[ 909 ] 



XCIX. The Use of the Telephone as an Oscillograph. 

 By Dr. Manne Siegbahn *. 



[Plate XVI.] 



1. Introduction. 



FOR the study of variable currents, instruments are often 

 required to continually register the intensity of the 

 current. It has frequently been proposed to use the telephone 

 for such purposes, as it seems probable that the telephone 

 membrane will reproduce the variations of the current most 

 accurately. Experience shows that even small peculiarities 

 in a person's voice or speech can be recognized through a 

 telephonic transference. There are also a multitude of 

 methods suggested for the registration of the membrane- 

 vibrations, but most of them are not sufficiently sensitive 

 for reproducing the normal amplitude of the telephone* 

 membrane. In a telephone, under ordinary conditions, the 

 amplitude is of the dimension of 0*0001 mm.; and our ex- 

 perience shows that the transference of speech is good only 

 under these circumstances. By increasing the amplitude to 

 such a value that the normal registrating apparatus can give 

 measurable deviations, the accuracy of the reproduction will 

 not be the same. 



Finally, it is to be remarked that a good acoustical trans- 

 ference does not guarantee that the mechanical vibrations of 

 the membrane are identical with the electrical variations, nor 

 do they even give a sufficiently good copy of them. My own 

 eyperiments on this subject prove that a very good telephone 

 in the reproduction of sounds, in many respects, causes 

 distortion of the microphone-vibrations. 



The following paper deals with the question as to whether 

 a telephone with a normally working current-intensity can 

 be used to register variable currents. 



2. Method for the Registration of the Membrane-vibi'ations. 



For the above purpose, it is necessary to have a method 

 which can register the vibrations of, say, 0"001 mm. at a 

 maximum with sufficient accuracy. In a paper in Annalen 

 der Physik, 1913, pp. 689-728, 1 have described a method by 

 which it is possible to reproduce the membrane-vibrations 

 with the required accuracy. Since the publication of the 



* Communicated hj the Author. 



