320 EXPERIMENT STATION RECORD. 



percentage of ca,lcium carbonate is read from the scale. If 0.4 or more of calcium 

 carbonate i?' thus found in the soil, application of lime is considered unnecessary. 



The determination of lime in soil samples, E. Hotter {Ztschr. Landw. Ver- 

 suchw. Oesterr. , 4 {1901 ) , No. 5, jjp. 632-636; ahs. in Chem. Ztg. , 25 ( 1901 ) , No. 5S, Repert. , 

 p. 208). — The method proposed by the author for the determination of available 

 lime (in form of carbonate) in soils is as follows: Digest 20 gm. of the dry soil, which 

 has been ground to pass a 15 mm. sieve, with 50 cc. of 20 per cent acetic acid for one- 

 half hour in a boiling water bath, make the volume to 1 liter, shake thoroughly, and 

 allow to stand for 12 hours for the solution to become clear. To 200 cc. of this solu- 

 tion add ammonium oxalate in excess to precipitate lime. Collect the precipitate on 

 a filter, wash with cold water, and heat gently at first and then strongly ignite, cool, 

 and weigh the CaO obtained. The ignited precipitate is generally colored gray or 

 brown by manganese oxid, but the amount of this substance is usually negligible. 

 It may be removed, however, by dissolving in dilute nitric acid, filtering, and rejire- 

 cipitating the lime. Twenty comparfsons of this method with the gravimetric 

 method based upon the determination of carbon dioxid are reported. The results 

 by the new method are as a rule higher than those given by the gravimetric method. 



The complete analysis of feeding materials, C. A. Browne, jr., and C. P. 

 Beistle {Jour. Arner. Cheui. Soc, 23 (1901), No. 4, pp. 229-236).— A sample of distil- 

 lery waste or mash analyzed by the usual method at the Pennsylvania Station 

 showed a nitrogen-free extract content of 42.93 per cent. Of this about 25 per cent 

 Avas pentosans, leaving some 15 per cent of material unaccounted for. In analyzing 

 the latter substance by the Sherman method (E. S. R., 8, p. 951) the sum of all the 

 ingredients in the feeding stuff was found to be 96.01, leaving a residuum of about 4 

 per cent. In order to determine the composition of this residuum a second series of 

 samples was carried through as in the first case, except that the pentosans in the 

 residues were estimated after each stage of the process. From these results it 

 appears that several per cent of a material of a pentose nature found its way into the 

 alcoholic or aqueous extracts, which because of its copper-reducing ' power would 

 affect the sugar and dextrin determinations. As no copper-reducing bodies were 

 removed by the alcohol, it seems that these pentoses were removed by the water. 

 The copper-reducing power of the aqueous extract the writers believe to be due to 

 bodies of a pentose nature, and if such be the case a serious error exists in the usual 

 process of determining dextrin. Further tests of the results found are presented, 

 and attention is called to the uncertainties in the analytical methods of feeding stuff 

 analysis, as well as to the more- or less arbitrary nature of the factors for calculating 

 protein and pentosans. 



The hydrochloric-acid-phloroglucin method in the determination of pento- 

 sans, E. Kkober {.lour. Lundir., 4^ (1901), No. 4, pp. ■157-384)- — The author made 

 an extended study for the purpose of testing the method, and in order to determine 

 factors for estimating furfurol and other substances from the phloroglucid obtained. 

 He found that the results are not influenced by allowing the precipitate to stand from 

 14 to 20 hours. A porcelain Gooch crucible is better adapted to collecting the pre- 

 cipitate, and the wash water should be put through in small portions, attention 

 being paid to the physical condition of the residue, otherwise the filtrate may show 

 no reaction for chlorin while the residue will contain it. Should the precipitate con- 

 tain chlorin, too high results will be obtained. No practical difference was found in 

 the results when pure phloroglucin and that containing diresorcin was used. Twice 

 as much phloroglucin as is expected of the furfurol should be used. Before adding 

 the phlor(jglucin to the furfurol solution it should be dissolved in warm dilated 

 hydrochloric acid (specific gravity 1.06). Drying the precipitate 4 hours at a tem- 

 perature of from 98.5 to 100° C. is recommended. Attention is called to the fact 

 that phloroglucid is very hydroscopic, and the author obtained lietter and more uni- 

 form results by cooling the Gooch crucible in a Aveighing bottle in the desiccator. 



