PLANT HYBRIDIZATION BEFORE MENDEL 331 



A cross was made between Trifolium pratense album and T. 

 pratense quinquefolium^ with the following results : 



Calc. 



19 



19 



56 



6 



These were for about 220 plants. 



De Vries closes with the important statement (p. 89) : 



"It is frequently possible, through the segregation experiments, to sepa- 

 rate simple characters into several factors. Thus, the color of flowers is 

 frequently composite and, after crossing, one obtains the individual fac- 

 tors, in part separated, in part in different mixtures." 



Experiments with Antirrhinum majus, Silene armeria, and 

 Prunella vulgaris^ are stated as confirming the above. 

 De Vries' final statement is {ib., p. 90) : 



"From these, and numerous further experiments, I conclude that the 

 law of segregation (Spaltungsgesetz), found by Mendel for peas, finds a 

 very general application, and that, for the study of the units out of 

 which the species-character is composed, it has a quite fundamental sig- 

 nificance." {lb., p. 90.) 



One of the most striking features of the above paper is the 

 anticipation by De Vries of the multiple factor explanation for 

 certain characters such as flower color, first reported upon ex- 

 perimentally by Bateson for the crossing of two white sweet peas 

 of the Emily Henderson variety, which gave purple in the F^. 



The next paper in the series to be considered is the third paper 

 of De Vries on "Erbungleiche Kreuzungen (Vorlaufige Mitthei- 

 lung)," constituting No. 53, in the Ber. d. d. Bot. Gesell., Novem- 

 ber 21, 1900. (ic.) 



De Vries in this article, briefly begins by summarizing the 

 Mendelian discovery. 



"In a paper published in these 'Berichte' concerning the Law of Segre- 

 gation of hybrids, I have shown that this law which Mendel had de- 

 rived from his investigations with peas, finds a very general application 

 in the plant kingdom, and is of capital importance for the theory of 

 hybridization. The since published important extensive investigations of 

 Correns, Tschermak, Webber and others, have established in part the 

 correctness of Mendel's results [Erfahrungen], and in part the justifica- 

 tion of this generalization. 



"Mendel had demonstrated for his peas-crosses, that their results could 

 be derived in simple manner from certain principles. In the first place 



