in a different set of characters which the parent possessed. This 

 principle will also explain some of the well'known cases of reversion 

 on crossing cited by Darwin and others. 



The Mendelian type of inheritance described by Professor Castle 

 appears also to be due to a process of analysis, and in maiiy cases, 

 as in mice and rabbits, coats of various colors seem to have arisen 

 by the loss of different factors which were present in the ancestral 

 form. This analytical process, however, appears to be simpler and 

 more easily understood than in the ease of mutations. 



It is probable that certain cases in man, such as a predisposition 

 to particular disea?es, can best be explained as due to the loss of 

 certain factors in the inheritance. 



Having thus very briefly and hastily presented before you a few 

 of the facts of mutation in the Evening Primroses, and a glimpse of 

 the general viewpoint growing out of these results, I should like for 

 a moment to direct j'our attention to an entirely different line of 

 work, which, however, led ir.o toward the same conclusion I have 

 already stntcd to yon, and in fact furnishes a further basis for it. 

 This is a str.dy of tbe structure of the germ cells of these forms, 

 to find out if there was any micro.scopic or cytological basis in their 

 nuclei, for the sudden origin of tlicse new types. I have been en- 

 gaged in this work fur several years and a discussion of the inter- 

 esting results niigjit well hii\c occupied all my time this evening. 

 Their technical cliaracter, however, renders them difficult to present 

 in a short time, and I will merely say that these studies have shown 

 that there is a possible basis for the sudden origin of the new types in 

 the behavior of the chromosomes in the germ cells at the time the latter 

 are formed. Owing to this peculiar behavior^ occasional irregulari- 

 ties occur in the distribution of the chromosomes at the time the 

 germ cells are produced. Ordinarily the germ cells contain a single 

 set of chromosomes, but cases will occur, owing to these irregular dis- 

 tributions, in which a germ cell will contain two chromosomes of 

 one pair and lack both representatives of another pair. The number 

 of chromosomes therefore will remain constant, but certain germ 

 cells will nevertheless be entirely deficient in a particular kind of 

 chromosome. Modern cytology has produced many lines of evi- 



1. Anyone wishing to read an account of this matter may refer 

 to the Botanical Gazette 46:1-34. 



