Inbreeding 371 



strain that may differ for the various strains, and after a few 

 generations will show only as much A'ariation as might be ex- 

 pected as the result of fluctuations. These strains are homozy- 

 gous and, since they are homozygous, they reach a fixed value, 

 or as we generally say, they breed true. Because they are homo- 

 zygous they show no genetic variation. 



A true-breeding homozygous strain is called a pure line, which 

 may be defined as the progeny produced solely by self-fertiliza- 

 tions from an original homozygous individual. Because each 

 pure line consists of only homozygous plants, the offspring from 

 large, intermediate, or small variants will all show the same 

 average size and the same amount of variation. Johannsen's 

 study of pure lines in beans showed that within the pure line 

 selection has no effect in producing any change in the character, 

 for the only variation found within a pure line except for occa- 

 sional mutations is environmental. When a mutation occurs in 

 a "pure line," the line does not become less pure, since no out- 

 cross has taken place. The individuals of such a pure line are 

 no longer homogeneous genotypically, however, as they now 

 consist of both the mutated and unmutated biotypes. It might 

 be mentioned here that in Johannsen's terminology "pure lines" 

 and "biotypes" are not synonymous terms. A biotype may be 

 heterozygous, such as the pink-flowered type of four-o'clock and 

 the blue Andalusian type of the domestic fowl. 



In a hypothetical problem in the last chapter, we assumed the 

 presence of three pairs of duplicate, cumulative, nondominant 

 genes (although exactly the same actual results could be attained 

 with dominant genes, some of which act in a plus direction and 

 others in a minus direction). Self-fertilization for several gen- 

 erations resulted in the segregation of homozygous lines which 

 were both larger and smaller than the Fi. The larger lines re- 

 sulted from the homozygous condition of some of the contribut- 

 ing genes; the smaller lines had many of the neutral alleles in 

 homozygous condition. If the Fi were therefore heterozygous for 

 a number of dominant genes, the subsequent generations would 

 differ from it phenotypically only to the extent that they were 

 homozygous for some of the recessive alleles. Homozygous 

 dominant lines would segregate out, to be sure, but because of 

 dominance, a line homozygous for all the dominants would be 

 phenotypically the same as one that was heterozygous. The two 



