VARIATION 



439 



It is quite impossible to say offhand, in any given case, 

 how much is due to variation in breed and how much is 

 due to modifications produced by surroundings. But every 

 farmer knows that, in addi- 

 tion to controlling the con- 

 ditions under which his 

 plants and animals develop, 

 he must also be careful to 

 select the right kinds of 

 seed or stock. The best 

 of care will not make an 

 ordinary white bean develop 

 into a plant bearing lima 

 beans, nor will extra feed- 

 ing make a scrub cow give 

 the kind of milk that may 

 be obtained from a good 

 Jersey cow. 



465. Improvement by 

 selection. All domestic 

 animals and plants have 

 been carefully watched for Forty thousand sugar bee ts, tested individually, 



Centuries for the purpose showed from 12 per cent to 19 per cent of 

 ,, , , i sugar. Beets containing 15.5 per cent of sugar 



were the most frequent, but there were almost 



sirable individuals as the a man y beets with J 5 p r cent or with 



, . ,. 1 6 per cent. As the percentage of sugar de- 



parents ot the succeeding parts more from the t y pical I5-5 per centj the 



Crops Or generations. The number of individuals with a given sugar content 



diminishes, so that the extremely poor and the 

 best heads Ot Wheat Were extremely rich beets are also fewest in number 



selected for seeding the 



following year ; the best beans and the best potatoes have 

 been set aside as the progenitors of the crops to come. And 

 the same principle has been applied in the raising of ani- 

 mals. The best milk cows were selected to be the mothers 

 of the calves, the swiftest mares were the mothers of the colts, 

 and so on. 



12 12.5 13 13.5 14 14.5 15 15.5 16 16.5 17 17.5 18 18.5 19 



FIG. 234. Variation in physiological 

 properties 



