58 The Third and Fourth Generation 



years before that the children of two very tall 

 individuals are not likely to be as tall as the 

 parents; the same tendency to return to 

 mediocrity is seen in the children of very short 

 parents. Johannsen found similarly that if 

 you plant large beans in a garden where cross- 

 pollination is certain, the offspring are prone 

 to be of less size on the average, but if you plant 

 a single large bean and let the blossoms on this 

 plant self-fertilize, the beans so produced are 

 very like the parent in size. That is, pure-line 

 stock is true to its type. Nilsson, the famous 

 director of the Swedish Experimental Station 

 at Svalov, found that the wheat produced 

 by planting the kernels of a single head 

 was remarkably uniform, displaying only the 

 parental qualities; but if seed from a number of 

 heads, all equally good to the eye, were sown 

 in the experimental plot, the resulting grain 

 was not at all uniform. Some might be good, 

 much indifferent, and some really bad. Nilsson 

 conceives that such grains as our oats and wheat 

 are resultants of the blending of many strains, 

 and that these separate out constantly, appear- 



