THE PRACTICAL VALUE OF GENETICS 355 



suggested by the size of the parent seed, and in the second, 

 parent seeds of the same weight in some cases produced beans 

 with very different ranges of variation. In spite of the fact that 

 the results were reasonably harmonious with Galton's law of 

 regression Johannsen concluded that he was dealing with mixed 

 stock and conducted further experiments to explain his discoveries. 



In these experiments plants produced from beans of known 

 weight were self-fertilized and their offspring for several genera- 

 tions were treated in the same way until it was certain that the 

 lines were homozygous. The result indicated that the original 

 stock was made up of nineteen different kinds of beans, each kind 

 varying between certain extremes of weight which might overlap 

 with others, but every one of the nineteen fluctuating about a 

 different mean. These nineteen groups were called pure lines. 

 Johannsen defined a pure line as the progeny of a self-fertilized 

 homozygous individual. An aggregation of pure lines such as 

 that with which he first dealt was called a population. Figure 196 

 illustrates graphically the difference between five of Johannsen's 

 pure lines and the population formed by mixing them. 



It is evident that intensive line selection in any plant may 

 bring about the isolation of pure lines (Fig. 197). The difficulty 

 of maintaining such lines under ordinary conditions is obvious, 

 however, and it is doubtful that true pure lines often occur in 

 nature. They must certainly be restricted to those plants in 

 which elaborate adaptations for self-fertilization are present. 



Equivalents of the Pure Line. The self-fertilization of a homo- 

 zygous individual is the same in result as the union of gametes 

 with the same complex of determiners from different individuals, 

 consequently homozygous crosses produce equally homogeneous 

 groups of offspring. It is also similar in effect to reproduction 

 without fertilization, since here there is no possibility of different 

 characters being brought in. Reproduction of the latter type 

 includes parthenogenesis and agamic reproduction, such as fission 

 and budding. The individuals descended from one ancestor 

 through a series of asexual generations collectively constitute a 

 clone. 



Under natural conditions none of these pure-line equivalents 

 are more likely to be maintained than the typical pure line, but 

 they may exist for a considerable time in many species. At the 

 end of a summer, for example, the offspring of each stem-mother 



