TRANSMISSION FEATURES OF NEW TELEPHONE SETS 369 



itself. Inasmuch as room noise also interferes directly with received 

 speech in the telephone ear by leakage under the receiver cap, the 

 contribution to the total noise of the sidetone pickup of the anti- 

 sidetone set is in most cases small. This is not true for the sidetone 

 set, where in many cases the sidetone noise may constitute the prin- 

 cipal interfering noise. The resultant net effective gain in receiving 

 is about 2 db compared with the sidetone set. 



This is a good illustration of the type of information which can be 

 obtained only from a field study. For example, the relationships 

 indicated on Fig. 5 are dependent on how far away from the mouth- 

 piece of the transmitter and at what level the speaker talks, and on 

 how tightly to his ear he holds the receiver. These in turn are result- 

 ants of all the conditions of the particular telephone conversation. 

 If incoming levels are so high as to be uncomfortable, the receiver may 

 well be held farther away from the ear. In that event, of course, the 

 sidetone conditions of the set become relatively less controlling. The 

 weight, size, and shape of the instrument in his hands may similarly 

 affect the subscriber's use of it, the results he gets, and the relative 

 importance of various factors of telephone design. 



For such reasons, not only must laboratory performance tests be 

 supplementary and subsidiary to field tests, but additional field tests 

 must be the basis for determining the effect of any major changes in 

 design, whether or not those changes are electrical, acoustical, or 

 purely mechanical. 



Considerations of this sort emphasize the importance of having 

 clearly and explicitly in mind the conditions and relationships of 

 direct conversation, as a general reference for the interpretation and 

 explanation of the effects of telephone design on telephone conversa- 

 tions. The sidetone ordinates of Fig. 5, for example, not only suggest 

 the difference in function of the anti-sidetone circuit in transmitting 

 and receiving, but also emphasize the fact that the overall sidetone 

 resulting from the combination of circuit, instruments, and method of 

 use, is the important factor rather than the sidetone circuit efficiency 

 only. Such matters are easily lost sight of, if design is not properly 

 coordinated in its correct perspective. 



The reduction of sidetone provided by the anti-sidetone sets is of 

 further advantage in two rather different ways. 



In attaching a transmitter (which is an amplifier) and a receiver, 

 to a common handle which mechanically couples the two, a condition 

 is set up in which the gain under certain conditions may exceed the 

 loss in the path made up of handle, air, and electrical sidetone circuit. 

 Sustained oscillation, or howling, will then result between transmitter 



