62 BRIDGMAN. 



duction. The results then obtained suggested a number of important 

 questions which could not then be answered because the data had not 

 been determined. In particular, the effect of pressure on the resistance 

 of only one liquid metal had been measured, mercury, and the com- 

 parative effect of pressiu"e on the resistance of the same metal in the 

 solid and the liquid state was not known for any substance. Further- 

 more, the effect of pressure on none of the alkali metals was known; 

 it is just here that the largest effects would be expected and the most 

 significant results for theory. The extension of the measurements to 

 these svibstances demanded certain changes and improvements in the 

 technique. I was soon able to make measurements on the alkali 

 metals, and to my very great surprise found that the resistance of 

 lithium increases under pressure, whereas that of the other alkali 

 metals sodium and potassium decreases greatly. It was obvious that 

 our pictures of the mechanism of conduction are not ^et so perfect 

 that we can reason by analogy even from one element to another 

 closely related in the periodic table. The importance of making 

 actual measurements on as many of the elements as possible, particu- 

 larly the rarer ones, was obvious. 



In the present work I have therefore attempted to extend the 

 previous measurements to as many new elements as the new resources 

 of my technique, and the availability of the elements themselves has 

 made possible, with especial reference to the question of the resistance 

 of the metals liquid within a moderate te^nperature range. Further- 

 more, a number of commercial alloys have been employed in other 

 work, and rough measurements of the effect of pressure on the re- 

 sistance of them have been necessary; it has been an easy matter to 

 make the measurements precise enough to be included here. Of 

 course the entire question of the effect of pressure on the resistance of 

 the alloys is a most complicated one, and these isolated results can- 

 not as yet have much significance. 



The results given here are for eighteen new elements and six com- 

 mercial alloys. In addition, better results than were possible in the 

 previous work are here given for tungsten and magnesium. Better 

 results on tungsten were possible because of the availability of purer 

 material, and for magnesium the new technique makes possible the 

 elimination of contact resistance, which introduced very large error 

 in the pre\dous work. 



