116 KENNELLY-TAYLOR— EXPLORATIONS OVER [April 22, 



two magnetic poles are indicated in dotted lines. It will be seen 

 that while the mode of motion is essentially fundamental, the ampli- 

 tude is not a maximum at the center, as in the ordinary acoustic 

 case. The maximum amphtude of 2.0 /a is reached in an elliptical 

 loop embracing the pole at the top. Inside this loop, and imme- 

 diately over the pole, the amplitude falls off to 1.8 /x. Over the 

 pole underneath, the amplitude is about 1.7 /a, but there appears to 

 be a slight diminution between the poles. If the geometrical and 

 magnetic conditions of the bipolar system were perfectly sym- 

 metrical, these dissymmetries would presumably disappear. 



The curves of mean amplitude against radial distance are pre- 

 sented in Fig. 14. The curve AAA corresponds to that found at 

 resonance, and shows that the amplitude is far from being a maxi- 

 mum at the center of the diaphragm, owing to the attractive forces 

 being estabhshed over polar areas on each side of the center. The 

 coefficient of equivalent mass for this curve is over 0.5. 



The curve ABB gives the corresponding distribution of mean 

 azimuthal amplitude for the frequency of 974'~'- The swelling 

 of the amplitude over the poles is less marked in this case, and does 

 not materially exceed that at the center. The equivalent mass co- 

 efficient for this curve is 0.36, or about double that for the Rayleigh- 

 Bessel curve case, which is indicated by ADD. The curve ACC 

 gives the distribution of mean amplitude in radial distance, for an- 

 other steel diaphragm (No. 3) in a bipolar telephone receiver, at 

 the resonant frequency of 1,020 '—'. 



For both steel diaphragms Nos. 2 and 3, a series of central 

 amplitude measurements were made, with the explorer, at constant 

 ahernating-current excitation, but adjustably varied frequency. 

 Simultaneous measurements were made by Mr. H. A. Affel, of the 

 resistance and inductance of the telephone-receiver coils, with the 

 diaphragm both free and damped. The explorer measurements in 

 both cases satisfactorily checked the electrically deduced velocity- 

 circle diagrams. It is proposed to report upon the electrical meas- 

 urements in another paper. Moreover, starting with the ampli- 

 tudes, measured at the center of the diaphragm, in curves A and C 

 of Fig. 14, the equivalent masses of the diaphragms, computed from 

 the electrical measurements, agreed, within a few per cent., with 

 those found by integrating curves A and C. 



