High Temperatures. 179 



below. If to t' — h the change in the oven temperature 8 be 

 added, the results contained in the column t — t' or the differ- 

 ence between the results obtained at the two pressures is 

 obtained. The mean value of this difference is d= 0*7° and 

 exceeds 1*0° only in a single instance. 



It follows from this that within the given limits of accuracy, 

 the measurements with the gas thermometer equipped with the 

 platin-iridium bulb were not disturbed by the pressure upon the 

 walls of the hot bulb. The initial pressure of the gas at 0° was 

 294 , 40 mm , nearly the same as in the earlier measurements, and 

 its coefficient of expansion O003666. 



It has been suggested that this method be employed for 

 obtaining the volume of the bulb at high known temperatures. 

 It would then be possible to arrive at its coefficient of expan- 

 sion in this way. In addition to the usual equation for the 

 calculation of the temperature with constant volume, 



^- = H„V„ 



a second relation is, to be sure, obtainable 



\l+at 1 + atJ 



= H„V, 



if t, be used to represent the temperature of the supplementary 

 volume Y r The equations are nevertheless not independent 

 of each other and are therefore not sufficient for the deter- 

 mination of the two unknown quantities Y and t. 



3. Comparison of the Thermo- element with the Gas Thermometer. 



Second Platin-iridium Bulb. — The second platin-iridium 

 bulb had nearly the same form and size as the first. It was, 

 however, made from the alloy 90 Pt. 10 Ir., and its walls (l mm ) 

 were twice as thick as those of the earlier bulb. Its volume at 

 0° was 195-87 cm3 . 



After being boiled with concentrated nitric acid and several 

 times rinsed with distilled water the bulb was carefully dried 

 and attached to the manometer. Then it was evacuated with 

 a mercury pump and maintained for several hours at a tem- 

 perature of 1300°, being "rinsed" from time to time with 

 fresh nitrogen and again evacuated. The gas for the final 

 filling was also admitted at the high temperature and the exact 

 pressure regulated after the bulb had cooled. 



The new bulb was used for several series of measurements 

 and proved to be quite as satisfactory as the old. The gas 

 pressure at 0° remained constant to within 0'l mm throughout. 



