±<o6 Dr. T. Carnelley on the 



weight of the gas. This holds good in the case of H 2 , CO, 

 N 2 , C0 2 , N 2 0, and S0 2 . But, for reasons explained in the 

 original paper, HBr and SiF 4 are exceptions to the rule. 



Conducting-power of Gases for Heat. — Magnus and, more 

 recently, Stefan (Chan. Centr. 1875, p. 529) have proved that 

 gases differ in their conducting-power for heat ; whilst Clausius 

 (Pogg. Ann. cxv. p. 1) and Mohr (Deut. chem. Ges. Ber. iv. 

 p. 85) have shown from mechanical considerations that these 

 conducting-powers are proportional to the velocity of the 

 molecules of the gas, and, further, that under equal pressure, 

 this velocity is inversely proportional to the square root of the 

 densities. 



Absorbing-power for Heat. — Aymonnet {Compt. Rend. 

 lxxxiii. p. 971) concludes from his researches that the coeffi- 

 cient of atomic absorption for heat is constant (1) for all sub- 

 stances dissolved in the same liquid, (2) for all substances 

 of similar chemical constitution. The coefficient of atomic 



ATT 

 absorption = ^r^, where A= coefficient of absorption, E = 



atomic weight, D= density, N= number of atoms. On com- 

 paring his conclusions, however, with his experimental num- 

 bers, they do not appear to be justified thereby to the extent 

 one might have hoped ; the above statements apply almost 

 equally well to the coefficient of absorption itself as to the co- 

 efficient of atomic absorption. If one may judge from the 

 few liquids submitted to experiment, the coefficient of absorp- 

 tion of compounds of analogous composition appears to be 

 proportional to the boiling-point of the liquid reckoned from 

 the absolute zero ; thus : — 



Boiling-point Coefficient of 



reckoned from absorption 



-273° 0. , . a 



W- o (C)> a' 



Methyl alcohol 839 '6879 -00202 



Ethyl „ 351 '6992 -00199 



Amyl „ 405 -7752 -00191 



Benzene 354 -4866 -00137 



Toluene 384 -4933 -00128 



Turpentine 438 '5805 -00133 



Latent Heat of Liquids. — In an important paper {Compt. 

 Bend, lxxxii. p. 260) Pictet has shown : — (1) That the latent 

 heat of all liquids brought to one and the same pressure, mul- 

 tiplied by the molecular weight, give a constant product. 

 (2) For all liquids the difference of the internal latent heats at 

 any two temperatures multiplied by the atomic weight is a 



