STUDIES IN RADIO BROADCAST TRANSMISSION 



15: 



from day to day there is often evidenced a modification of the general 

 character and the recurrence of these changes. An example of this 

 change in a short period of time is well illustrated by the oscillograms 

 in Fig. 10. Strips 1, 2 and 3 form a continuous record starting at 

 1:52 a.m.; strips 4, 5 and 6 start at 2:16 a.m.; and strips 7, 8 and 9 

 start at 2:37 a.m. These are three sections of a continuous record 



^ Time 



Fig. 10 — Single-frequency fading record, showing variation in rapidity of fading, 

 made at Riverhead, L. I., July 16, 1925, 1:52 a.m. Timing marks, on strip 10, 5 



seconds apart 



selected for the purpose of showing the decrease in the fading period, 

 in a 45-minute interval. The timing interval on strip 10 which ap- 

 plies to these records is 5 seconds. In this particular record only 

 half of the audio signal was recorded, the edge of the strip being the 

 zero line. 



These single frequency fading records do not offer very much to 

 work on. There is, however, just enough suggestion of regularity 

 about them to annoy one with the thought that perhaps they may 

 follow some definite combination of periodicities and with this in 

 mind we have taken sections of two different records and subjected 

 them to a harmonic analysis. 



So far we have been able to draw no more useful conclusions from 

 such harmonic analyses than that the heterogeneous scattering of 

 harmonic values is about what one would expect from the looks of 

 the curves. 



One significant thing about these oscillographic single frequency 

 fading records is that they show no high speed fading of important 



