REVIEWS 173 



The chapter on the limits of training in animals also calls for attention, 

 although no definite conclusions are reached in it. Clever Hans and the horses 

 of Elberfeld have attracted a good deal of public notice, and the investigations of 

 Prof. Stumpf on the former and Claparede on the latter are ably discussed. It 

 seems highly probable in both cases that a great number of the correct answers 

 to arithmetical problems, etc., are due to the trainer giving some sign to the 

 horses. The signs may not be given consciously or intentionally, but it is 

 certainly not to be regarded as accidental that the percentage of correct answers 

 drops from between eighty-five and ninety to between eight and twelve when the 

 questioner himself does not know the result. Making due allowance for this, 

 however, we can still see that the horse is capable of a fair amount of " intelligent " 

 work when dealt with sympathetically. Such is the case, through more marked, 

 in monkeys, and a very high state of training was reached in the case of the 

 well-known chimpanzee Peter. 



With the exception of those chapters that are intended for reference, and so 



indicated in the introduction, the book is very interesting. It is a thoughtful and 



stimulating work, and in it the biologist will find many matters dealt with from a 



point of view that is generally entirely overlooked in ordinary works of biology or 



natural history. 



C. H. O'D. 



The Determination of Sex. By L. Doncaster, Sc.D. [Pp. xii + 172, with 

 22 plates.] (Cambridge : at the University Press, 1914. Price 7s. 6d. net.) 



The fascination of the problems suggested by the title of this work is, to the 

 biologist, perhaps only equalled by that of the origin of living matter itself. Both 

 are constantly before him, and the present one meets him in some form or other 

 at practically every turn. 



One thing emerges quite clearly from reading this suggestive and useful book, 

 and that is, that at the present time there is no theory that will fit all the facts 

 now available. Thus we find in the Crustacea an alteration in the physiological 

 condition of the body appears to be able to bring about an alteration in the 

 primary sexual organs, while in the Vertebrata the reverse seems to be the case, 

 and it is probable that an alteration of the primary organ of sex by means of experi- 

 ments will bring about an alteration in the general condition of the body. The 

 removal of ovaries and substitution of testes in a mammal produce a marked 

 effect on the structure and behaviour of the individual, whereas the same operation 

 in insects makes no appreciable alteration in either. Furthermore, while in 

 certain cases the evidence undoubtedly points to the presence of two sorts of ova, 

 and so throws the main part of sex determination upon the mother, in other 

 cases the evidence just as strongly indicates two kinds of spermatozoa. All these 

 and many more points are lucidly dealt with in the present book. 



It is cautiously written ; indeed, perhaps too much attention is paid to the 

 exceptions, important though they be, for after all the main point is to find 

 general statements that will apply to the majority of cases. A generalisation is 

 not necessarily upset because of a few exceptions, and the reasons for their 

 apparent breach of uniformity may be discovered when they are investigated more 

 fully. 



A few statements call for note. On p. 105 we find : " If the hormone hypothesis 

 is correct, it should be possible to cause the appearance of the secondary sexual 

 characters in castrated animals by injecting extract of the testis or ovary." This 

 is not necessarily so, for it is conceivable that these organs contain only the mother 



