82 Mr. F. S. Spiers on Contact Electricity. 



I then again exhausted to a pressure of yJo- mm., when the 

 P.D. was 



0*95 volt, 



and again let in an atmosphere of hydrogen. The same 

 operation was repeated a third time, and finally the apparatus 

 was left to itself full of hydrogen for two days, so as to give 

 the gas a good chance of soaking into the metals and dis- 

 placing from them the original surface-layers of air. The 

 voltage was then 



0-87 volt, 



and remained constant at that value. (This diminution is in 

 accordance with Lord Kelvin's observation that soaking a 

 platinum plate in hydrogen makes it temporarily about 

 O'l volt positive to another platinum plate that has not been 

 so treated *.) 



I then again exhausted to -50V0 mm - an( ^ heated the tube 

 with a couple of bunsen-burners just as on former occasions. 

 When cold the P.D. was 



0-83 volt, 

 at a pressure of y^^ mm. 



From the practical constancy of this value, it seemed to 

 me that I had now really removed the last traces of air, and 

 that there was not sufficient oxygen present to oxidize the 

 aluminium plate as had happened before in an ordinary 

 vacuum, and that there was a true contact P.D. between 

 aluminium and platinum of about 0*8 of a volt. To subject 

 it to the severest possible test, I next heated the apparatus to 

 a far higher temperature with a blowpipe-flame. The outer 

 copper protecting sheath was kept red hot for about half an 

 hour, the pump working throughout the time. I then 

 allowed the tube to slowly cool. When it was quite cold 

 and the pressure was -^Wa mm v the P.D. between the plates 



WaS 0-09 volt. 



I let in a little hydrogen ; it had no immediate effect on 

 the P.D. I then let in dried air; the reading very slowly 

 fell to zero, and the nsxt morning it was 



-0-15 volt 



(?'. e. aluminium negative to platinum), at which value it 

 remained permanently. In appearance the aluminium was 

 unaltered, except for a very slight purple coloration just at 



* < Nature,' April 4, 1881. 



