CEKr.iix f.icTOKs .irrr.cTixc riu.Ec,u.\ru sri-.v.n xu 



cahli- is limited to a lUruiiti- v.iliio. Mon-Dvir, for (ritaiii reasons, llic 

 cable has an iiuiK-dancu associated wiili it at the sending end wliich 

 may make the voltage on the cable dilTer from the voltage applied 

 to the sending-cnd apparatus. Inasmuch as the limitation in thTs 

 case is voltage limitation at the cable, the ideal wave is one which 

 applies a rectangular wave to the cable rather than to the apparatus, 

 because it insures that the area under the curve should be the maxi- 

 mum consistent with the imposed limitations. It would be possible 

 to make the transmitting-end impedance approximately propor- 

 tional to the cable impedance throughout most of the important 

 range. This would insure that the wave applied to the cable would 

 have approximately the same shape as the wave applied to the ap- 

 paratus. It would probably be desirable for practical reasons to 

 make this impedance infinite for direct current. 



In connection with the submarine cable a special kind of inter- 

 ference is particularly important, namely, that due to imperfect 

 duplex balance. For a given degree of unbalance, the interference 

 due to this source may be reduced by putting networks either in the 

 path of the outgoing current or in the path of the incoming current. 

 These facts, together with the frequency distributions deduced above 

 for each of the several impressed waves as exhibited in P'ig. 2, make 

 it apparent that the beneficial reaction on the effect of duplex un- 

 balance, which can be obtained by the use of a half-cycle sine wave 

 instead of a rectangular wave, can be obtained more efTectively 

 by the use of a simple network, either in the path of the outgoing 

 or in the path of the incoming currents. Hither of these locations is 

 equally effective in reducing interferences from duplex unbalance, 

 but the location of the network in the path of the outgoing current 

 has the advantage that it decreases the interference into other circuits, 

 whereas the location in the path of the incoming current has the effect 

 of reducing the interference from other circuits. 



Before leaving the matter of submarine telegraphy, it may be well 

 to point out that it is common in practise to shorten the period during 

 which the battery is applied so as to make it less than the total period 

 allotted to the signal element in question. For instance, if it is 

 desired to transmit an e the battery may be applied for, say, 75 per 

 cent, of the time allotted to that e and during the remaining 25 per 

 cent, the circuit is grounded. The resulting voltage is shown in 

 Fig. 3F. From the foregoing, it is concluded that this method is less 

 advantageous than the application of the voltage for the whole period, 

 because while the shape of the received signal is substantially the same 

 in the two cases, the magnitude, being proportional to the area under 



