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



enough near the ground to produce any effect, becomes larger as the 

 distance from the earth increases. 



The extreme ultraviolet rays of the sun ionize the air, but these 

 rays are entirely absorbed during their penetration of the upper at- 

 mosphere. In the daytime, ionization in the higher levels is therefore 

 unquestionably much greater than at lower elevations. Fleming states 

 that the conductivity of the air near the earth's surface is 100,000 

 times too small to explain the observed absorption of radio waves. 

 The ionization of the upper atmosphere, however, is so very great that 

 it is probable that the waves which penetrate into the higher parts 

 of the air in the daytime are totally absorbed, and that the observed 

 absorption of waves received at a receiving station is to be explained 

 by the losses in the poorly conducting earth itself. 



While the variations from day to day in daytime transmission 

 may not be entirely a ground phenomenon, it seems probable that 

 they are attributable to variations in ground conditions. 



Intensity and fading of signals at night. — The short radio waves 

 on which signals are heard at great distances during the night must 

 travel detached from the earth's surface, for along the surface these 

 waves are highly absorbed. It is probable that they reach an upper 

 surface of the atmosphere, which is so highly ionized that its electrical 

 conductivity is far greater than that of the surface of the earth, and 

 hence they can travel with relatively little absorption. 



The idea of an upper conducting surface between which and the 

 earth's surface electrical waves would be propagated, antedates the 

 use of radio for long-distance communication since it was considered 

 by Fitzgerald in 1893 and by Heaviside in 1900. Considerations 

 largely independent of radio phenomena suggest the following structure 

 and boundaries of the atmosphere, as indicated in figure 1. 



(1) The earth's surface, a relatively poor conductor. 



(2) The troposphere, about 10 kilometers thick, within which 

 are the causes of our meteorological phenomena, and an atmosphere 

 similar to that which we breathe. 



(3) A radioactive layer, separating the troposphere from the region 

 above it. (The existence of this layer is not as well established as 

 the others, nor is its existence so important in explanation of the radio 

 phenomena.) 



(4) The stratosphere, or isothermal layer, having a thickness of 

 approximately 100 kilometers. The stratosphere is ionized in the 



