rKor.ic.mox or electric tr.irrs orEU run eaktii 217 



meters there is Rrcat irrcKnIarity in transmission so that rcliaMe 

 rommunication over land for distances as short as 100 miles is not 

 always (lossiMe even with larjje amounts of power. With decreasinR 

 wave length we tnul also variations in apparent direction of the wave. 

 On the other hand, as the wa\e length is decreased still further we 

 luul, sometimes, rather surprising increases in range and stability. 

 The nature of the fading changes, becoming more rapid, and the 

 absorption in many cases seems to decrease. This peculiarity of 

 wa\e transmission must be explained in a siitisfactory theory. In 

 addition to the apparent selective effect just mentionetl, some observa- 

 tions indicate that there are often differences between east and west 

 and north and south transmission at all wave lengths. 



The various irregularities in radio transmission, and particularly 

 the apparently erratic and anomalous behavior of electromagnetic 

 waxes occurring in the neighborhood of a few hundred meters wa\-e 

 length seem to indicate that as the wave length is decreased from a 

 value of several kilometers to a value of a few meters some kind of 

 selective effect occurs which changes the trend of the physical phe- 

 nomena. These considerations have suggested to us the possibility 

 of finding s<ime selective mechanism in the earth's surface or in the 

 atmosphere which becomes operati\e in the neighborhood of 200 

 meters. A rather superficial examination of the possibility that 

 such a selective mechanism ma\' be found in a possible distribution 

 of charged particles in the atmosphere has resulted in the conclusion 

 that a selective effect of the required kind cannot be produced by 

 such a physical mechanism. There is, however, in the earth's atmos- 

 phere — in .iddition to distributions of ions — a magnetic field due to 

 the earth, which in the presence of ions will have a disturbing effect 

 upon an electromagnetic wave. As is well known, a free ion moving 

 in a magnetic field has exerted upon it, due to the magnetic field, 

 a force at right angles to its velocity and to the magnetic field. If 

 the ion has impressed upon it a simple periodic electric force, it will 

 execute a free oscillation tf)gether with a forced oscillation whose 

 projection on a plane is an ellipse which is traversed in one period 

 of the applied force. The component velocities are linear functions 

 of the components of the electric field and at a certain frequency, 



depending onlv upon the magnetic field and the ratio — of the ion. 



Ill 



become very large unless limited by dissipation. This critical fre- 



Ile . . 



quency is ec|ual to if // is measured in electromagnetic units 



Zittnc 



and e in electrostatic units. It is the same as the frequency of free 



