338 



BRIDGMAN. 



tive maximum and decreases. The behavior of the Thomson heat is 

 compHcated. On the whole the course is, at every pressure, from 

 positive values at 0° to negative values at 100°, but near 40° the 

 normal course is arrested, the effect reversing in direction and passing 

 through a maximum between 50° and 60°, and then resuming its fall 

 to negative values. 



It will not pay to go into the behavior of the other two specimens in 

 such detail. Tables and diagrams are given showing the e.m.f. but 

 not the Peltier and the Thomson heats. Both showed qualitatively 

 the same behavior as the pure specimen; the effect was negative at 



TABLE XXVI. 



Iron (Pure, Annealed). 



Thermo-electromotive Force, volts X 10^. 



low temperatures (plotted against pressure), rising to normally high 

 positive values at 100°. The direction of curvature is the same as 

 for the annealed iron, but the anomaly in direction of hysteresis was 

 not repeated. The hysteresis effects were of about the same magni- 

 tude as for the first specimen, except that at 95° the hard drawn 

 specimen showed large permanent changes. This is not surprising. 

 The numerical values of thermal e.m.f. against temperature at con- 

 stant pressure are plotted in Figures 29 and 30. At low temperature 

 all three sets of curves for the three specimens are alike, but between 

 75° and 100° the hard drawn and the commercial iron do not show the 



