gray: temperature uniformity 137 



the principal use of the inner concentric heater is to hasten the 

 attainment of approximate thermal equilibrium. Experience has 

 shown that when the proper adjustment has been secured, the 

 steadying action of the insulation within the outer heater is 

 sufficient to wipe out fluctuations in both losses and supply of 

 heat.^ 



The temperatures and temperature distributions were deter- 

 mined by two thoroughly protected Pt-Pt Rh thermoelements of 

 the best quality in connection with a Diesselhorst potentiometer. 

 The sensitivity was such that one millimeter change in the gal- 

 vanometer deflection indicated a change in temperature of be- 

 tween 0.058° and 0.086°, depending on the temperature. 



Even before the flat end-heating coils were substituted for the 

 coils in the outer conducting blocks of the plugs, fair uniformity 

 throughout the central 30 cm. for which the furnace was designed 

 could be easily obtained by proper adjustment of the heating 

 currents. Good performance was not limited to some particular 

 temperature, as is the case with furnaces in which adjustment 

 of the distribution is attempted by the ordinary cut-and-try 

 method of crowding the windings near the ends. For example, 

 at 667° C. the mean temperature throughout the full 30 cm. was 

 0.37° lower than the temperature at the center, with a maximum 

 drop of one degree at the extreme" ends of the interval; at 145° 

 the mean temperature over the same region was 0.08° lower than 

 at the center, with a maximum drop of 0.25° at the extreme ends. 



As an indication of the constancy attainable, it might be men- 

 tioned that during the twenty minutes that passed while the 

 observations at 667° were being made the temperature at the 

 center rose 0.067°, or at the rate of 0.2° per hour. During this 

 period no attention whatever was paid to regulation of the heat- 

 ing currents. 



'' The use of two concentric heaters bears some resemblance to the cascade fur- 

 nace described by Harker (Proc. Roy. Soc. A 76: 237. 1905), for reaching a temper- 

 ature of about 2000°C in a heater-tube made of the same material as Nernst lamp 

 glowers. To avoid the necessity of having contacts on this tube capable of carrying 

 relatively large currents, and the consequent severe stress on the parts, it was sur- 

 rounded by a coil of nickel wire heated bj^ an independent circuit to about 1000°, 

 and insulated from the inner tube by a layer of zirconia. No precaution seems to 

 have been taken to secure uniformity of temperature within. 



