246 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 11, NO. 11 



are: (a) frequency selection, {b) selection of wave direction, and 

 (c) guiding the waves along a channel; for example, along a wire. 

 It is interesting to note that the first and last of these means for elim- 

 inating interference are also used in sound, while the second, the 

 selection of the direction, cannot be used in sound, since sound waves 

 are longitudinal while the electric waves are transverse. The matter 

 of station interference will not be discussed further in this paper. 



In regard to fading and to strays, a considerable literature now exists, 

 much of it being contradictory. The present authors have attempted 

 to coordinate some of the outstanding and fairly well established 

 facts, and to present recent results. Because of the complexity and 

 uncertainty of many of the phenomena, exceptions can be found to 

 almost all generalizations. 



Complexity of the problem. — The following four functions are 

 characteristic of radio reception: 



Intensity of received signals. 

 Fading of signals 

 Direction of signals. 

 Strays 



In order to determine the facts regarding these functions, from 

 the analysis of which the causes might be determined, research has 

 to be undertaken to determine their behavior with respect to the 

 following variables: 



Frequency of wave, 



Kind of wave (as continuous or damped). 



Distance, 



Place (e. g., sea, sea-shore, plain land, mountainous land, rivers). 



Time (e. g., day, night, sunrise, sunset, time of year), 



Occurrence of eclipse, 



State of solar activity. 



Atmospheric electricity, 



Terrestrial magnetism, 



Miscellaneous (as discontinuities observed in space and in time) . 



These variables are by no means independent; for instance, the 

 frequency and kind of wave affect differently the phenomena observed 

 at various places and distances. A complete study of the problem 

 would require about 500 separate researches, each of them on a large 

 scale. The completion of some of these researches would cost millions 

 of dollars. The entire mass of data at present available can be con- 

 sidered as completing only a very few of these researches. The known 

 facts about the variation of the four functions with a few of these 

 variables will now be summarized. Following that, a theory will 

 be given interpreting these facts. 



