478 BELL SYSTEM TECHNICAL JOURNAL 



range. Thiy rxtcnd later into the fall for the lower fretiiiencies, and 

 hence arc in e\idence on the date mi which these measurements 

 were taken, October-November. 



2. In genera! the noise is greater the lower the frequency. 



Noise as a Function of Frequency and of Receiving Location. The 

 distribution of static noise in the frequency range under consider- 

 ation is depicted in Fig. 15 for the case of reception at New Southgate, 

 England. The set of full-line curves is for daylight reception and the 

 set of dash-line curves for night-time reception. The values obtaining 

 during the transition period between day and night have been ex- 

 cluded. For both conditions three curves are shown, one the average 

 of the summer months, another the average of winter months and the 

 third, the heavj' line, the average for the entire year. The curves 

 represent averages for all of the measurements taken during both 

 1923 and 1924. In considering curves of this type it should be re- 

 membered that they represent an average of a wide range of condi- 

 tions and at any one time the distribution of static may differ widely 

 irom that indicated by the curves. Also it should be realized that 

 the extreme difference between winter and summer static is much 

 greater than the difference between the averages. 



A similar study of frequency distribution was made at two locations 

 in the United States, Belfast and Riverhead. The results obtained 

 at these two locations together with those for New Southgate, England, 

 are presented in Fig. 16 for a period during w'hich data were obtained 

 for all three places. The similarity of the three sets of curves shows 

 that there is an underlying cause common to both sides of the Atlantic 

 which may account for the ditference between the daytime and night- 

 time static on the longer waves. It will bee\ident from the curves that 

 for frequencies around 20 kc. there is not \ery much difference between 

 the day and night static noise but that at the higher freciuencies 

 in the range studied, the dajlight values become considerably less 

 than the night-time values. Actually the divergence betw-een the 

 night-time and the daytime noise curves up to about 40 kc. is an 

 exponential one. This suggests that the lowering of the daylight 

 values may be largely due to the higher absorjition which occurs in 

 the transmission medium during the day. There is a further inter- 

 esting point to be noted concerning both figures, namel>-, ihai the 

 night-time values decrease exponentially with increase in freiiiiency. 

 Since these night-time values are but little affected by absorption 

 in the transmitting medium, the ilistribution of the static energy as 

 received, alsfi roughly represents the distribution of the static power 

 generated. 



