484 BELL SYSTEM TECHNICAL JOURNAL 



Being discovered however some years before the war, it was and is 

 called a "fadeout." The apparatus was that which I mentioned on 

 page 471 ; accordingly each of the pictures was traced in fifteen minutes, 

 and as soon as each was finished the next was begun. See how the 

 pattern of (P',f) curves, familiar and sharp in the earlier pictures, 

 dissolves into fragments and then is completely wiped out ! Later on 

 it begins to come back piecemeal, and finally is restored as good as ever. 



Since attention was focussed on such events in 1935, they have been 

 reported by the scores in every year, varying in duration and in 

 severity. It requires no (h',f) curve to show them, since ordinarily 

 they cut off communication by radio, and with the sharpness of a 

 knife. Many an engineer, to quote from Dellinger, has "dissected 

 his receiving equipment in the vain effort to determine why it suddenly 

 went dead." Over broad areas the extinction is sudden and simul- 

 taneous in many fadeouts, more gradual in others; the restoration is 

 as a rule more gradual. 



Shall we interpret this strange and striking effect as a sudden 

 vanishing of the ionosphere and all the reflecting layers thereof, or as 

 a swallowing-up of the signals by something which is suddenly created 

 underneath the ionosphere? Against the first suggestion it is to be 

 said, that no one can image anything which might so suddenly frighten 

 all the electrons of the ionosphere back under cover, so to say — 

 drive them into the arms of their parent molecules in a few seconds 

 or minutes — when all day and even at night they manage to hold 

 their freedom. Such a graph as Fig. 15 speaks also against it forcibly. 

 Here in the upper part of the figure we see the critical frequencies ^ 

 of F2, Fi and E as located every fifteen minutes on (P',f) curves such 

 as those of Fig. 14. Each flock of data lies along a curve which, 

 intercepted though it is by the fadeout, resumes so nearly at the level 

 where it left off that one can hardly believe that the ionosphere totally 

 vanished in between. ^° 



As for the curve marked "/m/jv" in Fig. 15, it represents the lowest 

 frequency at which echoes are observed. I have said nothing as yet 

 about there being such a minimum-frequency. How indeed can there 

 be one, and why should signals of any frequency however low fail to 

 be echoed, since the mirror-density for any higher frequency is a fortiori 

 more than a mirror-density for any lower? 



' For the ordinary waves, as indicated by the superscript in symbols such as /"e- 



1° In violent magnetic storms the ionosphere is so convulsed that the echoes 



lose their sharpness entirely, and ih',f) curves like those of Fig. 7 are replaced by 



broad smudges; or echoes may vanish altogether. These are quite different from 



fadeouts. 



