430 EXPERIMENT STATION RECORD. 



modified, but does not disappear, when calcium sulphate is present. The rationale 

 of the application of gypsum or land plaster to soils containing 'black alkali,' or 

 sodium carbonate, resulting from the action of sodium chlorid ujjon calcium carbon- 

 ate, is made apparent. 



" Sodiuni sulphate has been shown to have an astonishingly great effect on the 

 soluljility of calidum carbonate, the resulting solutions containing not only hydrogen 

 carbonate, but normal carbonate as well. Further, there is no necessary precipita- 

 tion of the lime carbonate with increasing concentration with respect to sodium sul- 

 phate; but the solubility of the lime carbonate steadily increases with increasing 

 amounts of sodium sulphate in the solution, up to the saturation point of this latter 

 salt." 



A detailed description is given of the method of chemical examination of alkali 



soils worked out in the laboratory of the Division along the lines indicated by the 



results of the investigations above noted. The salient features of this method are — 



" (1) The aqueous solution of the soil alkali is made under similar conditions in 



all cases. 



" (2) The electrical resistance of the soil solution is used in facilitating the analyt- 

 ical proce,sses. 



" (3) A direct determination of the total solids in the soil solution is not made. 

 " (4) A unique, satisfactory, and accurate volumetric method is used for the deter- 

 mination of carbonates and bicarbonates (E. S. R., 12, p. 819). 



"(5) All the constituents, including sodium, are determined directly, and no 

 results are obtained by difference. 



"(6) The possible necessity for duplicate determinations is made evident l)y the 

 calculation of the results obtained. 



" (7) Finally, the report of the analysis includes the percentage of alkali present in 

 the soil and the relative percentage composition of this alkali; it is stated in terms of 

 the possible ions present in the solution as well as their combinations as electrolytes." 

 Comparative tests are reported which indicate that Congo red may in some cases 

 be substituted with advantage for methyl orange as an indicator in the method of 

 determining carbonates and bicarbonates referred to above. 



"For some eyes and under certain c'onditions Congo red is to be preferred to 

 methyl orange; for those who can use it, however, methyl orange seems to be the 

 more delicate indicator. 



"It is essential to the satisfactory use of the method that during the titration the 

 solution should not be in contact with carbonates, bicarbonates, or metallic hydroxids 

 in the solid j)hase." 



Soil and soil moisture investigations, J. D. Tinsley and J. J. Yerxon {Xeiv 

 Mexico Sta. Bui. S8, pp. 65-95, pis. 12). This bulletin reports the results of a study 

 of the physical character and of the fluctuations of moisture during the season of 

 1900 in 12 plats of soil planted to corn. The methods and apparatus employed 

 are described and the results of determinations of moisture on the different j)lats at 

 depths of from 3 in. to 10 ft. are tabulated. A form of auger for taking soil samples, 

 and a shaker for use in mechanical analysis are described. With these ext'eptions 

 the methods and apparatus were essentially those of the Division of Soils of this 

 Department. The soils of the experimental plats were found to be very uneven in 

 texture. Tliis is stated to be due to the fact that 4 different factors have entered 

 into their formation, (1) river deposit, (2) wash from the foothills, (3) sediment 

 from irrigating water, and (4) drift of soil by the winds. In studying the porosity 

 of the soils comparative tests were made of 4 methods, namely, Beeson's (E. S. R., 

 9, p. 429), Whitney's, A. MaVer's, and Wolff- Wahnschaffe's. The Wolff- Wahn- 

 schaffe method was found to give the most concordant results and was used in the 

 examinations reported. The studies on soil moisture were conducted on the same 

 plan as in previous years (E. S. R., 12, p. 425). Moisture determinations were made 



