148 BRIDGMAN. 



temperature coefficients of solid and liquid. With regard to the tem- 

 perature coefficients at atmospheric pressure it has long been known 

 that the coefficient of the liquid is less than that of the solid. This 

 is verified for all the metals measured here, except potassium. 



With regard to the pressure coefficient of resistance it is natural 

 to expect that of the liquid to be greater than that of the solid at the 

 same temperature. This is true for sodium. At 120° the pressure 

 coefficient of the liquid is about 7% greater than that of the solid extra- 

 polated to the same temperature. It is however, perhaps surprising 

 that the relative change of the pressure coefficient of solid sodium 

 brought about by an increase of pressure of 12000 kg. is greater than 

 that of the liquid under the same increase of pressure. The relative 

 decrease of the temperature coefficient under 12000 kg. is greater for 

 liquid sodium, howe^'er, than for the solid. 



The behavior of liquid potassium is not as we would expect. At the 

 melting point at atmospheric pressure the pressure coefficient of liquid 

 potassium is greater than that of the solid. Because of the abnormal 

 temperature coefficient of the pressure coefficient of the liquid, how- 

 ever, the coefficient of the solid would become greater than that of 

 the liquid if the solid could be superheated sufficiently. The relative 

 variation with pressure of the pressure coefficient is greater for the 

 solid than the liquid. This again is not w^hat we might expect. The 

 data for potassium do not cover a sufficient range to permit a com- 

 parison of the variation with pressure of the temperature coefficients 

 of solid and liquid. 



The pressure coefficient of solid mercury has been found to be con- 

 stant over the range from 7640 to 12000 kg. The coefficient of the 

 liquid, on the other hand, decreases with rising pressure. It has 

 already been mentioned as surprising that the coefficient of the solid 

 is greater than that of the liquid at 6500 kg. This difference would 

 become still more accentuated if the liquid could be carried in the 

 metastable state into the region of stability of the solid; in this range 

 its pressure coefficient would be found to vary considerably less than 

 that of the solid. The measurements were not accurate enough to 

 permit a comparison of the ^•ariations of the temperature coefficients 

 of the solid and liquid. It is known, however, that at atmospheric 

 pressure the temperature coefficient of the solid is normal, while that 

 of the liquid is abnormally low even for a liquid. 



The pressure coefficient of solid gallium is of the order of 2.5 less 

 than that of the liquid. The coefficient of the solid is independent 

 of the pressure, whereas that of the liquid decreases markedly with 



