An Examination of Weismannism

by George John Romanes

It is a well-known and admitted fact that if a bitch has two litters by two different dogs, the character of the first father is sure to be perpetuated in any litters she may afterwards have, no matter how pure-bred a dog may be the begetter.

After citing this testimony, Mr. Fookes goes on to give illustrations known to himself.

A friend of mine near this had a very valuable Dachshund bitch, which most unfortunately had a litter by a stray sheep-dog. The next year her owner sent her on a visit to a pure Dachshund dog, but the produce took quite as much of the first father as the second, and the next year he sent her to another Dachshund with the same result. Another case:—A friend of mine in Devizes had a litter of puppies, unsought for, by a setter from a favourite pointer bitch, and after this she never bred any true pointers, no matter of what the paternity was.

These further evidences, to which Mr. Fookes has since added others, render the general conclusion incontestable. Coming from remote places, from those who have no theory to support, and who are some of them astonished by the unexpected phenomena, the agreement dissipates all doubt. In four kinds of mammals, widely divergent in their natures—man, horse, dog, and pig—we have this same seemingly anomalous kind of heredity made visible under analogous conditions. We must take it as a demonstrated fact that, during gestation, traits of constitution inherited from the father produce effects upon the constitution of the mother; and that these communicated effects are transmitted by her to subsequent offspring. We are supplied with an absolute disproof of Professor Weismann’s doctrine that the reproductive cells are independent of, and uninfluenced by, the somatic cells; and there disappears absolutely the alleged obstacle to the transmission of acquired characters....

There is one other passage in Dr. Romanes’ criticism—that concerning the influence of a previous sire on progeny—which calls for comment. He sets down what he supposes Weismann will say in response to my argument. “First, he may question the fact.” Well, after the additional evidence given above, I think he is not likely to do that; unless, indeed, it be that along with readiness to base conclusions on things “it is easy to imagine” there goes reluctance to accept testimony which it is difficult to doubt. Second, he is supposed to reply that “the germ-plasm of the first sire has in some way or another become partly commingled with that of the immature ova”; and Dr. Romanes goes on to describe how there may be millions of spermatozoa and “thousands of millions” of their contained “ids” around the ovaries, to which these secondary effects are due. But, on the one hand, he does not explain why in such case each subsequent ovum, as it becomes matured, is not fertilized by the sperm-cells present, or their contained germ-plasm, rendering all subsequent fecundations needless; and, on the other hand, he does not explain why, if this does not happen, the potency of this remaining germ-plasm is nevertheless such as to affect not only the next succeeding offspring, but all subsequent offspring. The irreconcilability of these two implications would, I think, sufficiently dispose of the supposition, even had we not daily multitudinous proof that the surface of a mammalian ovarium is not a sperm-atheca. The third difficulty Dr. Romanes urges is the inconceivability of the process by which the germ-plasm of a preceding male parent affects the constitution of the female and her subsequent offspring. In response, I have to ask why he piles up a mountain of difficulties based on the assumption that Mr. Darwin’s explanation of heredity by “Pangenesis” is the only available explanation preceding that of Weismann? and why he presents these difficulties to me more especially, deliberately ignoring my own hypothesis of physiological units? It cannot be that he is ignorant of this hypothesis, since the work in which it is variously set forth (“Principles of Biology,” §§ 66-97) is one with which he is well acquainted: witness his “Scientific Evidences of Organic Evolution”; and he has had recent reminders of it in Weismann’s “Germ-plasm,” where it is repeatedly referred to. Why, then, does he assume that I abandon my own hypothesis and adopt that of Darwin, thereby entangling myself in difficulties which my own hypothesis avoids? If, as I have argued, the germ-plasm consists of substantially similar units (having only those minute differences expressive of individual and ancestral differences of structure), none of the complicated requirements which Dr. Romanes emphasises exists, and the alleged inconceivability disappears.

To this I responded, in the Contemporary Review for June:—

With regard to the influence of a previous sire, I ventured in my article to show that, even supposing it to be a fact, the phenomena concerned would not constitute any valid evidence against Weismann’s theory of germ-plasm, and, of course, still less would “they prove that while the reproductive cells multiply and arrange themselves during the evolution of the embryo, some of their germ-plasm passes into the mass of somatic cells constituting the parental body, and becomes a permanent component of it,” with the result that the phenomena in question “are simply fatal to Weismann’s hypothesis.” For a much simpler and more probable explanation is to be found in supposing that the unused germ-plasm of the first sire may survive the disintegration of its containing spermatozoa in the Fallopian tubes of the female, and thus gain access to the hitherto unripe ova directly, instead of first having to affect the whole maternal organism, and then being reflected from it to them. I showed, at some length, how immensely complex the mechanism of any such process would necessarily have to be; and for the purposes of exposition I employed the terminology of Darwin’s theory of Pangenesis. Mr. Spencer now says: “In response, I have to ask why he [I] piles up a mountain of difficulties based on the assumption that Mr. Darwin’s explanation of heredity by ‘Pangenesis’ is the only available explanation preceding that of Weismann? and why he presents these difficulties to me more expecially, deliberately ignoring my own hypothesis of physiological units?” Now my answer to this is very simple. I do not hold a brief for Weismann. On the contrary, I am in large measure an opponent of his views; and my only object in publishing my previous article was to save the theory of use-inheritance from what seemed to me the weaker parts of Mr. Spencer’s advocacy, while thus all the more emphasizing my acceptance of its stronger parts. Therefore, the impression which he seems to have gained from my attempts at impartiality is entirely erroneous. Far from “deliberately ignoring” any of his arguments or hypotheses which seemed to me at all available on the side of use-inheritance, I everywhere endeavoured to make the most of them. And, as regards this particular instance, I expressly used the term “gemmules,” instead of “physiological units,” simply because I could not see that, as far as my “mountain of difficulties” was concerned, it could make one atom of difference which term I employed. It now appears, however, that, in Mr. Spencer’s opinion, there is some very great difference. For, while he allows that the “mountain of difficulties” which I have “piled up” against his interpretation of the alleged phenomena would be valid on the supposition that the ultimate carriers of heredity are “gemmules,” he denies that such is the case if we suppose these ultimate carriers to be “physiological units.” For this statement, however, he gives no justification; and, as I am unable to conceive wherein the difference lies, I sincerely hope that in any subsequent editions of his pamphlet Mr. Spencer will furnish the requisite explanation. Gladly substituting the words “physiological units” wherever I have used the word “gemmules,” I am genuinely anxious to ascertain how he would overcome the “mountain of difficulties” in question. For I do not regard the subject as one of mere dialectics. It is a subject of no small importance to the general issue, Weismann versus Lamarck; and, therefore, if Mr. Spencer could show that the phenomena in question make exclusively in favour of the latter, as he alleges, he might profitably inform us in what way he supposes them to do so.

In conclusion, I would like to take this opportunity of explaining that my former article was written in Madeira, where I did not receive a copy of Weismann’s most recent work, entitled The Germ-plasm, until the Contemporary Review for April was being printed off. Thus, I was not then aware that in this work Professor Weismann had fully anticipated several of Mr. Spencer’s criticisms—including this matter of the influence of a previous sire. Here he adopts exactly the position which in my article I surmised that he would; so that, to all who have read The Germ-plasm, it must have appeared that I was prophesying after the event. Hence the need of this explanation.

Lastly, in the same issue of the Contemporary Review, Mr. Spencer explained:—