16 BELL SYSTEM TECHNICAL JOURNAL 



4. The amplitude of variation of the sunset cycle apparently is 

 reduced greatly at frequencies below 17 kilocycles and on paths shorter 

 than 200 km. 



5. Evidence of interference fringes on some of the observations 

 suggests the possibility that the main sunset minimum may be the 

 result of interference phenomena. 



6. The fact that the phases of the radio transmission cycle closely 

 follow the optical eclipse circumstances indicates that the radio 

 phenomena must be related to solar radiation of velocity similar to 

 that of light. 



7. On transatlantic paths during the spring and summer months 

 the average sunset minimum occurred when the sun was approximately 

 6° below the horizon at one-sixth of the total path length from the 

 eastern terminal. Since these paths varied considerably in latitude 

 and length, the phenomena may be related to effects occurring at the 

 most easterly apex of a three-reflection path. The data obtained in 

 this investigation, however, are not sufificient to definitely establish the 

 generality of this hypothesis. 



8. Empirical rules may be formulated for the prediction of the time 

 of occurrence of various phases of the sunset cycle on short transmission 

 paths. For example, the beginning of the drop in field on the 60-kc. 

 Rocky Point-Houlton path occurs at mid-path surface sunset, and the 

 minimum occurs at approximately 23 minutes after surface sunset 

 at Rocky Point. On longer paths larger fortuitous variations occur 

 and available data fail to connect the time of the minimum with sunset 

 at any point on the transmission path. Representative curves drawn 

 from the data, although subject to random errors, provide an empirical 

 method for the prediction of the approximate time of occurrence of 

 the phenomena and are of service in traffic and power scheduling. 



Acknowledgments 



The authors are indebted to the engineers of the British Post Office 

 and the operating personnel of the American Telephone and Telegraph 

 Company, for their generous cooperation in measurement schedules, 

 and to their colleagues in Bell Telephone Laboratories for their 

 assistance and constructive criticism in the preparation of this paper. 



References 



1. "Transatlantic Radio Telephony," H. D. Arnold and Lloyd Espenschied, Jour. 



A.I.E.E., August 1923; Bell Sys. Tech. Jour., October 1923. 



2. 24th Kelvin Lecture, "The Travel of Wireless Waves," by Sir Frank E. Smith, 



Inst. E.E., Vol. 9, No. 25, March 1934. 



3. "The Propagation of Radio Waves," by P. O. Pedersen. 



4. "Transatlantic Radio Transmission and Solar Activity," C. N. Anderson, Proc. 



I.R.E., \ol. 16, pp. 297-347, March 1928. 



