How far dos Mendel's Law apply in Crosses of various kinds? The relationship of the Mendelian principle to crosses of different kinds is of such great importance that it is necessary to consider it in some detail, even though we may agree with Prof. Bateson (p. 285) that any safe conchsin is still far off. De Vries holds strongly that the difference between elementary species, or, as they are often called, smaller or sub-species, on the one hand, and varieties on the other, is quite a marked one' (p. 128). Among other points of distinction, he maintains that a ross between a species and its variety obeys Mendel's law, while a cross between one sub-species and another does not. Before dealing with this important distinction it is convenient to give some account of the other differences which de Vries recognises between varieties and elementary species. He tells us that the latter arise by the acquisition of entirely new characters'; while varieties arise by the loss of existing qualities or by the gain of such peculiarities as may already be seen in closely allied species' (p. 141). In nearly every instance, where true varieties and not elementary species are concerned, a single term expresses the whole character.... On the contrary we find elementary species in different genera based on the greatest possible diversity of features' (p. 126). Thus de Vries says that his new mutations or large and sudden variations of Enothera lamarckiana, pigns and rubrinervis, oblonga and albida, obviously bear the characters of progressive elementary species. They are not diferentiated from [the parent] lamarckiana by one or two features. They diverge from it in nearly all organs, and in all in a definite though small degree' (p. 565). Comparing these mutations with the two hundred 'elementary species of Draba and other similar instances the same as the sub-species' or 'geographical varieties' recognised by It must be borne in mind that de Vries' 'elementary species' are not magists in general. De Vries splits up the well-known Linnean species d plants into numerous elementary species which commonly occur interized together, freely interbreeding in the same locality. The zoologist pes the term 'sub-species' to a local race occupying a different geographical area from that of allied sub-species (see p. 15). we find the same main feature, the minute differences in nearly all points' (pp. 565, 566). De Vries' varieties are of two kinds, negative or retrograde, formed by the loss of an old character; positive or progressive, formed by the addition of a new one; 'And as it is more easy to lose what one has than to obtain something new, negative varieties are much more common than are positive ones (p. 130). Whether this interpretation be valid or not, de Vries finds that the negative form is the one which prevails nearly everywhere; . . . positive aberrations are: in a general sense so rare that they might even be taken for exceptions to the rule' (p. 134). Examples of the single characters by which such negative or retrograde varieties are distinguished are whiteness in place of colour in flowers, smoothness in place of hairiness, etc. Two or more such single characters may be combined in cultivation, such as the dwarf habit and the whiteness of flowers. Elementary species, on the other hand, have already been stated to diverge from the parent species in a multiplicity of features. De Vries finds in their respective behaviour when hybridised 'the means of a still better distinction between elementary species and varieties. I will try to show that these two contrasting groups behave in quite a different manner, when subjected to crossing experiments; and that the hope is justified that some day crosses may become the means of deciding, in any given instance, what is to be called species, and what variety, on physiological grounds' (p. 251). When, however, this distinguished botanist proceeds to explain the different results produced by such crossings, we are met at the outset with a painful inconsistency. In spite of all that de Vries has said about elementary species differing in nearly all points,' he introduces his account of hybridising as follows (pp. 252, 253): Elementary species differ from their nearest allies by progressive changes, that is, by the acquirement of some new character. The derivative species has one unit more than the parent. All other qualities are the same as in the parent.' The words I have italicised entirely contradict every statement previously made by de Vries on the same subject. The further development of de Vries' hybrid isation test is consistent with the above contradiction In crosses between species and varieties, on the other band, the differentiating character is not unpaired, being, as de Vries brings much evidence to show, present, alform of variety, distinguished by the loss of a character, though in a latent state, in the variety. In the common de Vries infers that this latter, which is present and active in the species,' is also present but dormant in the variety (p. 254). He therefore concludes that in the Crossing of varieties no unpaired remainder is left, all tion (p. 255). 'Summarising this discussion, we may units combining in pairs exactly as in ordinary fertilisaconclude that in normal fertilisation and in the interCrossing of varieties all characters are paired, while in crosses between elementary species the differentiating marks are not mated' (p. 255). For these two types the terms 'balanced' and 'un balanced crosses' are proposed. Unbalanced crosses between elementary species) commonly show a diminutim of fertility, the sterility being,' broadly speaking, the greater, the less the affinity between the parents' A21). Such crosses also lead to 'a constant offspring, as far as experience now goes' (p. 275). Balanced crosses (between a species and its variety), on the other hand, Dever show any tendency to diminished fertility. 'Hence there can be little doubt that the unpaired units are the se of this decrease in reproductive power' (p. 262). on the other hand, 'there is no reason for a diminution la the balanced cross between variety and species, of the fertility, as all characters are paired in the hybrid, and no disturbance whatever ensues in its interna structure' (p. 277). The paired varietal character will according to de Vries, exhibit the appearance of the active unit of the pair and not the latent; in other words. the hybrid formed by the first cross will resemble the species and not the retrograde variety. In Mendelian terminology the active unit of the species is 'dominant, while the corresponding latent unit in the variety is ' recessive.' When the hybrids of the first filial genera tion (F. 1) are interbred, the offspring do not exhibit the constancy of those produced by the unbalanced crosses, but split up into the well-known Mendelian proportions— one-fourth pure recessive, one-fourth pure dominant, and one-half hybrid like their parents, and therefore bearing the same appearance as the pure dominants and as the parent species. as a We may note that Prof. Bateson, in his latest work, comes to the following conclusion about dominance : 'The dominance of certain characters is often an important but never an essential feature of Mendelian heredity. Those who first treated of Mendel's work most unfortunately fell into the error of enunciating a "Law of Dominance proposition comparable with the discovery of segregation. Mendel himself enunciates no such law . . . it is only a subordinate incident of special cases; and Mendel's principles of inheritance apply equally to cases where there is no dominance and the heterozygous [hybrid] type is intermediate in character between the two pure types' (pp. 13, 14). A good deal of unnecessary discussion has been raised about the progressive quality which has been supposed to characterise Mendelian dominance. Progress, if it means anything, means onward evolution; and this is determined by the relationship at the moment between the organism and its environment, especially its organic environment. The acquisition of a new structure may be progress in one environment; its loss may equally be progress in a different environment. The loss of limbs by snakes and snake-like lizards was progress, as also was the reduction of digits in the ancestors of the horse, and the endless simplifications of parasitic existence. Omitting reference to progress, which, as shown above, is a question of the environment, we have already seen (see p. 31) that de Vries concludes-and his conclusion is based on a very broad experience-that the presence of a character in the species is dominant over its loss, or rather later, in a 'retrogressive variety.' Although convinced of the high importance and interest of Mendel's principle, de Vries firmly maintains that it isngidly limited in its application. It is only of late years' (he says) 'that it has assumed a high place in scientific literature, and attained the first rank as an instation on fundamental questions of heredity. Read in thelight of modern ideas on unit-characters it is now one of de most important works on heredity and has already widegread and abiding influence on the philosophy of hybridism in general. But from its very nature and from the choice of the material made by Mendel, it is restricted to balanced or varietal crosses' (pp. 293, 294). 'Ordinary species. . . differ from each other partly in specific, partly in varietal characters. As to the first, they give in their hybrids stable peculiarities, while as to the latter, they split up according to Mendel's law' (p. 307). these De Although Prof. Bateson, probably rightly, looks upon The results were entirely negative so far as Mendelian segre gation is concerned. . . . The authors consider that the be- Vol. 211.-No. 420. D |