324 RADIO WAVE PROPAGATION EXPERIMENTS 
ured angles of arrival at the Beer’s Hill receiver and 
the angles of arrival computed from meteorological 
data. A grand synthesis is presented in Figure 8 of 
all days for which both BTL propagation measure- 
ments and Signal Corps tower meteorological sound- 
ings were available for comparison. Since simultaneity 
in radio and meteorological measurements was a rare 
occurrence, the angle of arrival evaluations from both 
sources of data are plotted along a time scale (in 
hours) for each day of the period. The angle evalua- 
tions from radio data are represented by circles; 
angles calculated from the meteorological data are 
represented by crosses of time length equal to the 
duration of sampling of the layer between transmitter 
and receiver (including both ascent and descent of 
the tower). Equipment limitations set the accuracy 
of BTL angle of arrival measurements at +0.04°. 
It is believed that generalized, overall conclusions 
for the entire period of comparison can be made as 
follows : 
1. That occasions of true bearing and “near true 
bearing” (say —0.11 to 0°) could have been pre- 
dicted from the meteorological data. 
2. That the occurrence of extreme deviations from 
true bearing would have been predicted from mete- 
orological data nearest in time to the radio measure- 
ments. 
3. That the magnitude of the most extreme meas- 
ured deviation (0.46°) from true bearing can also be 
calculated from observed meteorological data, though 
not simultaneously observed. 
Conclusions 
The propagation path of microwave radiation can 
be fairly well specified, given only a knowledge of the 
temperature and water vapor pressure distribution 
in the lewer atmosphere and the positions in space of 
transmitter and receiver. The equations of motion of 
the propagation of the individual wave fronts have 
been written in a form such that the angles of de- 
parture from the transmitter and the angles of arrival 
at the receiver can be evaluated directly from the 
meteorological stratification. Application of the theory 
to certain angle-of-arrival radio propagation experi- 
ments conducted by Bell Telephone Laboratories dur- 
ing the summer of 1944 has resulted in the following 
conclusions: 
1. A surprisingly good correlation exists between 
angles of arrival computed from meteorological and 
survey data only and the angles of arrival deter- 
mined experimentally. 
2. The extreme deviations (0.46°) from rectilinear 
propagation measured experimentally by BTL are con- 
firmed as plausible on the basis of observed meteoro- 
logical stratification. 
3. The meteorological analysis indicates that de- 
viations from rectilinear propagation and the fluctua- 
tion of the deviations about a mean value are as much 
a function of position of transmitter and receiver 
as they are a function of the existing meteorological 
structure. 
It is strongly recommended that low-level meteoro- 
logical soundings be considered an indispensable part 
of any experimental angle-of-arrival measurements. 
The good correlations secured between evaluation of 
angles of arrival from meteorological data and angles 
of arrival measured experimentally suggest that de- 
viations from, rectilinear propagation can be ac- 
counted for by measurable atmospheric conditions 
and that, further, these deviations can be reasonably 
well predicted. 
