2d2 Mr. R. T. Glazebrook on the Measurement 



of AD, Q that of D C (P and Q include the resistances added 

 to the ends of the bridge-wire), AF the resistance K to be 

 measured, F C the standard S with which 11 is to be com- 

 pared. At K in the battery circuit is a key. 



In one position of the key the battery-current has to pass 

 through the resistance p ; in the other the ends of the coil p 

 are connected, and it is thrown out of the circuit. 



The experiment is as follows : — Place the key so that the 

 resistance p is shunted, and adjust the position of D until 

 there is no current through the galvanometer. Raise the 

 key, and so throw p into the battery circuit ; then the galvano- 

 meter-needle is considerably deflected, if the other resistances 

 be small compared with p. 



The experiment may of course be conducted in the reverse 

 way, the position of equilibrium being found when p is in 

 circuit. On depressing the key and shunting p the needle 

 is deflected. 



The deflection in this case is much greater than in the 

 former; but then the battery-current is increased many times 

 by the operation, and the apparatus therefore is rendered more 

 sensitive; so that part of the effect may be due to a slight 

 error in the original position of D. 



The effect was reversed by changing the direction of the bat- 

 tery-current. This last fact showed that it could not spring from 

 any heating of the coils. Moreover any heating-effect would 

 be due to the difference in the temperature-coefficients of 

 R and S or P and Q respectively ; but R and S were through- 

 out as nearly as possible equal and similar, as also were P 

 and Q. 



But it is well known that a thermoelectric effect is produced 

 by the contact at D. This, as will shortly be shown, will ex- 

 plain the phenomenon; and the effect may be made a means of 

 measuring, approximately at least, the electromotive force set 

 up at the junction. For let E be the E.M.F. of the battery, e 

 that at the junction ; then we may show that the current 

 through the galvanometer is 



„ E(QR-SP) + g{B(Q + B + 8 + P) + (P + Q)(S + R)f 



D W 



B being the resistance in the battery circuit, and D a 

 function of the resistances with which we are not at present 

 concerned. 



The condition for no current, therefore, is 



0=E(QR-.SP) + ^B(Q + R+S + P) + (P-fQ)(S+R)} 5 



