190 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
skillfully made by my assistant, Mr. Dyke, proved very useful. I 
was thus able to confirm Mr. Marconi’s observations and my own 
theory of them, and furthermore noticed that when the nonvertical 
part of the transmitting antenna was bent so that it was not hori- 
zontal but pointed downwards, a 
very remarkable nonsymmetry of 
radiation occurred, quite, however, 
accounted for by theory. (See 
fig. 20.) Mr. Marconi has made 
very effective practical use of the 
bent receiving antenna to locate the 
position of a ship or station send- 
ing out  electric-wave messages 
when so far off as to be below the 
horizon. 
In this case he arranges the re- 
ceiving antenna so that a very short 
part is vertical and the greater 
part horizontal, and furthermore 
permits the horizontal part to be 
swiveled round the vertical part as a center. In the vertical portion 
he places his magnetic or some other detector. If, then, there be a 
distant station in correspondence with this receiver, the direction in 
which the transmitter lies can be determined within a few degrees 
by swiveling round the 
recelving antenna and not- 
ing the position in which 
it picks up signals or picks 
them up best from this 
transmitter. The trans- 
mitter then lies in the di- 
rection opposite to that in 
which the free end of the 
receiver wire points. If it 
is not convenient to swivel 
round the horizontal por- 
tion, then Marconi ar- 
ranges a number of hori- 
zontal receiving antenne 
like the spokes of a wheel, 
all having a common 
shorter vertical part as their center. (See fig. 21.) In the vertical 
part a magnetic detector is inserted, and by means of a switch any one 
of the horizontal radial antennz can be put in connection with it. By 
finding which radial gives the strongest signals, the direction of the 
a 
Fire. 19.—Thermal detector. 
CiMMUMU YIP A Cia 
Receiver 21 Ft. vertical 
Distance 138 Fe. 
180° 
Fig. 20. 
