accepting the same position, but differ in methods of reaching it: the conclusion of Descartes being the starting point with Reid (p. 106).
The fourth article consists of emendations of the French translation of Kant's Prolegomena published by Hachette, Paris, 1891.
G. M. Duncan.
Yale University.
When the author's 'Mutationstheorie' appeared (Vol. I, 1901, Vol. II, 1903), it was characterized as the most important contribution which had been made to the theory of evolution since the time of Darwin. Following that appearance, it has been shown that a number of independent lines of investigation converge to support and confirm the hypothesis advanced therein.
The 'Mutationstheorie' presented the complete detailed evidence, obtained from trustworthy historical records and from experimental researches carried on for twenty years, for this new theory of the origin of species. The present volume repeats some of these descriptions, but the results of new experiments have been added and a wider choice of material has been made from recent current literature on the subject. It is, however, the more important phases that are here emphasized.
At the beginning, it is important to note that, in the author's own opinion, the results of his work are in almost full accord with the principles laid down by Darwin. In one point, however, they differ. If Darwin's theory is a theory of selection, it is not necessarily a theory of descent, and that of de Vries is this. All evolutionary science is now based on the general idea of descent with modification, but this is quite independent of the modus in single instances of the change of one species into another.
The present work consists of twenty-eight lectures arranged in six groups. The first lecture is an introduction dealing with 'Theories of Evolution' and 'Methods of Investigation.' Natural selection is only a sieve, and is not a force of nature or a direct cause of improvement. It presupposes such a change. Darwin recognized two methods of change, the one, mutations, the other, variations or 'fluctuations.' Wallace and the Neo-Lamarckians reject the first, but in de Vries's opinion "species are not known to originate in any other way" than
