REVIEWS 789 



be thrown away, but it has been recognised for long that their work will have to 

 be followed by some attempt to delimit what have been called " master species," 

 the real units into which the forms of life are thrown, and that their named species 

 and museum types will at the most serve as index numbers. But at the least it 

 may be said that the large series of individuals, with localities and dates carefully 

 marked, that are now being collected in museums, will be of great service when 

 the work of systematists comes to be translated into science. 



The second chapter opens with a statement so astonishing that it is difficult to 

 qualify it with any other term than the term " perverse." " Twenty years ago," 

 declares Mr. Bateson, " in describing the facts of variation, argument was necessary 

 to show that these phenomena had a special value in the sciences of zoology and 

 botany." I cannot conceive how this proposition could be justified, and it is an 

 introduction entirely unnecessary for the extremely interesting discussion to which 

 it leads. Mr. Bateson offers as a preliminary classification of the facts of varia- 

 tion, the distinction between those which are the results of changes in the mode 

 of division, and those which relate to differentiation in the substances divided. 

 He suggests that the first set of variations are possibly in the last resort dependent 

 on the second, and offers valuable comments on the nature of the two processes. 

 He is inclined to think, in this differing from perhaps a majority of modern 

 writers, that substantive variations, depending on differentiation of the substances 

 divided, are more easy to understand than meristic variations. The former may 

 be due to some kind of chemical process ; the latter, and, indeed, the nature of 

 division itself, so far remain nothing but observed facts of life. Two chapters are 

 occupied with a clear description, illustrated by many examples, of the different 

 kinds of merism and segmentation, and of the recovery of symmetry in dividing 

 parts. In Chapter IV. substantive variations are discussed, and the attempt 

 is made to correlate them with the unit factors of Mendelian analysis. The 

 suggestion is developed that recessive characters are due to the omission of a 

 factor and dominant characters to the addition of a factor, but this distinction is 

 abandoned in an appendix, and we are left with the tentative picture that all 

 substantive variation is due to the loss of a pre-existing factor — a conception that 

 does not appear to carry the argument on to any very useful plane. In Chapter V. 

 there is a very fair account of the failure of Mendelian analysis to account for the 

 phenomena of mutation in Oenothera. 



In three interesting and detailed chapters the relations between geographical 

 distribution and variation are discussed, with, however, the result of showing that 

 Mendelian methods are as yet no more satisfactory than any other methods in 

 explaining the relation of geographical races to species. It is obvious, as Mr. 

 Bateson suggests, that no great advance can be made in this direction until 

 extensive breeding experiments have been undertaken. In Chapters IX. and X., 

 under the title " Adaptation," Mr. Bateson discusses recent evidence for the 

 inheritance of acquired characters, and dismisses it partly on the ground that it 

 is unconfirmed, but even more cogently because in the alleged cases the normal 

 course of inheritance under undisturbed conditions is not sufficiently known. The 

 utmost length to which Mr. Bateson thinks the evidence can be stretched is to 

 suppose that in some parthenogenetic forms, variations, produced in response to 

 special conditions, recur in one or two generations after the removal of those 

 conditions, and that violent maltreatment may in rare cases so affect the germ 

 cells contained in the parents as to bring about in the offspring, resulting from 

 the fertilisation of these germ cells, an arrest of development similar to that which 

 their parents underwent. Examination of the present condition of knowledge as 



