94 PROCEEDINGS OF THE AMERICAN ACADEMY 



of Cross and Hayes, but their results are in substantial accordance 

 with his for the same conditions of experiment, although the range 

 of thickness in the diaphragms used by them was probably too small 

 to produce the peculiar alternations of effect noticed by the former 

 observer. 



It seemed to the present writers desirable to verify these conclu- 

 sions by direct observations upon the motion of the diaphragm itself, 

 and the investigations detailed were undertaken with this end in view. 

 We have thus far considered only the first of the two propositions 

 just stated, leaving the second for future study. The experimental 

 work was completed during the spring of 1891. 



As it is very desirable not to have the free motion of the diaphragm 

 interfered with in any way, we decided to make use of the stroboscopic 

 method of observing its vibrations. 



The telephone which we employed as a receiver had for its polar- 

 izing magnet a bar of Norway iron three fourths of an inch in diameter, 

 and eight inches long, which was surrounded by a magnetizing coil 

 consisting of 2750 turns of No. 18 (B. & S.) copper wire, whose 

 resistance was approximately eight ohms. The line coil surrounding 

 the end of the core was made of No. 36 (B. & S.) copper wire, and 

 had a resistance of 99.5 ohms. The core was capable of being moved 

 longitudinally by means of a screw, so that the distance of its end 

 from the diaphragm could be adjusted. In all of our experiments 

 the distance between the core and diaphragm was kept at ^ of an 

 inch. The strength of the magnet was varied by means of a resist- 

 ance which altered the current passing through the magnetizing coil. 



Since it seemed very doubtful whether the excursion of the dia- 

 phragm of the receiver when a current so feeble as that of a telephone 

 was used would be sufficiently great to allow of satisfactory measure- 

 ment, we decided to employ the alternating current from the secondary 

 of a transformer through whose primary was passed the current from 

 an ordinary alternating current dynamo making 128 complete alterna- 

 tions per second. By the introduction of a suitable wire resistance of 

 variable amount into the secondary circuit, the current flowing in it 

 could be reduced to a convenient strength, so that when the magneto- 

 telephone was placed in a derived circuit running from extreme points 

 on this wire the coils were traversed by a current comparable in 

 magnitude with the ordinary telephone current. By operating with 

 a current thus produced, and passing through the coils of the receiver, 

 we hoped to be able to employ a current somewhat, but not very 

 greatly, larger than the ordinary telephone current, so that we might 



