CHEMISTRY. 417 



with hydrochloric acid, platinum chlorid added, and the solution evap- 

 orated on a water bath to a sirupy consistency. After completely 

 cooling, the residue is covered with a mixture of 9 parts of 95 per cent 

 alcohol and 1 part of ether and allowed to stand several hours. It 

 is then brought upon a filter and washed with the same mixture. The 

 precipitate of potassium-platinum chlorid is dissolved on the filter with 

 hot water, and the platinum salt is reduced by pouring this hot solu- 

 tion into a boiling solution of sodium formate. Heating the solution 

 for a short time causes the metallic platinum to flocculate so that it 

 is easily collected on the filter and washed. The metal is brought upon 

 the filter by means of cold, slightly acidified water, and finally washed 

 with boiling water. It is then dried, ignited, and weighed. Ninety 

 per cent alcohol has been substituted for the mixture of alcohol and 

 ether in the author's laboratory vrith very satisfactory results. 



On the use of sodium superoxid for separating iron oxid and 

 alumina, C. Glaser (Ghem. Ztg., 21 {1897), No. 69, p. 678). — If the 

 solution does not contain phosphate of lime it is simply necessary to 

 nearly neutralize with ammonia before employing the method of sepa- 

 ration of iron and alumina. If phosphate of lime is present the greater 

 portion of it should be removed by means of ammonium acetate or by 

 Glaser's method, the lime-free solution then being nearly neutralized. 

 To the cool solution should be added sufficient dry sodium superoxid to 

 produce a clear solution, from 3 to 6 gm. of superoxid to each 0.4 gm. of 

 substance being usually sufficient. The solution should be heated for 

 a short time at boiling temperature, allowed to se.ttle, and filtered 

 through a thick filter. The alumina in the filtrate is determined after 

 acidifying in the usual way. The author recommends precipitating 

 twice as aluminum phosphate by adding phosphoric acid and ammonium 

 acetate. The iron on the filtrate is dissolved in hydrochloric acid and 

 precipitated again in order to purify it. 



The results obtained by this method on a number of phosphates 

 agreed quite closely with those obtained by the fusion method. 



On the determination of iron oxid and alumina in phosphates, 

 N. Blattner and J. Brasseur (Bui. Soc. Chim. Paris, 3. ser.,17 (1897), 

 No. 15, pp. 760, 761). — The author compared the acetic acid, Glaser, 

 Lasne, von Griiber, Gladding, and Thomson methods, with the fol- 

 lowing results: The first gave results entirely too low, the acetic acid 

 retaining a considerable amount of aluminum in solution. The Glaser 

 method was sufficiently accurate when manganese was absent and was 

 rapid and easy. The method of Lasne 1 gave scientifically exact results 

 when all the details of the method, as described by the author, were 

 carefully observed. Von Griiber's 2 method is considered simply an 

 abridged or, more properly, a mutilated form of Lasne's method, and 

 gave inaccurate results. Gladding's 3 method, using caustic potash, is 



1 Bui. Soc. Chim. Paris, 15 (1896), p. 118 (E. S. E., 7, p. 915). 

 2 Zts< hr. angew. Chern., 1896, p. 741 (E. S. R., 8, p. 559). 

 3 Jour. Amer. Cheui. Soc, 18 (1896), p. 721. 



