88 



THE BELL SYSTEM TECHNICAL JOURNAL, JANUARY 1952 



of short duration sometimes observed when the average signal level was 

 depressed by the mechanisms of Figs. 3(b) or 3(d) were found to be fre- 

 quency selective. 



Some of the studies described in this paper were made with vertically 

 polarized waves and some with horizontally polarized waves; at times, 

 45° polarization was used. In so far as it was possible to determine, the 

 propagation characteristics of both paths were independent of the polari- 

 zation used. 



No meteorological soundings were made in connection with this work. 

 Considering the rapid changes usually observed with the angle-of -arrival 

 and frequency sweep apparatus, it is doubtful that meteorological meas- 

 urements made in the usual manner would show much correlation with 

 the radio observations except, perhaps, in a general way. The sequence 

 of pictures in Fig. 10 is included to show how the angle-of-arrival and 



0.8 



0.6 

 0.5 

 0.4 



0.3 



O 0.01 

 £ 0.008 



0.006 

 0.005 



-5 -to -15 -20 -25 -30 -35 -40 



SIGNAL LEVEL RELATIVE TO NORMAL DAYTIME VALUE 



IN DECIBELS 



Fig. 13 — Time distribution curves of the signal levels observed on the Murray 

 Hill-Crawford Hill path. Data of 1947, 1948, 1949 and 1950. 



