628 EXPERIMENT STATION RECORD. 



The results for Ul varieties of millets belonging to the species Pani- 

 cum crus-galli, P. miliaceum, and P. italicum, grown on plats 1 sq. rod 

 in size, are given in a table. Owing to the small size of the plats, no 

 conclusions as to the relative merits of the different varieties are given. 

 Japanese barnyard millet ( P. crus-galli), Japanese broom-corn millet 

 (P. miliaceum), and Japanese millet (P. italicum) yielded at the rate of 

 about 40 bu. of seed per acre. The early white soy bean yielded 18.7 

 bu. of beans per acre, medium black 16 bu., and medium green 31.7 bu. 

 The test of medium red, mammoth, alsike, and crimson clover was 

 continued from last year (E. S. It., 9, p. 310). Mammoth clover gave the 

 best results. The other kinds were injured to a considerable extent 

 during the winter. The results of a comparison of sulphate and 

 muriate of potash as fertilizers for clover were without special signifi- 

 cance. Sweet clover (Melilotus alba), grown on the same plat, made a 

 better growth this year than the year before, the yield being at the rate 

 of about 12£ tons per acre. This is believed to be due to a larger sup- 

 ply of the appropriate nodular bacteria in the soil. The crop is recom- 

 mended for green manuring rather than for fodder. 



Experiments with alfalfa, saccaline, crimson clover, winter vetch, 

 Idaho field or coffee pea (Gicer arietiuum), Brazilian stooling flour corn, 

 and black chaff or African millet are briefly noted. Alfalfa seed treated 

 with Nitragin showed no advantage over untreated seed. Saccaline, 

 crimson clover, Idaho field or coffee pea, and Brazilian stooling flour 

 corn were found unprofitable as fodder crops. Black chaff or African 

 millet resembles Kaffir corn, and both are regarded as inferior to maize 

 for that climate. 



Report of the director, H. Benton (Alabama Ganebrake Sta. Rpt. 

 1897, pp. 5, 6, 8-11). — The experiments described comprise culture, 

 variety, and fertilizer tests of cotton ; variety and culture tests of corn, 

 and a comparison of standard varieties of potatoes. 



Experiments icith cotton. — Fertilizer tests were made on a sandy cal- 

 careous soil with application of sulphate of ammonia, sulphate of pot- 

 ash, acid phosphate, tankage, cotton-seed meal, and barnyard manure. 

 The results show that acid phosphate applied singly or in combination 

 did not increase the yield. Sulphate of ammonia and sulphate of potash 

 applied alone or in combination gave good results. Sulphate of potash 

 and tankage was found the most profitable combination on these 

 soils. It is concluded from the results of the experiments that the sandy 

 calcareous soils of that region are deficient in potash and ammonia. 

 Shallow preparation of land with the cotton planted on beds gave 

 better results than deep preparation with the cotton planted on the level. 

 Among 7 varieties Truitt gave the best yield of seed cotton. 



Experiments with corn. — Land plowed 4 in. deep with the corn planted 

 in beds yielded 24.1 bu. per acre; land plowed 10 in. deep with corn 

 planted on the level yielded 24.2 bu. There was practically no differ- 

 ence in the yields of Huffman Early and Welborn Conscience corn. 



