ELECTiaiCAL RESISTANCE UNDER PRESSURE. 7? 



a capillary of an alloy of high specific resistance, which is not attacked 

 by the lithium. For this purpose I used the allow known under the 

 trade name of " ^ 193 Alloy," manufactured by the Driver Harris 

 Co. The specific resistance is about 50 times that of copper. I suc- 

 ceeded in drawing this into a fine capillary 0.045 inches outside diame- 

 ter, and 0.032 inside diameter. An attempt to treat Nichrome simi- 

 larly did not meet with such ready success, it being much harder to 

 draw. The capillary was plugged at one end with a pin of iron, and 

 at the other end an iron cup was silver soldered, no silver solder 

 coming in contact with the lithium. Four copper leads were soldered 

 to the outside of the capillary, two at each end, for use as current and 

 potential leads, and the resistance was measured in the regular way by 

 the potentiometer method. The capillary was filled by melting the 

 lithium into it in vacuum. The resistance of the capillary was from 

 five to ten times the resistance of the lithium which filled it, varying 

 with the pressure and temperature. Preliminary measurements were 

 made on the resistance of the capillary when empty, and on its temper- 

 ature and pressure coefficients. These values for the capillary are 

 given for themselves under the heading # 193 Alloy. The tempera- 

 ture and pressure coefficients are both small compared with those of 

 lithium. 



The capillary and the lithium inside it constitute two resistances in 

 parallel. The various resistances are connected by the relation 



1 = 1 + 1 



Ro Re Rl 



where Rl is the resistance of the lithium filling the capillary, Re is 

 the resistance of the capillary alone, and Ro is the observed resistance 

 of the capillary and lithium in parallel. By differentiating this ex- 

 pression with respect to the pressure we obtain 



1 dRj. Rl /I dRo\ Rl I 1 clR 



Rl dp Ro \Ro dp / Re \Rc dp 



From the pressure coefficient as observed, and the pressure coefficient 

 of the capillary separately determined, it is therefore possible to 

 obtain the pressure coefficient of the lithium alone which fills the 

 capillary. The term involving the pressure coefficient of the capillary 

 is seen to be small, so that this coefficient need not be known with 

 great accuracy, and the ratio of the resistance of the lithium to the 

 observed resistance may be found with any accuracy desired, so that 



