1881.] On the Absorption of Gases by Solids. 407 



It would appear from this table that the phases of the various 

 magnetic inequalities occur at Toronto nearly two days (more strictly 

 1*6 days) before the advent of the corresponding phases at Kew. 



The result already obtained for Kew and Prague shows that the 

 phases of magnetic inequalities occur at Kew nearly one day (more 

 strictly 0*7 day) before the advent of the corresponding phases at 

 Prague. Thus the two results agree together in representing a pro- 

 gress of magnetic weather from west to east, and agree also with a 

 result obtained by Balfour Stewart and Morisabro Hiraoka (see " Proc. 

 Roy. Soc," vol. 28, p. 288), showing that magnetic weather changes 

 occur at Trevandrum, in India, 9*7 days later than at Kew. 



It ought, however, to be borne in mind that in the intercomparison 

 of Toronto, Kew, and Prague, the observations include disturbances, 

 while in the intercomparison of Kew and Trevandrum the undisturbed 

 observations at Kew are compared with the whole body of observations 

 at Trevandrum, this latter being a tropical station in which the effect 

 of disturbance is extremely small. 



XIV. "On the Absorption of Gases by Solids." By J. B. 

 Hannay, F.R.S.E., F.C.S. Communicated by Professor G. 

 G. Stokes, D.C.L., &c, Sec. U.S. Received June 4, 1881. 



During the progress of the investigations which I have from time to 

 time had the honour of bringing under the notice of the Royal Society, 

 I have again and again noticed the apparent disappearance of gases 

 inclosed in vessels of various materials when the disappearance could 

 not be accounted for upon the assumption of ordinary leakage. After 

 a careful examination of the subject I found that the solids absorbed 

 or dissolved the gases, giving rise to a striking example of the fixation 

 of a gas in a solid without chemical action. 



In carrying out that most troublesome investigation, the crystalline 

 separation of carbon from its compounds, the tubes used for experi- 

 ment have been in nine cases out of ten found to be empty on opening 

 them, and in most cases a careful testing by hydraulic press showed 

 no leakage. The gases seemed to go through the solid iron, although 

 it was 2 inches thick. A series of experiments with various linings 

 were tried. The tube was electro-plated with copper, silver, and gold, 

 but with no greater success. Siliceous linings were tried — fusible 

 enamels and glass — but still the tubes refused to hold the contents. 

 Out of thirty-four experiments made since my last results were pub- 

 lished, only four contained any liquid or condensed gaseous matter 

 after the furnacing. I became convinced that the solid matter at the 

 very high pressure and temperature used must be pervious to gases. 



