﻿Manchester Memoirs, Vol. xlix. (1905), No. 10. 3 



platinum thermometer placed as close to it as possible, 

 its ends being soldered to two similar copper leads. 



Various methods have been suggested for the main- 

 tenance of low temperatures, one of the most recent 

 depends upon the fact that a solenoid of wire with its ends 

 dipping into liquid air encloses a space whose average 

 temperature exceeds that of the liquid air by an amount 

 depending on the length of the solenoid. Evidently, 

 although the temperature at any point may be constant, 

 it is a very great drawback to have the temperature 

 varying from point to point. 



The first method adopted was that used by Dr. C. H. 

 Lees in his work on thermal conductivity at low tempera- 

 tures. The carbon and platinum were immersed in liquid 

 air, which is a non-conductor of electricity, and observations 

 taken alternately of the resistances of the platinum and 

 carbon, the time at which each observation was completed 

 being noted. The resistance box used, w r hich was kindly 

 lent by Dr. Beattie, was made by Nalder Bros., and was 

 provided with dial switches instead of plugs for varying 

 the resistances. With this we were able to make deter- 

 minations of resistance very quickly. 



The liquid air rapidly boils away and the apparatus 

 heats up to the temperature of the room. Curves are 

 then drawn, connecting the alteration of resistance of 

 both platinum and carbon with the time, and by inter- 

 polation the resistance of the platinum can be found at 

 the instants when the resistance of the carbon was 

 determined. Knowing the resistance of the platinum at 

 ioo°C. and o°C, the temperature on the platinum 

 thermometric scale corresponding to a resistance R of the 

 platinum spiral, is found from the equation — 



pt -= io % R :~R 



