Response of Fish to Direct Current — MORGAN 
strongly than smaller fish, so an attempt was 
made to control this factor by selecting fish 
of approximately the same size for the ex- 
periments. The range in size was from 9-5 
cm. to 15.0 cm., the mean size was 11.8 cm., 
and the standard deviation was 1.04 cm. 
Procedure 
The collection of quantitative data on the 
reaction of fish to electrical stimuli was con- 
sidered highly desirable, and an experimental 
plan was formulated with this in mind. The 
main object of the experimental work was to 
cause the fish to move, involuntarily, to the 
positive pole, where it would be held, either 
by a strong attractive force or by passing into 
a condition of electronarcosis. To eliminate, 
as far as possible, subjective observations and 
to measure this effect under different elec- 
trical conditions, the experimental design dis- 
cussed in the following paragraphs was used. 
The tank was considered to consist of two 
equal sections, positive and negative. The fish 
were held in the center of the tank between 
two net barriers. When the power was turned 
on, the barriers were removed. The power 
remained on for 60 seconds. During this pe- 
riod, the time the fish spent in the positive 
section was recorded. Under optimum elec- 
trical influence, the fish, when released from 
the center area, would move directly to the 
positive pole and become immobilized, thus 
spending the full period of 60 seconds in the 
positive section. When no influence was in 
effect, the fish (over an average) would be 
expected to spend 30 seconds in the negative 
section and 30 seconds in the positive section. 
A two-pole, double-throw switch was placed 
in the circuit so the polarity could be reversed. 
This obviated the possibility that some in- 
fluence other than electrical stimulation might 
cause the fish to seek one end of the tank 
more consistently. Each fish was given two 
timed trials, the polarity being reversed in the 
second trial. To determine whether the move- 
ments of the fish in the tank were random, 
preliminary trials were made with the power 
487 
off. Ten tests were made using individual fish. 
They spent an average of 31.8 seconds in the 
positive section and 28.2 seconds in the nega- 
tive section. These average times did not de- 
viate significantly from the expected times of 
30 seconds in each end of the tank, so it was 
concluded that movement throughout the 
tank was random when the current was off. 
Timed trials were also made with several fish 
in the tank at the same time. However, in- 
dividual variation made accurate timing of 
each fish difficult, so it was decided that fish 
would be used individually rather than in 
groups. 
During the experiments, the voltage re- 
mained relatively constant (220-230 volts). 
The depth of water in the tank was 12 inches, 
and the distance between electrodes was 11 
feet. These factors were constant for all tests. 
One group of experiments was conducted 
with continuously flowing (uninterrupted) 
current. In it were included a series of experi- 
ments at amperages of 1 to 14, inclusive. The 
current in amperes may be converted to den- 
sity in milliamperes per square inch by mul- 
tiplying the amperage by the factor 3.62. 
Thus, 14 amperes is equivalent to a current 
density of 51 milliamperes per square inch. 
Three other groups of experiments were 
conducted with interrupted current using dif- 
ferent periods of impulse duration (pulse 
time). The variation in impulse duration was 
obtained by using rotating discs with different 
bronze-Bakelite combinations as already de- 
scribed. In the first group, the disc used was 
half bronze and half Bakelite. Thus, during 
each revolution the power was on for one half 
the time and off for one half the time (1:1 
on-off ratio). In the second group of experi- 
ments, the disc was three-fourths bronze and 
one-fourth Bakelite. During each revolution 
the power would then be on three times as 
long as it would be off (3:1 on-off ratio). In 
the third group of experiments one fourth of 
the interrupting disc was bronze while three 
fourths was Bakelite, In this case the power 
was on one fourth of the time and off three 
