1905.] 



Sampling of Seeds. 



223 



For eleven years in succession an experiment has also been 

 conducted in breeding oats by means of selection. Three grades 

 of oats were chosen, viz., (1) large, plump, well-developed seeds, 

 (2) light-weighing and light-coloured seeds, and (3) seeds from 

 which the hulls had been removed. The test was commenced 

 in the spring of 1893, by selecting seed from the general crop of 

 Joanette oats of the previous year. The selection made in each 

 of the following years has been from the product of the selected 

 seed of the previous year. The number of grains used on each 

 plot was carefully counted, and an equal number was used of 

 each description each year. In the crop produced in 1903, it 

 was found that the large plump seed produced y&g bushels ; 

 the light seed 577 bushels ; and the hulled seed 72*2 bushels 

 per acre. Only the best quality of seed is hulled, and it will be 

 seen that it gave nearly as good results as the carefully-selected 

 large seed. In weight per measured bushel, the crop produced 

 from the large plump seed weighed y\ lb. more than that from 

 the light^seed, and about one-third of a pound more than that 

 from the hulled seed. The difference throughout between the 

 large, plump, well-developed seeds and the light-weighing and 

 light-coloured seeds is very marked, and shows the great 

 importance of sowing the former and discarding the latter. 

 It is interesting to notice that the crop produced from the large 

 plump seed required only 1,208 grains to weigh an ounce, while 

 the crop produced from the light seed required 1,586 grains 

 to make the same weight. 



The United States Treasury Department, at the instance of 

 the Secretary of Agriculture, have directed that for one year, 

 from July 1st, 1905, 2-oz. samples of all 

 Sampling of importations of 100 lb. or more of grass, 



Seeds Imported c } over anc j forage-plant seeds be prepared 



into the • ' . . - j / * 



United States. at the earliest practicable date after entry, 



and forwarded to the Seed Laboratory 



Department of Agriculture, Washington, labelled with names 



and addresses of consignors and consignees, name of seed as 



given in the invoice, and quantity of the consignment. 



