SELECTION WITHOUT AETIFICIAL CROSSING. 



37 



Bulletin 132 of the same station gives the results of six years' selec- 

 tion of corn for high and low ears. See Table Y. 



Table V. — Difference between two strains of corn selected for ears high or low on stalk. 



Years. 



Difference 

 in height 

 of ears. 



Difference 

 in nuniber 

 of inter- 

 nodes in 

 stalks. 





Inches. 



Number. 



1903 



13.6 



1.5 



1904 



12.0 



1.5 



1905 



21.7 



1.8 



1906 



31.1 



4. 1 



1907 



39.2 



3.3 



1908 



34.2 



4.0 



Another instance of the effect produced on a species which is par- 

 tially cross-fertilized is seen in some interesting work of Professor 

 von Riimker at the Breslau Experiment Station in Germany. This 

 work was done on rye, in which species more or less cross-fertilization 

 occurs. By continued individual selection for color of seed Pro- 

 fessor von Riimker finally obtained several strains of markedly dif- 

 ferent color. Yellow color was more difficult to fix than green. 

 This is probably due to the compound nature of the yellow color 

 from the Mendelian standpoint. Some interesting cases of correla- 

 tion were found in this work. Green-colored seeds produced stronger 

 stalks; brown seeds were less mnter hardy. It was found that the 

 selection must be continued in order to maintain the characters for 

 which the selections were made. It is doubtful if these results could 

 have been obtained in a strictly self-fertilized species. 



The reason why selection produces these effects on cross-fertilized 

 plants is seen in the follomng: Suppose we start with a corn plant 

 that is heterozygote for yellow and white corn and for starchy and 

 sweet corn characters. The presence of yellow may be represented 

 by Y, the absence of yellow (that is, white) by y; the presence of 

 starch-forming character by S, and its absence (that is, sweet-corn 

 character) by s. Figure 2 shows the nine different types of corn 

 which would be produced by the individuals. If, now, we plant all 

 the seed produced by these nine types of corn and plant them where 

 they can freely cross-fertilize, but where they will not cross with 

 other kinds of corn, the next year we shall again get these same nine 

 types, but not in exactly the same proportion. If corn were com- 

 pletely cross-fertilized the proportion of these nine types would be 

 approximately the same the second year, and each year thereafter, 

 assuming, of course, that all the types are equally productive. This 

 is illustrated in figure 2. On the other hand, if corn were completely 

 self-fertilized, these nine types would behave as those sho^vn in fig- 



165 



