356 BRIDGMAN. 



unannealed. This produced a large permanent zero reading, but 

 should not introduce any error. 



At atmospheric pressure the average temperature coefficient of 

 resistance between 0° and 100° was found to be 0.00461 ; the relation 

 between temperature and resistance was sensibly linear. This coeffi- 

 cient is considerably higher than that of the piece whose pressure 

 coefficient was previously measured, for which the mean temperature 

 coefficient was 0.00434. Both molybdenum and tungsten are known 

 to vary considerably in properties with the amount of mechanical 

 working, so that the difference between these two specimens may 

 have been accountable for by the difference of size, and not by a 

 difference of chemical constitution. 



The thermo-electric behavior at atmospheric pressure against 

 lead is given by the formulas: 



E = (5.892 f+ 0.02167/2-0.000025^3) X lO'^ volts, 



P = (5.892 + 0.04334/ -0.000075/2) (/ + 273) X 10-« volts, 



(7 = (0.04334-0.000150/) (/ + 273) X lO'^ volts/°C. 



For the pressure measurements the wire was seasoned by two 

 applications of 12000 kg. at room temperature. The e.m.f. under 

 pressure is small, reaching at the most at 12000 kg. and 100° only 

 + 0.8 X 10"^ volts. Individual readings showed large variations; 

 the zero readings did not vary so much as those at high pressures. 

 32 readings in all were made; of these 6 were bad, showing variations 

 from 7 to 50% of the total effect. The remaining 26 readings were 

 not so bad, the maximum departure from a smooth curve being 5% 

 of the total effect, and the average numerical departure 1.2%. It 

 was not necessary to make any readjustment in passing from the 

 e.m.f. curves at constant temperature to those at constant pressure. 



The numerical results are shown in Tables XXXV and XXXVI 

 and Figures 36 and 37. At constant temperature the relation between 

 e.m.f. and pressure is linear; at constant pressure the curves against 

 temperature pass through a point of inflection. This means a com- 

 plicated behavior of both Peltier and Thomson heats. The Peltier 

 heat is throughout positive, but at constant pressure passes through 

 a pronounced maximum near 40°. The Thomson heat is both posi- 

 tive and negative, passing tlirough pronounced positive maxima near 

 25°, and negative minima near 50°. 



There are no previous results for comparison. 



Tungsten. This, like molybdenum, was a fresh sample from Dr. 

 \Y. D. Coolidge of the General Electric Company, the specimen 



