1917] FIELD CROPS. 635 



Five methods for sampling a rice crop In the field to estimate the total yield 

 are described and were tested, and the results presented in tabular form. In 

 order to place the results on some basis for comparison, the estimated standard 

 error has been calculated in each case according to the following formula: 



when n is the total number of plants sampled, <r the real standard error, and 300 

 a constant. From these results the author concludes that either the " diagonal " 

 or " scattered " method of sampling is superior to the other methods tested. 



[Report of jfield crop work], J. M. Scott {Florida Sta. Rpt. 1916, pp. 23- 

 29). — Fertilizer tests with Japanese cane, velvet beans, and sweet potatoes, 

 variety tests with cowi)eas and velvet beans, and field tests with cotton and 

 sorghum are briefly noted. 



The 1915 yield of Japanese cane showed a decrease, as compared with that 

 of 1914, of more than 50 per cent on some plats, while on others the decrease 

 varied from 10 to 20 per cent. In the spring of 1915 land grown continuously 

 with Japanese cane under varying fertilizer treatments since 1908 was plowed 

 and replanted to cane, using the same fertilizers. The results are interpreted 

 as strongly indicating the advisability of replanting Japanese cane every three 

 or four years. The highest yield, 31.9 tons of green forage per acre, was se- 

 cured from the plats receiving 84 lbs. of muriate of potash and 224 lbs. of acid 

 phosphate. The lowest yield, 18 tons per acre, was obtained from the plat 

 receiving 112 lbs. of dried blood and 224 lbs. of acid phosphate. Applications 

 of ground limestone failed to show any substantial benefit after the first appli- 

 cation in 1909. 



Fertilizer tests with velvet beans gave the highest average yield, 658.4 lbs. 

 of pods per acre, with an application of 360 lbs. of Thomas slag, while the 

 lowest yield, 583.9 lbs., was secured from a 400-lb. application of acid phosphate. 

 The untreated checks yielded 616 and 642.8 lbs. per acre, respectively. 



An application of 112 lbs. of dried blood, 224 lbs. of acid phosphate, 84 lbs. 

 of sulphate of potash, and 2,000 lbs. of ground limestone showed the highest 

 average yield of sweet potatoes, estimated at 269.6 bu. per acre. The lowest 

 yield, estimated at 99.6 bu., was obtained from a plat receiving 112 lbs. of dried 

 blood and 224 lbs. of acid phosphate. 



Of the four varieties of co^vpeas tested for forage production, Monette S. P. I. 

 No. 1541 was highest with a yield of 1,705.1 lbs. per acre and Brabham second 

 with a yield of 1,577.9 lbs. In seed production Monette S. P. I. No. 1541 

 yielded 531.7, Brabham 517.3, and S. P. I. No. 27863 258.2 lbs. of seed pods per 

 acre. 



Tests with velvet beans included the Yokohama, Osceola, Florida, Chinese, 

 and Wakulla varieties, which yielded 1,893, 1,394.6, 1,320, 1,229.5, and 856 lbs. 

 of pods per acre, respectively. 



The average acre yield of seed cotton amounted to 349.5 lbs. per acre. Selec- 

 tion work with cotton is being continued. 



The seed heads of Sumac sorghum grown on the station farm amounted to 

 1,236.25 lbs. per acre green weight and 1,129 lbs. dry weight. The average 

 yields in green and dry forage amounted to 9,512 and 3,037 lbs. per acre, respec- 

 tively, based on a 2-acre yield. 



[Experiments with field crops in Maine], C. D. Woods (Maine Sta. Bui. 

 260 {1917), pp. 102-120). — Variety and rate-of-seeding tests with oats were 

 continued as previously noted (E. S. R., 35, p. 33), and fertilizer experiments 

 with oats and potatoes reported. 



