by August Weismann (memoirs translated by Meldola) by Edward B. Poulton (see his addresses and memoirs published in the Transactions of the Entomological Society and elsewhere), and by Abbot Thayer (Concealing Coloration in the Animal Kingdom, Macmillan & Co. 1910). In the branch of bionomics, however, concerned with the laws of variation and heredity (thremmatology), there has been considerable progress. In the first place, the continued study of human population has thrown additional light on some of the questions involved, whilst the progress of microscopical research has given us a clear foundation as to the structural facts connected with the origin of the egg-cell and sperm-cell and the process of fertilization.
Great attention has been given lately to the important experiments upon the results of hybridizing certain cultivated varieties of plants which were published so long ago as 1865, by the Abbé Mendel, but failed to attract notice until thirty-five years later, sixteen years after his death (see Mendelism). Mendelism. Mendel’s object was to gain further knowledge as to result of mixing by cross-fertilization or interbreeding two strains exhibiting diverse characters or structural features. The whole question as to the mixture of characters in offspring thus produced was—and remains—very imperfectly observed. Mendel’s observations constitute an ingenious attempt to throw light on the matter, and in the opinion of some biologists have led to the discovery of an important principle. Mendel made his chief experiments with cultivated varieties of the self-fertilizing edible pea. He selected a variety with some one marked structural feature and crossed it with another variety in which that feature was absent. Instances of his selected varieties are the tall variety which he hybridized with a dwarf variety, a yellow-seeded variety which he hybridized with a green-seeded variety, and again a smooth-seeded variety which he hybridized with a wrinkle-seeded variety. In each set of experiments he concentrated his attention on the one character selected for observation. Having obtained a first hybrid generation, he allowed the hybrids to self-fertilize, and recorded the result in a large number of instances (a thousand or more) as to the number of individuals in the first, second, third and fourth generations in which the character selected for experiment made its appearance. In the first hybrid generation formed by the union of the reproductive germs of the positive variety (that possessing the structural character selected for observation) with those of the negative variety, it is not surprising that all or nearly all the individuals were found to exhibit, as a result of the mixture, the positive character. In subsequent generations produced by self-fertilization of the hybrids it was found that the positive character was not present in all the individuals, but that a result was obtained showing that in the formation of the reproductive cells (ova and sperms) of the hybrid, half were endowed with the positive character and half with the negative. Consequently the result of the haphazard pairing of a large number of these two groups of reproductive cells was to yield, according to the regular law of chance combination, the proportion 1PP, 2PN, 1NN, where P stands for the positive character and N for its absence or negative character—the positive character being accordingly present in three-fourths of the offspring and absent from one fourth. The fact that in the formation of the reproductive cells of the hybrid generation the material which carries the positive quality is not subdivided so as to give a half-quantity to each reproductive cell, but on the contrary is apparently distributed as an undivided whole to half only of the reproductive cells and not at all to the remainder, is the important inference from Mendel’s experiments. Whether this inference is applicable to other classes of cases than those studied by Mendel and his followers is a question which is still under investigation. The failure of the material carrying a positive character to divide so as to distribute itself among all the reproductive cells of a hybrid individual, and the limitation of its distribution to half only of those cells, must prevent the “swamping” of a newly appearing character in the course of the inter-breeding of those individuals possessed of the character with those which do not possess it. The tendency of the proportions in the offspring of 1PP, 2PN, 1NN is to give in a series of generations a regular reversion from the hybrid form PN to the two pure races, viz. the race with the positive character simply and the race with the total absence of it. It has been maintained that this tendency to a severance of the hybrid stock into its components must favour the persistence of a new character of large volume suddenly appearing in a stock, and the observations of Mendel have been held to favour in this way the views of those who hold that the variations upon which natural selection has acted in the production of new species are not small variations but large and “discontinuous.” It does not, however, appear that “large” variations would thus be favoured any more than small ones, nor that the eliminating action of natural selection upon an unfavourable variation could be checked.
A good deal of confusion has arisen in the discussions of this latter topic, owing to defective nomenclature. By some writers the word “mutation” is applied only to large and suddenly appearing variations which are found to be capable of hereditary transmission, whilst the term “fluctuation” is applied to small variations whether capable of transmission or not. By others the word “fluctuation” is apparently applied only to those small “acquired” variations due to the direct action of changes in food, moisture and other features of the environment. It is no discovery that this latter kind of variation is not hereditable, and it is not the fact that the small variations, to which Darwin attached great but not exclusive importance as the material upon which natural selection operates, are of this latter kind. The most instructive classification of the “variations” exhibited by fully formed organisms consists in the separation in the first place of those which arise from antecedent congenital, innate, constitutional or germinal variations from those which arise merely from the operation of variation of the environment or the food-supply upon normally constituted individuals. The former are “innate” variations, the latter are “superimposed” variations (so-called “acquired variations”). Both innate and superimposed variations are capable of division into those which are more and those which are less obvious to the human eye. Scarcely perceptible variations of the innate class are regularly and invariably present in every new generation of every species of living thing. Their greatness or smallness so far as human perception goes is not of much significance; their real importance in regard to the origin of new species depends on whether they are of value to the organism and therefore capable of selection in the struggle for existence. An absolutely imperceptible physiological difference arising as a variation may be of selective value, and it may carry with it correlated variations which appeal to the human eye but are of no selective value themselves. The present writer has, for many years, urged the importance of this consideration.
The views of de Vries and others as to the importance of “saltatory variation,” the soundness of which was still by no means generally accepted in 1910, may be gathered from the articles Mendelism and Variation. A due appreciation of the far-reaching results of “correlated variation” must, it appears, give a new and distinct explanation to the phenomena which are referred to as “large mutations,” “discontinuous variation” and “saltatory evolution." Whatever value is to be attached to Mendel’s observation of the breaking up of self-fertilized hybrids of cultivated varieties into the two original parent forms according to the formula “1PP, 2PN, 1NN,” it cannot be considered as more than a contribution to the extensive investigation of heredity which still remains to be carried out. The analysis of the specific variations of organic form so as to determine what is really the nature and limitation of a single “character” or “individual variation,” and whether two such true and strictly defined single variations of a single structural unit can actually “blend” when one is transmitted by the male parent and the other by the female
