Sudan Grass. 21 animal, the lot fed on alfalia showed an average gain of 5 pounds per animal, and -the lot fed on kafir stover showed an average loss of 50 pounds per head.

The Fort Hays experiment station also found Sudan grass hay an efficient feed for carrying stock cattle through the winter. When fed with a small supplementary ration of silage and linseed or cottonseed meal, steady gains in weight were obtained during the winter at a reasonable cost.

For milk cows Sudan grass hay was slightly less efficient than alfalfa in a test carried out at the Kansas experiment station. Cows fed on Sudan grass hay with a supplementary ration of silage and grain produced 97 per cent as much milk as when fed alfalfa hay in place of the Sudan grass hay.

SEED PRODUCTION.

Sudan grass produces seed freely in a loose, open panicle which is held nearly erect by the stem, and can be harvested easily. The seed is retained fairly well, and thus the loss from shattering is much less than in other wild forms of sorghum. However, on the Great Plains, high winds sometimes shatter out a large percentage of the seed.

At the present time most of the commercial Sudan grass seed is produced in Texas, Oklahoma, and Kansas, but western Missouri and eastern Colorado and New Mexico also produce more seed than is needed for local consumption. Seed yields per acre are largest in the irrigated regions of California and Arizona, but only limited acreages are devoted to Sudan grass in these localities because of the profitable returns from other crops. The yields of seed per acre obtained in different sections of the United States are shown in Table III.

TABLE III.— Yields of Sudan grass seed under différent climatic conditions in cultivated rows and clone drills.

Yield of seed per acre (pounds).

..._ Climatic conditions. Broadcast or dose drills.

Cultivated rows 18 to 24 Inches apart.

Cultlvüted rows 30 to 44 Inches apart.

Number of localities.

Uumld 328 224 1,420 417 290 1,210 354 270 1,110 10 Dry (not irrifiated) 13 I>ry (irrigated) 6 i The yields for the humid reliions are averages for tests at Manhattan, Kans.; Stlllwater, Olda.; Beevllle and Temple Tex,; Jackson, Tcnn.; St. Paul, Minn.; Madison, Wi».; College Park, Md.: and Arlington Farm and lîlai-kslmrg, Va.; for dry regions not Irrigated, averages of tests at Colby, Hays, Tribune, (larden City, and Dodge City, Kans.; Chillicolhe, Spur, Lubbook, Amarillo, and Dalhart, Tex.; Bitiville and Wenatchee, Wash.; ami Davis, Calif.; for dry regions Irrigated, averages of tests at San Antonio, Tex., Phoenix, Ariz.; Bard, Davis, and Chico, Calif.; and Umatllla, Greg.