194 THE MENDEL JOURNAL 



scientific impartiality had some reference to the later 

 Mendelian work (Rep. Evol. Commt. iv. 1908, page 21) 

 been made either by the authors or by the Editor of 

 " Biometrika." This later Mendelian evidence suggests 

 that these " intermediate " pea combs are not due to im- 

 perfect dominance or to continuous variation, but to a 

 specific difference in the genetic nature of the single- 

 comb base upon which the pea constitutent of it rests. 



To ignore deductions derived from experimental 

 evidence and to persist in maintaining conclusions based 

 on observations that are eight years old, and which have 

 been superseded by more recent and extended work, 

 cannot we think help the advance of knowledge. The 

 Biometricians still persist in believing that all single- 

 combs are genetically alike, in spite of their own standard 

 deviations and coefficients of variation, whereas the 

 Mendelian has already good evidence to show that 

 there exist several distinct gametic types. And the 

 existence of these explains the intermediate pea-single 

 type of comb. 



NOTE. — After tliis article was in press, we sought the permission of 

 the PubUshers and the Editor of " Biometrika," to the reproduction in our 

 pages, of four selected figures from Dr. Pearl's three plates. The Editor, 

 while unable to see his way to grant permission for the reproduction of 

 the four figures selected by us, was willing to allow us to reproduce 

 any one of the three plates as a whole. At our further request the Editor 

 of " Biometrika " very kindly permitted us to reproduce one of Dr. Pearl's 

 plates together with the four figures. Hence the appearance of one 

 whole plate, to which there is no reference in the text of our article. 

 Doubtless, it is the desire of the Editor of " Biometrika." in fairness to Dr. 

 Pearl's conclusions, that one series, at least, of his figures should be seen by 

 our readers, side by side, with the four selected by us. We very gladly 

 pubhsh this plate. It is intended to show the existence of continuous 

 variation. To us there appear to be perceptible gaps. But even though 

 the blends or grades formed a perfect series, the fact still remains, that 

 the only reliable way to test whether gametic blending exists or not, is to 

 breed the extremes of a n-puted lilended character, and note the results. 



