.\ii;i:rri;v STANDARDS OF RESISTANCE, ETC. 



85 



The ice box was a double one, with cork dust as a lagging. A double lid was also 

 fitted, and under ordinary conditions the ice did not necessitate renewal for several 

 days. In practice, however, a little was added each morning, the water being 

 drained away at the same time. Two wooden rails were fixed to the base of the 

 inner box, and corresponding grooves made in the wooden blocks of the framework 

 supporting the tul>e. The latter could therefore be moved in the direction of its 

 length only. This was found to be very convenient. 



The leads of the standard passed to insulated terminals screwed on to one half of 

 the hollow cover of the box. From these terminals thick copper leads made the 

 necessary connections to the Kelvin double bridge. In addition, other wires were 

 stretched to four terminals, the latter being conveniently situated for the measure- 

 ment of the standard's resistance by the potentiometer method. 



Measurement of Resistance, (a) Kelvin Double Bridge (fig. 8). 



The circuit being closed, the condition for the potential difference at the galvano- 

 meter terminals to be zero is that P shall be equal to 



\ f R (1 + S/W')1 

 d) IS(1 + R/W")J 



da. 



+ Q/W ' . 



When the value of QR/S is approximately one ohm, it may be determined liy the 

 substitution of a standard coil for P. If so found, the probable error is diminished, 



W 



w 



Fig. 8. 

 since accurate knowledge of the temperature coefficients of the coils n. I!, and S is 



unnecessary. 



I'lir substitution <>f a men-my standard for P causes the value of d to IK- much 

 greater than usually attains. With standard coils, d has an average value of 

 U'OOOOG ohm, but with the mercury tubes it approximated to 0'15 ohm. 



