SEEDS AXD SEED-GBOWTH. 7 



germination. The writer's plan has been to plant at usual 

 depth and then mounding a ridge of earth two or three 

 inches deep over the line of each row. This defines the 

 line of row, and in spring, with a pronged hoe, the mound- 

 ing is raked off, leaving a mellow seed-bed. When small 

 seeds, such as those of the grape, raspberry, blackberry, 

 currant, gooseberry, strawberry, and Juneberry, are 

 planted in the fall, the raking off in spring should about 

 reach the seeds. In the relatively dry air of the prairie 

 and arid States the line oL row after raking should be 

 covered with boards or mulch until the roots of the delicate 

 seeds start, when the covering is removed. 



8. Seed-testing. — Seed-testing indoors is not wholly sat- 

 isfactory, as the conditions are more favorable than in the 

 soil of the open field or garden. The best test for the 

 amateur or commercial planter is probably in quite deep 

 earthen dishes placed in the greenhouse or a warm living- 

 room or kitchen. But the test should be continued until 

 the plantlet shows the true leaves and a system of roots 

 for taking up nutriment from the soil. The simple 

 sprouting of seeds is no satisfactory test of their vitality, 

 as many seeds will sprout that are not capable of forming 

 perfect plants. What is commercially called seed-testing 

 is simply sprouting the samples. The number of seeds 

 that sprout are counted. The experiment stations and 

 seed-dealers have several kinds of apparatus for this use, 

 in which the moisture and temperature are so controlled 

 that a seed with low vitality will sprout feebly that never 

 can develop true leaves or roots. As Bailey says: "The 

 sproating-test is almost wholly an attempt to arrive at a 

 numerical estimate of the sample rather than an effort to 

 determine the relative strength of germinating power." 

 With home-saved seeds kept as outlined under the head of 

 Seed-saving (4) tests of vitality are not needed. But with 



