RATING TRANSMISSION PERFORMANCE 129 



rating' method here described, by using as a reference a practical circuit, 

 it is possible and practicable to make direct comparisons of the service 

 performance of the reference circuit, or circuits having closely similar 

 characteristics, with the performances of various commercial circuits. 

 The maintenance of the first advantage will require, however, 

 changes in the working reference circuit as material improvements are 

 made in the transmission characteristics of the commercial circuits. 

 To obtain the second advantage means the use at present of carbon 

 transmitters in the working reference circuit. These are open to the 

 same objection here as they were in the Standard Cable Reference 

 System, namely, the difficulty of exactly specifying their performance 

 raises questions as to the reproducibility of their performance from 

 time to time. This was one of the major reasons for the replacement 

 of the Standard Cable Reference System as the basis for volume ratings 

 by the Master Reference System for Telephone Transmission with its 

 specifiable performance. To preserve the first advantage mentioned 

 and at the same time to obtain a reference system whose reproducibility 

 can be assured, it is the purpose, as more complete correlations are ob- 

 tained between transmission characteristics and service performance, 

 to associate with the Master Reference System, the means to make its 

 transmission characteristics meet the requirements necessary' to retain 

 the first advantage. Meanwhile the Master Reference System will 

 continue its function as a reference for volume ratings. 



Determination of Ratings 



To provide the basis for such a system of effective transmission 

 ratings as has been outlined, several series of tests hiive been made, the 

 most comprehensive of which has been under way for more than a year 

 between several hundred stations in the American Telephone and 

 Telegraph Company headquarters building and a similar number of 

 stations of the Bell Telephone Laboratories, between which there is a 

 large amount of intercommunication. The connections between these 

 stations are handled over special trunks in which the attenuation, cut- 

 off frequency and line noise can be varied. At the stations, different 

 types of instruments and station circuits have been employed. Ob- 

 servers are connected to each of these trunks who monitor the conversa- 

 tions over them and note the number of repetitions requested in each 

 conversation and also the duration of the conversation. In this way is 

 determined the repetition rate for a number of conversations between a 

 number of different people for the various combinations of circuit 

 characteristics so provided. Thus ratings are established directly of 

 such effects as those of trunk cutoff, noise on the trunk, different types 



