164 



BELL SYSTEM TECHNICAL JOURNAL 



and minima and as the paths change the plane section ot the pattern 

 taken by the surface of the earth wanders so that at any one point 

 the field is continually fading in and out as the maxima and minima 

 glide by it. Each frequency has its own pattern differing from those 

 of its neighboring frequencies in such a way that at any given point 

 the relation between amplitude and frequency is that just discussed 



16 n le 13 10 !i n n !< a K n it> i9 x) v 52 53 » 55 » Ji 58 59 40 



Fig. 14 — Plotted curves of signal amplitudes condensing a long fading record, part of 

 which is shown in Fig. 13. Numbers along time axis correspond to successive 25 



second timing interruptions 



above. Thus as the paths change and the patterns shift the different 

 frequencies fade not simultaneously but progressively. 



In the above analysis of wave interference it has been assumed 

 that all frequencies traveled from transmitter to receiver over a 

 given path in the same elapsed time. This does not mean that they 

 necessarily follow exactly the same route on this path since they 

 might follow somewhat different routes of equal length or if their 

 transmission velocities were different they might follow different 

 routes of unequal length and still come within the definition of a 

 "path." It seems reasonable to assume that over the width of an 



