WE WILL GO TO WORK. 183 



the seed, and the former were large but useless. 

 If, however, the seed-stalk had been cut away 

 and the bulbs used in their green state, they 

 would have been most excellent. I was thus 

 satisfied that a very early and profitable crop of 

 onions could be raised from seed sown in Au- 

 gust. . 



My next experiment was to sow the seed late 

 in September, so that the plants would not be 

 large enough in spring to run up to seed, but 

 develop large bulbs, as in the case with seed 

 sown in April. If the plants would winter over, 

 no matter how small they were, the crop would 

 be far earlier than any that could be started with 

 the opening season, unless it be from what are 

 termed " sets " or little onions put out as soon 

 as the frost is gone. But I found it would not 

 answer. Unless the seed was sown in August, 

 the plants did not gain size, vigor, and rout- 

 power enough to resist the winter. Farther to 



