1906.] 071 Hoiv to Improve Telepliony. 27b 



trate how very complex the movements of the air particles become in 

 the case of speech. 



The problem in telephony is the accurate reproduction at a distance 

 of these complex vibrations of the air : the more nearly the movements 

 of the air at the receiving point correspond to those at the trans- 

 mitting point, the better will be the quality of the telephony. If the 

 movements at the receiving station are similar in pattern but of less 

 amplitude than tliose at the transmitting station, then we have simply 

 attenuation of the sound ; if, however, the sound pattern at the 

 receiving point is distorted, then loss in articulation takes place. It 

 happens, very luckily, that the ear has a wonderful power of recog- 

 nising a sound pattern even when considerably distorted ; if it were 

 not for this latitude, as we may call it, in transmitting our sound 

 pattern, telephony would not be practicable over anything like the 

 distance already attained. 



Let us now come back to electric telephony, and examine what takes 

 place between the original air movements and their final reproduction 

 at a distance. The movements of the air are first converted into 

 movements of a diaphragm, which movements are again mechanically 

 transmitted to the carbon grains in the microphone, thus altering 

 its electrical resistance. The varying resistance of the microphone 

 causes the current through it and the connected transformer to vary 

 and so induces varying cmTents in the secondary of the transformer. 

 This secondary is connected to the line, so that the currents are 

 conveyed to the receiver at the distant place. The cuiTents are here 

 transformed into a varying magnetic field which acts on a diaphragm 

 and causes it to vibrate and thus start the air around it in movement. 

 In all this long train of transmissions and transformations the 

 character of the original sound pattern must be preserved sufficiently 

 well to enable the ear to recognise it in its final form. When we 

 consider that at every one of these steps distortion and loss of energy 

 must take place, it is not surprising that there are difficulties in the 

 way of telephony. In fact, it is a matter for wonder that electric 

 telephony is possible at all. 



Telephony has already reached a very high degree of excellence, 

 how can it be improved ? At every step in the long train of trans- 

 formations we must inquire : What are the losses ? What are the 

 distortions introduced ? How can we avoid them ? Answers to 

 these questions can only be given by systematic accurate measure- 

 ments and experiments. 



If one consults the literature on telephony, one is surprised how 

 little quantitative data is available on any given point in comparison 

 with that in other branches of engineering. Experimenters of 

 late seem to have avoided telephony ; dozens of investigations are 

 published on the efficiency of induction motors, and hardly one on the 

 efficiency of telephone induction coils ; yet while the former are made 

 in thousands, the latter are made in millions. Telephone engineers 



T 2 



