146 THE BREEDING OF ANIMALS 



result is the same as in the second generation. The off- 

 spring will be 25 per cent pure polled, 25 per cent pure 

 horned, and 50 per cent hybrid polled. (See Fig. 17.) 



This principle has been repeatedly used in the produc- 

 tion of polled breeds of cattle. The only practical diffi- 

 culty is encountered in determining which of the second 

 generation are pure polled and which are hybrid polled. 

 This difficulty can be overcome only by repeated matings 

 and by observation of the offspring. 



139. Theory of pure lines. The practical breeder 

 of animals has depended almost entirely upon methodical 

 selection for the improvement of domestic animals. It 

 was Darwin's opinion that the improvement accom- 

 plished in the desirable qualities of animals and plants 

 was due to the persistent selection of desirable continuous 

 variations. In bringing about improvement, therefore, 

 the breeder only required keen powers of observation 

 to detect any variations in a standard sort which were 

 better than the qualities of the ancestors. By selecting 

 these varieties and continuing this process for many gen- 

 erations, highly improved sorts were ultimately developed 

 which came true when bred together. 



In the practical application of this theory, it has been 

 frequently discovered that the limits of improvement 

 through selection were quickly reached. Apparently 

 the degree of improvement could not be carried beyond 

 a certain definite point. In a more careful analysis of 

 the fundamental basis of improvement by selection, 

 Johanssen l has demonstrated that very many of the 

 domestic plants are not possessed of single characters 

 only, but are a mixture. The selection exercised by 

 man in such cases is essentially a process of selecting out 



1 Johanssen, 1909, "Elemente der exakten Erblichkeitslehre." 



