20 
Colorado Experiment Station 
After marking off the Canton flannel in 2-inch squares with ink or in¬ 
delible pencil it should be well moistened and placed over the wet 
sawdust. The squares may be numbered to correspond with the num¬ 
bers on the ears being tested. Kernels from different parts of the 
ear, usually six to ten in number, are placed in each square, and the 
whole is then covered with a second layer of Canton flannel. The 
final count in this case will come after six days. 
In making a germination test there are, in brief, several items of 
importance: (1) Count out the seeds accurately; (2) keep blotters 
Fig. 7.—The soil flat, especially recommended for testing germination of corn, 
beans, peas, and other large seeds. 
and cloths moist, but not saturated; (3) keep at a uniform tempera¬ 
ture, and (4) count the sprouts at periods stated in the table. 
The ease with which germination tests can be made, from the 
standpoint of apparatus and labor, means that home germination 
tests will rapidly become popular among farmers. To them it repre¬ 
sents a guarantee not only of their own seed but all seed that is being 
offered on the market, and the consequent lessening of risk of great 
losses. 
III. METHODS OF SEED TREATMENT FOR THE PREVENTION 
OF DISEASES IN CERTAIN FARM CROPS. 
One of the chief reasons why so little attention is given to the 
sowing of healthy seed is that the damage resulting from various plant 
diseases is invariably underestimated by the crop grower. In a goodly 
number of instances the purity and germinative power of seed have 
been carefully determined, but the failure to go one step farther and 
make sure that it is wholly free from disease has brought quite un- 
