In sexual reproduction, with the union of ovum and sperm, we seem to have a fertile source of variation. The parents are not precisely alike, and their individual differences are, ex hypothesi, germinal products. In the union of ovum and sperm, therefore, we see the union of somewhat dissimilar germs. And in sexual reproduction we have a constantly varying series of experiments in germinal combinations, some of which, we may fairly suppose, will be successful in giving rise to new or favourable variations. This view, however, would seem to involve an hypothesis which may be true, but which, in any case, should be indicated. For it is clear that if new or favourable variations arise in this way, the germinal union cannot be a mere mixture, but an organic combination.

An analogy will serve to indicate the distinction implied in these phrases. It is well known that if oxygen and hydrogen be mixed together, at a temperature over 100°C., there will result a gaseous substance with characters intermediate between those of the two several gases which are thus commingled. But if they are made to combine, there will result a gas, water-vapour, with quite new properties and characters. In like manner, if, in sexual union, there is a mere mixture, a mere commingling of hereditary characters, it is quite impossible that new characters should result, or any intensification of existing characters be produced beyond the mean of those of ovum and sperm. If, for example, it be true, as breeders believe, that when an organ is strongly developed in both parents it is likely to be even more strongly developed in the offspring, and that weakly parts tend to become still weaker, this cannot be the result of germinal mixture. Let us suppose, for the sake of illustration, that a pair of organisms have each an available store of forty units of growth-force, and that these are distributed among five sets of organs, a to e, as in the first two columns. Then the offspring will show the organs as arranged in the third column.[BP]

Parents.Offspring.
a 10 10 10
b 8 10 9
c 9 5 7
d 7 9 8
e 6 6 6
— — —
40 40 40

There is no increase in the set of organs a, which are strongly developed in both parents; and no decrease in the set of organs e, which are weakly developed in both parents. By sexual admixture alone there can be no increase or decrease beyond the mean of the two parental forms. If, then, the union of sperm and ovum be the source of new or more favourable variations other than or stronger than those of either parent, this must be due to the fact that the hereditary tendencies not merely commingle, but under favourable conditions combine, in some way different indeed from, but perhaps analogous to, that exemplified in chemical combination.

Such organic combination, as opposed to mere commixture, is altogether hypothetical, but it may be worth while to glance at some of its implications. If it be analogous to chemical combination, the products would be of a definite nature; in other words, the variations would be in definite directions. Selection and elimination would not have to deal with variations in any and all directions, but would have presented to them variations specially directed along certain lines determined by the laws of organic combination. As Professor Huxley has said, "It is quite conceivable that every species tends to produce varieties of a limited number and kind, and that the effect of natural selection is to favour the development of some of these, while it opposes the development of others along their predetermined line of modification." Mr. Gulick[BQ] and others have been led to believe in a tendency to divergent evolution residing in organic life-forms. Such a tendency might be due to special modes of organic combination giving rise to particular lines of divergence. Again, we have seen that some naturalists believe that specific characters are not always of utilitarian significance. But, as was before pointed out, on the hypothesis of all-round variation, there is nothing to give these non-useful specific characters fixity and stability, nothing to prevent their being swamped by intercrossing. If, however, on the hypothesis of combination, we have definite organic compounds, instead of, or as well as, mere hereditary mixtures; if, in other words, variations take definite lines determined by the laws of organic combination (as the nature and properties of chemical compounds are determined by the laws of chemical combination), then this difficulty disappears. There is no reason why a neutral divergence—one neither useful nor deleterious—should be selected or eliminated. And if its direction is predetermined, there is no reason why it should not persist, though, of course, it will not be kept at a high standard by elimination. It has again and again been pointed out as a difficulty in the path of natural selection that, in their first inception, certain characters or structures cannot yet be of sufficient utility to give the possessor much advantage in the struggle for existence. If, however, these be definite products of organic combination, this difficulty also disappears. So long as they are not harmful, they will not be eliminated, and by fortunate combinations will progress slowly until natural selection gets a hold on them and pushes them forward, developing to the full the inherent tendency. Finally, we must notice that, on this hypothesis, our conception of panmixia, or intercrossing, would have to be modified. As generally held, this doctrine is based upon hereditary mixture, not organic combination. It is a doctrine of means and averages. There is a good deal of evidence that intercrossing does not, at least in all cases, produce mean or average results. And according to the hypothesis of organic combination, it need not always do so. According to this hypothesis, then, divergent modifications might arise and be perpetuated without the necessity of isolation. Sterility might result from the fact that divergence had been carried so far that organic combination was no longer possible; reversion, due to intercrossing, from the fact that combinations long rendered impossible by the isolation of the necessary factors in distinct varieties, are again rendered possible when these varieties interbreed.

On this hypothesis of organic combination, to which we shall recur in the chapter on "Organic Evolution," the varied forms of animal life are the outcome of definite organic products with definite organic structure, analogous to the definite chemical compounds with definite crystalline and molecular structure; and the analogy between the regeneration of hydra and the reconstruction of a crystal is carried on a step further. I do not say that I am myself at present prepared to adopt the hypothesis, at least in this crude form; but it is, perhaps, worth a passing consideration. Its connection with Mr. Herbert Spencer's doctrine of physiological units is obvious. The analogy there is with crystallization; here it is with chemical combination.

We must now return to the point which gave rise to this digression, and repeat that mere hereditary commixture in the union of ovum and sperm cannot give rise to new characters or raise existing structures (1) where there is free intercrossing beyond the mean of the species, and (2) where there is rigorous elimination beyond the existing maximum of the species. Variations beyond this existing maximum must be due to some other cause.

Professor Weismann has suggested, as a cause of variation, the extrusion of the polar cells from the ovum. It has before been mentioned that, generally previous to fertilization, the ripe ovum buds off two minute polar bodies. The nucleus of the ovum divides, and one half is extruded in the first polar cell; the nucleus then (except in parthenogenetic[BR] forms, where there is no union of ovum and sperm) again divides, and a second polar cell is extruded. In accordance with his special view of the absolute distinction between the body-plasm and the germ-plasm, the first polar cell is formed to carry off the body-plasm of the ovum-nucleus. For the ovum, besides being a germ-bearer, is a specialized cell, and its special form is determined by the body-plasm it contains. This is got rid of in the first polar cell, and nothing but germ-plasm remains. Now, if nothing further took place, all the ova of this same individual containing similar germ-plasm would be identical, and similarly with all the sperms from the same parent. The union of these similar ova from one parent with similar sperms from another should therefore give rise to similar offspring. But the offspring are not all similar; they vary. Professor Weismann here makes use of the second polar cell.[BS] "A reduction of the germ-plasm," he says, "is brought about by its formation, a reduction not only in quantity, but above all, in the complexity of its constitution. By means of the second nuclear division, the excessive accumulation of different kinds of hereditary tendencies or germ-plasms is prevented. With the nucleus of the second polar body, as many different kinds of plasm are removed from the egg as will be afterwards introduced by the sperm-nucleus." "If, therefore, every egg expels half the number of its ancestral germ-plasms during maturation, the germ-cells of the same mother cannot contain the same hereditary tendencies, unless we make the supposition that corresponding ancestral germ-plasms chance to be retained by all eggs—a supposition that cannot be sustained."

The two polar cells are therefore, on this view, of totally different character; and the nuclear division in each case of a special kind and sui generis. I do not think that the evidence afforded by observation lends much support to this view. But with that we are not here specially concerned. We have to consider how this reduction of the number of ancestral germ-plasms can further the kind of variation required. Now, it is difficult to see, and Professor Weismann does not explain, how the getting rid of certain ancestral tendencies can give rise to new characters or the enhancement of old characters. One can understand how this "reducing division," as Dr. Weismann calls it, can reduce the level of now one and now another character. But how it can raise the level beyond that attained by either parent is not obvious. It is perhaps possible, though Professor Weismann does not, I think, suggest it, that, by a kind of compensation,[BT] the reduction of certain characters may lead to the enhancement of others. Let us revert to the illustration on [p. 150], where each individual has an available store of forty units of growth-force; and let us express by the minus sign the units lost in the parents by the extrusion of the polar cell and an analogous process which may occur in the genesis of the sperm. Then the units of growth-force which may thus be lost by a "reducing division" in b, c, and e may be, in the offspring, applied to the further growth of a; thus—

Parents.Offspring.
a 10 10 14
b 8-110-37
c9-1 5-1 6
d7 9 8
e6-2 6 5