A Nl'GGESTION REGARDING- HEAVY AND LIGHT 

 SEED GRAIN 1 



L. R. WALDRON 

 Dickixsox Sl'i:-kxi'ei;imkxt Station, Dickinson, N. Dak. 



A consideeable amount of work has been done by in- 

 vestigators of cereals, regarding the comparative value 

 of heavy and light grains used as seed. The major por- 

 tion of the experiments have been conducted with wheat, 

 oats and barley. The problem appeared simple at the 

 beginning, but has developed many complications. 



In many cases bulk grain has been graded into various 

 classes without determining the relative number of grains 

 per measured quantity. Thus the experiments have been 

 vitiated by the failure to consider the different rates of 

 seeding, as regards the number of grains per area, which 

 would naturally ensue. Even if allowance for seeding 

 wore made, and the number of grains per area deter- 

 mined as accurately as possible, it would be an excel- 

 lent thing to conduct also a rate of sowing test. Such a 

 test might throw light upon results induced by climatic 

 conditions. 



Some of the workers have made no distinction between 

 shriveled grains and small plump grains. To eliminate 

 the factor of shriveled grains would be virtually doing 

 away with fanning mill methods of grading, which would 

 seem to be necessary if we are to simplify the problem 

 and to obviate the conflicting factors. Zavitz, of Ontario, 

 has worked with small, hand-picked samples of grain and 

 has evidently succeeded in overcoming the difficulties 

 mentioned. Mis results, extending over a series of years, 

 are remarkably consistent and worthy of careful study. 



Despite some of the errors one can not fail to be im- 

 pressed, as the literature is studied with the preponder- 



1 Contribution No. Ill, Laboratory Experimental Plant-breeding, Cornell 

 University. 



48 



