485 
Response of Fish to Direct Current ^ — Morgan 
when not actually under observation (Fig. 2). 
Current Interrupter 
After overloading had been controlled, pre- 
liminary experiments showed that the full 
load of the generator, approximately 20 am- 
peres and 230 volts, would not attract and 
hold fish about 4 inches long in a concrete 
tank 10 feet square filled with water to a depth 
of 2 feet. When the fish swam near the elec- 
trodes, they were effectively held, but they 
would remain at the greatest distance possible 
from the electrodes or in the corners of the 
tank near the bottom. 
Thus it became obvious that, if a steadily 
flowing current was to be used, a much greater 
source of power would be needed. A similar 
observation was made by Peglow (1949). To 
attract the fish to the positive pole a much 
more efficient use of the available power would 
be necessary. The best possibility of accom- 
plishing this, according to the general reports 
of previous investigators, would be through 
the use of interrupted current (Smolian, 1944; 
Anon., 1950). 
Various mechanical devices were con- 
structed, mainly from war surplus material, 
for the interruption of direct current. First 
attempts centered around the use of relays, 
but for various reasons these did not prove 
practicable. It was difficult to determine the 
exact duration of the power-on period com- 
pared with the power-off period. When opera- 
ting at higher interruption frequencies and 
with the higher power values, the arc became 
very difficult to control, even with condensers. 
Observation of the wave form on an oscillo- 
scope showed that, if the power-off period 
was short, the arc often lasted over the entire 
period. The relay points frequently became 
heated and failed to separate. It was also dif- 
ficult to determine the exact number of in- 
terruptions per second. 
A revolving disc-type interrupter was fi- 
nally decided upon. Although not satisfactory 
■Revolving 
Detail of i nterrupfri ng disks , 
Ra kel ita -N ^T3ronz.a 
Fig. 3. Diagram of current-interrupting equipment 
used in experiments. 
in every respect, it at least overcame most of 
the objectionable features of previous equip- 
ment (Fig. 3). 
Carbon brushes bearing on either side of 
the disc provided a constant connection when 
the bronze portion of the disc was between 
them. When the disc revolved, the connec- 
tion was definitely broken when the Bakelite 
portion was between the brushes. Some arc 
was unavoidable, but an attempt was made 
to minimize this by three condensers (10 
mfd,, 600 volts, d.c.) arranged in parallel. 
A variable speed motor controlled by a ”ve- 
riac” was used to turn the disc. 
A small generator was connected directly 
to the shaft of the motor. The voltage pro- 
duced by the generator was recorded on a 
voltmeter. The relationship between voltage 
and speed was determined so that, by reading 
the voltage on the voltmeter, the speed of the 
generator and, therefore, of the motor could 
be calculated. 
