12 



"weight and percentage are given of the different sizes of lettuce seed 

 which were passed through a series of millimeter sieves. In table 

 IV are given the results obtained from the use of a number of varie- 

 ties of seeds, and in this series of experiments 200 seeds were used 

 from each sifting with the exception of lettuce and radish, the latter 

 representing an average of two experiments and the former three. 



In these experiments different sizes of sieves with round perfora- 

 tions were used, some of which were cut to millimeters and some to 

 inches. In every instance the percentage of gain was based on the 

 lightest seed. Undoubtedly in some cases sieves of other sizes 

 might be employed to advantage. All sweet pea seed passing 

 through a sieve with y^^ inch perforations, and carrot seed through a 

 i|- millimeter sieve should be discarded. In the case of cucumbers 

 there is a difference of 35^^ in the two sizes of seed, but cucumber 

 seed might better be separated by a slot perforation. A i inch sieve 

 could be used for tomatoes and spinach, a 2 millimeter sieve for cab- 

 bage, I millimeter for lettuce, .j\ inch for radishes and either a -^''^^ or 

 Jy inch for millet. The percentage of germination in some cases is 

 the highest with the largest seed, and in others with those of medium 

 size, but with the exception of the carrot and tomato seed the great- 

 est weight of seedlings was obtained from the largest seed. In 

 the case of the tomato there was practically no difference in the 

 average weight of the seedlings. 



From a sample of onion seed separated by the air method we 

 obtained ;^^^/o of light seed. 200 seeds were selected from this sam- 

 ple and germinated with the following results : 



I'he heavy seed gave 92.5^ and the light seed 55'/c of germination. 



In our experiments in the separation of celery seed by means of 

 an air blower there was a marked difference in the number of seeds 

 which germinated and the size of the seedlings, the larger seeds giv- 

 ing the best results. A sieve with perforations about | millimeter 

 or g\; inch in diameter would be suitable for separating celery seed. 

 From our experiments we find that it would well pay one to separate 

 celery seed, planting the separated seed in sterilized soil or soil free 

 from the germs of diseases peculiar to celery. If these precautions 

 were taken we believe much better crops of celery, less subject to 

 infection, could be raised. 



