410 Mr. Ervin S. Ferry on the Specific 



in the sides of a hard rubber cell, with a fixed distance of 

 about 0*3 centim. separating the disks. 



The first adjustment of the apparatus required is to secure 

 resonance of the two secondary circuits with the primary. 

 With circuits of fixed self-induction placed in inductive 

 relation with each other, resonance can be produced by simply 

 altering the capacity in one of the circuits. The adjustment 

 is known to be that giving the most perfect resonance when 

 the current oscillating back and forth in the secondary is a 

 maximum. In this experiment the current flowing in the 

 secondary circuits is determined by measuring the change in 

 resistance of a portion of the circuit produced by the heating- 

 effect of the current. If the two secondaries are singly in 

 resonance with the primary, they are in resonance with each 

 other. The condenser that is to contain the specimen is con- 

 nected to one primary circuit, the condenser removed from 

 the other secondary , and then the capacity in the primary 

 circuit adjusted till the galvanometer- needle gives a maximum 

 deflexion. Now this secondary is in resonance with the 

 primary. Then the condenser is removed from this secondary 

 circuit and inserted in the other secondary. If the two 

 secondaries have the same self-induction, this latter arrange- 

 ment will also give a maximum galvanometer deflexion and 

 in the opposite direction to that in the former arrangement, 

 but the deflexions in the two cases may not be exactly equal. 

 These maximum deflexions can be made equal by slightly 

 adjusting the distance between the rectangular circuits. In 

 these experiments the three rectangular circuits were placed 

 about 10 cm. apart. The two condensers now in circuit are 

 allowed to remain unaltered, the standard air-condenser is 

 inserted in the remaining secondary circuit and its capacity 

 altered until the galvanometer deflexion is zero. The con- 

 densers in the secondary circuits may now be interchanged 

 without altering the galvanometer deflexion. The capacity 

 of the standard condenser is now the capacity of the condenser 

 in the other secondary circuit. 



All parts of the bolometer must be carefully screened from 

 heating effects. Air-draughts and similar sudden changes 

 can be guarded against by thick coverings of cotton-wool. 

 If for any reason the temperature of different parts of the 

 bolometer-circuit is different at one time from that at another 

 time, the galvanometer-needle may be brought back to the 

 same zero-point by shunting a large resistance around one of 

 the balancing arms. In these experiments the shunt was 

 seldom less than fifteen thousand ohms. 



The diameter of the plates and the distance separating the 



