The Dissociation of Water Vapour. 715 



found desirable to make an experiment on an iron salt at 

 great dilution. Solutions of iron alum, Fe 2 .3S04.K 2 S0 4 . 

 24 H 2 0, having strengths in the salt of 1 in 100,000 and 1 in 

 1,000,000, were made up, and the absorption ratios were 

 respectively *679 and *924. The latter solution is l-20th the 

 strength in salt of the weakest of the four principal con- 

 stituents, and thus it is evident that a number of such 

 impurities are quite likely to be able to bridge over the gap 

 between the natural and made-up tap-water. 



5. From the above it might be concluded that the absorp- 

 tion of ultra-violet light might be a delicate enough test to 

 distinguish between different kinds of distilled water. But I 

 have been unable to detect any difference. The specimens of 

 distilled water available in the laboratory gave identical 

 results, within the limits of experimental error, with two 

 specimens of greater purity. One of these was of conduc- 

 tivity '8 x 10~ 6 , and was also sterilized. 



It need hardly be said that no mathematical formula can 

 be given for the form of the curves in rig. 1. Apart from 

 the fact that the non-homogeneity of the source of light 

 prevents the simple exponential law of intensity from 

 holding, it has been shown by Griffith ** that in any case the 

 intensity of the transmitted light is not proportional to the 

 electrometer deflexion. 



During the experiments I have been much helped by the 

 advice of Professor Dreyer, and he has been kind enough to 

 sterilize some of the solutions. Mr. D. H. Nagel has very 

 kindly furnished me with some very pure distilled water. 

 Lastly, I have to offer my thanks to Professor Townsend, in 

 whose laboratory the work was performed, for his encourage- 

 ment and suggestions during its progress. 



LXVI. The Dissociation of Water Vapour. 

 By Alfred Holt, Junr.| 



TWO years ago the author published in this Journal a 

 short account of some experiments on the dissociation 

 of water vapour by an incandescent platinum wire (Phil. 

 Mag. 1907, p. 630). The percentage decomposition at a 

 temperature of about 1000° (abs.) { was in close agreement 

 with the values previously published by Nernst and Warten- 

 burg (Gottingen Nachrichten, 1905), and Langmuir (Journ. 



* I. O. Griffith, Phil. Mag., Aug. 1907. 

 t Communicated by the Author. 



| All temperatures mentioned in this paper are absolute, unless stated 

 to the contrary. 



