THE PRACTICAL VALUE OF GENETICS 



357 



phila, Paramecium and many plants have been the basis for the 

 conclusion b}'- many biologists that selection within the pure hne 

 is without effect. 



The Pure Line as a Limit of Selection. The fact that selection 

 has proved a valuable method of improving organisms according 

 to most experimental evidence, rests upon the possibility of isolat- 



Average of_ 

 all progeny 

 Weight of parent seed] 



Centigrams 



70 



jEU 



45. U I 



45.0 I 



im 



'''■''' ' ' ' ' L 



I I t 



10 2U 30 40 5U 60 70 10 20 30 4U 60 60 70 10 20 30 40 60 6070 10 20 30 40 60 60 70 10 20 30 40 60 60 70 



Pure line number >II VII XV XVIII 



Fig. 198. — The result of selection in four pure lines of beans. The vertical 

 columns, representing the average progeny from different sizes of parents 

 all derived from the same pure lines, contain groups nearer alike than the 

 horizontal columns, repre.senting progeny from parents of the same size 

 but of different pure lines. All of the numbers indicate weight in centi- 

 grams. (From Walter, after data from Johannsen.) 



ing those pure lines which best express the desired character. It 

 has been found that environmental conditions may bring about 

 differences in the development of individuals within the pure line, 

 hence this factor may also enter into improvement by selection, 

 and finally supplementary MendeUan factors may act upon char- 

 acters in a pure line to produce an entirely different result. The 

 pure line is therefore only a partial limitation of selection and 

 even without the inconstant modifications mentioned there is a 

 lack of absolute evidence that it is permanently fixed in nature. 

 Mutations occur to change pure lines, and there is a growing 

 feeling that the environment may be important in the modification 

 of inherent qualities. 



