822 



EXPERIMEI^T STATION RECORD. 



(Vol. 37 



in natural habitats and that the action of the photoelectric cell in light is 

 more nearly parallel to that of the organism than that of any other light- 

 measTiring instruments hitherto available. 



FIELD CHOPS. 



Effect of fall irrigation on crop jdelds at Belle Fourche, S. Dak., F. D. 

 Farkell and B. Aune (U. S. Dept Agr. Bui. 546 (1917), pp. IS, fig. 1).— The 

 favorable results secured at Scottsbluff. Nebr. (E. S. R., 32, p. 36), with fall 

 irrigation for spring-planted crops led to a repetition of the experiments under 

 different soil conditions at the Belle Fourche Experiment Farm. The experi- 

 ments were begun in 1913, and involved the use of oats, sugar beets, flax, po- 

 tatoes, barley, corn, and wheat gro\\'n on duplicate check plats receiving the 

 usual summer irrigation and on duplicate plats receiving in addition a fall irri- 

 gation. The results secured in 1014, 1915, and 1916 are reported and discussed. 



The soil upon which these experiments were conducted is described as an 

 extremely heavy clay, popularly knowm as " gumbo " and classified by the 

 Bureau of Soils as Pierre clay. Mathews (E. S. R.. 36, p. 210) reported that 

 this soil woTild carry approximately 30 per cent moisture, about half of which 

 would be available to crops, and that the wilting coeflicient of the soil approxi- 

 mates 17 per cent. During the nine years 1908-1916, inclusive, the annual 

 rainfall varied from 6.64 in. in 1911 to 21.02 in. in 1915, the mean being 14.05 in. 

 The mean precipitation for the fall period (August to October, inclusive) for 

 the 9-year period was 3.75 in. and for the winter period (November to March, 

 inclusive) 2.15 in. 



The average crop yields secured on the fall-irrigated plats and on the check 

 plats are reported in tabular form and the summarized statement below given 

 to show the probable errors of the average yields. 



Summary of crop yields ahoicxng the probable errors of (he averages. 



I 



Since none of the differences in favor of fall irrigation was as great as the 

 probable error, all were regarded as insignificant. The lower average yields of 

 the fall-irrigated plats are attributed to the relatively low productivity of 

 one of the two fall-irrigated scries, the 3-year average yield of which was 16 

 per cent lower than that of the duplicate series, while the corresponding aver- 

 ages of the two check series were identical. This low production was thought 

 to have been associated with a heavy infestation of gumbo weed, Iva axiHaris, 

 on the low-yielding series. 



Soil moisture determinations made in the spring and early summer of each 

 year are reported and the data tabulated for each year of the experiment. In 

 1914 the first two samplings showed more moisture in the first 3 ft. of the 



