"The marvellous phenomena initiated by the meeting of sperm-cell and germ-cell, or rather of their nuclei, naturally suggest the conception of some quite special and peculiar properties possessed by these cells. It seems obvious that this mysterious power which they display of originating a new and complex organism, distinguishes them in the broadest way from portions of organic substance in general. Nevertheless, the more we study the evidence the more are we led towards the conclusion that these cells are not fundamentally different from other cells." The evidence he gives is: (1) that small fragments of tissue in many plants and inferior animals may develop into entire organisms; (2) that the reproductive organs producing eggs and sperms are organs of low organisation, with no specialities of structure "which might be looked for, did sperm-cells and germ-cells need endowing with properties unlike those of all other organic agents." "Thus, there is no warrant for the assumption that sperm-cells and germ-cells possess powers fundamentally unlike those of other cells."
To this it must be answered: (1) though sperm-cells and egg-cells, being living units, cannot be "fundamentally unlike" other living units, such as ordinary body-cells, yet they may be very unlike them; (2) that the germ-cells are very unlike ordinary body-cells is shown by the fact that they can do what no single body-cell can do, build up a whole organism; (3) so specific are germ-cells that in certain cases and in favourable conditions a small fraction of an egg, bereft of its own nucleus, may, if fertilised, develop into an entire and normal larva; (4) it is quite consistent with the idea of evolution that in lower organisms the contrast between body-cells and germ-cells should be less pronounced than in higher forms. But the fundamental answer is found when we inquire into the history of the germ-cells. In many cases, and the list is being added to, the future reproductive cells are segregated off at an early stage in embryonic development. Even before differentiation sets in, the future reproductive cells may be set apart from the body-forming cells. The latter develop in manifold variety into skin and nerve, muscle and blood, gut and gland; they differentiate, and may lose almost all protoplasmic likeness to the mother ovum. But the reproductive cells are set apart; they take no share in the differentiation, but remain virtually unchanged, continuing unaltered the protoplasmic tradition of the original fertilised ovum. After a while their division-products will be liberated as functional reproductive cells or germ-cells, handing on the tradition intact to the next generation.
An early isolation of the reproductive cells has been observed in the harlequin fly (Chironomus) and in some other insects, in the aberrant worm-type Sagitta, in leeches, in thread-worms, in some Polyzoa, in some small Crustaceans known as Cladocera, in the water-flea Moina, in some Arachnoids (Phalangidæ), in the bony fish Micrometrus aggregatus, and in other cases. In the development of the threadworm of the horse according to Boveri, the very first cleavage of the ovum establishes a distinction between somatic and reproductive cells. One of the first two cells is the ancestor of all the cells of the body; the other is the ancestor of all the germ-cells. "Moreover, from the outset the progenitor of the germ-cells differs from the somatic cells not only in the greater size and richness of the chromatin of its nucleus, but also in its mode of mitosis (division), for in all those blastomeres (segmentation-cells) destined to produce somatic cells a portion of the chromatin is cast out into the cytoplasm, where it degenerates, and only in the germ-cells is the sum-total of the chromatin retained" (E. B. Wilson, The Cell in Development and Inheritance, 1896, p. 111).
In the majority of cases, we admit, the reproductive cells are not to be seen in early segregation, and the continuous lineage from the fertilised ovum cannot be traced. In the majority of cases, the germ-cells are seen as such after considerable differentiation has gone on, and although they are linear descendants of the ovum, their special lineage cannot be traced. But it seems legitimate to argue from the clear cases to the obscure cases, and to say that the germ-cells are those cells which retain the complete complement of heritable qualities. Adopting the conception of the germ-plasm as the material within the nucleus which bears all the properties transmitted in inheritance, we may still say, in Weismann's words, "In every development a portion of this specific germ-plasm, which the parental ovum contains, is unused in the upbuilding of the offspring's body, and is reserved unchanged to form the germ-cells of the next generation.... The germ-cells no longer appear as products of the body, at least not in their more essential part—the specific germ-plasm; they appear rather as something opposed to the sum-total of body-cells; and the germ-cells of successive generations are related to one another like generations of Protozoa." In terms of this conception, which fits many facts, we may say that in plants and lower animals the distinction between germ-plasm and somato-plasm has not been much accentuated, and that in some organisms the body-cells retain enough undifferentiated germ-plasm to enable them in small or large companies to regrow an entire organism.
It may be said that Spencer must also have regarded the germ-cells as containing the whole complement of hereditary qualities. It must be so. The point is that he rejected the theory which gives a rational account of how the germ-cells have this content and their power of developing into an organism, like from like. The sentence in which he points out that the reproductive organs have "none of the specialities of structure which might be looked for, did the sperm-cells and germ-cells need endowing with properties unlike those of all other organic agents," shows how far he deliberately stood from the conception we have outlined.
Here we may note that the "Inductions" regarding Heredity are discussed in our eleventh chapter, and those regarding Variation in our twelfth chapter. We have not dealt with the suggestive concrete sections which deal with structural and functional evolution, partly because they are too concrete to be dealt with briefly, and partly because they are saturated with the hypothesis of the transmission of acquired characters. Spencer's most important conclusion in regard to the Laws of Multiplication is referred to under the heading Population.