148 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1911. 
The line circuit itself was readily tuned to resonance for the par- 
ticular frequency of the dynamo by noting the maximum reading of 
the hot wire ammeter A, in the line itself. This maximum is readily 
found by varying either the capacity C or the inductance L, or both. 
At the receiving end of the line, coil L’ and the condenser C’, as well 
as the coil L’’ and the condenser C’’, were tuned to give a maximum 
intensity of signals in the receiving telephone of the audion. 
The audion, a detector of the so-called vacuum type, consists of an 
exhausted bulb containing (a) a tungsten filament maintained at in- 
candescence by a current from a local battery of 6 volts and (6) two 
platinum electrodes insulated from the filament and from each other. 
To these electrodes, one of which isa platinum plate and the other a 
platinum grid, there are applied through the high resistance receivers 
about 35 to 45 volts from a local battery. The brilliancy of the fila- 
ment is controlled by a small series rheostat, and the voltage applied 
to the insulated terminals by a local potentiometer. 
The gases in the bulb, becoming ionized by contact with the glowing 
electrode, serve as a conductor of electricity, having a high unilateral 
conductivity. If the platinum wire grid is close to the hot filament 
and the plate at some greater distance, the direction of greater con- 
ductivity is from the plate through the gas by the ionic path to the 
erid, so that if the positive terminal of the telephone battery is ap- 
plied at the plate terminal and the negative at the grid terminal, a 
sufficient current to operate the telephone will flow. 
If the terminals of the condenser of a resonant receiving circuit are 
connected to the grid and to one terminal of the filament the high 
frequency e.m‘f. impressed from this resonant circuit will cause a 
greater current to flow through the gas in one direction than in the 
other, as in the case of the direct-current potential applied through 
the telephone receiver. This rectifying effect will be reproduced in 
the telephone receivers, causing them to make audible the received 
signals. 
By changing the coefficient of coupling or the potential across the 
audion, which is adjustable, or the amount of ionization of the gases 
in the tube by adjusting the current through the filament, or any 
combination of these, it was found that the receiving operator could 
bring out the speech to suit his particular fancy. 
As stated above, the dynamo operated regularly at ranges from 
160,000 cycles per second down to 20,000 cycles per second. It was 
therefore possible to try the effect of a comparatively wide range of 
frequencies in these experiments, covering three octaves, the induc- 
tances and capacities being chosen to correspond to each particular 
frequency. It was found that more energy was delivered over this 
particular type and length of circuit by using the lower frequencies of 
