I could say something in answer to most of Delpino’s ingenious criticisms, but I will confine myself to one point; remarking, however, that I cannot see the force of his objection to the belief that the gemmules have the power of self-division.² The analogy of the multiplication within the body of an infected person of the contagious atoms of small-pox or scarlet-fever still seems to me fair.³ The criticism which at first struck me as much the most forcible, relates to the re-growth of an amputated limb or other organ.⁴ I have assumed, from what we know of the ordinary laws of growth, that gemmules become developed only when united with the nascent or very young cells which precede them in the due order of successive development. Now, Delpino remarks that when the limb of a mature salamander is cut off, nascent cells would not exist on the stump, but only mature ones; consequently, as he urges, the gemmules, according to the hypothesis, could not be there developed and form a new limb. But have we any reason for assuming that with animals the same cells or organic units endure throughout life and are not replaced by new cells? All physiologists admit that the whole frame is being constantly renovated. The bones, which it might be thought would be least capable of renewal, undergo before maturity great changes in form and size and in the diameter of the medullary canal; even after maturity their shape does not remain the same. If, then, it be admitted that the tissues throughout the body are constantly being renovated by means of the old cells giving birth through proliferation or division to new cells, the difficulty of the gemmules being developed at any point at which a limb may be cut off, disappears; for some nascent cells would exist there; and we must bear in mind that gemmules may possibly become developed in union with cells which are not strictly the proper ones, as shown in the extreme case of nails occasionally growing on the stumps of the amputated fingers of man. It is probable that the renovation of the tissues by the formation of new cells becomes less and less in old age, and so does the power of regrowth. The curious fact that insects which live but a short time after their final metamorphosis, have singularly little power of regrowth, may be partly explained by the deficiency in their tissues of nascent cells.

In the case of plants, the cells composing the tissues endure for life, and are not renewed. The woody cells in the trunk of a tree originally existed as delicate cells in the young shoot. Now, it is remarkable, considering how low plants stand in the scale of life, what little power of re-growth most of their organs or parts possess, not withstanding their high vitality. Cut off a young shoot, and the next bud below will form a new shoot; cut off an old stem, and hundreds of adventitious buds will be formed. No one, I believe, ever saw a mutilated leaf repaired in the same manner as the limb of a salamander; yet on the cut margins of certain leaves innumerable buds will be quickly developed. In the bark, on the other hand, new cells must be continually forming, for the outer layers are continually being thrown off, and bark can certainly reproduce itself on a decorticated surface (see Professor Oliver in Transactions of Tyneside Club, vol. iii. 1855, p. 67),⁵ from an isolated point not connected with the surrounding uninjured bark; so that this case is fairly analogous with that of the re-growth of an amputated limb.

I will add only one other remark. Delpino insists that, according to the hypothesis, the same cells must throw off gemmules at many successive periods of their development.⁶ But I can see no such necessity. Certain epidermic cells, for instance, form horn; but their conversion into this substance is apparently due to the chemical nature of the contents of the cells; and of the changes consequent on the absorption of certain elements. An atom or gemmule thrown off at an early period from one of these cells would, when properly nourished, first form a cell like the parent cell, and ultimately be converted into horn, without the necessity of throwing off successive gemmules. As, however, a cell is a complex structure, with its investing membrane, nucleus, and nucleolus, a gemmule, as Mr. G. H. Lewes has remarked in his interesting discussion on this subject (Fortnightly Review, Nov. 1, 1868, p. 508), must, perhaps, be a compound one, so as to reproduce all the parts.⁷ The attack by Delpino on the weak point of the hypothesis, namely, that it requires the support of several subordinate hypotheses, is perfectly just, though perhaps pushed to an extreme. But I can speak from recent experience, that he who has to consider complex cases of inheritance, as limited either separately or conjointly by sex, age, and season, with the inherited characters themselves and the form of inheritance liable to change from crossing and variability, will be able to disentangle the phenomena much more clearly, if he admits for the time our hypothesis with all its imperfections. He will then have fixed in his mind that transmission and development are quite distinct phenomena,—the gemmules being thrown off from all parts of the organization, and transmitted from both parents to both sexes at the earliest age, their development alone depending on the nature of the nascent tissues of the individual, whether permanently or temporarily modified by sex, age, season of the year, or other conditions.