Mr. 0. Heaviside on Duplex Telegraphy. 429 



cated instruments (as, for instance, Wheatstone's automatic 

 transmitter) is not at first sight so evident; but I have done it 

 in a very simple manner, which it is unnecessary to describe. 

 On long lines, or with high-speed instruments, an attention to 

 these minutm is desirable ; but on short lines and with common 

 Morse apparatus they are superfluous. 



It is not essential, though sometimes desirable, to use differen- 

 tially wound instruments. Most telegraph instruments are 

 constructed with two separate coils of wire, each on its own core. 

 By connecting the battery to the wire joining these coils, we 

 have a differential arrangement, and frequently all that is needed. 

 In fact, if the armature is polarized, as in most relays, the result 

 is the same as if they were differentially wound. With an un- 

 polarized Morse direct-writer, however, the effect of the out- 

 going currents would not be completely neutralized. This is of 

 little consequence, as the spring which draws the armature 

 from the electromagnets may have a tension given it that only 

 the received currents can overcome. The rheostats r and r! 

 (fig. 1) may even be dispensed with and a direct earth-con- 

 nexion substituted, provided the external resistance be not too 

 great. 



Quite recently another system has been brought forward, un- 

 deniably the most perfect, which may be called the bridge duplex, 

 its principle being that of Wheatstone's bridge. To whom the 

 idea first occurred of using this arrangement for duplex telegraphy 

 is unknown to me. It has been claimed by Mr. Eden, of Edin- 

 burgh ; but it has been patented by Mr. Stearns, of Boston, 

 U.S., who also patents a number of plans, all depending on the 

 differential system before described. 



The arrangement for the bridge duplex is shown theoretically 

 in fig. 2. a, b, c and a!, b 1 , c' are resistances, g an&g' the re- 

 ceiving instruments, and / and /' the batteries. By the well- 

 known law of the balance, when a : b = c : d, where d is the whole 

 external resistance between station A and the earth at B, the 

 electromotive force E will cause no current in g; and simi- 

 larly for station B. The circumstance that the out-going cur- 

 rents do not pass through the receiving instruments is very im- 

 portant, as it allows any description of existing instruments to 

 be used, and without any alteration. As in the differential plan, 

 it is not always indispensable to adhere rigidly to the conditions 

 which give theoretical perfection. 



Although the signals sent by station A are only received at 

 B, and vice versa, and it is convenient to assume that the cur- 

 rents producing these signals actually come from the opposite 

 station, yet it does not always happen that such is the fact. To 

 take an extreme case. Let all the apparatus at each station, and 



