276 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1955 



Apparently it was thought by some that the use of higher voltages 

 would tend to ovei'come the capacity effect. This may have been part 

 of the reason why very high voltages were applied to the first trans- 

 atlantic cable. The insulation was seriously affected, making the cable 

 unworkable and discouraging further efforts for several years. How- 

 ever, as noted, by a more complete understanding of the physics of 

 the problem, under the leadership of Lord Kelvin, terminal equipment 

 and operating teclmiques were devised which imposed lesser stresses 

 on the cable insulation. Reliable transoceanic cable telegraphy then 

 came into being (1866). 



Even then the operating speeds were only of the order of tlu-ee words 

 per minute — very modest, but nevertheless an achievement, considered 

 in the light of the fact that the alternative at the time was no direct 

 communication, only a delay of days, if not weeks, in getting messages 

 across the Atlantic via ship. 



How far such a telegraph cable falls short of serving for telephony 

 will be appreciated from the fact that the current fluctuations, or range 

 of frequencies, required for a single speech path are roughly one thou- 

 sand times as great as the slow-speed telegraph signals. It is not sur- 

 prising therefore that it took many decades to achieve performance of 

 this order. 



In the meantime, the communication needs across the Atlantic were 

 being augmented from time to time, until there are now some 20 tele- 

 graph cables, as well as numerous radio-telegraph and more than 15 

 radio-telephone circuits, between the two continents. 



STEPS TOWARD THE ULTIMATE TELEPHONE CABLE 



Beginning in 1919 a small development group was organized under 

 Dr. Oliver E. Buckley at the Bell Telephone Laboratories to study the 

 problem of increasing the transmission capacity of long deep-sea 

 cables. These studies led to several important developments. 



1. — Experiments were made and cables successfully laid using "load- 

 ing" in order to improve their transmission capabilities. Loading is 

 a process of adding "inductance" along the cable in order to 

 offset tlie effect of the capacity effects which have been previously 

 noted. Inductance may be added by inserting coils of wire at intervals 

 in the cable, or by surrounding the conductor with tapes of some mag- 

 netic material. A special magnetic material called perminvar was de- 

 vised for this purpose. 



The addition of loading to cables greatly reduces the effect of ca- 

 pacity and makes it possible to use much longer cables for a given 

 amount of distortion, or it may permit higher signaling speeds for a 

 given cable. However, one difficulty with loading is that it imposes 

 an upper limit on the speed of operation. In other words, it limits the 



