( % 111 istry <inil Phyeii 8 1 3 



solid, is id trod ucec! Into the bulb, previously filled with hydrogen, 

 in the form of Bmall cast Bticks. li liquid, it, is placed in small 

 tubes of Wood's fusible metal when the temperature to be em- 

 ployed is above 60° lo 80°, the fusing point of the metal, <>r in 

 tubes of glass, stoppered, when the temperature is below tins; 

 the stopper being shaken out by tapping with the finger. The 

 bulb should have a capacity of. about 1(H)', and a diameter of 

 three centimeters; and the bottom which must not be too thin, is 

 somewhat flattened to facilitate the volatilization of the substance. 

 The neck is to have a diameter of not over 4 or 5 millimeters ; and 

 the quantity of the substance taken should be such as to produce 

 an expulsion of to ll co of gas from the vessel. It is not desira- 

 ble with this method of working to cover the bottom with sand 

 or asbestos, since these Bub Stance 8 retard the production of vapor; 

 but the bulb may be protected by a small spiral of platinum wire, 

 or some mercury may be placed in it. Results are given, in the 

 paper, of the determination of the vapor-density of xylene at 100°, 

 (40° below its boiling point), of nitrobenzene 23° and 30° below, 

 of naphthalene 35° below and of para-nitrotoluene 33° below the 

 boiling point. Also of ethyl ether at the ordinary tempera- 

 ture of the laboratory, 17°, its boiling point being 35°. These re- 

 sults bdow the method to give satisfactory results. — Ber. Berl. 

 Chem. Ges., xxiii, 311, Feb., 1890. G. f. b. 



2. On the Vapor-density of Aluminum Chloride. — In conse- 

 quence of the results obtained by Friedel and Crafts,* Nilson 

 and Petteusson have repeated and extended their experiments 

 on the vapor-density of aluminum chloride. In meeting the criti- 

 cism of the French chemists upon their method, they observe that 

 for every volatile substance, there exists a certain range of tem- 

 perature within which the vapor-density is a function not only of 

 the temperature but also of the pressure. Within this interval, 

 which they call a region of gas-dissociation, they concede that the 

 method of Dumas gives the true vapor-density, that of Meyer (or 

 of Dulong, as they call it) giving values which are always too 

 low. But as soon as the temperature is reached at which the den- 

 sity of the vapor becomes constant, the same results are obtained 

 whichever of the two methods be. employed. Since the law of 

 Avogadro is true only for bodies in the condition of true gases, 

 and since in the state of a perfect gas the density is independent 

 of the temperature and the pressure, and the two methods then 

 give identical results, it would seem that Friedel and Crafts were 

 hardly justified in giving the preference to one of these methods, 

 rather than to the other. Accordingly the authors undertook a 

 re-determination of the vapor-density of aluminum chloride, em- 

 ploying both methods for the purpose. For the Meyer method 

 they employed a protecting tube of Bayeux porcelain placed in a 

 Perrot furnace for the higher and a tube of Thuringian glass for 

 the lower temperatures. The density obtained, referred to air as 



*This Journal, xxxvi, 465, Dec, 1888. 



