WATER SOILS. 565 



ing the necessity of maintaining a certain ratio between calcium and magnesium 

 compounds in the soil. The results indicate that plants grow well in media contain- 

 ing more magnesium salts than lime salts. The highest yields were obtained in 

 water culture with 0.4 part of lime (CaO) to 1 part of magnesia (MgO) and hence 

 the results did not bear out Loew's hypothesis. 



Fertility of soil according- to the most recent data, S. Bogdanov {Selsk. 

 Klioz. i L)/esov., 210 {1903), Sept., j)p. 628-667). — The author attempts a systematic 

 discussion of the question of the fertility of the soil, giving a summary of his previ- 

 ous articles ^on the subject (E. S. R., 12, p. 725) and supplementing it with results 

 obtained by other investigators. He gives especial attention to the sulphur require- 

 ments of i)lants, which he believes have heretofore been underrated, and reports 

 experiments made in 1898 and 1899 with special sulphate fertilizers in comparison 

 with other more common fertilizing materials on white mustard grown in pots of 

 clayey chernozem. The api)lication of sulphuric acid in form of sodium sulphate 

 greatly increased the yield in lx)th years. — p. firkman. 



The green sandstone soils of Lower Bavaria and the Oberpfalz, Puchner 

 ( Vrtljschr. Bayer. Landw. Rath., 8 {1903), No. 3, pp. 530-533). — Physical and chem- 

 ical examinations of several of these typical soils are briefly reported. The results 

 show that the soils are as a rule light and have only medium retentive power for 

 moisture. They need phosphoric acid most of all, occasionally potash, and usually 

 nitrogen. 



Some investigations on the physical properties of soils, W. Bagger ( Ugeskr. 

 Lnndm., 48 {1903), Xo^. 12, pp. 133-135; 13, pp. 152, 153). 



Mechanical and physico-chemical analyses of the soils of the provinces of 

 Forli and Cesena, Italy {A)m. li. Staz. Agr. Forli {1902), No. 31, p>p. 65-74). 



A new theory of the soil, A. D. Hall {Nature [London], 69 {1903), No. 1777, 

 pp. 58, 59). — A critical review of Bulletin 22 of the Bureau of Soils of this Department 

 (E. S. R., 15, p. 457). 



On the origin, properties, and applicability of Swedish moor soils, R. Tolp 

 {Svemk. Mosskult. TUMr., 17 {1903), No. 4, App., pp. 32). 



Worn-out farms, W. M. Munson {Maine [Univ.'] Bid., 6 {1903), No. 2, pp. 9-24).— 

 A general discussion of the causes of soil exhaustion and of practical means for the 

 restoration of worn-out lands. 



Studies in soil bacteriology, F. D. Chester {Delaware Sta. Ept. 1902, pp. 46-78,^ 

 figs. 6). — Studies are reported on the variation in the number of bacteria in soils, the 

 effect of stirring and pulverizing soil on its bacterial content, the effect of mineral 

 fertilizers on the development of soil bacteria, the predominating bacteria in a soil 

 sample, and on the nitrogen-assimilating bacteria of soils. 



In the previous report (E. S. R., 14, p. 232) the author showed the wide variation 

 in the number of bacteria in soils of adjacent plats bearing different cover crops. The 

 variation within narrow horizontal limits was investigated, samples being taken at 

 intervals of 12 in. and the number of bacteria per gram of dry soil determined. The 

 determinations, which are given in tables, show that within a range of 5 ft. the num- 

 ber of bacteria in the soil do not vary greatly. In order to test further the limit of 

 variation, plats were planted with peas and vetches and at intervals of 10 ft. soil 

 samples were removed for examination, the sample being taken to a depth of 8 in. 

 The determinations made in these experiments indicated that a fairly reliable aver- 

 age of the number of bacteria in the soil of the field can be obtained by taking 

 samples at intervals of 10 to 20 ft. and thoroughly mixing them. 



The effect of stirring and pulverizing soil on its bacterial content was investigated 

 to see what relation tillage would bear to the development of soil bacteria. Samples 

 of soil were collected and thoroughly mixed, all lumps being reduced to uniform size. 

 This mixing and pulverizing was assumed to be equivalent to thorough tillage. The 

 bacterial content of the soil was determined immediately after filling the pots and at 



