300 Scientific Intelligence. 



For most of the other substances the ratios range between 2*2 

 and 3-0, thus: 



Br, 



2-11 



2-44 



2-22 



2-25 



HO 



2-34 



SO 



SnCL 



CHCI 



C.H,.. 



-..-- 2-47 



2-62 



2-75 



The value for carbon dioxide is only 1-41, while a few bodies 

 give ratios higher than three, ether 3-06, OS, 3*36, PCI3 3-47, 

 alcohol 3-59. 



The author offers no explanation of these ratios, but believes 

 that they are not meaningless. — Am. Chem. Journal^ xviii, 618. 



H. L. w. 



2. The Quantitative Separation of Chlorine^ Bromine and 

 Iodine. — Bennett and Placeway have applied the qualitative 

 method, described by Prof. Edward Hart a number of years ago, 

 to the quantitative separation of the halogens with satisfactory 

 results. The method recommended is briefly as follows : To a 

 solution of the alkali salts in a distilling-flask, a solution of 10^ 

 of ammonium ferric alum is added with water enough to make a 

 total volume of about 200^*^, and the mixture is distilled for about 

 25 minutes. The iodine is absorbed in a receiver with a solu- 

 tion of potassium iodide and titrated with sodium thiosulphate. 

 To determine bromine, 35*^° of a saturated solution of potassium 

 permanganate are added to the residue in the distilling-flask, 

 water is added to make the volume 200°*^ as at first, and a distil- 

 lation is made exactly as in the previous case. Chlorine can be 

 determined in an aliquot part of the final residue in the distilling- 

 flask either by a volumetric or a gravimetric method. — Jour. Am. 

 Chem. Soc, xviii, 688. h. l. w. 



3. The homogeneity of Argon and of Helium. — Ramsay and 

 Collie have studied this interesting question by submitting the 

 two gases to systematic diffusion, dividing each gas into two 

 parts by a method analogous to fractional distillation. The gases 

 were caused to traverse a duct of porous pipe-clay, submitted on 

 one of its surfaces to the action of a vacuum. The authors 

 satisfied themselves that the separation of hydrogen and helium 

 and that of oxygen and carbonic acid could be effected in this 

 way, and that, by measuring the descent of a column of mercury 

 connected with the apparatus, it is possible to arrive at a good 

 determination of the molecular weight of various gases. 



When argon was submitted to this fractionation, the two 

 extreme portions had densities of 19-93 and 20-01. The separa- 

 tion, if any, in this case was very slight. 



With helium the results were different. A more diffusible part 

 with a density of 1*874 was obtained, while, at the other end, the 

 gas had a density of 2*133. A great number of further frac- 

 tionations did not change these densities. The two fractions, 

 however, showed absolutely identical spectra. Considering what 

 happens when a mixture of two gases is diffused^ as, for example, 



