188 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1907. 
researches were not pressed to such logical issue as to disclose any 
definite general scientific principle, whilst in some cases the results 
said to have been obtained are clearly in contradiction to well ascer- 
tained facts. 
Time will not permit further reference to these early and inconclu- 
sive observations. 
In March last year Mr. Marconi communicated to the Royal Society 
a paper on the radiation 
A from an antenna having 
Le a short part of its length 
Coil vertical and the greater 
part horizontal, and on 
the receptive powers of a 
similar antenna in vari- 
ous azimuths. (See fig. 
16.) He found that such a bent antenna emits a less intense radiation 
at any given distance in the direction in which the free end points than 
in the opposite direction. Also, since the law of exchanges holds good 
for electric radiators, a similar form of antenna receives or absorbs 
best electric waves which reach it from a direction opposite to that 
to which the free end points.* Hence two similar bent antenne, when 
set up back to back, that is, with their free ends pointing away from 
each other, form a system 
of radiator and receiver 
Fig. 16.—Mareconi bent antenna. 
: BE 
which has greater range 
in that position than in 
I ‘+ H+h/ +H-h 
any other for the same =O) )-@--- 
distance, and hence has 
directive qualities not 
possessed by the ordinary c DF 
vertical antenne. 
Although I have given 
the mathematical expla- 
nation of the reasons for 
this in another place,’ it 
is not difficult to trans- bese eseeecee 
late the common sense of 
it into nonsymbolic lan- 
guage. Imagine a square circuit of wire half buried vertically in 
Fig. 17.—Theory of Marconi bent antenna. 
“This is an extension to electric radiation of the principle known as Prevost’s 
Theory of Exchanges, as amplified by Balfour Stewart and Kirchhoff, which 
forms the basis of spectrum analysis laid down by Stokes, Kirchhoff, Bunsen, 
and others. 
+See “A Note on the Theory of Directive Antenne,” Proc. Roy. Soc. Lond., 
Vol. LXXVIII4, 1906, p. 1. 
