LUTHER BURBANK 



related species of plants, but that equally fertile 

 hybrids may be produced by the union of a good 

 many species that are so widely separated as to be 

 classified in different genera. 



They showed that the first-generation hybrids 

 may resemble one parent or the other pretty 

 closely or may show a blending of qualities; and 

 that in the second generation, with rare excep- 

 tions, there is a segregation and recombination of 

 the racial qualities of the original parent species, 

 in which the extreme forms will more or less 

 closely duplicate one parent or the other, and the 

 intermediate forms may show almost every con- 

 ceivable gradation between the two. 



They showed, further, that it is possible by 

 selecting among the second-generation hybrids the 

 individuals that show any desired combination of 

 qualities, to develop, in the course of a few gen- 

 erations of inbreeding, races in which this com- 

 bination of qualities is so accentuated and fixed as 

 to constitute a distinguishing characteristic of a 

 new variety quite unlike the original forms. 



Moreover, the later-generation hybrids might 

 reveal racial traits that were not observable in 

 either of the parent species. 



The segregation and redistribution of charac- 

 ters often gave opportunity for the appearance of 

 qualities that have long been submerged. 



[134] 



