The Occlusion of Gases. 455 



open end being placed in connection with SprengePs Mercurial 

 Exhauster. The metallic tubes were placed within porcelain 

 tubes ; the gas under examination was allowed to circulate 

 through the annular space between the two, and the arrange- 

 ment admitted the subsequent application of heat. 



With a platinum tube, and air circulating outside it, the 

 vacuum remained undisturbed, even when the temperature rose 

 to a bright red heat : but when dry hydrogen was made to 

 pass through the annular space, the platinum allowed the 

 hydrogen to pass through into the vacuum, as soon as, but not 

 until, the metal was raised to a red heat. In seven minutes the 

 Sprengel tube delivered 15*47 cubic centimetres of gas, of 

 which 15"27 cubic centimetres were pure hydrogen. The pla- 

 tinum tube employed was 1*1 millimetre in thickness, with an 

 internal diameter of 12 millimetres. 



The surface actually heated was 200 millimetres (8 inches). 

 The rate of passage of the hydrogen was therefore 489 # 2 cubic 

 centimetres, through a square metre of platinum 1*1 millimetres 

 thick in one minute. The passage of other gases through the 

 same tube was next examined, the experiments being conducted 

 in exactly the same way, the metal being heated to bright 

 redness. The most interesting fact developed itself, that while 

 hydrogen could penetrate at the above rate, the following 

 gases were incapable of passing, even to the extent of 0*2 

 cubic centimetres in one hour : — Oxygen, nitrogen, chlorine, 

 hydrochloric acid, steam, carbonic acid, carbonic oxide, marsh 

 gas, olefiant gas, hydrosulphuric acid, and ammonia. 



It was, however, with a tube of palladium that the most 

 remarkable results were obtained, that metal permitting the 

 permeation of hydrogen with far greater facility than platinum, 

 and at a temperature short of redness. The closed palladium 

 tube remained perfectly tight when connected with the Mer- 

 curial Exhauster, with air (or carbonic acid) outside the tube, 

 both at the ordinary temperature, and at a temperature near 

 low redness. 



When dry hydrogen was allowed to circulate in the annular 

 space, none passed through in three hours at a temperature of 

 100° C, but when the temperature was raised to 240° C, the 

 hydrogen began to come through at a gradually increasing 

 rate, until 265° C. was reached, when the exhauster delivered 

 1T2 cubic centimetres in five minutes, or 423 cubic centimetres 

 for a square metre of palladium one millimetre thick. 



It is interesting to compare the passage of hydrogen 

 through the palladium with the penetration of the same gas 

 into vacuum through a septum of india-rubber. The palladium, 

 one millimetre in thickness, was seventy times the thickness 

 of the rubber sheet. 



