BULLETIN 235. 



STUDIES ON OAT BREEDING. II. SETECTION 

 WITHIN PURE TINES.^ 



By Frank M. Surface and Raymond Peari,. 



Previous to 1910 it was almost universally assumed that 

 small fluctuating variations were, to some extent at least, in- 

 herited. It was further believed that such variations were 

 cumulative in effect and that substantial progress in breeding 

 in a desired direction could be made by selecting, in successive 

 generations, those individuals showing the given character in 

 the most pronounced fashion. Since the appearance of de 

 Vries's Mutation Theory and the great impetus given to genetic 

 studies by the rediscovery of Mendel's Law this conception of 

 the process of inheritance has been materially changed. 



De Vries produced a large amount of experimental evidence 

 tending to show that there are two sharply defined classes of 

 variation. The one called fluctuating or continuous variation 

 {Modifikation of the Germans) is due entirely to differences 

 in the environmental influences. These variations, caused by 

 external conditions, he believed were not transmitted, in any 

 degree, from one generation to the next. The second class 

 called discontinuous variations, mutations, etc., have their origin 

 in variations in the germinal substance. These variations are 

 transmitted from generation to generation. The distinction, 

 then, is between variations which arise in the germ plasm and 

 hence are inherited and variations which arise in the soma and 

 are not inherited. This distinction was pointed out years before 

 by Weismann on purely theoretical grounds. 



Mendelian results in countless experiments with various plants 

 and animals have shown that individual characters are inherited 



^Papers from the Biological Laboratory of the Maine Agricuiltural 

 Experiment Station, No. 79. 



