430 ME. T. GRAHAM ON THE ABSORPTION AND 



degree which will admit liquid but not gaseous molecules. Now the numerous liquid 

 compounds of carbon and hydrogen have all a nearly similar density, generally a little 

 under that of water. There is no reason to suppose that the density of liquid hydrogen 

 would differ greatly from the hydrocarbon class ; but then the surprising lightness of 

 hydrogen gas must cause liquid hydrogen to yield a volume of vapour disproportionately 

 large when compared with the former class of substances, or, indeed, with any other 

 substance whatever. The absorption of hydrogen by palladium will appear, then, less 

 extravagantly great when viewed as the absorption of a highly volatile liquid capable of 

 yielding an exceedingly light vapour, rather than that of a gas. 



An excellent opportunity of observing the penetration by hydrogen of a compact plate 

 of palladium, 1 millim. in thickness, was afforded by a tube of that metal constructed 

 by Mr. Matthey. This tube was said to have been welded from palladium near the point 

 of fusion of the metal. The length of the tube was 115 millims., its internal diameter 

 12 millims., thickness 1 millim., and external surface 0-0053 of a square metre. It was 

 closed by thick plates of platinum soldered at both ends, one of the plates being perfo- 

 rated by a long small tube of platinum, by which the cavity of the palladium tube could 

 be exhausted of air. 



Now the closed palladium tube remained air-tight, when exhausted by the Sprengel 

 tube, at the ordinary temperature, at 260°, and at a temperature verging on low redness, 

 the gas without being atmospheric air. Hydrogen being then substituted as the ex- 

 ternal gas, the walls of the palladium tube still remained impermeable at a low tempe- 

 rature. No hydrogen gained the interior in three hours at 100°. But the temperature 

 being gradually raised by means of an oil-bath to 240°, hydrogen then began to come 

 through, and at a gradually increasing rate to 265°. The hydrogen then entered 

 steadily at the rate of 8 "6 7 cub. centims. in five minutes. This gives a rate of 327 cub. 

 centims. for a square metre of surface per minute. Heated to a temperature just short 

 of redness, the passage of hydrogen was increased to 11*2 cub. centims. in five minutes, 

 or 423 cub. centims. for a square metre per minute. 



With coal-gas as the external atmosphere the penetration of the palladium began 

 about the same temperature, and was continued at 270° at the rate of 57 cub. centims. 

 for a square metre of surface per minute. The penetrating gas had no odour of coal-gas, 

 contained no trace of carbon, and appeared to be absolutely pure hydrogen. The exact 

 isolation of the latter gas by septa of both platinum and palladium appears most extra- 

 ordinary. 



A quantitative determination of the hydrogen in a gaseous mixture could probably 

 be effected by means of the hollow cylinder of palladium. 



Is the power to penetrate the metals in question confined to hydrogen 1 It has been 

 lately concluded by Dr. C. Wetherill that the turgescence of the ammonium amalgam 

 depends entirely upon the retention of hydrogen gas-bubbles*; hydrogen, then, appears 

 to exhibit an attraction of a peculiar kind for mercury. The ready liquefaction of the 

 same gas by the platinum metals evinces also a powerful mutual attraction. The only 



* Americ6in Journal of Science, vol. xlii. No. 124. 



