log p = 5*5():-)S — 



1904.] oa Liquid Hf/drof/e/i Calorimetry. 595 



of the meriii specitic heat in (jnestioii being in reality about 0*4 

 instead of (>. This correetion removes hydrogen from the list of 

 substances which follow Dnlong and Petit's law, its atomic heat 

 being only about half the required amount. In an early communi- 

 cation to the Royal Society of Edinburgh,* I showed that the specific 

 heat of hydrogen absorbed by palladium was about ;)-5. It seems 

 therefore that liydrogen in the gaseous, the occluded, and the liquid 

 condition, has substantially the same specific heat. 



It is of interest to see what support can be got for the value of 

 the latent heat of hydrogen being about 121 or 122. From some 

 early observations of mine with the helium thermometer on the 

 vapour pressure of hydrogen below the l)oiling point, a Rankine 

 formula was obtained, 



58 -l;] 

 T ' 



which, combined with Clapeyron's equation, where account was taken 

 of the difference of the specific volumes of hydrogen in the liquid 

 and the gaseous states at the boiling point, gave 120-:) as the latent 

 heat of liquid hydrogen. Similar treatment of Travel's' t smoothed 

 results gave 110. "Two Willard Gibbs' formulae calculated from 

 results evenly selected from these actual observations gave 12:5 '4 and 

 117*5, or a mean of 120*5. 



For nitrogen, the value of the latent heat was found to be about 



50 * 4 gramme-calories at the boiling point. Other oliservers (Fischer 



and Alt, Alt, Shearer) give the values of this quantity as 4S-9, 48*7 



(at 71.S mm. pressure), 52*07 (at OB mm. pressure), and 41)*<s. The 



application of a Rankine formula deduced from my own observations, 



•>92 

 log p = 6* 6402 - ^, 



gave 4.S • (j:> ; while the same process applied to observations of 

 Fischer and Alt gave 49 * 65, and a Willard Gibbs' formula gave 51 '4. 



In the case of oxygen its latent heat was found to be 51*15 

 gramme-calories. Several experimenters have found the value of this 

 quantity to be as high as 5<S • 0, 60 * 9, 61 * 0. A careful direct deter- 

 mination of Alt's gives 52*07 (725 mm.) and 58*85 (68 mm.). By 

 means of Rankine and Willard Gibbs' formulae, calculated from the 

 observations of Olzewski, Estreicher, and Travers, I have found the 

 values 51*4, 52 * 5:), 5;> * 7s. 



These indirect methods of determining the latent heat depend on 

 formulae of which only the principal terms are retained. We cannot, 

 therefore, expect more than approximate values from them, so that it 

 is sufficient at present to show that the direct experimental values 

 found in the above series of observations substantially agree with 

 those obtained by indirect and approximate calculations. 



* "The Physical Constants of Hydrogenium," Trans. Eoy. Soc. Ed. 1873. 

 t Phil. Trans., 1902, A., vol. cc. p. 169. 



