The author's meaning would probably have been better expressed if he had written 'ce qui paraît empêcher.' By 'modification reversible' he means a change in the ovum which will produce in the next generation a somatic modification similar to that by which it was produced. It seems natural to think of the influence of the ovum on the body and of the body on the ovum as of similar kind but in opposite directions, but it must be remembered always that the development of the body from the ovum Is not an influence at all but a direct conversion by cell-division and differentiation of the ovum into the body.
Delage argues that if the egg contains the substances characteristic of certain categories of cells of the organism it ought to be affected at the same time as those cells and by the same agents. He thinks that the egg only contains the substances or the arrangements characteristic of certain general functions (nervous, muscular, perhaps glandular of divers kinds) but without attribution to localised organs. In his view there is no representation of parts or of functions in the ovum, but a simple qualitative conformity of constitution between the egg and the categories of cells which in the body are charged with the accomplishment of the principal functions. Thus mutilations of organs formed of tissues occurring also elsewhere in the body cannot be hereditary, but if the organ affected contains the whole of a certain kind of tissue such as liver, spleen, kidney, then the blood undergoes a qualitative modification which reacts on the constitution of the egg.
Suppose the internal secretion of a gland (e.g. glucose for the liver, glycolytic for the ferment for the pancreas) is the physiological excitant for the gland. If the gland is removed in whole or in part the proportion of its internal secretion in the blood will be diminished. Then the gland, if the suppression is partial, will undergo a new diminution of activity But in, the egg the specific substance of the gland will also be less stimulated, and in the next generation a diminution of the gland may result. Thus Delage states Massin found that partial removal of the liver in rabbits had an inherited effect. In the case of excretory glands the contrary will be the case, for their removal causes increase in the blood of the exciting urea and uric acid.
The effects of disuse are similar to those of mutilations and of use vice versa. Delage, as seen above, does not consider that increase or decrease of particular muscles can be inherited, but only the muscular system in general. If, however, in consequence of the disuse of a group of muscles there was a general diminution of the inherited muscular system, the special group would remain diminished while the rest were developed by use in the individual: there would thus be a heredity produced indirectly. With regard to general conditions of life, Delage states that there are only two of which we know anything—namely, climate and alimentation—and he merely suggests that temperature and food act at the same time on the cells of the body and on the similar substances in the egg.
H. M. Vernon (Variation in Animals and Plants, 1903, pp. 351 seq.) cites instances of the cumulative effects of changed conditions of life, and points out that they are not really instances of the inheritance of acquired characters, but merely of the germ-plasm and the body tissues being simultaneously affected. He then asks, Through what agency is the environment enabled to act on the germ-plasm? And answers that the only conceivable one is a chemical influence through products of metabolism and specific internal secretions. He cites several cases of specific internal secretions, making one statement in particular which seems unintelligible, viz. that extirpation of the total kidney substance of a dog leads not to a diminished secretion of urine but to a largely increased secretion accompanied by a rapid wasting away which soon ends fatally.
Whenever a changed environment acts upon the organism, therefore, it to some extent affects the normal excretions and secretions of some or all of the various tissues, and these react not only on the tissues themselves, but also to a less degree upon the determinants representing them in the germ-plasm. Thus the relative size of the brain has decreased in the tame rabbit. This may be due to disuse; the excretions and secretions of the nervous tissues would be diminished, and the corresponding determinants less stimulated. Another instance is afforded by pigmentation of the skin in man; which varies with the amount of light and heat from the sun to which the skin is exposed. Specific excretory products of pigment in the skin may stimulate the pigment determinants in the germ-plasm to vigour. But only those characters of which the corresponding tissues possess a specific secretion or excretion could become hereditary in this way. For instance, the brawny arm of the blacksmith could not be transmitted, as it is scarcely possible that the arm muscles can have a secretion different from that of the other muscles.
In 1904, P. Schiefferdecker [Footnote: P. Schiefferdecker, Ueber Symbiose. S.B. d. Niederrhein. Gesellsch. zu Bonn. Sitzung der Medicinischen Sektion, 13 Juni 1904.] made the definite suggestion that the presence of specific internal secretions could be very well used for the explanation of the inheritance of acquired characters. When particular parts of the body were changed, these modifications must change the mixture of materials in the blood by the substances secreted by the changed parts. Thereby would be found a connexion between the modified parts of the body and the germ-cells, the only connexion in existence. It is to be assumed, according to this author, that only a qualitative change in the nutritive fluid of the germ-cells could produce an effect: a quantitative change would only cause increased or decreased nourishment of the entire germ cells.
In my own volume on Sexual Dimorphism in the Animal Kingdom, published in 1900, I attempted to explain the limitation of secondary sexual characters not only to one sex, but usually to one period of the individual life, namely, that of sexual maturity; and in some cases, as in male Cervidae, to one season of the year in which alone the sexual organs are active. It had been known for centuries that the normal development of male sexual characters did not take place in castrated animals, but the exact nature of the influence of the male generative organs on that development was not known till a year or two later than 1900, when it was shown to be due to an internal secretion. My argument was that all selection theories failed to account for the limitation of secondary sexual characters in heredity, whereas the Lamarckian theory would explain them if the assumption were made that the effects of stimulation having been originally produced when the body and tissues were under the influence of the sexual organs in functional activity, these effects were only developed in heredity when the body was in the same condition.
About the year 1906, when preparing two special lectures in London University on the same subject, I became acquainted with the work of Starling and others on internal secretions or hormones, and saw at once that the hormone from the testes was the actual agent which constituted the 'influence' assumed by me in 1900. In these lectures I elaborated a definite Lamarckian theory of the origin of Secondary Sexual Characters in relation to Hormones, extending the theory also to ordinary adaptive structures and characters which are not related to sex. Having met with many obstacles in endeavouring to get a paper founded on the original lectures published in England, I finally sent it to Professor Wilhelm Roux, the editor of the Archiv für Entwicklungsmechanik der Organismen, in which it was published in 1908.
In his volume on the Embryology of the Invertebrata, 1914 (Text-Book of Embryology, edited by Walter Heape, vol. i.), Professor E. W. MacBride in his general summary (chapter xviii.) puts forward suggestions concerning hormones without any reference to those who have discussed the subject previously. He considers the matter from the point of view of development, and after indicating the probability that hormones are given off by all the tissues of the body, gives instances of organs being formed in regeneration (eye of shrimp) or larvae (common sea-urchin) as the result of the presence of neighbouring organs, an influence which he thinks can only be due to a hormone given off by the organ already present. He then states that Professor Langley had pointed out to him in correspondence that if an animal changes its structure in response to a changed environment, the hormones produced by the altered organs will be changed. The altered hormones will circulate in the blood and bathe the growing and maturing genital cells. Sooner or later, he assumes, some of these hormones may become incorporated in the nuclear matter of the genital cells, and when these cells develop into embryos the hormones will be set free at the corresponding period of development at which they were originally formed, and reinforce the action of the environment. In this way MacBride attempts to explain recapitulation in development and the tendency to precocity in the development of ancestral structures. His idea that the hormones act by 'incorporation' in the genital cells is different from that of stimulation of determinants put forward by myself and others, but it is surprising that he should refer to unpublished suggestions of Professor Langley, and not to the publications of authors who had previously discussed the possible action of hormones in connexion with the heredity of somatic modifications.