Intelligence and Miscellaneous Articles. 479 



vibrations being neglected, we shall obtain, if not the exact value 



C 

 of the ratio — , at least a value more closely approximate than -§-. 

 c 



That is the idea which has led me to undertake this research. 



To calculate the vis viva due to the rotations of the molecules of 

 a gas, recourse must necessarily be had to certain hypotheses. 

 Without introducing any new ones ; I adopt those which appear to 

 me the most plausible and are most generally accepted. I call 

 perfect gases those which follow the laws of Mariotte and Gay- 

 Lussac, and in which the internal work is nil, so that their mole- 

 cules have velocities of translation of which the squares are propor- 

 tional to the absolute temperature. I call simple and tetratomic the 

 gases whose molecules are composed of four smaller molecules 

 identical with each other — not necessarily atoms, but which may 

 be regarded as atoms in the present state of science : such are (or 

 appear to be) hydrogen, oxygen, nitrogen, &c. I imagine that, in 

 such a gas, the four atoms occupy the apices of a regular tetrahedron, 

 the edge of which is longer than the diameter of each of them, and 

 that the interior of the tetrahedron is filled with aether free or 

 condensed. JNow, taking account of the rotation of each elementary 

 tetrahedron about its centre of gravity, and regarding the vibra- 

 tions of the atoms as none or insensible, I have found the ratio of 

 the two specific heats to be exactly -J or 1*4, while experiment 

 has given, for the gases above-mentioned, values between 1*39 

 and 1-42. 



From this accordance between calculation and experiment it 

 seems to follow that the internal vibrations, neglected in the cal- 

 culation, are negligible in reality. I am thus led to think that, in 

 simple gases, the physical molecules remain sensibly invariable in 

 form and dimensions as long as no electrical or chemical action 

 is produced. — Comptes Rendus de VAcademie des Sciences, tome 

 lxxxiii. pp. 726-728. 



A NEW PROCESS FOR THE SEPARATION OF GALLIUM. 

 BY LECOQ DE BOISBAUDRAN. 



I have recently simplified and much shortened the preparation 

 of gallium by operating as follows : — 



1. The ore is dissolved, according to its nature, in aqua regia, 

 chlorhydric acid, or sulphuric acid. The liquid is treated, cold, 

 with plates of zinc, and then filtered while the liberation of hydro- 

 gen is still pretty considerable ; the liquid is then heated with a 

 large excess of zinc. The gelatinous precipitate is washed, and 

 taken up again by chlorhydric acid. The new solution is heated 

 with an excess of zinc, and a second gelatinous precipitate obtained. 

 To this point the course is identical with that of the former pro- 

 cess {Comptes Rendus, May 8, 1876, p. 1098). 



2. Into the chlorhydric solution of the second precipitate formed 



