But in demonstrating against the Neo-Lamarckians that somatic modifications unrepresented in the germ plasm could have no significance in the process of racial evolution, Weismann had _proved too much_. His argument was no less telling against Darwinism than it was against Lamarckism. Darwin’s “individual differences” or “slight variations,” now spoken of as fluctuations, were quite as unrepresented and unrecorded in the germ cells as Lamarck’s “acquired adaptations.” There can be no “summation of individual differences” for the simple reason that fluctuations have no germinal basis and are therefore uninheritable—“We must bear in mind the fact,” says Prof. Edmund Wilson, “that Darwin often failed to distinguish between non-inheritable fluctuations and hereditary mutations of small degree.” (Smithson. Inst. Rpt. for 1915, p. 406.) Fluctuations, as we have seen, are due to variability in the environmental conditions, _e.g._ in access to soil nutrients, etc. As an instance of fluctuational variation the seeds of the ragweed may be cited. Normally these seeds have six spines, but around this average there is considerable fluctuation in individual seeds, some having as many as nine spines and others no more than one. Yet the plants reared from nine-spine seeds, even when similarly mated, show no greater tendency to produce nine-spine seeds than do plants reared from one-spine seeds. To meet the difficulty presented by the non-inheritability of the Lamarckian adaptation and the Darwinian fluctuation, De Vries substituted for them those rare and abruptly-appearing inheritable variations, which he called mutations[2] and regarded as elementary steps in the evolutionary process. This new version of transformism was announced by De Vries in 1901, and more fully explained in his “Die Mutations-Theorie” (Leipzig, 1902-1903). Renner has shown that De Vries’ new forms of Œnothera were cases of complex hybridization rather than real mutants, as the forms produced by mutation are now called. Nevertheless, the work of Morgan, Bateson, and others leaves little doubt as to the actual occurrence of _factorial_ mutants, while Dr. Albert F. Blakeslee has demonstrated the existence of _chromosomal_ mutants. When unqualified, the term mutant usually denotes the factorial mutant, which arises from a change in one or more of the concatenated genes (hereditary factors) of a single chromosome (nuclear thread) in the germinal (_i.e._ gametic) complex. All such changes are called factorial mutations. They are hereditarily transmissible, and affect the somatic characters of the race permanently, although, in rare cases, such as that of the bar-eyed Drosophila mutant, the phenomenon of _reversion_ has been observed. The chromosomal mutant, on the contrary, is not due to changes in the single factors or genes, but to duplication of one or more entire chromosomes (linkage-groups) in the gametic complex. Like the factorial mutant, it produces a