Distinction Between Species and Varieties. 585 



The truth of this conclusion has become more obvious 

 owing to the attention which has been paid to the subject 

 since the re-discovery of AIendel's work. The paral- 

 lehsni between the two groups of hybridization and tlie 

 two types of systematic subdivisions has been most ex- 

 haustively dealt with by Tschermak, who attempted to 

 base upon it a principle of distinction between the spe- 

 cific hybrids and varietal mongrels.^ 



We w'ill, then, regard the principle in its new f(jrni 

 as demonstrated, and examine the question why the cri- 

 teria which it supplies are not sufficient for universal 

 application. In doing so I shall, for various reasons, 

 leave the mutation crosses, which Tschermak also re- 

 gards as specific hybrids, out of consideration ; and shall 

 denote the Mendelian hybrids as bi-sexual in accordance 

 with AIacfarlane's terminology, employing also the 

 term uni-sexual in the sense in which it is used by that 

 author. Expressed very briefly, therefore, bi-sexual 

 crosses produce varietal hybrids, unisexual ones spe- 

 cific hybrids. 



But some limitation is necessary ; and herein lies the 

 difficulty of the question, which is felt by every one who 

 endeavors to apply the conclusions drawn from the study 

 of hybrids to taxinomic problems. This limitation is, that 

 the criterion really applies only to monohybrids ; for di- 

 polyhybrids, however, only in so far as they can be com- 

 pared with these. 



We have given the name of monohybrids to those 

 mongrels whose parents differ from one another in a 

 single elementary character only. Obviously they occur 

 both in uni-sexual and in bi-sexual crosses. But, of 



^E. Tschermak. in the third appendix to his edition of ^Ies- 

 Del's VersHche iibcr PflanzcnJiybridcti. p. 58. 



