1912] 



Very High Temperatures. 



321 



Some time ago we endeavoured at the National Physical Labora- 

 tory to make a furnace for very high temperatures without employ- 

 ing carbon. The introduction of the Nernst lamp was suggestive. 

 Itwas found that a great number of substances could be made to 

 act like a Nernst filament, e.g. a piece of the stem of a churchwarden 

 pipe, if sufficiently strongly heated, can be made to conduct electricity 

 well enough to become incandescent. Carborundum crystal behaves 

 similarly, and requires no initial heating (experiment shown) ; in this 



Pig. 2.— Small Model Stbaight Caebon-Tube Furnace. 



case the temperature can be raised high enough to volatilize off the 

 silicon, which burns, forming a cloud of silica. A cascade furnace 

 was constructed on these lines : a tube made up of zirconia and a 

 little yttria was raised by means of an insulated nickel winding to 

 500° or 600°, at which temperature the tube conducts sufficiently 

 well to enable a heating current to be passed through it. There is 

 no difficulty in melting platinum, for example, in such a furnace 

 using a quite small heating current (about 2 amps.) A zirconia 

 tube from such a furnace was taken out after it had been run for six 



