SOILS FERTILIZERS. 415 



and 0.17 per cent nitrogen. Potash was not determined. Other papers on the 

 snbject are reviewed. 



Predicting water supply for the farmer, J. C. Alter (Sci. Amer. Sni)., 73 

 {1912), No. 190.'f, pp. 413, 414, figs. 6). — A brief account is given of a snow 

 survey of the Maple Creek watershed near Springfield, Utah, by the Weather 

 Bureau of this Department. 



" The actual work of the snow survey consisted in going up the bottom of 

 each gulch or canyon and back and forth along the slopes, measuring the snow 

 depth and density every thousand feet or so, depending upon the variability of 

 the snow deposit, the general topography, and the brush or forest cover, and 

 mapping the snow area on a field map. ... 



" Stream flow measurements at a weir, located some distance above the high- 

 est farm land, have been made daily since the completion of the survey, and 

 will be continued throughout this year (1912) and probably indefinitely in the 

 future. Eain and snowfall observations have also been made from a precipita- 

 tion gage located near the weir, and are to be continued indefinitely. 



" The first year's records of run-off, following the survey, are not directly 

 comparable with the amount of water in the form of snow, shown by the sur- 

 vey, even when corrected for precipitation gain, and evaporation loss, as there 

 is a seepage loss in this particular canyon, the amount of which can not be 

 determined from one year's observations." 



SOILS— FERTILIZERS. 



Investigations on soils from crystalline rocks in process of weathering, K. 

 BuscH (Untcrsuchurigen iiber Venciitciiingsboden kristaUinischcr Gesteine. 

 Inaug. Diss., Univ. Halle, 1911, pp. 67; Kiihn Arch., 1 (1911), pt. 2, pp. 357-389, 

 fiy. 1; abs. in Zentbl. Agr. Chem., J,l {1912), No. 3, pp. l-'i5-147).—lt is pointed 

 out that previous investigations on soil particles, such as those by Ramann, 

 Keilhack, Meyer, Sachse, Borzuchowski, and Puchner, have had to do with soil 

 material of a heterogeneous character. In his work, therefore, the author used 

 soils derived from known sources, that is, those formed in place from diabase, 

 granite, and basalt rocks. With them he studied (1) the relation between the 

 mechanical composition as determined by Hilgard's elutriator method and the 

 hygroscopicity of the soil by the Mitscherlich method (E. S. R., 24, p. 419), and 

 (2) the proportion of plant-food constituents contained in the different sized 

 groups of soil particles. The surface soil derived from the diabase rock was 

 a porous, fertile loam with a very porous, slightly sandy subsoil. The granite 

 soil was a shallow, sandy loam with a rocky subsoil, and the basalt soil was a 

 deep, heavy loam. 



It was found that the individual soil particles of the different groups as 

 separated out by sedimentation were not of the same size. This the author 

 attributes to a difference in the surface area of the particles and a consequent 

 variation in the resistance to the current of water. The variation in results of 

 repeated mechanical analyses of the same soil is attributed to errors in sam- 

 pling. Contrary to the conclusions of Mitscherlich, the hygi-oscopicity of these 

 stuls was not found to be proportional to the outer surface area. This may 

 be explained by the fact that the larger soil particles contained hygroscopic 

 water in the weathered cracks and crevices. Soils of the same mechanical com- 

 position differed in their hygroscopicity, this difference being determined by the 

 mineral composition and the stage of weathering of the soil. 



A relation between the size of the soil particles and the plant-food constitu- 

 ents was determined only- for the iron and the calcium and magnesium car- 

 bonates. The iron was associated mostly with the finest particles. The dis- 



