1886.] A Theory of Voltaic Action. 305 



the mere oxidation by free oxygen, bat is associated with the decom- 

 position of the condensed film on the metal. 



24. The apparatus, fig. 1, was admittedly not sufficiently imper- 

 vious to gaseous diffusion to permit of a crucial experiment on the 

 effect of drying the atmosphere surrounding the metals. However, 

 in the hope of partial success, the following experiment was begun in 

 1884. The copper and zinc quadrants were adjusted in position, the 

 water vessel having been removed, and a weight with paper vanes 

 substituted. The difference of potential was found 0'68D. Two 

 small porcelain capsules containing phosphoric anhydride were placed 

 inside the instrument, and the joint between the glass front and the 

 metal facing made as good as possible with grease. The following 

 measurements were then made in terms of standard Daniell ; t in this 

 case standing for days after closing up the instrument. 



t=0 1 2 3 5 7 9 11 13 

 P/D = 0-64 0-62 0-64 061 0"59 0*59 058 0"58 057 



t=20 35 78 124 134 304 305 



P/D= 56 0-56 0557 0'54 0'50 0'52 0'51 



The phosphoric anhydride was now taken out, and the instrument 

 allowed to remain open for twenty minutes, after which a very care- 

 fully made and satisfactory observation showed that the difference of 

 potential had increased to 0*646D. Fresh phosphoric anhydride, 

 was now placed in the instrument, which was closed as before, and 

 the value measured immediately afterwards found to be 0659D. 



A similar series of observations was then made for a period of 

 173 days, during which the observed difference of potential fell more 

 or less regularly to about 0'5D. Then on allowing the instrument to 

 remain open for one hour it rose to about 067D. These results 

 were quite as marked as I had hoped for with such unsuitable appa- 

 ratus, and seem to show very clearly that even a partial drying of the 

 metallic surfaces alters the difference of potential near them in a 

 decided manner. 



25. In 1881 I had endeavoured to construct a copper-zinc quadrant 

 electrometer hermetically sealed in a glass vessel, from which I 

 could exhaust the air and absorb the oxygen compounds by potas- 

 sium. Repeated failures in the construction of the apparatus have 

 hitherto debarred the experiment. A similar idea occurred to Herr 

 Von Zahn about the same time (' Untersuchungen uber Contact- 

 Elektricitat,' p. 48). 



He enclosed a platinum-zinc condenser in a hermetically sealed 

 tube, together with metallic sodium. The difference of potential was 

 reduced in this case to about 05D, which the author attributes, in 

 part at least, to the absence of moisture. He considers the experiment 



