26 APPLICATION OF PEINCIPLES OF HEREDITY TO BEEEDING. 



such is the case. This is especially true of varieties of wheat. A 

 single variety seldom retains its supremacy in any given locality for 

 half a century. Yet it is far from demonstrated that careful selec- 

 tion of wheat varieties would not maintain vigor almost indefinitely. 

 This whole question of the running out of varieties needs much 

 further study before the last word can be said on the subject. 



Having outlined the main principles with which we have to deal 

 in plant breeding, we may now proceed to a consideration of the 

 different methods of breeding and selection and the application of 

 the principles involved. 



SELECTION WITHOUT ARTIFICIAL CROSSING. 



We have already seen that close-fertilized and cross-fertilized 

 species behave difi^erently under selection. In addition to self- 

 fertilized and cross-fertilized species we must also consider the effect 

 of selection on those varieties which are propagated vegetatively^* 

 that is, from cuttings, grafts, tubers, etc., including all methods of 

 propagation other than from seed. 



VEGETATIVE PEOPAGATION. 



In plants propagated vegetatively we have several kinds of varia- 

 tion to consider, for the effect of selection on each of these is different. 

 First, we have those fluctuating variations which are due wholly to 

 environment, such as difference of food supply, moisture condi- 

 tions, etc., which modify the individuals of a generation but which 

 are not hereditary. As we have already seen, such investigations 

 as have been made on this subject indicate that in nearly all cases 

 of fluctuating variation due to environment selection is entirely 

 without permanent effect in changing the plant from year to year. 

 In Bulletin No. 127 of the Illinois Agricultural Experiment Station 

 Dr. E. M. East made a careful survey of all the literature he could 

 find relating to the effect of selection on these fluctuating variations 

 in potatoes. He concluded that it is not proved that selection can 

 change these variations, though the question is left in some doubt. 



Prof. H. S. Jennings, of Johns Hopkins University, has during 

 the past few years made some investigations on the unicellular animal 

 Paramecium which must rank among the most important biological 

 work that has been done, at least in the field of experimental evo- 

 lution. While Paramecium is an animal, it propagates for hun- 

 dreds of generations by simple division, and hence there is every 

 reason to suppose that the principles applicable to Paramecium are 

 applicable to plants which are propagated vegetatively. Jennings 

 gives an excellent summary of his work in the American Naturalist 

 for June, 1909, where reference will be found to the original technical 



165 



