REVIEWS 521 



to evolution. As far as can be judged from the few remarks in his Introduction, 

 his view is that the only important problem is the problem of species, that 

 Darwin's solution of that problem has proved a failure, and that for the present 

 we do well to confine our attention to the immediate problems of genetic 

 physiology. Indications, however, are not wanting that in thus concentrating 

 his attention Prof. Bateson is running the risk of misunderstanding other 

 phenomena with which his researches bring him into contact. 



The essential phenomenon of Mendelism is the segregation of characters 

 in the offspring of hybrids or crosses, a segregation which must be due to the 

 segregation in the gametes of the elements that determine the characters. 

 These elements are therefore regarded as units which may be combined but 

 cannot be divided. When a pair of characters are united one may be dominant, 

 but dominance is not necessary or universal. Modern investigations have gone 

 far beyond the simple cases studied by Mendel, and the mere assumption of 

 units in the gametes corresponding to the visible characters has been found 

 insufficient to explain the results. It has been found necessary to invent purely 

 hypothetical units of whose real nature nothing is known, and which are mere 

 names for elemental factors whose existence is inferred from their visible com- 

 binations. Although dominance is not universal, the evidence leads to the general 

 conclusion that allelomorphic characters differ only in the presence or absence 

 of some single factor. For example, the rose comb in fowls is allelomorphic 

 and dominant to the single comb ; the interpretation being that the rose comb 

 is due to the presence of a factor which is absent from the single. In segregation 

 the present factor segregates from the absent. By the aid of this method the 

 remarkable results of crossing the different types of combs are elucidated. When 

 pure rose and pure pea are crossed, the type known as walnut is produced ; when 

 these are bred together four types are produced, namely, walnut, rose, pea, and 

 single. The extraordinary thing here is the appearance of single when no single 

 occurred in the parentage. The explanation adopted is that the pure rose 

 contains the recessive p, that is, the absence of pea, and the pure pea the 

 absence of rose. When the two absences meet in fertilisation the result is single. 

 Although the theoretical results given by this scheme agree with the observed 

 results, it must be evident to any one but a Mendelian that the explanation is 

 a mere juggling with words and symbols that have no real meaning. In the 

 terms rose and absence of rose, obviously absence of rose means single comb ; 

 in fact, the author explicitly says so : " The rose comb is a single comb modified 

 by the presence of R." Similarly, absence of pea means the same thing, namely, 

 single comb. Yet for the purpose of the formula, absence of rose and absence 

 of pea are treated as two different factors. The rose used in the experiment 

 breeds true and never yields single, but for the purpose of the formula it is 

 supposed to contain the recessive p or absence of pea, which separates in the 

 breeding of the hybrid walnut. Either p means nothing at all or it means 

 the factor for single comb, and therefore the rose itself ought to segregate and 

 give single, which it does not. According to Prof. Bateson, the absence of rose 

 is one thing, the absence of pea is another, but absences by themselves are 

 nothing. In mathematirs it is legitimate to deal with negative quantities and 

 to add together —a and - b, but Prof. Bateson deals with zeros and makes 

 a -a one quantity and b-b another; it might be maintained that absence of 

 rose could mean pea comb as well as single, but then the pea comb would be 

 merely PP instead of Pr, and this would not improve matters. 



A great part of the book deals with the heredity of colour characters, and 



