exhibit the same antispermotoxic power. Experiments have, however, proved that this is not the case. The serum of castrated rabbits that have been injected several times with normal guinea-pig’s serum becomes distinctly antispermotoxic, but its power to protect the spermatozoa of the rabbit against being deprived of motility by the guinea-pig’s spermotoxin is greatly inferior to that which is developed in the serum of other castrated rabbits that I injected with spermotoxic guinea-pig’s serum. Of course all the other conditions of the experiment were the same for the two groups of rabbits.

Several series of facts, then, focus to this fundamental point, that the organism of an animal that has been deprived of its male sexual organs is in a condition to produce antispermofixative. Against the argument that we have drawn from the fact that the antispermotoxic serum of castrated rabbits that have been treated with spermotoxic serum acts without being heated, might be cited certain experiments made by Ehrlich and Morgenroth. The antispermotoxic action in this case, as already stated, demonstrates that the serum of prepared rabbits contains antifixative. Otherwise, had the fixative not been neutralised, it would have allowed the macrocytase of the rabbit’s serum to arrest the movements of the spermatozoa. Now the two above-named observers have demonstrated^[173] that the injection of different serums into animals is capable of exciting in their blood the development of anticytases. The macrocytase of castrated rabbits which, before treatment with the spermotoxin, was capable of arresting the movements of rabbits’ spermatozoa acted upon by a fixative, might become inert after the injections of spermotoxic serum of guinea-pigs. To clear up this point I asked M. Weichardt^[174], who has carried out work on this subject in my laboratory, to try by means of unheated serums of normal animals, to restore the activity of spermotoxin that had been mixed with antispermotoxic serum. Spermatozoa of rabbits were put into a definite mixture of spermotoxic guinea-pig’s serum, heated to 56° C., and antispermotoxic serum, also heated to 56° C., obtained from castrated rabbits that had been treated with spermotoxin. The spermatozoa remained very active in this mixture which contains specific fixative (in the spermotoxic guinea-pig’s serum) and antispermotoxin. To this mixture is added a little normal rabbit’s or horse’s serum, unheated. These serums contain cytases and would be quite capable of arresting the movements of the spermatozoa if there was found in the mixture any free fixative that would enable the macrocytase to be linked to the spermatozoa. Under these conditions the spermatozoa remain motile for a long time. The fixative, then, was no longer active; it was neutralised by the antifixative of the antispermotoxic serum of castrated rabbits. A control experiment was made with the same substances; but the castrated rabbits’ serum that had been treated with spermotoxic serum was replaced by the serum of other castrated rabbits treated with normal guinea-pig’s serum. In these latter mixtures the spermatozoa became motionless at the end of a very short time; the fixative, not being neutralised, readily allowed the rabbit’s and horse’s cytases to affect the spermatozoa.

It follows from all this that the antispermotoxic serum of castrated male rabbits, when treated with normal guinea-pig’s serum, contains anticytase only; whilst the serum of castrated male rabbits, treated with specific and spermotoxic guinea-pig’s serum, contains anticytase and antifixative. The latter, then, has been produced independently of the sensitive elements,—the spermatozoa.

Having established the fact that antispermotoxin does not come from the male organs, it was necessary to try to ascertain its true source. With this object in view we injected spermotoxic serum into young rabbits (quite capable of producing antispermotoxin) and tried to follow the fate of the spermotoxin in the organism. When spermotoxic guinea-pig’s serum is injected into the peritoneal cavity of the rabbit a notable amount of spermotoxin is found in the thickened portion of the omentum made up of lymphoid tissue. But the greater portion of the poison passes into the circulation whence it goes to fix itself in various organs, especially the spleen. At the moment when the spermotoxin is found in the blood a certain quantity of this fluid was drawn off into tubes containing some drops of extract of leeches’ heads. After the blood thus treated had been centrifugalised the plasma was decanted and its power of arresting the movements of spermatozoa was compared with that of serum of the same blood prepared in the usual way. From these researches it results that the plasma is always richer in spermotoxin than is the corresponding serum. Sometimes the difference in favour of the plasma is very great.

A part of the spermotoxin passes into the kidneys and the suprarenal capsules. It is probable that, as is the case with so many soluble poisons, a certain proportion of the spermotoxin may be eliminated by the uropoietic organs. A small quantity of this poison is found also in the male and female sexual glands of young non-castrated rabbits.

The search for some main centre of origin for the production of antispermotoxin has as yet led to no positive result. The power of arresting the movements of spermatozoa first appears in the blood plasma, and it is this same fluid which, later, is more antispermotoxic than is any organ. Amongst the tissues which fix spermotoxin the genital organs play not the slightest part in the production of antispermotoxin. The experiments with castrated rabbits afford sufficient proof of this. On the other hand it becomes more and more probable that the phagocytic system, disseminated in many organs, and especially the leucocytes, furnish the antispermotoxic substance. The fixation of the spermotoxin by the leucocytes of the blood, such as the cells of the omentum and of the spleen, already offers us a valuable indication. The absence of any particular organ that might have the monopoly of fixing the spermotoxin and which should later be found charged with a predominant amount of antispermotoxin also speaks in favour of the phagocytic origin of this antitoxin.

After a single intraperitoneal injection of spermotoxic guinea-pig’s serum into young rabbits, the blood of the latter is distinctly spermotoxic for several days; later it becomes indifferent, but eight or ten days after the commencement of the experiment the blood begins to exhibit an antispermotoxic power. In these cases the plasma shows itself more active than the serum. When the rabbits are killed at this stage of commencing antitoxic production, it is found that an extract of the organs is not antispermotoxic or only feebly so. In all cases this power, when it exists, is more feeble than that of the blood fluid. The results obtained with extracts of organs are not constant. Sometimes the spleen possesses more antitoxic activity, whilst the liver, thymus, omentum, lymphatic glands and genital glands exhibit none of this property. In other cases the survival of the spermatozoa that are influenced by the spermotoxin has been longest in the extract of the suprarenal capsules. Sometimes the extract of the omentum exhibits the greatest antispermotoxic power. This great variability in the development of the property of protecting the spermatozoa accords well with the idea that the elements which produce antispermotoxin are wandering cells which, under diverse influences, may be localised in very diverse points of the organism.

We must not deceive ourselves. The facts which have been collected up to the present do not allow us as yet to form a final opinion on the origin of anticytotoxins, but we are quite justified in regarding as very probable the hypothesis that the phagocytes play a most important part in the process. It is in all cases beyond doubt that the amoeboid cells which resorb the formed elements play a very important part in the resorption of fluids of very complex molecular composition.