GENETIC CHANGES 217 



farm practice elimination from the seed planted of all but 

 the best pure hues, may greatly increase the total yield. 

 This in many cases has actually been done. The work of 

 determining what are the best pure lines and of increasing 

 these to the exclusion of all others is the main work of the 

 plant breeder. It is a work which will have to be done over 

 again generation after generation, because impurities will 

 creep in from accidental crossing with inferior ^'a^ieties or 

 from the occasional origin within the pure line of new genetic 

 changes for the worse, for quite as many of this sort occur, 

 probably, as of those which are for the better. Besides dis- 

 covering and isolating, as pure lines, homozygous strains of 

 favorable genetic variations, as they occur in commercial 

 field crops, the plant breeder has a second important function 

 to perform in connection with plants normally self -fertilized. 

 He may, by artificially crossing varieties, combine the excel- 

 lent qualities which they severally possess. It often happens 

 that favorable variations which have arisen in the crops of 

 one country are unknown in those of another country. The 

 plant breeder may bring together the best varieties of all 

 countries, determine the good qualities of each and then by 

 suitable crosses combine these in new varieties adapted to 

 special conditions or particular regions. Never in the history 

 of the world has this been done on so extensive a scale or 

 with greater success than in the United States at the present 

 time. 



To return to the point of our departure, how common in 

 occurrence are genetic changes in self -fertilizing plants.^ An 

 answer to this question can be made only in relative terms. 

 It is scarcely safe to assume that they never occur. The very 

 existence of numerous genetically different pure lines in 

 every self -fertilized crop shows that genetic changes have 

 occurred in the past, and if so they are doubtless occurring 

 today. Some refer all such multiplicity of varieties to past 

 hybridizations of species genetically different, but this is 

 only referring to a more remote period the genetic changes 

 which are uivolved in the origin of the hypothetical species 



