AGRICULTURAL CHEMISTRY AGROTECHNY, 611 



with water. The acidity remaining in 50 cc. of this mixture is titrated with one- 

 half normal potassium hydroxid, using phenolphthalein as an indicator. One 

 ce. of one-half normal acid is then added and boiled for 2 minutes in order to 

 remove the remaining carbon dioxid, and titrated back with one-half nor- 

 mal potassium hydroxid. The calculation, which considers both the lime 

 and magnesia as calcium hydroxid, is, for 1 gm. of titrated substance, 



( 51— rt. ^ KOH j 1.4, a signifying the total amount of one-half normal potassivmi 



hydroxid used. 



The second method consists of a modification of the Balthasar method, but 

 uses only one-half as strong a titrated potassium permanganate solution and a 

 larger amount of substance than the original method and allows of the deter- 

 mination of the lime alone. The difference between the calcium oxid determina- 

 tions of the two methods, multijilied by five-sevenths, gives the magnesia content 

 of the substance. The presence of silica does not influence the results in any 

 way, but a large amount of phosphorus does in the case of the second method. 



Volumetric determination of small amounts of arsenic, L. W. Andrews 

 and H. V. Farr {Ztsclir. Anorgan. Clicm., IJ.i (1909). A o. J. p. 123-128; abs. in 

 Ztschr. Untersuch. IS^aht: it. Gemissmtl., 19 (1910), No. 1. pp. 31, 32).— The 

 material, if in solution, is concentrated to a bulk of 15 to 20 cc. There is then 

 added 2.5 times its bulk of a solution consisting of 20 gm. of stannous chlorid 

 crystals and 40 gm. of tartaric acid, made up to the liter with 40 per cent 

 hydrochloric acid. 



The mixture is then allowed to stand for 2 to 3 hours in a stoppered glass 

 bottle (capacity 80 to 100 cc.) in a warm place until the supernatant liquid is 

 cleai-. The precipitate formed is collected on an asbestos filter with the aid 

 of hydrochloric acid, washed with water, and with the asbestos put back in 

 the original bottle and shaken with a known amount of a one-hundredth or one- 

 tenth normal iodin solution and a sufficient quantity of 5 per cent of sodium 

 bicarbonate or sodium phosphate solution to obtain a neutral reaction. The 

 shaking process is continued until all the arsenic is dissolved, when the excess 

 iodin is titrated with a one-hundredth or one-thousandth normal arsenite solu- 

 tion. For arsenic under 0.5 mg. one-thousandth normal and for 10 to 100 mg., 

 one-tenth normal solution can be used. In the first instance, however, a cor- 

 rection of 0.6 cc. of a one-thousandth normal solution for every 50 cc. of solu- 

 tion em])loyed must be applied. 



Aids to the analysis of food and drug's, C. G. Moor and W. Partridge 

 (London. 1909, 3. ed., pp. 2'i9). — This is a small handy volume, dealing with 

 the general analysis of foods and drugs, and containing much matter of interest 

 to the public analyst. 



A source of error in the examination of foods for salicylic acid, H. C. 

 Sherman (Jour. Indus, and Engin. Chem., 2 (1910), No. 1, pp. 24, 25). — It is 

 commonly assumed in testing foods for preservatives that a body soluble in 

 ether, volatile with steam, sublimable, and crystallizable, and giving a violet 

 reaction with ferric chlorid, is salicylic acid. The author draws attention to 

 the fact that in the case of cereal foods this coloration is probably due to the 

 maltol of Brand, as no reaction could be obtained with Jorissen's reagent. A 

 recommendation is made to adopt Jorissen's reaction for such work. 



See also Backe's work in this regard ( E. S. R., 22, p. 412). 



Detection of formaldehyde in chopped meat, W. Bremer and R. Beythien 

 (Ztschr. Untersuch. Nahr. u. Genussmtl., 18 (1909), No. 12, pp. 733-737).— 

 Comfiarative tests were made between the Rideal and Hehner methods, the 

 latter being considered the more satisfactory. 



