160 Referate. 



only to the advantages of giving to a single offspring the nutritional and 

 spatial relations that would otherwise be divided among several. Another 

 irrelevant feature which is discussed at some length, is the early isolation of 

 the germ-cells. In plants no such early isolation occurs although it is always 

 theoretically jiossible of course, to trace a direct cell-lineage from one gener- 

 ation of germ-cells to the next. 



The third lecture develops the Mendelian principles of heredity in the 

 briefest possible manner and then proceeds to the application of these principles 

 to the determination of sex as illustrated by sex-limited characters. The 

 author attempts to substitute the term "sex-linked" for the older and more 

 generally used term ''sex-limited", and applies the latter term to secondary 

 sexual characters, but to the reviewer this does not seem a desirable 

 change. As nearly all cases to which the term "sex-limited" have been applied 

 are of the kind which Morgan calls "sex-linked" it would be a more desir- 

 able reform to simplj- drop the usage of "sex-limited" in referring to second- 

 ary sexual characters. 



The fourth, fifth and sixth lectures deal mainly with secondary sexual 

 relations; these include secondary sexual characters, sexual selection, the 

 effects of castration and the transplantation of ovaries, gynandromorphism, 

 hermaphroditism and parthenogenesis. About two pages of these chapters 

 are devoted to hermaphroditism in plants. Here the terms "hermaiihroditic" 

 and "sporophytic" are used as if they were synonymous, and at another place 

 "sexual" and "hermaphroditic" are used as antonyms, but hermaphrodites are 

 sexual, too, and sporophytes may be either hermaphroditic or dioecious. 

 The classic case of Bryonia, in which crosses between a hermaphroditic and 

 a dioecious species led Correns to the first clear statement of the Mendelian 

 interpretation of sex-inheritance which is now generally accepted, was worthy 

 of more than a single short paragraph closed with the statement that it is 

 difficult to bring this case into harmony with other theories of sex deter- 

 mination. The author's difficulty here seems to be that he limits himself to 

 a single genotypic formulation for the sexes in those cases in which the 

 female is homozygous: he assumes that the female is FF, and the hetero- 

 zygous male Ff. The homozygous female could be FF or iiiiii and the hetero- 

 zygous male Ff. FM or Mm. With the recognition of these available for- 

 mulae the results in Bryonia may be easily harmonized with other cases of 

 sex-inheritance. The mixture of chromosome-terminology with genotypic 

 formulae tends to confusion. Morgan i) has shown plausibly that the sex- 

 determiner is not the X-chromosome but is only a part of it, or associated 

 with it. It is not proper, therefore, to indicate the sex-determiner by X, as 

 he has done in these lectures. It would be much better to substitute F for 

 this X. 



Morgan assumes that all Mendelian determiners lie in the chromosomes, 

 but that they need not remain in the same chromosome. As a working 

 hypothesis his adoption of .1 ohannsens" "chiasmatype" to explain the coupling 

 and repulsion of characters and the breaking of correlations between coupled 

 characters, is j)roving fruitful, in that it is leading to interesting discoveries 

 regarding the relative frequency of such "cross overs", or broken correlations, 

 in different sets of coupled characters. It seems that two sex-limited cha- 

 acters carried by the unpaired chromosome of the male are incapable of 



•) T. H. Morgan, An attempt to analyze the constitution of the chromosomes 

 on the hasis of sex-limited inheritance in Drosophil/i . Jour. Exper. Zool. 11, 365 bis 

 411, lull. 



