478 BELL SYSTEM TECHNICAL JOURNAL 



on a certain day of early spring (1933) Appleton found 1.2-10^ at the 

 crown of E and 3.8-10^ at the crown of F, the F-layer being at that 

 time and place not differentiated into Fi and F^. These figures are 

 not far apart, if E be compared with E and Fi with F; but with a 

 little search I could have found plenty of values differing much more 

 greatly, as is attested by Figs. 9 and 10. From the former of these 

 we have already deduced that critical frequencies vary as the sunspot 

 cycle proceeds: I now add that from minimum to maximum of the 

 cycle just ending, TV at the crown of the £-Iayer increased by three- 

 fifths while N at the crown of F2 went up no less than fourfold ! From 

 Fig. 10 we infer, by squaring the values of critical frequencies, how 

 great is the change of these iV-values with hour of day. Sudden 

 unaccountable changes also occur; one evening over Cambridge 

 (Massachusetts) the iV-value for the E-layer was more than tenfold 

 the valuies given above, being ascertained by Mimno as 2.8-10^! 



I therefore summarize, as precisely as seems justifiable: the N- 

 values at the crowns of the layers vary with hour of day and time of 

 year and year of the sunspot-cycle very markedly, not to speak of 

 sudden unexplained fluctuations; and 10^ to 10^ electrons per cc. is a 

 good figure to keep in mind for the order of magnitude thereof. 



To terminate this section I show Fig. 12, in which the delay of the 

 echo for a certain frequency (2 mc.) is plotted against time during the 

 hours preceding and following dawn. Interpreting with the aid of 

 Fig. 3A : during the night the -E-peak was too low to echo back the 

 signals of this frequency, which accordingly climbed farther and found 

 their mirror in F; but at 6:35 a.m. very sharply, the E-peak increased 

 in height to just the extent needed to intercept them. Or since 

 confusion may arise from using the word "height" in two senses, 

 I express what went on in an exacter way: during the night the electron- 

 density at the crown of the £-layer was inferior to the mirror-density 

 for/ = 2 mc, but with the oncoming of day it rose, and at 6:35 a.m. 

 very sharply it attained and overpassed that mirror-density. 



I recall that Fig. 1 1 exhibits how the virtual altitudes of the layer- 

 crowns vary with hour of day and season of the year in the sky over 

 Washington. It is evident that £ is a fixture of the ionosphere with 

 a virtual height surprisingly steady at the close neighborhood of 120 

 km, while F rises and falls in the course of a winter day, rises and 

 falls and divides and merges again during a day of summer. 



Now we must take brief notice of a difficult subject: the forkings 

 and the doublings of the iP',f) curves, and the theory which finds 

 their source in the earth's magnetic field. 



