38 EXPERIMENT STATION RECORD. [7ol.36 



garding the soil and season at the several farms and stations for 1914 are 

 compiled. 



Sudan grass in Kansas, G. E. Thompson {Kansas Sta. Bui. 212 {.1916), pp. 

 3-29, figs. 8). — This bulletin presents a general discussion of Sudan grass cul- 

 ture, and reports the results of experiments relating to date and rate of seeding, 

 yielding capacity with and without irrigation, comparative feeding value, and 

 the usefulness of the grass for pasture. The chemical analysis of Sudan grass 

 seed as compared with wheat, corn, oats, and Kafir corn, and the composition 

 of Sudan gi-ass hay as compared with alfalfa and prairie hay and sorghum fodder 

 are given in tables. The experimental work was carried on at the main station 

 and the various substations and partly in cooperation with the United States 

 Department of Agriculture. The principal diseases attacking Sudan grass 

 together with control measures are briefly described. 



The results indicated in general that Sudan grass is primarily a hay crop 

 which is adapted to practically all soils of the State excepting those very wet 

 and poorly drained, extremely alkaline, or extremely sandy. It is stated that 

 the crop may be planted and handled in much the same manner as the ordinary 

 sorghums and that, when planted in cultivated rows, 2 to 4 lbs. of seed per 

 acre are needed, while when broadcasted or drilled, 15 to 35 lbs. per acre 

 give the best i-esults. It was found that 2 to 8 cuttings for hay or one seed 

 crop may be secured each season and that 400 lbs. of seed per acre or 2 

 to 4 tons of hay constitute an average crop. Comparative analyses showed 

 that Sudan grass hay, while richer in protein than prairie hay, is not so rich 

 as alfalfa hay. In feeding tests with dairy cows and with beef cattle and work 

 horses and mules carried through the winter uniformly good results were 

 secured. The results of pasture tests indicated this grass to be valuable in a 

 limited way for grazing purposes and that it may have an important place on 

 many farms in furnishing grazing for hogs. 



The time to seed wheat in Kansas, L. E. Call, S. C. Salmon, and C. O. 

 Cunningham {Kansas Sta. Bui. 213 {1916), pp. 3-16, fig. 1). — Experiments to 

 determine the best time to seed wheat were begun in 1911 at the station and 

 at six places in as many counties in the main wheat-producing area of the State 

 where the Hessian fly is likely to cause serious loss. In each test a series of 

 plats was sown at weekly intervals for six or seven weeks, beginning the second 

 week in September. The outcome of the tests is given in tables and is briefly 

 discussed. 



It was found that in central and eastern Kansas wheat may be sown late 

 enough to avoid most of the injury from Hessian fly and yet early enough on 

 well-prepared ground to secure a good growth before winter. The dates given 

 as best for seeding are as follows : Northeastern Kansas from about September 

 25 to October 3, southern central Kansas from about September 25 to October 7, 

 and northern central Kansas from about September 20 to October 1. It is 

 pointed out that as the time for seeding western Kansas depends on the rainfall, 

 the land should be prepared early and seeded when in proper condition to insure 

 germination and good growth. It is further stated that wheat may be sown 

 later in rich soil on well-prepared ground than in soil that is poor or poorly 

 prepared, as early and thorough preparation not only destroys many of the flies 

 but enables the wheat to make a more rapid, vigorous, and resistant growth. 



Tests to observe the elfect of the time of seeding on the tillering of wheat 

 showed that the rate of seeding made very little difference in the yield when 

 the sowing was done before September 28. After this date the higher rates] 

 of seeding gave the best yield. 



Liming the wheat crop, C. E. Thobne {Mo. Bui. Ohio Sta., 1 {1916), No. 5,1 

 pp. 277-281, figs. S). — This article presents some results obtained from experi-j 



