CHEMISTRY. 711 



Occlusion of magnesium oxalate by calcium oxalate. Solubility of cal- 

 cium oxalate, T. W. Hicmakdjs, C. F. McCakfkey, and 11. I5ishek {Ztschr. Anorgan. 

 Chem., 28 {1901), ]>p, 71-89; abs. in Jour. Oiem. Soc. [London], SO {1901), No. 468, 

 II, p. 6^4; Jour. Aiiu't: Chem. Soc, 23 {1901), No. 12, Rev. Chem., p. ;i?Oi).— Investi- 

 gations are reported which showed that the occlusion of magnesium oxalate was 

 dependent upon the distribution of undissociated magnesium oxalate between the 

 sokition and the soUd sul)stance, and that as tlie proportion of undissociated mag- 

 nesium oxalate in solution was diminished the amount found in the precipitated 

 calcium oxalate was reduced. 



"The authors recommend tiie following process for the separation of calcium from 

 magnesium, in solutions which should not be more than 150 normal with respect to 

 magnesium: To the solution, an amount of ammonium chlorid equivalent to 10 

 times the amount of magnesium present is added, and sufficient oxalic acid, to 

 which some hydrochloric acid has previously been added, to completely precipitate 

 the calcium. The solution is then boiled, and very dilute ammonia is added in 

 small ((uantities at a time until the whole is exactly neutral to methyl-orange. A 

 large excess of ammonium oxalate is now added, and the mixture, after standing for 



4 hours, is filtered and washed with very dilute ammonium oxalate. 



"The solubility of calcium oxalate in pure water (0.0068 gm. per liter at 25°, 

 0.00955 gm. at 50°, and 0.014 gm. at 95°) is so large as to cause an appreciable 

 error in exact analysis; in ammonium oxalate solution it is very considerably less 

 soluble." 



On the determination of perchlorate in nitrate of soda, Grimm ( Chem. Ind., 

 24 {1901), p. 476; ahs. in Chem. Ztrj., 25 {1901), No. 76, p. ;.>'7.5).— In the method 

 proposed 20 gm. of the nitrate is strongly heated in a platinum, nickel, or iron dish 

 on an asbestos plate, and covered with a large platinum dish, for 1 hour, with 3 gm. 

 of manganese dioxid, the temperature being gradually raised. The difference 

 between the amount of chlorin in the material l)efore and after treatment is taken 

 to represent perchlorate. 



Detection of potassium by means of sodium picrate, C. Reichard {Ztschr. 

 AnaUiL Chem., 40 {1901), pp. 877-384). — If an excess of a saturated (10 percent) solu- 

 tion of sodium picrate be added to a solution of potassium chlorid (not weaker than 

 1 per cent) a precipitate of acicular crystals is formed. Other potassium salts (except 

 carbonate and cyanid) must be of at least a 2 per cent strength. The reaction is not 

 interfered with by sodium salts (except the carbonate), but the solution must be 

 free from ammonium salts and free acids. 



Agronometry, A. Lonay {Bui. Assoc. Beige Chim., 15 {1901), No. 4-5, pp. 186- 

 196; tihs. in Bnl. Soc. Chim. Paris, 3. ser., 25 {1901), No. 22, p. 991).—T\\\ii is an 

 account mainly of methods followed by the author in the analysis of soils. 



The determination of biological arsenic, B. (TALLi-VALERioand C. Strzvzowski 

 {Pharm. Post, 33 {1900), pp. 637-639, 649-651; abs. in Ztschr. Untermch. Nahr. u. 

 (lenussmtl., 4 {1901), No. 12, pp. 54'), 546). — Methods of determining minute portions 

 of arsenic contained in organic matter. 



A modification of the sulphuric acid test for formaldehyde in milk, A. G. 

 LrEHEHT {Jonr. Anirr. Chem. Soc, 23 {1901), No. 9, pj). 682, 683). — In estimating 

 nitrogen in milk by the Kjeldahl method the author noted that in the presence of a 

 small quantity of formaldehyde there was a peculiar violet coloration of the potas- 

 sium sulphate cry.yfals and the sulphuric acid surrounding them. This led to trials 

 from which the following method was devised for detecting formaldehyde in milk: 



5 gm. of coarsely jKjwdered potassium sulphate is placed in a 100 cc. flask, 5 cc. of 

 the suspected milk distributed over it with a pipette, and 10 cc. of sulphuric acid 

 (specific gravity 1.84) carefully jMJured down the side of the flask, and the whole 

 allowe<l to stand. If formaldehyde is present, the violet coloration of the potassium 



21164— No. 8— 02 2 



