436 EXPERIMENT STATION RECORD. 



.solution, which is run into the flask from a burette so graduated that its readings 

 indicate directly the percentage of calcium carbonate. 



On the question of the separation of iron and aluminum from manganese, 

 calcium, and magnesium in the analysis of plant ashes, P. Kaschinsky (Jour. 

 Landw., 53 (1905), pp. 179-185). — A series of comparative tests of Konig's method 

 (precipitation of iron phosphate in presence of ammonium acetate) and Tollens'a 

 ii let hod (precipitation in presence of acetic acid) is reported. From the results the 

 author concludes that neither of these methods gives accurate results, due to the fact 

 that considerable amounts of calcium phosphate are precipitated with the iron 

 phosphate. 



Determination of the loss on ignition in soil analysis, H. Mehring (Jour. 

 Landw., 53 (1905), No. 3, pp. 229-237).— It is claimed that incineration with ammo- 

 nium carbonate gives a greater weight than that which can be ascribed to the restora- 

 tion of carbon dioxid which would be lost by ignition without the use of carbonate. 



The loss on ignition is a very indefinite factor, and is made up of three main ele- 

 ments, carbon dioxid driven off, chemically combined water eliminated, and organic 

 matter destroyed. The use of ammonium carbonate in ignition increases rather than 

 decreases the error of the determination. It is claimed that the determination of 

 humus by combustion with copper oxid or chromic acid is more reliable and useful 

 than incineration. 



On the difficulty of detecting small variations in the nitrogen content of 

 soils by means of the Kjeldahl method, R. Thiele (Mitt. Landw. Inst. Breslau, 

 3 (1905), No. 2, pp. 157-178, figs. 4)- — This article reviews very thoroughly the litera- 

 ture relating to the sampling of soils and discusses the difficulty of securing repre- 

 sentative samples showing with accuracy the small differences occurring in the plant- 

 food constituents, particularly nitrogen, in soils. 



A series of determinations of nitrogen (5 for each sample) by the Gunning- Atterberg 

 method in 24 samples of soil, composites of samples from 10 different places at regu- 

 lar intervals on 200 square meter plats, are reported, and the results are platted in 

 comparison with the curve for rainfall for the period during which samples were 

 taken. Examinations were also made of 10 samples of soil made up from portions 

 of 10 spadefuls of soil taken around the 10 selected points. 



The results indicate that it is not possible to get a strictly representative sample of 

 a large area by present methods of sampling, but that it is possible to detect by 

 chemical analysis much smaller variations in nitrogen content than are likely to be 

 of significance in experimental work. 



Determination of nitric acid in the presence of organic substances, B. Pfyl 

 (Ztschr. Untersuch. Nahr. u. GenussmtL, 10(1905), No. 1-2, pp. 101-104, fig- 1; abs. in 

 Analyst, 30 (1905), No. 355, pp. 345, 346, fig. I).— The method described is based 

 upon the reduction of the nitric acid to nitric oxid by means of ferrous chlorid and 

 hydrochloric acid, the nitric oxid being washed by passing it through sodium 

 hydroxid solution and absorbed by one-tenth normal potassium permanganate solu- 

 tion. The excess permanganate solution used is titrated with tenth normal ferrous 

 chlorid solution. A special form of apparatus for use in the reduction is described. 



Methods of examination of potable waters and the interpretation of the 

 results, F. Dienert (Ann. Inst. Pasteur, 19 (1905), No. 9, pp. 541-563, figs. 4). — 

 Various methods of examining potable waters are discussed in this article, especial 

 attention being given to a method based on electric conductivity, which is described 

 in detail, and which it is claimed is well adapted to a study of variations in under- 

 ground waters. This method with quantitative determinations of Bacillus coli com- 

 munis is believed to furnish an adequate basis for judging of the sanitary quality of 

 waters. 



The estimation of cellulose and lignin in feeding stuffs and foods, J. Konig 

 (5. Internal. Kong. Angew. Chem., 5 (1903), Ber. 3, pp. 1052-1060) .—The author 



