COIL BRIDGE. 361 



galvanometer and of the battery, may be opened or closed at 

 pleasure, either by two distinct keys, or by a double contact key. 

 By means of a special plug K the circuit of the dials may be 

 opened, and thus a resistance be introduced which is practically 

 infinite. 



The galvanometer is usually one of high resistance with an 

 astatic needle. It is provided with a shunt, and the sensitiveness 

 may be modified by an auxiliary magnet. 



The beam CAD is a strong bar of copper of about 2 centimetres 

 in the side, divided in several pieces, which may be connected by 

 plugs, and between which the coils are placed symmetrically in 

 pairs. In order that the two coils of the same order shall be at 

 the same temperature, the corresponding wires are coiled simul- 

 taneously on the same core. The resistances of these coils are 

 respectively equal to 10, 100, 1000, and 10,000 ohms. By re- 

 moving a plug on each side, equal resistances are brought into 

 the two arms if the plugs are symmetrical, or, if not, resistances 

 are introduced in the ratio of one of the values 10, 100, 1000. 

 If all the plugs are left, the special resistance of the beam would 

 be so slight that the galvanometer would give no indication. If 

 only one plug is removed, the resistance of one branch would 

 be virtually zero compared with the other. 



The box gives directly resistances in whole numbers from i to 

 10,000 ohms ; but by the play of the beam by which the two arms 

 a and a' may be put in the ratio i, 10, 100, and 1000, or their 

 reciprocals, we may measure all resistances from o'ooi to 10,000,000. 



The balance is only arranged to about a unit with a number of 

 four figures given by the four dials. But we may push the approxi- 

 mation further. We may observe the displacements 8 and 8' on 

 either side (permanent deflections or swings) which the needle 

 experiences for two successive numbers n and n+i of the box 

 between which is comprised the value corresponding to the resist- 

 ance sought. The displacements of the needle being proportional 

 to the error of adjustment, we may take for x the closely 

 approximate value 



8 



In the case of mean resistances, we begin by getting a balance 

 to within a unit, with equal arms and shunting the galvanometer 



to - ; then the ratio of the arms of the balance is multiplied 

 1000 



