636 Dr. W. Wilson on the Discharge of 



whose positive pole was connected through w to p. When p 

 had been raised to the desired temperature in the manner 



Fiflr. 2. 



<uT 



0£ 



already described, the other terminal of the galvanometer 

 was connected with the mercury in M and the ensuing 

 current from p to D measured. The experiments were con- 

 ducted in air, at atmospheric pressure, which had been freed 

 from dust and moisture by passing through tubes containing 

 cotton-wool, calcium chloride, caustic potash, and phosphorus 

 pentoxide. 



Diminution of Positive Leak from Hot Platinum not due to 

 Heating only. — It is a well-known fact that the positive leak 

 from hot platinum diminishes with continued heating. My 

 experiments point to the conclusion that this phenomenon is 

 not due to the temperature alone, but is rather a consequence 

 of the discharge of positive electricity. The following obser- 

 vations illustrate this: — 



In the initial observation the current was represented by 

 186 scale-divisions, the potential-difference between the wire, 

 p, and the cylinder, D (fig. 2), being 100 volts. The tempe- 

 rature of the wire was about 1 100° C. After this measurement 

 the cylinder, D, was completely insulated by disconnecting 

 at M, the wire being maintained at exactly the same tempe- 

 rature as before. At the end of 10 minutes the connexion 



