Chemical Affinity in terms of Electromotive Force. 179 



jointly with Prof. W. C. Roberts, has shown (Chern. Soc. 

 Journal, 1873, p. 112) that the specific heat of the portion of 

 hydrogen first occluded by palladium is much greater than 

 that of the later portions, the limiting values being near to 

 9 on the one side and 3*4 on the other — or values respec- 

 tively representing a condensation to something approaching 

 the liquid state (judged by analogy with liquid bromine, the 

 product of the combining number and specific heat of which 

 is near 9), and a condensation but little exceeding that due 

 to ordinary atmospheric pressure. Presumably when a film 

 of gas is attracted to the surface of a solid, the portions first 

 attracted are very highly condensed, the next ones less so, and 

 so on, a kind of atmosphere or aura surrounding the body 

 being thus produced of density varying from the maximum 

 amount of condensation to the minimum, i. e. the pressure of 

 the surrounding atmosphere. The total quantity of gas thus 

 condensed is not inconsiderable : thus an ordinary porcelain 

 or, better still, a platinum crucible, if ignited and cooled in 

 dry air, will weigh less by from one to several milligrammes, 

 according to its size, if weighed immediately it is cold, and 

 before the film or aura of condensed gas expelled by the heat 

 is completely re-formed, than it will weigh if allowed to remain 

 some time in the dry air, and weighed after the aura has been 

 attracted to saturation. The inequality in the deviations from 

 Boyle's law exhibited by gases under high pressures, but when 

 still far removed from their condensing or from their critical 

 points, has been considered by many physicists to be probably 

 largely due to the difference in the attracting or aura-forming 

 capacity exhibited by the tubes and measuring-vessels used, 

 according as the nature of the gas varies. Not only gases, but 

 also liquids, are attracted by solids, forming a layer or aura of 

 highly condensed liquid adherent to the solid : Schliermacher 

 has recently calculated (Dingier, Pol. Journal, pp. 224, 471; 

 Chemical- Society Journal, Abstracts, 1880, p. 363) that this 

 causes an appreciable error in the determinations of the 

 weight of a unit volume of water hitherto made. 



63. When water is electrolysed with non-oxidizable recently 

 ignited electrodes {e.g. of platinum), the length of time which 

 must elapse before the electrodes become saturated with gases 

 will depend on circumstances, and amongst others on the fact 

 that the water will tend to take up oxygen at the + electrode 

 and hydrogen at the — electrode, and the solutions thus pro- 

 duced will diffuse towards the opposite poles, thus tending to 

 set up a double form of " electrolytic convection." 



Suppose that a current pass through acidulated water, so 

 that gases are slowly evolved ; if by saturating with oxygen 



