30 EXPERIMENT STATION. [Jan. 



suininer sowing the best conditions wiil usually occur about 

 Julj 25 to August 10. If sowing is deferred much later than 

 the middle of August, the alfalfa does not get sufficiently Avell 

 rooted nor make sutficicnl growth to go through the winter suc- 

 cessfully. When sown about the last of July the crop fre- 

 quently attains a height of a foot before cold weather, and 

 should on no account be cut, as this growth is needed for winter 

 pi'otcction. 



Alfalfa Seed. — No pains essential to procuring the best pos- 

 sible seed shoidd be spared. Numerous varieties of alfalfa 

 have been introduced and extensively tried in ditl'erent parts 

 of the country, and a mnnber of them have been under trial 

 in this experiment station. So far, however, as the experiments 

 of the writer enable him to judge, none of the newer varieties 

 appears to l>e superior to the common type of alfalfa as culti- 

 vated in our northwestern States. Northern-grown seed, bright, 

 heavy, free from impurities of all kinds, and especially free 

 from dodder (a parasite which, if present in any amount, will 

 destroy the crop), should be looked for. The seeds of dodder 

 are extremely minute, and will be easily overlooked by the cas- 

 ual observer. The dealer should be asked to guarantee fret^lom 

 from seed of this parasite, and in all cases of doubt samples 

 should be sent to the experiment station for examination. 



Qvantity of Seed. — As is the case with all farm seeds, the 

 quantity which may wisely be sown varies with conditions, and 

 should be greater in proportion as these are unfavorable. With 

 the best conditicms for germination, and with seed of the best 

 quality, about -."> to '30 ]-ounds per acre should be sufficient. 

 A thick stand of plants is, however, of the very highest impor- 

 tance. When the alfalfa is thin, weeds, grasses and clovers 

 come in, and the latter especially tend to crowd the alfalfa 

 out. Tt is the belief of the writer that the quantity of seed 

 should therefore, even under the best soil conditions, be not 

 less than 30 pounds to the acre, and in seeding with corn it 

 would seem to be advisable to use from 5 to 8 pounds more. 

 J voculnilon. — Tt is uow generally understood that l(\ii'umes 

 have the capacity to take nitrogen from the air only in pai-tner- 

 ship with bacteria which live in uocbdes on their roots. These 



