THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FIFTH SERIES.] 



DECEMBER 1877. 



LY. On the Thermal Conductivity and Diathermancy of Air 

 and Hydrogen. By Dr. Henry Buff, Professor of Physics 

 in the University of Giessen *. 



MORE than fourteen years ago Magnus published expe- 

 rimental researches t, the results of which led him to 

 conclude that hydrogen possesses a thermal conductivity re- 

 sembling that of the metals. This conclusion, though not 

 entirely unobjected to, has since passed into most physical 

 textbooks. 



Doubts as to the correctness of this assumption have of 

 course no reference to the capacity possessed by hydrogen, as 

 well as other gases, of conducting heat in a small degree inde- 

 pendent of accompanying currents. These doubts also are not 

 removed by the observations of Kundt and Warburg J, which 

 prove the thermal conductivity of hydrogen to be, indeed, con- 

 siderably greater than that of other gases ; for this is quite pos- 

 sible without assuming that the conducting-power of hydro- 

 gen approximates as nearly to that of the metals as Magnus 

 thought necessary to assume, in order to explain such pheno- 

 mena as the incandescence of platinum wire, first observed by 

 Grove. True, Magnus did not think himself justified in sup- 

 posing that in this experiment a greater decrease of the tem- 

 perature of the platinum wire in the hydrogen was caused by 

 a more rapid current of the gas. '^ It is not evident," he said, 

 " why currents, produced by differences of temperature, should 



* Communicated bv the Author. 



t Pogg. Ann. Bd. cxii. S. 497. [Phil. Mag. July 1861, p. 1.1 



X Ibid. Bd. civ. S. 337. [Phil, Mag. July 1875, p. 53.] 



Phil. Mag. S. 5. Vol. 4. No. 27. Dec. 1877. 2 D 



