SINGLE COMBS OF FOWLS 199 



Another objection which was urged against the homo- 

 geneity of his material was " that we cannot by mere 

 somatic selection, however long it may be continued, 

 obtain a homogeneous race of smgle combs in respect 

 to any of their characters, except as the result of a happy 

 and exceedingly rare accident." Now Dr. Pearl admits the 

 justice of this criticism, for he says, as we have ahead}'- 

 pointed out, in his reply (p. 192) " that the somatic 

 condition of a character is, generally speaking, an un- 

 reliable criterion of its behaviour m inheritance." And 

 this race of hens, selected on the old somatic basis in 

 vogue twenty years ago, and which has been continued to 

 the present, can therefore have no claim of any sort, 

 on Dr. Pearl's own admission, to be regarded as homo- 

 geneous. 



Thus, on all the essential points which we criticised, 

 Dr. Pearl virtually admits the justice and validity of the 

 criticism. 



There is one further point in Dr. Pearl's reply which 

 calls for notice. He says (p. 192) " it seemed to him 

 that within the broad categorv of ' single comb ' there 

 might exist more than one gametic type." Now, we do not 

 question that Dr. Pearl conceived, even before the 

 appearance of our criticism, the possibility of there 

 existing more than one gametic type " within the category 

 of smgle comb." But we do know, as the note appended 

 to the article in which that criticism is expressed shows, 

 the Editor of " Biometrika " declined to allow us to 

 reproduce the figures of four particular combs which we 

 had selected from Dr. Pearl's illustrations. But he was 

 willing that we should reproduce any one of Dr. Pearl's 

 three plates complete. As a concession, and upon our 

 suggestion, he subsequently kindly allowed us to reproduce 

 both the four selected figures and one whole plate 

 of figures. Now there can be only one object m 

 desirmg this reproduction of a whole set of figures, 

 as they were arranged by Dr. Pearl, and which were quite 

 immaterial to our purpose. And that object is to demon- 

 strate what the Editor of " Biometrika " conceives to be a 

 case of contmuous or fluctuating variation. And, if these 

 figures conceivably show continuous variation, where, 

 then, we may ask, is the conception of gametic types ? 



