But it may be asked, whence come the injurious results of inbreeding, if not from the union of two nearly related germ-plasms? They certainly do arise from that cause, but it is not through a 'formative stimulus,' too slight in this case, exercising a direct formative chemical effect upon the two hereditary substances, but through the indirect influences exerted by these too similar hereditary contributions during the development of the new individual. Lest it be imagined that I am tilting against windmills, I will refer to one of the numerous examples of the evil effects of inbreeding which have been submitted to me as specially corroborative of the conception of amphimixis as a 'formative stimulus' whose strength depends upon the difference between the germ-substances. The renowned breeder, Nathusius, allowed the progeny of a sow of the large Yorkshire breed, imported from England when with young, to reproduce by inbreeding for three generations. The result was unfavourable, for the young were weakly in constitution and were not prolific. One of the last female animals, for instance, when paired with its own uncle—known to be fertile with sows of a different breed—produced a litter of six, and a second litter of five weakly piglings. But when Nathusius paired the same sow with a boar of a small black breed, which boar had begotten seven to nine young when paired with sows of his own breed, the sow of the large Yorkshire breed produced in the first litter twenty-one and in the second eighteen piglings.

How could this really remarkable difference in the fertility of the sow in question be the result of a formative stimulus, exercised by the sperm-cells of the unrelated boar upon the ova of the female animal? If the progeny of the sow had been more fertile than herself, then we should have been at least logically justified in concluding that this was the case, but it is not intelligible that the egg-cells of this mother sow should be increased twice or three times because they were fertilized by a new kind of sperm as they glided from the ovary. The number of ova which are liberated from the ovary depends in the first instance upon the number of mature ova contained in it; and unless we are to make the highly improbable assumption that the crossing with the strange boar had as an immediate result the maturing of a large number of ova, we must look elsewhere than in the ovary of the animal for the cause of this sudden fertility, possibly in chance circumstances which we are unaware of and which make the ovary occasionally more productive, possibly however in the fact that inbreeding may have brought about various slight structural variations in the animal, and among these some which made the fertilization of the abundantly produced ova by the sperm of the related boar less easy, and caused it to fail more frequently. As will be readily understood, I cannot say anything definite on this point, but we know that very slight variations in the sperm-cell or the ovum may make fertilization difficult, or may even prevent it. I need only remind you of the interesting experiments in hybridization which Pflüger and Born made with Batrachians nearly thirty years ago, which showed that in two nearly related species of frog the ova of the species A were frequently fertilized by the sperms of the species B, but not conversely, the ova of the species B by the sperms of A. This is the case, for instance, with the green edible frog (Rana esculenta), and the brown grass-frog (Rana fusca), and the reason of this dissimilarity in the effectiveness of the sperm lies simply in 'rough mechanical conditions,' in the width of the micropyle of the ovum, and the thickness of the head of the spermatozoon. If each species possesses a micropyle which is exactly wide enough to admit of the passage of the spermatozoon of its own species, another species will only be able to fertilize these eggs if the head of its spermatozoon be not larger than that of the first species. Thus, as experiment has proved, the spermatozoa of Rana fusca fertilize the ova of almost all other related species, for they have the thinnest head and it is at the same time very pointed. In this case, therefore, it depends upon the microscopic structure of the ovum whether fertilization can take place or not, and we can imagine that similar or perhaps other minute variations had taken place in the ova in the case of Nathusius's sow, and that these made it difficult for the sperms of boars of the same family to effect fertilization. These variations may have arisen as a result of the continued inbreeding, because the same ids were constantly being brought together in the fertilized ova, and thus any unfavourable directions of variations which existed were strengthened.

It seems to me that in this way alone can the injurious effect of inbreeding be made intelligible. From both parents identical ids meet in the fertilized ovum, in greater numbers the longer inbreeding continues, for at the maturation of every germ-cell the number of different ids is diminished by a few, and their number must therefore gradually decrease, and it is conceivable that ultimately it may sink to one kind of id, that is, that the germ-plasm may then consist entirely of identical ids. If chance variations of certain determinants in unfavourable directions occur in some of the ids composing the germ-plasm, these are brought together in the offspring from both the maternal and the paternal side, and will occur in an increasing number of ids the longer the inbreeding has gone on, that is, the smaller the number of different ids has become. The unfavourable variation-tendency is therefore persisted in, and its influence upon the development of a new descendant will be the greater the larger the number of identical ids with these unfavourable variations. It is obvious that the crossing of an animal, which is thus, so to speak, degenerating slightly, with a member of an unrelated family must immediately have a good effect upon the descendants, for in this way quite different ids with other variations of their determinants are introduced into the inbred germ-plasm which had become too monotonous.

From this theoretical interpretation of the injurious consequences of inbreeding we may at once infer that not every inbreeding necessarily implies degeneration, for the occurrence of unfavourable variational tendencies in the germ-plasm is presupposed as the starting-point of degeneration, and if these do not exist there can be no degeneration. This harmonizes with the fact that the evil effects of inbreeding are observed to vary greatly in amount, and may not occur at all. But they are greatest in breeds artificially selected by man, which have long been under unnatural, directly influential conditions, and are also removed from the purifying influence of natural selection. In such cases, therefore, there is every probability that diverse unfavourable variational tendencies in the determinants will occur.

But how are we to understand the fact that pure parthenogenesis may last through innumerable generations, and yet no degeneration sets in? I believe very simply. In this case too, the same ids which were peculiar to the mother of the race are contained in the descendants, but they do not diminish in number, for in pure and normal parthenogenesis, such as that of Cypris reptans, the second maturation-division of the ovum does not take place, and this is precisely the nuclear division which effects reduction. In addition, the introduction of identical ids, which must take place in the case of inbreeding at every amphimixis, does not occur, and, what is certainly of great importance, all these cases are old species, living under natural conditions—the same conditions under which they lived as amphigonous species, and not newly formed breeds under artificial conditions, as has probably always been the case in the experiments in inbreeding.

It is true that even in old species, living in a state of nature, unfavourable variations may arise in the germ-plasm, and may go on increasing during purely parthenogenetic multiplication, for the ids with unfavourably varying determinants will no longer be set aside by means of reducing division. But those individuals in which the unfavourable variational tendency increases until it has attained selection-value will be subject to selection and will be gradually eliminated; indeed, the weeding out of the inferior individuals will be more drastic here than where amphigony obtains, because in this case all the offspring of one mother are nearly alike, so that the whole progeny is exterminated if the mother varies unfavourably.

On the other hand, a transformation in a favourable direction, an adaptation to new conditions of life, as far at least as that implies the simultaneous variation and harmonious co-adaptation of many parts, cannot, as far as I can see, be effected in the course of purely parthenogenetic reproduction, nor can a degeneration of complicated parts which have become superfluous. For both these changes, in my opinion, require that the ids of the germ-plasm should be frequently mingled afresh, since apart from this there cannot be a harmonious readjustment of complicated structures, nor can a uniform degeneration affecting all parts set in. As an example of this last case we may take that organ which became functionless in the purely parthenogenetic species of Ostracods when amphigonous reproduction was given up—the sperm-pocket or receptaculum of the female. All these species still possess an unaltered receptaculum seminis, a large pear-shaped bladder with a long, narrow, spirally coiled entrance-duct, very well adapted for allowing the enormous spermatozoa of the males to make their way in singly, and to arrange themselves within the receptacle side by side in the most beautiful order, like a long ribbon, and finally to migrate out again singly to fertilize the liberated ova. In Cypris reptans and several other species, however, no males have been found in any of the places which have been carefully searched, and the receptaculum of the female is always found to be empty. Nevertheless it shows no hint of degeneration. It is possible enough that, as in Apus cancriformis, which is of similar habit, the males have become extinct in most colonies of these species, but that nevertheless they do occur here and there from time to time in the area inhabited by the species, and if this should prove to be the case, it would confirm the conclusion, which is very probable on other grounds, that the pure parthenogenesis of these species has not existed in most of their habitats for a long time, speaking phylogenetically. For this reason we must not over-estimate the significance of the complete persistence of the receptaculum even with exclusively parthenogenetic reproduction. It proves, however, that degeneration of a superfluous organ does not necessarily set in even after hundreds of generations, and in this fact there is certainly a corroboration of the view that it is 'chance' germinal variations which give the impulse to degeneration. These first induce a downgrade variation through germinal selection, and this, if it concerns an organ of no importance to the survival of the species, is not hindered in its progress by personal selection. Whether degeneration of the receptaculum would have occurred in these parthenogenetic species if they had retained even a periodic sexual reproduction, as is actually the case in the generations of the alternately parthenogenetic and sexually reproducing Aphides, we cannot decide, since we know nothing in either case as to the length of time that parthenogenesis has prevailed among them, nor have we any method of computing the number of generations that must elapse before a superfluous organ begins to vacillate. We only know that the parthenogenetic generations of Aphides no longer possess a receptaculum, while other forms with alternating bi-sexual and parthenogenetic modes of reproduction, which are in this respect possibly more modern, e. g. some of the gall-wasps, possess one similar to that of the Ostracods.

I must refer to one other case of parthenogenesis, since it has been hitherto regarded as a formidable puzzle for the germ-plasm theory, and has only recently found its solution, I mean the facultative parthenogenesis of the queen-bee. As the 'male' eggs of the bee remain unfertilized, and yet undergo two reducing divisions, which must diminish the number of ids in the ovum-nucleus by a half, the number of ids in the germ-plasm of the bee must be steadily decreasing, and this state of things has therefore been regarded by some English biologists as convincing evidence of the untenability of the conception of ids and of the whole germ-plasm theory. Apparently, indeed, it is contradictory to the theory, and we must inquire whether the contradiction is merely an apparent one, disappearing when the facts are more precisely known. It was mainly on this ground that I instituted the researches carried out by Dr. Petrunkewitsch, the results of which I have already in part communicated in a former lecture. These results confirmed the previous conclusions that the 'male' eggs of the queen-bee remain unfertilized, that two reducing divisions occur, and that in consequence the ovum-nucleus only contains half the normal number of chromosomes. That these increase again by division to the normal number does not save the theory, for only identical ids can arise in this way, while the significance of the multiplicity of the ids lies mainly in their difference. The halving of the number of ids in each 'male' ovum would necessarily lead, if not to a permanent diminution in the number of ids, at least to a monotony of the germ-plasm, since the number of different ids would be steadily decreasing and the number of identical ids as steadily increasing. This too would be a contradiction of the theory. But Dr. Petrunkewitsch's investigations have shown that, of the four nuclei which are formed by the two reducing divisions, the two middle ones (Fig. 79, K 2 and K 3) recombine with one another, and fuse into a single nucleus, and that from this copulation-nucleus in the course of development the primitive germ-cells of the embryo arise. Now all the ids which were originally present in the nucleus of the immature ovum may be reunited in this 'polar copulation-nucleus' if the two nuclei K 2 and K 3 turned towards each other in Fig. 79 contain different ids. That this is the case cannot of course be seen from the ids themselves, but it seems to me extremely probable, since it is dissimilar poles of the two nuclear spindles which here unite, namely, the lower pole (daughter-nucleus) of the upper spindle and the upper pole of the lower spindle. In the first directive or polar spindle there lay thirty-two chromosomes, which had increased by duplication from sixteen, and of these sixteen passed over into the first polar nucleus, while sixteen formed the basis of the second directive spindle. These two sets of sixteen chromosomes must have been quite similar, since the two sets arose by division of the sixteen mother-chromosomes. Let us call the chromosomes a, b, c, d-q, then similar sets of chromosomes must have been contained in the two nuclear spindle figures depicted in Fig. 79 at the beginning of the division, and eight of these went to each daughter-nucleus. Now, if a-k migrated to the upper pole of the spindle and l-q to the lower pole, then the union of K 2 with K 3 would bring together again all the ids that had before been present. In consideration of this I predicted to Dr. Petrunkewitsch that this copulation-product might be the basis of the formation of the germ-cells in the drone-bee, and his painstaking and difficult researches have confirmed this prediction, strange though it may seem, that the male germ-cells have a different origin from the female germ-cells. But this discovery gives a strong support to the germ-plasm theory. It may, of course, be objected that the assumed regular distribution of the ids in the two daughter-nuclei cannot be proved, but we know already that this dividing apparatus does very exact work, and we are at liberty to assume it in an even higher degree. Moreover, what other interpretation of the unexpected development of the germ-cells discovered by Petrunkewitsch could be given if this had to be rejected? A clearer proof of the individual differences of ids and of their essential importance could not be desired, than lies in the fact that in the 'male' eggs of the queen-bee a different and novel mode of germ-cell formation is instituted, after half the ids have been irrecoverably withdrawn from the ovum-nucleus. We see from this that for individual development a duplication of individual ids may suffice, but that for the further development of the species a retention of the diversity of the ids is important.