The progress of Genetics since the rediscovery of Mendel's papers. 411 



in question remains unpaired in the hybrid. The latter description 

 is applied to characters which breed true in the cross-bred. This 

 sugg'estion is the out-come of an attempt to incorporate the facts of 

 Mendelian inheritance with the conclusions previously drawn from 

 mutations. But as knowledg'e of Mendelian cases has progressed, and 

 as it has been found possible to describe so large a proportion of 

 those cases in terms of the presence and absence of a definite factor, 

 the doubt may be expressed whether it is not rather in the Mendelian 

 phenomena that the determining- factors should be regarded as un- 

 paired. 



De Vries further distinguishes the characters which segregate 

 as "retrogressive" and "degressive" in contradistinction to "pro- 

 gressive" characters which are genuine species- characters and are 

 declared not to segregate. It seems to me that such a distinction 

 is mainly one of definition, and I am not convinced that on a wider 

 survey of the facts it could be maintained as a valuable or natural 

 grouping of the phenomena. 



In the preparation of this paper I have had the benefit of several 

 criticisms and suggestions of value from Mr. E. C. Pun nett. I take 

 also the opportunity of acknowledging the large share which during 

 the past three years he has contributed towards the elucidation of 

 the problems which have been the subject of our joint investigations. 

 Not only has he taken an active part in the practical business of 

 experiment; but to his ingenuity and determination is very largely 

 due such success as we have attained in unravelling some of the 

 more complicated phenomena, especially those of the heredity of colour 

 in the Sweet Pea. 



August 1906. 



