August 14, 1908] 



SCIENCE 



211 



however, under the flag of Weismannism and 

 nails this flag firmly to the mast, refusing to 

 believe, for example, that any explanation of 

 the regeneration of a Stentor or a Hydra is 

 possible " apart from the postulate of diffusely 

 distributed ' specific units.' " True, he admits 

 that determinants are "scientific fictions," 

 that they are elements of a " symbolic nota- 

 tion " to be discarded so soon as it is shown 

 to be inconsistent with demonstrable facts, 

 but nevertheless he believes that heredity can 

 be discussed and understood at present only 

 on the assumption of the existence of such 

 material bases of inheritance. But that such 

 a concept is merely carrying the difiiculty a 

 step farther back is not considered. The de- 

 terminants are living entities which grow and 

 reproduce, vary and inherit, even as the cell, 

 and, admitting their existence, we are still 

 far from understanding the ultimate causes of 

 the phenomena of heredity. Indeed, it would 

 seem that the problem of heredity and the 

 problem of life are fundamentally the same, 

 or, at all events, that the solution of the one 

 is dependent on that of the other, for, as Pro- 

 fessor Thomson puts it, " the organism and its 

 inheritance are, to begin with, one and the 

 same." But a complete comprehension of the 

 ultimate causes of life, of organization and of 

 inheritance is yet of the future, and it must 

 be admitted that the concept of determinants 

 or specific material units furnishes a con- 

 venient " notation " for the discussion of cer- 

 tain phenomena of inheritance ; it is not, how- 

 ever, the only concept possible, and it is to be 

 regretted that its dominance in Professor 

 Thomson has rendered him somewhat in- 

 tolerant of epigenetie possibilities. 



The book starts with definitions of heredity 

 and inheritance, and proceeds to discuss the 

 physical basis of inheritance, the germ cells, 

 their maturation and fertilization. Then fol- 

 lows a discussion of variation, which is 

 classified as fluctuating and discontinuous, a 

 consideration of the latter leading to a criti- 

 cism of the mutation theory of DeVries ; and, 

 similarly, a chapter is devoted to modes of 

 inheritance, which are classified as blended, 

 exclusive (preponderant, although discarded 



by the author, seems preferable) and particu- 

 late. A remarkably sane discussion of rever- 

 sion follows, the phenomenon being regarded 

 as of rare occurrence, and many of the cases 

 usually referred to it, such, for instance, as 

 the classic one of supernumerary mammse, are 

 properly excluded. Eeversion is defined as in- 

 cluding " all cases where, through inheritance, 

 there appears in an individual some character 

 or combination of characters which was not 

 expressed in his immediate lineage, but which 

 had occurred in a remoter but not hypothetical 

 ancestor." With a discussion of telegony and 

 maternal impressions, in which again a thor- 

 oughly judicious position is taken, and of the 

 inheritance of acquired characters, concern- 

 ing which the author's Weismannian convic- 

 tions determine his position, the exposition of 

 what may be termed the illustrative side of 

 the question is concluded, this having oc- 

 cupied approximately one half of the volume. 



A most interesting chapter on heredity and 

 disease precedes one on the statistical study 

 of inheritance, and this is followed by a con- 

 sideration of the results of experimental 

 studies, in which the Mendelian phenomena 

 are discussed. Then follows an account of the 

 theories of heredity, a relatively brief his- 

 torical review preceding an exposition of 

 Weismann's theories, and after a considera- 

 tion of heredity and sex and of sex determina- 

 tion, the book concludes with a most sug- 

 gestive and admirably expressed discussion of 

 the social aspects of the question. 



This summary of the contents of the book 

 will indicate the breadth of its scope and the 

 thoroughness with which its subject is dis- 

 cussed. Criticism is, of course, possible, but 

 the errors against which it must be directed 

 are rather of omission than of commission. 

 One would have wished, for instance, a fuller 

 statement of the method and results of the 

 statistical study of heredity, the chapter upon 

 that topic being principally a discussion of 

 Galton's views of inheritance, and space for 

 the enlargement of the chapter could have 

 been obtained by a curtailment of that devoted 

 to Weismann's theories, which have already 

 been made familiar to English readers. A 



