722 The Dissociation of Water Vapour. 



observed by Nernst and Wartenburg, for an alteration of 0*01 

 in the value of K makes 2*3 per cent, change in x. 



The author has employed the same formula in calculating 

 his values for K, and obtained the mean value for all his 

 experiments of 3*791, which is identical with that of 

 Langmuir. 



The values in Table II. are, however, probably more correct 

 than those in Table I., since they cover a temperature range 

 of 400°. 



The two other formula? proposed by Nernst and Warten- 

 burg were published three years ago {Zeit. Phijsik. Chem. 

 1906, li. p. 534). 



The first of these, 



1 ^ -H OO 25030 , O lAl T 



g h * Yi ~M 2 ' =1 38 — r- +2 ' m °zim 



"Tooj 



-0'00016(T-1000), 



WAX- 



gave results which, when compared with those obtained by 

 their experiments, were somewhat too high at high tempera- 

 ture, but were in good agreement at 1370°. 

 The second formula, 



, 2x* 11y| . 25030 ._. T 



l0g / 9 , * V.jlY — *" +2381og iooo 



\ w + iooA 100/ 



-1-38 x 10- 4 (T-1000) -0-685 x 10- 7 (T 2 -1000 2 ), 



gave results which were too high at about 1370°, but became 

 more and more in agreement with the experimentally deter- 

 mined values as the temperature increased. If in both these 

 formulae x compared to 1 is neglected, and then the whole 

 formula is divided by 3, the constant K has the value 3*79 

 or 3*82; and as one of these formulae gives too high results 

 where the other gives too low, and vice versa, it is evident 

 that if the experimental values of Nernst and Wartenburg 

 are correct, the constant must lie between them. Their mean 

 3' 805 will be seen to be practically the same as that from 

 the experiments given in Table II. 



The values obtained by Lowenstein (Zeit. Physik. Chem. 

 1906, liv. p. 715) for the decomposition of water vapour at 

 very high temperatures (up to 1968° abs.) are in agreement 

 with the values calculated for K = 3*80. 



As these experiments were performed in a totally different 

 manner from those of Nernst and Wartenburg, or Langmuir, 



