TRANSATLANTIC RADIO TELEPHONY 251 



with the time of day, being stronger at night. Second, the amount 

 of radio noise present is usually less in the morning and increases 

 towards the evening and well into the night. It is not the absolute 

 strength of the signal which is controlling, but the extent to which 

 it dominates the noise, therefore the ratio between the signal and the 

 noise is the thing which indicates the satisfactoriness with which 

 communication can be carried on. While signal transmission does 

 not change widely between summer and winter, the amount of noise 

 present in the summer time is usually very much greater than that 

 in the winter time, so that the difficulties of communication in the 

 summer are greatly increased. 



It soon became apparent that the amount of increase in the signal- 

 to-noise ratio which would be necessary to put conversation on any- 

 thing like a practical basis would be so great that to hope to get it 

 by increasing the power at the transmitting end was quite out of the 

 question. Thus improvement had to be looked for at the receiving 

 end and the problem became one not of increasing the signal strength, 

 but of decreasing the amount of noise which was allowed to get into 

 the receiving set along with the signal. There are three known ways 

 of decreasing the effect of static in a case of this kind. 



Since static is distributed over the entire frequency range, the first 

 and most obvious one is to use such high selectivity that only the signal 

 frequencies are permitted to enter the receiving set. This reduces the 

 amount of static to that which is encompassed by the frequency band 

 occupied by the signal.^ If suitable band filters are employed it is 

 possible to obtain a degree of selectivity such that practically all static 

 noise which can be eliminated by this method is prevented. 



A second method of reducing the amount of static is to employ re- 

 ceiving antenna systems which are directional, in other words, systems 

 which are receptive only to signals coming from the direction of the 

 transmitting station and are blind to interfering signals or interfering 

 static coming from other directions. The most practical system which 

 has so far been developed for doing this at long wave-lengths is the so- 

 called wave antenna.^ This consists of an open wire line three or 

 four miles long which is grounded at both ends in the characteristic 

 impedance of the wire-to-ground circuit. Thus it is substantially an 

 aperiodic system, there being no reflections at the terminals. Radio 

 waves which approach this line from the side produce relatively very 



* "Selective Circuits and Static Interference," John R. Carson: Bell System 

 Technical Journal, Vol. IV. No. 2, April 1925, pp. 265-279. 



^"Wave AntennjE," H. H. Beverage, Chester W. Rice and E. W, Kellogg: 

 Journal of the American Institute of Electrical Engineers, Vol. 42, March 1923, pp. 

 258-269; May 1923, pp. 510-519; and July 1923, pp. 728-738. 

 17 



