2 BELL SYSTEM TECHNICAL JOURNAL 



behavior of the long-wave circuit during the sunset period was needed, 

 and it was for this purpose that many of the observations reported in 

 this paper were made. 



An analysis of these observations indicated that, aside from their 

 practical application, some rather fundamental information concerning 

 the probable mechanism of transmission on long waves could be 

 obtained. To these observations other data were added, and all of 

 this available material was systematically studied to determine the 

 effect of the cessation of the sun's active rays on radio transmission 

 at long wave-lengths. Although the results alone are rather incon- 

 clusive, they do provide sufficient evidence to indicate that the mecha- 

 nism of long-wave transmission is in some ways the same as that of 

 short-wave transmission and for the longer paths depends primarily 

 on waves returned to the earth by layers in the atmosphere. 



No attempt will be made in this paper to review the present status 

 of radio transmission theory or of related geophysical phenomena, 

 except in special cases where required to show the significance of our 

 results. A background of the theory has been provided by many 

 investigators, among whom Smith,^ Pedersen,^ Anderson," Appleton,^ 

 Green,^ Namba,' Yokoyama and Tanimura,^ Hollingworth ^ and 

 Heising ^° should be mentioned. 



The analysis of the data taken during the present investigation indi- 

 cated that the connection between solar and radio phenomena is 

 effected through the agency of electromagnetic radiation, and not by 

 means of low velocity corpuscular solar emission. A partial corrobora- 

 tion of this conclusion was secured by Schafer and Goodall,"- ^^- ^^ the 

 U. S. Bureau of Standards,!" and others during several solar eclipses. 

 The rather complete analysis of data taken during eclipses which has 

 been made by Appleton and Chapman ^^ also confirms this view. 



Method of Measurement and Estimated Precision 

 The field strength of special 60-kc. test dashes transmitted from 

 WNL, Rocky Point, New York, was measured at Houlton, Maine, 

 Chatham, New Jersey, and Cupar, Scotland. The Houlton and 

 Chatham measurements were made by means of meter comparisons 

 with a calibrated local oscillator and it is believed that their relative 

 accuracy is of the order of ±0.1 db, although the absolute accuracy 

 probably falls short of this figure by a considerable margin. Due to 

 the comparatively slow rate of long-wave field variation, no effort was 

 exerted to obtain better timing than ± 10 seconds. Since weak fields 

 and high noise are common during the late afternoon hours on the 

 transatlantic path, it is believed that the precision of the Cupar 

 measurements necessarily falls short of that possible for local tests. 



