PRESSURE ON RESISTANCE OF METALS. 637 



Theoretical Bearings. 



Before discussing the bearing of these results on electron theories 

 of metals, it will pay to emphasize two points. The first is that the 

 coefficients tabulated are the actual observed coefficients, measured 

 by the ordinary methods with electrodes permanently fixed to determi- 

 nate parts of the surface. But in theoretical discussion we are more 

 inclined to be interested in the variation of specific resistance. To 

 get this, the observed results must be corrected by a factor equal to 

 the change of linear dimensions. It is easy to see that for normal 

 metals the temperature coefficient of observed resistance is numeri- 

 cally smaller than the temperature coefficient of specific resistance by 

 the linear thermal dilatation, and the pressure coefficient of observed 

 resistance is numerically less than the pressure coefficient of specific 

 resistance by the linear compressibility. This factor is not important 

 for the temperature coefficient, rising in the extreme case above (In) 

 to 1%, but for the pressure coefficient it may amount to 10% in some 

 cases. In making correction for the compressibility we are confronted 

 by the difficulty that only one or two compressibilities have been 

 measured over any extensive pressure range. For the initial com- 

 pressibility we have the data of Richards ;^^ at higher pressures the 

 best that we can do is to neglect the change of compressibility with 

 pressure. For the less compressible metals any such change is proba- 

 bly slight. A number of years ago I measured the compressibility 

 of iron up to 10000 kg. and of aluminum to 6500 kg., and could find 

 no variation over this range. ^° Of course in any discussion of the 

 pressure coefficient of specific resistance at atmospheric pressure this 

 source of uncertainty does not enter. 



The second observation is concerning the magnitude of the effects. 

 It has been obvious enough that the data have presented no spec- 

 tacular features, and I must confess to a sense of disappointment that 

 an extension of the pressure range to at least four fold that of previous 

 measurements has brought out no striking new facts to reward the 

 extra effort. It is true that as far as I am aware the independence of 

 temperature coefficient and pressure was not previously known, or at 

 least was never emphasized, but it might have been discovered by meas- 

 urements to only 3000 kg. if one had been willing to take the trouble. 



29 T. W. Richards, Jour. Amer. Chem. 8oc. 37, 1643-1656 (1915). 



30 P. W. Bridgman, Proc. Amer. Acad. 44, 255-279 (1909) and 47, 866 (1911). 



