FIEI.D CROPS, 547 



rt is st.-itcd tlKit siiKill cultivators use seed of coiunioii (|ualit.v. roi,Mnlless of 

 ori^'iii :iu(I piu-ity. aud tiiat to oveiroiue tliis dilliculty tlic Klu'divial Ayrieul- 

 tural Socit'ty. in conut'ctiou with tlio asrirultural iiaulc. distriliutcs annually tlu' 

 host seed ohtainalde at cost price, the value of which is collected at the end of 

 the following cotton season. This system does nothing; to actually improve the 

 seed and foi' this reason the country is in urgent need of a more effective 

 method of si'lection and distrihution. A description of cotton seed selection, 

 as given in the Yearbook of this Department for 1902, is reproduced. It is 

 liointed out that a variety of cotton known as Hindi is found mixed with the 

 inii)roved varieties, esi)eci:illy Aliili. and that tlH> elimination of this sort from 

 the better kinds is one of the itrincii)al problems connected with seed selection. 



Experiments with oats, .T. F. Duggar (Mnhania Collcyc tSta. liiil. J.il, pp- 

 ■')9-i)-'i. fi<j'<. S). — This bulletin snnmiarizes the experiments with oats conducted 

 by the station over a period of 10 years. 



In the tests of varieties the yields of Red Rust Proof, Api)ler, and Culberson. 

 when sown in the fall, were practically equal. When sown after Christmas, 

 Burt or May oats averaged 7 per cent less grain than Red Rust I'roof. Turf, 

 or Gray winter oats, sown in November produced only 59 per cent as nmch 

 grain as Red Rust Proof sown at the sanie time. The order of ripening of the 

 principal varieties sown in the fall w'a.s Burt, Red Rust Proof, and Turf. Red 

 Rust I'roof and its related varieties or sti-ains. Appier and Culberson, are re- 

 garded as the best general-purpose varieties for that region. Burt oats was 

 found chiefly valuable for its earliness. but it was apparently not as hardy as 

 some of the other sorts. 



The experiments conducted to determine methods for the reduction or pre- 

 vention of winterkilling indicate that sowing in dee;) drills in October, growing 

 a hardy variety, and using the land roller after the plants have been heaved 

 and their roots exposed, are the most effective means. As compared with 

 broadcasting, oats drilled in deep furrows about 2 ft. apart yielded .3.2 bu. per 

 acre more when the deep furrows were only partly tilled and 2..'! bu. more when 

 the furrows were almost completely filled. 



In general, fall sowing gave much better results than s])ring sowing. In 7 

 experiments made in 7 different years. Red Rust I'roof sown in November gave 

 an average of 11.?. bu. per acre more than when sown in February. From the 

 results secured, it is advised to sow the seed in the fall in October and in the 

 spring, during the first few days of February. 



The results of the fertilizer tests show that nitrogenous fertilizers were more 

 profitable than phosphate or ])otash on the sandy and loamy soils of the sta- 

 tion, l)ut they also indicate that on such soils the application of 100 lbs. of acid 

 phosphate at the time of sowing may be recommended. Nitrogen in the form 

 of nitrate of soda was more effective than in any other forms, and was also 

 somewhat more effective in the form of cotton-s(>ed meal than of cotton seed. 

 It is advised to apply cotton seed or cotton-seed meal .it the time of sowing and 

 nitrate of soda as a top dressing in March after growth begins. Barnyard 

 manure greatly increased the yield of oats and exerted some effect on the next 

 crop. lu one experiment 4.S.1 lbs. of nitrate of soda and 10.3 lbs. of acid phos- 

 phate were required to produce the same increase as 1 ton of fine, fresh, un- 

 bleached horse manure. In 1.3 experiments with nitrate of soda the yield and 

 total profit increased with the <iuantity applied up to 200 lbs. iier acre, but the 

 cost of nitrate of soda recpiired to produce one additional bushel of oats was 

 14.5 cts. from the use of 03 lbs. per acre, 17.7 cts. from the use of 100 lbs., and 

 21.1 cts. from the use of 200 lbs. Cowpeas, velvet beans, peanuts, or soy beans, 

 whether the entire growth was plowed under the fertilizer or only the stubble, 

 gave an increase in the succeeding oat crop of from 6.2 to 33.G bu. per acre. 



