SELECTION AND SAMPLING OF SEED. 155 



appeared to be in favor of large seed, but toward the 

 end of the season, certain changes occurred and no 

 difference could be noticed. The area of experiment 

 was small. 



These results appear to be in direct contradiction 

 to his early experiments. Walkhoff is decidedly in 

 favor of large seed, as he declares that the resulting 

 beets are more hardy. 



It seems to us that the strong argument in favor 

 of large seed is, that the young plants, the outcome 

 from them, can better resist the variations of the 

 weather than the small. This is explained in various 

 ways: Hollrung argues that small seeds mature 

 early, owing to their greater facility to germinate. The 

 average yield per acre is evidently greater with large 

 than with small seed, owing, as we have just said, to 

 there being a larger number of sprouts or germs per 

 individual seed used. The pericarp is necessarily very 

 much greater for large than for small seed. From 

 this fact, Knauer concludes that the actual weight of 

 seed proper is very much greater in small than in large 

 seed. This outer covering for large seed represents 

 75 per cent, of its total weight, while for the small seed 

 only 72 per cent. If the large and small seed be put to 

 a germinating test, the argument appears to be in favor 

 of small seed. With five grams of large seeds may be 

 obtained 283 sprouts, while with the same weight of 

 small seeds 469 sprouts are obtained. 



Briem has also given this subject more than usual 

 attention and his conclusions are worth recording. 

 He admits that seed may be divided into three classes, 

 large, medium and small. These all gave very great 



