596 



BRIDGMAN. 



The mean temperature coefficient of this wire, 0.00416, is distinctly 

 higher than the best wire of Jaeger and Diesselhorst,^ for which they 

 give 0.00402 (uncorrected for range). The zinc of Jaeger and Diessel- 

 horst was from the same source as this and showed on analysis not 

 more than 0.01% of Pb, Cd, or Fe. The initial pressure coefficient 

 at 0° is given as — O.OsSO by Lisell,^ against —0.05540 found graphi- 

 cally from the deviation curve above. Lisell does not mention the 

 source of his zinc, nor does he give its analysis or temperature coeffi- 

 cient. He did not regard it as one of his purer materials, and says that 

 he estimates the error in his value to be not over 10%. 



i.0030 



Q.0025 



|.0020q, 



20° 40° 60° 80° 100° 

 Temperature 



i ''^^ 



20° 40° 60° 80° 



Temperature 



Zinc 



100° 



Figure 8. Zinc, results for the measured resistance. The deviations from 

 linearity are given as fractions of the resistance at kg. and 0°C. The devia- 

 tions are symmetrical about the mean pressure, so that it is not necessary to 

 give the detailed deviation curves as in the preceding figures. The pressure 

 coefficient is the average coefficient between and 12000 kg. 



The general character of the results is as follows. When the resist- 

 ance-pressure curves for the different temperatures are scaled to the 

 same initial resistance the curves at higher temperatures are less steep, 

 but show very little difference of curvature. 



Magnesium. The material available was commercial magnesium 

 from Eimer and Amend, for which I have no analysis. It was 

 extruded hot into wire 0.007 inch diameter and wound loosely on a 

 bone core. The initial resistance was 5.6 ohms. It is a matter of 

 great difficulty to make any sort of electrical connections ; no commer- 

 cial method is known. I was able to make a connection with alumi- 

 num solder which appeared good to the eye and gave satisfactory 

 results up to SOOO kg. But beyond this the solder l)roke loose, proba- 

 bly because of unequal compression, and no further results of any 

 constancy were possible. I also made indifferent connections with a 



