306 



Mr. J. Brown. 



[Nov. 18, 



would be a crucial one in favour of the chemical theory, if, after 

 opening the tube, the difference of potential was found to have 

 increased. He does not, however, open it. 



There are two objections to this particular experiment. The 

 tube, &c, was washed out with water which tarnished the zinc. 

 The action of the sodium on moisture would evolve hydrogen, which 

 would alloy with the platinum. 



The former would reduce- the difference of potential permanently, 

 and the latter perhaps temporarily, but neither would, I think, much 

 affect the conclusion to be drawn from a measurement made imme- 

 diately on opening the tube. 



26. Von Zahn also experimented qualitatively on the difference of 

 potential between a copper plate and a flat spiral of glass tube con- 

 taining sodium used as a Volta condenser plate, but does not seem to 

 have tested the effect after the lapse of any considerable time, 

 although he states that the surface of the sodium remained bright 

 after standing some years. 



If Bunsen's hypothesis ('Phil. Mag.,' vol. 17, 1884, p. 172) that 

 glass may gradually, year after year,, absorb a part of the liquid film 

 in contact with it, be true, it would probably be necessary to allow 

 such apparatus to stand a long time before the absorbed water had 

 freed itself from the glass. The same criticism applies to Von Zahn's 

 experiment with a sodium copper condenser m vacuo, where the tube 

 broke after the first observation. 



27. The foregoing observations, the fact that gases under ordinary 

 conditions are non-conductors, and the much greater simplicity of the 

 theory, all point to the view that the so-called " contact " effects are 

 due to the action of condensed films on the surface of the metals. 

 When the copper and zinc quadrants are wet with water, the 

 difference of potential near them is practically that of the terminals of 

 a cell formed of the same metals dipping in water. If instead of the 

 water we use copper sulphate saturated solution on the copper 

 quadrant and zinc sulphate solution on the zinc, the difference of 

 potential is almost exactly that of the Daniell cell, of which this 

 arrangement is the analogue. In fact whether' we break such simple 

 galvanic circuits in their metallic parts, or in their liquid parts, we 

 obtain the same value at the terminals in either case. 



28. Considering then the two metals of a copper^zinc Volta con- 

 denser as the elements of a cell, and the moisture film on them as the 

 electrolyte, we have a simple copper-water-zinc cell, divided on its 

 electrolyte, and showing, when the metals are clean, approximately 

 the difference of potential appropriate to such a cell. 



The well-known experiment of Sir William Thomson, where a 

 water-drop was placed between the previously disconnected metal 

 quadrants, is thus easily explained. The water-drop connexion 



