430 RECORD OF SCIENCE FOR 1887 AND 1888. 



Paterno and Nasini with aqueous solutious flud that citraconic, mes- 

 aconic, and itacouic acids have the same coefficieuts of depression, and 

 are therefore strictly isomeric. A like statement holds of malic and 

 fumaric acids. (Berichte, xxi, 2156.) 



Louise and Roux, in order to determine the constitution of aluminum 

 compounds, investigated aluminum ethyl, propyl, and isoamyl by 

 Raoult's method, with ethylene bromide as the solvent. The results 

 agree with the formulte AI2X6, in which X stands for the alcohoiradicle. 

 Compare the diverse results obtainedby other chemists with the vapor- 

 density method. (Compt. Rend., cvii,j£00.) 



The depression proluced in the freezing point of benzene by various 

 phenolic substances was studied by Paterno. Phenol bebaves abnor- 

 mally, but ethylphenol, acetylphenol, two nitrophenols, tribromphenol, 

 picric acid, paracresol, methyl salicylate, thymol, nitro thymol, nitro- 

 sotbymol, naphthol, and benzylphenol all gave normal results, both iu 

 benzene and in acetic acid. (Berichte, xxi, 3178.) 



Density of gases and vapors. — In discussing the atomic weight of oxy- 

 gen attention was called to Lord Rayleigh's observation that the volume 

 of a glass globe when full differs from the volume of the same globe 

 when empty. This observation affects all of Regnault's determinations 

 of the weight of gases, and the results of the latter have been cor- 

 rected by Crafts, who had access to Regnault's original apparatus, ex- 

 cepting the glass globe actually used, which bad been destroyed. An- 

 other globe of the same lot, however, was available, and presumably 

 identical for all practical purposes with it. The following are the spe- 

 cific gravities as given by Reguault and as corrected by Crafts, air being 

 taken as unity. 



Regnault. Crafts. 



N 97137 .97138 



H 069-27 .06949 



1.105(34 1.10562 



CO, 1.52910 1.52897 



The actual correction was 3 milligrammes added to the weight of 

 each gas as given by Reguault at 760 millimeters pressure. (Compt. 

 Rend., cvi, 1662.) 



J. Mensching and Victor Meyer find that at high temperatures the 

 vapors of phosphorus and arsenic are not iu agreement with the sup- 

 posed molecules P4 and AS4, but that at a white heat they approximate 

 to P3 and As2. For antimony the molecule Sbi does not exist, the 

 probable molecule being either 81)2 or Sbi. (Berichte, xx, 1833.) 



Biltz, redetermining the vapor density of sulphur, under various con- 

 ditions, obtains values ranging from 7.937 down to 4.734. He concludes, 

 hence, that the commonly accepted view that free sulphur in vapor is 

 Sg is not correct, and that the true formula is S^. (Berichte, xxi, 2013.) 



With reference to the chlorides of inui, (chromium, indium, gallium, 

 and aluminum, various determinations liave been made, with varying 

 results. First, us to aluminum, whose chloride has hitherto been writ- 



