DR. GEORGE A. CAMPBELL 555 



telephone repeater. Here, an entirely new line problem arose,— 

 namely, that of avoiding singing when repeaters are adapted to two- 

 way amplification. Up to 1912, the only type of repeater circuit used 

 was the so-called 21-type, in which a single repeater element ampli- 

 fies messages which reach it from both directions and which requires 

 that the two associated sections of line have very similar character- 

 istics. A well-known limitation of the 21-type repeater is its tend- 

 ency to "sing" when line unbalance or amplification exceed certain 

 rather low limits. 



On the other hand, the 22-type repeater has two amplifying elements 

 and two artificial lines, one to balance each associated section of actual 

 line. While the basic idea of the 22-type was old, it remained for 

 Campbell, in a memorandum dated March 7, 1912, to reveal its prop- 

 erties of inherent stability. He points out that "singing will not be 

 introduced by any possible unbalance however large, in either of the 

 lines, provided the unbalance of the other line does not exceed a certain 

 critical magnitude." Also his words, "the use of a compensating 

 device such as an artificial line, to reduce the amplification at the 

 resonant frequencies to the level of the amplification at other telephonic 

 frequencies" suggest broadly the idea of equalizing for amplitude- 

 frequency distortion which is brought in by the selective characteristics 

 of the line circuits or other apparatus in a long system. Furthermore, 

 "if it became necessary merely to eliminate certain frequencies lying 

 outside of the range required for telephony, the use of an artificial 

 selecting circuit" is definite anticipation of the use, subsequently 

 common in all repeaters, of low-pass filters to cut off frequencies outside 

 of the band transmitted and thus minimize line balance difficulties. 



Campbell also indicated that the stability of a circuit as regards 

 singing could be improved if the amplification were distributed among 

 a number of properly spaced points along the line rather than concen- 

 trated at a single point. 



Moreover, the great amplification made possible in telephone circuits 

 by the perfection of the vacuum tube and its associated circuits 

 permitted the use of cables for long distances with manifest advantages 

 for congested routes. The great amplifications required for the longer 

 cable circuits, however, could most effectively be handled by the use 

 of "four-wire" circuits both for voice frequencies and later for carrier 

 systems. Campbell was the originator of this type of circuit. In the 

 same memorandum of 1912, which discussed the 22-type repeater, he 

 suggested it as the logical extension of the one-way paths in the 22-type 

 repeater, each path containing as many one-way amplifiers and line 

 sections as desired. 



