May 5. 1910] 



NA TURE 



273 



n amongst the depths of the subject they are 

 ounding, and so the book lacks that invigorating 

 >hness that can only be given bj- men who are for 

 r probing the secrets of nature and working 

 ing the things they talk about — its atmosphere 

 \rong. 

 U'e cannot help thinking that the authors would 

 . e obtained much better results had the}' workeo 

 the model set many years ago by Johnson, one 

 the best writers on agricultural chemistry the 

 ■ted States, or, for that matter, any other countn.', 

 . ^r produced. He sets before the student accounts 

 -of the investigations that have made the subject, 

 shows pictures of the apparatus used in the classical 

 experiments, and gives some details of the actual 

 working. The result as a book that after forty years 

 still retains its freshness and its power of inspira- 

 tion, because it shows how men have wrestled with 

 mature to win her secrets. If when a second edition 

 of the present book is called for the authors would, 

 in a similar way, make room for some of the classical 

 work of the great masters, without extending the size 

 of the book, it is certain that their industn,- and 

 painstaking efforts would meet with a more fitting 

 reward. E. J. R. 



MECHANICS OF HEREDITY. 

 Vas Vererbungsproblem im Lichte der Entwicklungs- 

 mechatiik betrachtet. By Prof. E. Godlewski, 

 Jun. (Being Part ix. of Roux's "Vortrage und 

 Aufsatze iiber Entwickelungsmechanik.") Pp. 

 ,01. (Leipzig : \V. Engelmann, 1909.) Price 

 7 marks. 



IN this book the author attempts a critical review 

 of our knowledge of the mechanism by which 

 hereditary characters are transmitted, and makes 

 it his chief object to distinguish clearly be- 

 tween ascertained facts and the inferences based 

 upon them. Part i. contains an outline of the facts 

 of heredit}-, only so much being given as is necessary 

 to an understanding of their relation with develop- 

 inental mechanics. The possible inheritance of 

 ^"acquired" characters is discussed, inheritance in 

 non-sexual reproduction, including an account of 

 W'inckler's recent work on graft-hybrids, and, finally, 

 inheritance in sexual reproduction. This is classified 

 under the heads of blended, mosaic, and alternative. 

 Under mosaic heredity, cases like Toyama's g}nan- 

 dromorphic silkworm are included, which seems 

 scarcely justifiable. Under alternative inheritance 

 and Mendel's law the author seems not thoroughly 

 to grasp the independence of the facts of dominance 

 and segregation, and the same want of clearness in 

 this respect reappears in the general summary. Also 

 in discussing the relations of the different forms of 

 heredity we note a couple of slips in his account of 

 Galton's and Pearson's statements of the law of 

 ancestral heredity. 



In part ii. the author is more completely master 

 of his subject. Essentially the problem to be solved 

 is whether a substance which determines the appear- 

 ance of inherited characters exists and is transmitted 

 from generation to generation ; if so, where it is 

 NO. 2 1 14, VOL. 83] 



localised and how it acts, so as to produce the 

 different kinds of heredity found in different characters 

 in the same or different organisms. The theory 

 originally made familiar by Roux and Weismann, 

 that the nucleus, and especially the chromosomes, are 

 the "bearers of heredity," is first discussed, and the 

 wor^ of Driesch and his hypothesis of " Entelechy " 

 are explained, and, finally, the theories of Semon and 

 others are reviewed. It is pointed out that writers 

 are divided into two schools — those who believe in a 

 transmitted substance as the basis of inheritance, and 

 those who regard such a substance as a medium 

 for the action of inherent properties. The next four 

 sections deal with work which seeks to discover a 

 transmitted substance, and especially with work on 

 the nucleus. Recent work on amitosis, the structure 

 of the germ-cells, and the facts of fertilisation leads 

 the author to conclude that from these phenomena no 

 '■ nuclear monopoly '" in inheritance can be deduced. 

 A review of recent work on the chromosomes leads 

 to the same conclusion ; although their constancy in 

 number and form for each species is admitted, pick's 

 " Manovrierhypothese " is regarded as equally con- 

 sistent with the facts with Boveri's theor>- of indi- 

 viduality. Perhaps insufficient weight is attached to 

 the work, especially of .\merican cytolog^sts, on the 

 behaviour of the chromosomes in the maturation 

 divisions, and in the general summary the author 

 admits that he regards the appearance of their con- 

 jugation, and of the relative independence of paternal 

 and maternal elements, as illuson.-. Since the reality 

 and cause of Mendelian segregation is nowhere fully 

 discussed, this question might have been treated more 

 completely with advantage. 



Perhaps the most important section of the book deals 

 with experimental work on hybridisation and fertilisa- 

 tion of non-nucleated fragments (Boveri, Seeliger, 

 Delage, &c.), combination of artificial parthenogenesis 

 and hybridisation (Herbst), and the results of poly- 

 spermy (Boveri). These experiments are very 

 thoroughly described, and the conclusion is arrived 

 at that from them also no evidence for nuclear mono- 

 poly is obtainable. That the nucleus is of primary 

 importance is proved, but the relations between 

 nucleus and cytoplasm, rather than the nucleus itself, 

 are regarded as the basis of heredity. At the end of 

 this section a summary- of the evidence for the action 

 of the cytoplasm is given ; although relatively little 

 work has been done in this field, yet one feels that 

 the treatment is rather meagre compared with that 

 devoted to the nucleus. 



In the last sections are discussed the nature of the 

 determining substances (possibly enz\-mes, &c.), and 

 work on the influences of external factors on chang- 

 ing the " Vererbungsrichtung." The work of Guthrie 

 and Magnus on transplantation of ovaries is men- 

 tioned, but the author does not seem to know the 

 paper of Tower on the beetle Leptinotarsa. 



The book, on the whole, is eminently readable, and 

 succeeds well in its difficult task of summarising the 

 results of recent work and of disentangling ascer- 

 tained fact from deduction. The author regards 

 experimental facts as the only legitimate proofs of 

 the basis of heredity, and a tendency to be perhaps 



