GENETICS 



arbitrary standard agreed upon, the better they illus- 

 trate the stock-breeder's idea of a pure line. For ex- 

 ample, in "The Standard of Perfection," a book pub- 

 lished by the American Poultry Association, there are 

 recognized 42 breeds and 121 varieties of chickens. 

 To belong to any particular breed in this gallinaceous 

 Blue Book the chicken must look the part regardless of 

 its germinal derivation. 



To the biologist, on the contrary, the pure line is 

 like an imaginary mathematical concept depending en- 

 tirely upon similarity of the determining hereditary 

 complex. The biologist's pure line is genotypic. The 

 stock-breeder's is phenotypic, a difference of definition 

 which has given rise to considerable confusion. 



In a certain general way it will be seen that the pure 

 line stands over against mutation, since it is concerned 

 with the conservative maintenance of type while muta- 

 tion attempts to change it. 



The inevitable monotony of a pure line may be con- 

 siderably masked by individual somatic modification. 

 DeVries has said paradoxically, "The pure line is com- 

 pletely constant and extremely variable." That is, it 

 is "completely constant" except for mutations, and 

 it is "extremely variable" in the somatic development 

 that may be attained by separate individuals. 



3. JOHANNSEN'S NINETEEN BEANS 



To return to experiments with beans, Johannsen 

 found out that the progeny of every one of his pure 

 lines varied around its own mean, which was different 



