214 Scientific Intelligence. 



carbonate, neutralized this with pure hydrobromic acid, and have 

 then found the amounts of silver and silver bromide corresponding 

 to the sodium carbonate. It was found that 29*43501 g. of 

 sodium carbonate were equivalent to 59*91676 g. of silver and 

 also to a weight of hydrobromic acid solution which could pre- 

 cipitate 104*3023 g. of silver bromide. From these results, if 

 silver is 107*88, sodium carbonate becomes 105*995 and carbon 

 12*005. This work, by connecting various ratios in a new way, 

 shows the consistency of a great variety of earlier work done by 

 and under the direction of Professor Richards. — Jour. Amer. 

 Chem. iSoc, xxxvii, 95. h. l. w. 



2. The Molecular Weight of Sodium Sulphate and the Atomic 

 Weight of Sulphur, — Having worked out a method for the pre- 

 paration of pure sodium carbonate in connection with the investi- 

 gation referred to in the previous notice, Richakds and Hooveb 

 have extended the work by neutralizing sodium carbonate with 

 sulphuric acid and comparing the weights of the two salts. The 

 ratio of the carbonate to the sulphate was found to be 1*00000 : 

 1*340155. Assuming Na 2 C0 3 =: 105*995, as found in the preceding 

 research (Ag = 107*880) the molecular weight of sodium sulphate 

 becomes 142*050, and the atomic weight of sulphur becomes 

 32*060. This is very close to the value 32*069 previously obtained 

 by Richards, which is practically the value (32*07) of the Inter- 

 national Table. The authors say that the mean value 32 065 may 

 perhaps be taken as the most trustworthy value for the atomic 

 weight of sulphur thus far recorded. — Jour. Amer. Chem. Soc, 

 xxvii, 108. h. l. w. 



3. The Separation of Tungsten and Molybdenum. — The 

 analytical separation of these comparatively rare metals which 

 frequently occur together has become more important of late on 

 account of their extensive technical application. E. E. Maebakee 

 has studied this somewhat difficult problem and has worked out 

 an apparently satisfactory method for the determination of the 

 two elements, chiefly by the combination of previously known 

 processes. He prefers to start with sodium tungstate and 

 molybdate rather than with the ammonium salts. The tungsten 

 is precipitated, according to Mdivani's method, with a carefully 

 regulated quantity of stannous chloride in the presence of hydro- 

 chloric acid in the form of the blue oxide W 2 5 . This is filtered 

 off, washed with very dilute hydrochloric acid, ignited and 

 weighed as W0 3 . The filtrate is concentrated, treated with 

 granulated zinc, whereby the tin is removed as metal. Then the 

 filtrate from this precipitation is passed through a Jones 

 "reductor" into a solution of ferric sulphate and titrated in the 

 presence of phosphoric acid, manganese sulphate, and sulphuric 

 acid by means of potassium permanganate solution, according to 

 the method for the determination of molybdenum devised by 

 Randall and published in this Journal in 1907. The test analyses 

 show remarkably accurate results. — Jour. Amer. Chem. Soc, 

 xxvii, 86. h. l. w. 



