494 



Mr. W. Craig Henderson on 



as the warm vapour might condense on these supports and 

 so spoil the insulation it was necessary to have all insulators 

 outside the case A containing the water-vessel. The arrange- 

 ment made is shown in tig. 2. 



The metal disk B supporting the water-vessel V was fixed 



Fis 2. 



BUNSEN BURNER 



at the end of a stout bar of brass (of breadth 1 cm.) which 

 passed into the case A, of galvanized iron, by a narrow slot 

 in the side. Outside the case A this bar passed along the 

 axis of a metal tube D, fixed in clamps, and was supported 

 by D, and insulated from it, by plugs of sulphur at the ends. 

 These sulphur plugs had cork centres through which the bar 

 passed, in order to prevent melting of the sulphur by heat 

 conducted along the rod. The other end of this bar was 

 connected as before by means of a mercury cup to the insu- 

 lated quadrants of the electrometer. A piece of ebonite was 

 fixed above the lid of the case A at the end of a short brass 

 angle-piece rivetted to the lid, and through this ebonite 

 passed a stout wire turning down through the hole in the 

 lid, and bearing the disk C at its end inside the case. AH 

 other connexions were as before. The water-vessel now used 

 was of tinned iron and of the same radius as the lead vessel 

 used in fig. 1. A large Bunsen-burner was placed under the 

 case A, and a metal screen was attached to the case to prevent 

 as much as possible any of the hot gases from reaching the 

 rod attached to the electrometer. 



In actual experiments it was found that as long as the 

 Bunsen flame was present there was always an electrical 

 effect, due to this flame, shown on the electrometer. As 

 soon as the flame was extinguished this effect ceased. Read- 

 ings were therefore taken while the liquid was cooling. In 



