504 Mr. T. Graham on the Absorption and 



sentially repulsive. A degree of minute porosity is conceivable, 

 which will admit a liquid, but may be impassable to a gas,. even 

 under its molecular movement of diffusion. 



Finally, there is presented to us a bold and original conjecture 

 by M. Deville, in explanation of his own observations. It is 

 clearly expressed in the following quotation taken from the last 

 publication of M. Deville on this subject : — 



" La permeabilite de la matiere est d'une nature toute dif- 

 ferente dans les corps homogenes, comme le fer et le platine, et 

 dans des pates plus ou moins discontinues, resserrees par la 

 cuisson ou la pression, comme la terre a creuset, la plombagine, 

 dont M. Graham s'est servi dans ses memorables experiences. 

 Dans les metaux, la porosite resulte de la dilatation que la cha- 

 leur fait eprouver aux espaces intermoleculaires ; elle est en rela- 

 tion avec la forme des molecules que Ton peut toujours supposer 

 regulieres, et avec leur alignement qui determine le clivage ou 

 les plans de facile fracture des masses cristallisees. C'est cet 

 intervalle intermoleculaire que le phenomene de la porosite des 

 metaux purs et fondus accuse avec une evidence eclatante, c'est 

 aussi par ce phenomene qu'on peut esperer de calculer la dis- 

 tance des molecules solides aux temperatures elevees ou les gaz 

 peuvent s'y introcluire." 



A new kind of porosity in metals is imagined, of a greater de- 

 gree of minuteness than the porosity of graphite and earthen- 

 ware. This is an intermolecular porosity due entirely to dilata- 

 tion. The intermolecular porosity of platinum and iron is not 

 sufficient to admit any passage of gas at low temperatures, but 

 is supposed by M. Deville to be developed by the expansive 

 agency of heat upon the metals, and to become sensible at the 

 temperature of ignition. Such a species of porosity, if it exists, 

 may well be expected to throw light on the distances of solid 

 molecules at elevated temperatures, when gases introduce them- 

 selves. The ready passage through platinum of some gases/ 

 particularly of hydrogen, and the difficult passage of others ren- 

 der such molecular views the more remarkable. 



The passage of hydrogen through the substance of heated pla- 

 tinum appears in its most simple aspect when the gas is allowed 

 to make its way through the metal into a vacuous space. The 

 experiment of M. Deville, where a tube of platinum charged 

 with nitrogen is placed within a large porcelain tube charged 

 with hydrogen*, was modified by placing the platinum tube, 

 closed at one end, in communication by the other (open) extre- 

 mity with the Sprengel pump, so that a vacuum was substituted 

 for the nitrogen. It was then easy to observe that a vacuum in 

 the platinum tube was preserved for hours when the external 

 * Comptes Rendus, vol. lvii. p. %5. 



