380 



PROCEEDINGS OF THE AMERICAN ACADEMY. 



resonance with the alternating voltage impressed on the system by the 

 periodic impact of the sound waves. This adjustment was easily made 

 experimentally. 



The proper choice of the transformer PS and the telephone T was a 

 more difficult problem. A theoretical solution of this problem was not 

 at hand, on account of lack of knowledge of the characteristics of the 

 telephone when used as a generator of oscillatory currents and on ac- 

 count of the fact that the current through the crystal in the secondary 

 is not a simple function of the voltage in this circuit (see Figure 6). 

 Some aid in the choice was had in the following considerations, which 

 served to point vaguely the direction in which experiment was to be 

 made : 



1. Since the primary circuit was to be brought to resonance with the 

 oscillations, the inductance of the primary circuit is negligible, if we 



(DM 



AV 



Figure 2. — Electric circuit. 



may neglect the reaction of the secondary circuit on the primary. With 

 this approximation it follows from elementary considerations that the 

 resistance of the primary coil should be eciual to the resistance of the 

 telephone. Experiment soon showed that the reaction of the secondary 

 circuit was not negligible, and since the effect of the reaction of the 

 secondary is to increase the apparent resistance of the primary, it fol- 

 lows that the resistance of the primary coil should be somewhat less 

 than that of the telephone. 



2. The iron core of the transformer should be such as to be properly 

 magnetizable by the current generated by the telephone, which in fre- 

 quency and intensity approaches to the current used in telephony. 

 Whence it seemed probable that the small terminal transformers used 

 in telephony would have about the proper amount of iron for use in 

 the present experiments. 



