STUDIES ON OAT BREEDING. 143 



ber of undesirable, low yielding plants. Such a variety is said 

 to be "run out." This condition undoubtedly results from the 

 chance preservation and multiplication of germinal variations 

 in the direction of low yield. With care to avoid mixing and 

 the occasional selection of seed in the field, it is possible to 

 avoid the undue multiplication of these undesirable types and 

 hence to avoid the so-called "running out" of varieties. 



That germinal variations similar to the above do occur in 

 the case of morphological characters has been clearly shown by 

 Nilsson-Ehle 30 . He was able to show that in pure breeding 

 strains and under conditions in which there was no chance of 

 mixing, grains appear which differed either in their color or 

 in the character of their grain. These variations either bred 

 true at once or after one or two generations the great majority 

 of their progeny bred true to their new characters. The exist- 

 ence of these well authenticated cases of germinal variation in 

 morphological characters makes it almost certain that similarly 

 inherited variation may occur in respect to physiological char- 

 acters such as yield. 



The frequency with which such variations occur is unknown 

 but experiments indicate that they are relatively very rare. 

 Likewise the underlying cause of these germinal variations is 

 unknown. The evidence indicates that they are probably not 

 due to hybridization. They probably belong to the class of mu- 

 tations as defined by de Vries. 



In the case of self-fertilized plants such as oats, wheat, beans, 

 etc., there is a large amount of evidence to indicate that when 

 once acquired these new characters will breed true. It is these 

 germinal variations in respect to yield and to strength of straw 

 that we have attempted to isolate. The evidence from the 

 three year tests indicates that we have been successful with 

 these twelve pure lines, or with the majority of them at least. 

 At the same time it is clear that these variations in yield have 

 not been accompanied by any marked morphological changes ill 

 the characters studied. 



30 Nilsson-Ehle, H. Uber Falle spontanen Wegfallens eines Hemmungs- 

 faktors beim Hafer. Zeit. f. indukt. Abst. — u. Vererbungslehre, Bd. 



5- PP- 1-37, 19". 



. Spontanes Wegfallen eines Farbenfaktors 



beim Hafer. Verh. d. naturf. Vereins in Briinn, Bd. 49, pp. 139-156. 



