Characterization of Electric Shock
in Torpedo californica
by
Suellen Lynn
Hopkins Marine Station
Pacific Grove, CA 93950
Advisor: Dr. Stuart Thompson
June 9, 1989
Abstract
This paper is devoted to characterizing the electric organs
in the Pacific electric ray, Torpedo californica. These rays do
not show signs of conditioning to electric stimuli. It is
possible to record the amplitude of the voltage generated by the
ays' electric organs and to record post-synaptic potentials from
isolated living electric organ tissue. The neuromuscular junction
potentials are capable of following nerve stimulation at
frequencies in exc
ess of 80 Hertz. These isolated column
recordings show that facilitation occurs in the neurotransmitting
mechanism that produces the shock.
Introduction
he possession of electric organs by fresh water as well as
salt water fish is well documented. Electric fields are used by
fish for various purposes, including electrolocation and prey
capture (Kalmijn 1966). The Electric eel, (Electrophorus
electricus) is a common species that is able to use its electric
organ to capture prey (Bennett, 1971). A similar type of electric
organ is found in the Pacific electric ray, Torpedo californica.
The organs are composed of specialized muscle cells, called
electrocytes, arranged in columns (Bennett, et al. 1961). Each
cell in a column is innervated on the ventral side by numerous
nerves which simultaneously discharge, releasing the
neurotransmitter acetylcholine into the synapse. This opens ion
channels in the cell wall, creating a massive influx of ions into
the cell. Simultaneously, the resistance of the dorsal membrane is
lowered. The sum of all these mechanisms generates current
(Aidley 1978, Bennett 1970). Shocks as large as 50 Volts and 50
amps have been recorded (Fong, 1981).
Torpedos can use their electric organs for two purposes, as a
warning to intruders, and as a means of capturing prey. Although
these seem to be separate activities, they may in fact be two
different aspects of the same action. Each electric organ is
innervated by three large trunks, as in Narcine, which is also a
torpedinid (Bennett, 1971). These large trunks branch into
smaller nerves which branch profusely and terminate on the ventral
membrane of the electrocytes. Each electrocyte is innervated by
-2.
multiple nerves. In T. nobiliana, Bennett, et al., (1961)
observed facilitation, the augmenting of shock voltage due to
increased transmitter release, possibly in response to build up of
excess calcium in the terminal site. Since the organs of the two
species are similar, it is conceivable that facilitation also
occurs in the Pacific electric ray. This study focuses on the
warning shock, its accurate measurement, and the characteristics
that make up this uniquely evolved organ in the torpedo ray.
Materials and Methods
A total of fourteen rays were used in these experiments.
The first six rays were placed in a 109 x 225 x 44 am tank with
fresh, unfiltered flowing sea water run through a de-embolizing
column. This tank was enclosed in a tarpaulin shed designed to
emulate the lighting at thirty feet below the surface, the natural
habitat of the Torpedo. The bottom of the tank was covered with a
The other eight rays were placed in a large
layer of fine sand.
round holding tank with no sand and no covering.
Each ray was marked with small multicolored beads sewn to the
tail, in order to identify individuals. The rays were left alone
for one week and then small Boccaccio rock fish were introduced to
the tank for feeding purposes. The intention was to wait for the
ays to begin feeding and by that action, exhibit habituation to
the tank.
Upon expiring each ray was dissected in order to familiarize
the experimenters with the anatomy of the rays.
-3-
Conditioning
In this experiment, the goal was to condition the ray to an
electric stimulus (performed in collaboration with Lara
Rosenthal). A Grass S44 stimulator was used to generate a 10 Volt
impulse at 100 Hertz for 1 second. An electric probe was
constructed from a long wooden pole and two silver/silver chloride
electrodes were taped to the end. Immediately upon cessation of
the electric stimulus (conditioned stimulus or CS), the ray was
thumped on the wing with the probe (unconditioned stimulus or
UCS). These episodes were repeated every two minutes for a period
of thirty minutes and then CS was presented alone. The intent was
to ascertain if the ray would learn to associate the conditioned
stimulus with the unconditioned stimulus. On later trials, the
S was applied to the ground next to the ray, to provide a
control.
Alarm Shock Measurement:
This experiment was performed as an attempt to measure the
voltage and current generated by the Torpedo during alarm shocks.
A 31 x 38 om table of aluminum screening was constructed and
soldered to a length of Teflon coated 26 gauge wire which was
soldered to a 47 ohm resistor. Another electrode was constructed
with a 13 x 20 om piece of copper tape coated circuit board, with
an insulating wooden handle. This was soldered to an identical
length of wire without the resistor and both were run to a Nicolet
3091 digital oscilloscope.
Signals were recorded differentially
across the resistor on channel B of the oscilloscope. The
-4-
oscilloscope was set to save the screen when triggered by the
rising phase of the shock.
The experiment was run at least five times on each of three
rays in the tarpaulin covered tank by placing the ray on the
aluminum screen and applying the copper paddle to one electric
organ. Pictures were taken of the saved screen with a Tektronix
C-5B Oscilloscope camera. Measurements were taken from the rays
suspended over the water on the screen so that the current would
not be lost to the relatively highly conductive sea water.
In order to record the current generated by the ray in one
electric organ, the stimulus pole was rerigged so that one
electrode was coiled at the tip of the pole while the second
electrode was taped 1.35 cm above the tip. The signal was
recorded differentially on the Nicolet oscilloscope.
Single Column Recording:
In this experiment, the electric organs were dissected out of
the ray and single columns of electrocytes were isolated with
their activating nerves. A suction electrode was applied to one
end of the column and run to the Nicolet 3091 digital
oscilloscope. Another suction electrode was applied to a large
nerve innervating the electrocyte column and run to the Grass S44
stimulator. Stimuli was applied to the nerve at 12 Volts for 0.5
ms and the summed neuromuscular junctional potentials were
recorded on the oscilloscope.
Upon establishing a response, a train of stimuli (0.5 ms, 12
Volts) was applied to the nerve at 80 Hertz for a brief interval,
-5-
254.5153.3 ms, followed by a short rest, 109.03t42.3 ms, and then
another brief train, 195.8149.6 ms, was applied at the same
frequency. The amplitudes of the first two peaks and last peak
during each train was recorded in order-to ascertain whether there
was any change in the amplitude of the post-synaptic potential.
Results
Conditioning
After pairing the conditioned stimulus (CS) and unconditioned
stimulus (UCS) for approximately 30 minutes, a single CS without
the USC elicited no response. When the UCS was applied to the
ground beside the ray, a shock was elicited similar to those
elicited by thumps to the wing. This only occurred on the
stimulus immediately following a true thump, and on no subsequent
stimuli. At times, the ray did not respond to the CS/UCS pairing
with a characteristic shock train. This seemed to happen toward
the end of the testing period.
Alarm Shock Masurement:
This experiment produced accurate voltage amplitude
measurements on the three rays.
Warning shocks produced by the
rays in these trials usually consisted of long trains of brief
spikes (Figure 4), the trains averaging 19.36t14.3 spikes in
length with a latency between the first two spikes of 13.112.1 ms.
The time between individual peaks in a single shock train was very
regular although some trains consisted of two phases of differing
-6-
regular latencies. Length of shock strings varied from ray to ray
and seemed to correlate with the size of the ray (Figure 1).
The first peak of each series of impulses, in all but one
case, reached the highest voltage. The voltage amplitude
subsequently dropped in all cases in a smooth curve, although in
trains greater than ten impulses, the voltage fluctuated later in
the train (Figure 4,5).
Peak voltage varied between rays but there was also some
relation to the size of the ray (Figure 1). The largest voltage
recorded was 36.74 Volts, from the largest ray. It also seems
that the amplitude of the peak voltage is correlated with the
length of the train, within the trials of each ray (Figure 2).
For example, the smallest ray, 16 om measured from wing tip to
wing tip, generated a train of only three spikes with a peak of
18.84 Volts, while the longest train from that ray, 28 spikes,
generated only 9.98 Volts.
Amperage recorded from animals under water averaged
0.038t0.025 amps for the 16 cm ray and 0.13710.117 amps for the 17
E ray.
Single Column Recording:
Upon ascertaining that the tissue was alive , recordings were
taken during 80 Hertz trains of stimuli (0.5 ms in duration, 12 V
amplitude) applied to the nerve. The resulting post-synaptic
response frequently displayed marked increase in peak amplitude
during the stimulus (Figure 8). The increase in amplitude of the
post-synaptic potential increased by between -6.6 percent to 700
-7-
percent, as in figure 8, and averaged at 18.319.2 percent.
Although there were measurements of very large facilitation, the
majority of recordings were smaller, thus the small average.
During long trains of stimuli, amplitudes tended to decrease
again toward the end of the train. Also, the second of the two
consecutive trains always began with a post-synaptic potential of
lower amplitude than the last response of the preceding train,
except in cases toward the end of the life of the tissue when the
train would begin with a particularly large spike and then
continue normally from there (Figure 9). Average decrease in the
amplitude of the spikes as measured from the last peak of the
first train to the first peak of the second train was 0.025t0.012
V/s.
Latency was recorded from the stimulus point to the beginning
of the resulting spike and averaged 8.410.6 ms. The latency of
the first spike was consistently at least 0.6 ms faster than the
remaining train, which had uniform latencies.
Two drops of 0.2mM curare, a neurotransmitter block, in a
50/50 mix of distilled water and sea water were applied to the
column. Recordings were taken at intervals of 30 seconds
beginning immediately after applying the drug for 14 minutes. The
signal emitted by the electroplaque column dropped sharply within
minutes after the curare was dripped into the ringer solution
(Figure 10a,b,c).
Conduction velocity of the nerve was also recorded by placing
the two electrodes 1.5 mm apart on a living nerve and then
activating the nerve with the stimulus and recording the latency
-8-
from the stimulus to the action potential. The measured value is
2.27 m/s.
Discussion
Conditioning:
It is apparent that the ray used in the conditioning
experiment did not become conditioned to the electrical stimulus.
It did, however, become sensitized to the mechanical thump of the
stimulus pole. It may be that the visual presence and the
actually mechanical thump of the pole where overwhelming stimuli,
such that the ray became conditioned to the UCS rather than the
CS. The ray also would periodically not respond to the stimulus
at all, i.e. it would emit no shock. In those cases, the ray may
have become habituated to the UCS, and the visual presence of the
pole. However, this was not tested substantially.
It is unlikely that the ray cannot detect the electric
stimulus at all, since it is well documented that elasmobranchs
are capable of detecting electric fields. Further testing of the
ability of these rays to detect electric fields is warranted.
Alarm Shock Measurement
It is apparent that Torpedo is capable of generating at least
36 Volts of electricity with its electric organs. Upon viewing
figure 1, it seems that the average peak voltage of each ray is
linearly related to the size of the ray. The sampling pool for
this experiment is rather small due to a dearth in experimental
-9-
subjects and it is not appropriate to derive a formula relating
the two parameters.
If one thinks of each electrocyte in a column as a battery
then it is apparent that the cells are in series in an electric
circuit. The sea water completes the circuit. Since potential
adds in a series arrangement of batteries, it follows that the
more batteries in a series, the more voltage would be generated.
Following this logic, voltage will depend on the number of
electrocytes in the column, and presumably, the thickness of the
electric organ. The thickness of the animal certainly increases
with length from wing to wing, so it is possible to speculate that
the voltage in a very large ray, say one meter in diameter, would
be very large indeed.
Sea water has a low resistance, about 25 ohm/cm, so that a
thin film of water surrounding the ray would probably short
circuit the current so that most of the electricity would not
travel through the resistor. Therefore, the voltage was measured
across a known distance in sea water and the amperage was
calculated fran these values using the standard equation: VIR
(voltage-currentxresistance). This is not altogether accurate
since the area over the electric organ from which the current was
recorded is not known. Therefore, only a small amount of current
generated by the electric organ was recorded. After experiencing
the shock first hand, I guess that the amperage is something on
the scale of household electrical systems, 5-15 Amperes, at least
two orders of magnitude greater than that measured.
-10-
As the data indicate, it is possible for an individual ray to
generate shocks of varying voltages, within a train and between
trains. Although this seems intuitive for a living organism, as
opposed to a signal of mechanical origin, it might be expected
from the isolated column experiment that the shock would follow
the pattern of increasing voltage along the spike train. However,
as the data show, this is not the case.
While it is possible that the method of recording the voltage
may have caused same errors, it is unlikely that the voltage is
inaccurate since each section of the electric organ should be
giving out equal voltages, providing that the width of the
electric organs is uniform. Therefore, the measurements were
always accurate for at least some part of the electric organ. It
is possible that the variances within the trains is indicative of
the activation of different parts of the electric organ in
sequence, or it is possible that the ray is able to turn off some
of the cells in a stack in order to produce a smaller voltage.
Perhaps this turning off of cells is involuntary, due to synaptic
fatique or some unknown factor.
Likewise with the varied trains, it is unclear whether the
ray voluntarily controls the train length, or whether the cells
shut down at some threshold of fatigue.
Single Column Recording
It is clear that synaptic facilitation occurs in the isolated
columns of the electric organ. As seen in the results, the first
peak of a spike train is smaller than the second, and the
-11:
subsequent peaks increase in height. It seems, however, that with
long trains of stimuli, facilitation is countered by synaptic
fatigue, possibly resulting from depletion of neurotransmitter.
Therefore, the size of the peaks does not increase steadily to the
end of the train, but instead it grows to a maximum and then
decreases slightly until the end of the train. The first peak of
the next train always has a lower amplitude than the last peak of
the first train, in reliable tests. This indicates that the
facilitation does not carry over short breaks in stimuli bursts.
We also see less facilitation in the second train, indicating that
the synaptic fatigue carries over the short interval between
stimulus trains.
Firing a train of spikes may have advantages over giving off
one single blast with the mechanism used in the ray. Since it
seems that facilitation is an integral part of the shock
mechanism, it would stand to reason that one isolated spike could
not develop enough voltage to stun a fish or warn an intruder.
Facilitation may be necessary to build up enough voltage in the
electric organ itself so that the external spike string is as
potent as it is. On the other hand, perhaps facilitation is
necessary to compensate for synaptic fatigue.
It is interesting to note that the latency from the stimulus
to the peak is very regular. This is in contrast to the shock
pattern elicited from the whole animal. The time intervals
between peaks in the animal are variable, and therefore probably
reflect voluntary control by the ray.
-12-
he curare experiment was performed for interest sake,
substantiating the biology/viability of the tissue. It identifies
that the mechanisms operating during the electric shock are
synaptic in origin and involving the neurotransmitter,
acetylcholine. The sharp fall in voltage amplitude with no data
points in the curare test is the result of mechanical failure and
was cured by jiggling a cord. However, the points on either side
of the fall are accurate and therefore, I accept the validity of
the test.
Summary
It is to be concluded, therefore, that 1) within the scope of
this experiment, rays cannot be conditioned to react to an
electric stimulus, 2) all shocks emitted by viable Torpedos are
made up of strings of spikes, 3) voltages can be recorded from a
Torpedo suspended in the air, 4) isolated columns of living
electric organ tissue can be stimulated to generate voltage
strings at a frequency of at least 80 Hertz, 5) facilitation and
depletion of neurotransmitter do occur in this tissue and have a
cancelling effect on each other.
-13.
Bibliography
Aidley, D.J. (1978) The Physiology of Excitable Cells, 2nd ed.
Cambridge Univ. Press, Cambridge. pp. 336-347.
Bennett, Michael V.L. (1970) Comparative Physiology, Electric
Organs. Ann. Rev. of Physiol. 32:471-528.
Bennett, M.V.L. (1971) Electric Organs. Eish Physiology. 5:347-
491.
Bennett, M.V.L., Wurzel, M., and Grundfest, H. (1961) The Electro¬
physiology of the Electric Organs of Marine Electric Fishes.
I. Properties of the Electroplaques of T. nobiliana. J. of
Gen. Physiol. 44:757-804.
Bray
R.N. (1978) Night Shocker: Predatory Behavior of the Pacific
Electric Ray (T. californica). Science. 200:333-334.
Fong, H. Loren (1981) Predatory Behavior of the Pacific Electric
Ray Toredo californica. Stanford Honors Thesis, Hopkins
Marine Station, Pacific Grove, CA.
Hopkins, Carl D. (1983) Functions and Mechanisms in
Electroreception, in Ch. 6 of Fish Neurobiology. Univ. of
215-259.
Mich. pp. 21
Kalmijn, A.J. (1966) Electroreception in Sharks and Rays. Nature.
212:1232-1233.
Lissman, H.W. (1958) On the Function and Evolution of Electric
Organs in Fish. J. Exp. Biol. 35:156-191.
Murray, R.W. (1962) The Response of the Ampullae of Lorenzini of
Elasmo-branchs to Electrical Stimulation. J. of Exp. Biol.
39:119-128.
Murray, R.W. (19650 Electroreceptor Mechanisms: the Relation of
Impulse Frequency to Stimuli in the Ampullae of Lorenzini of
Elasmobranchs. J. Physiol. 180:592-606.
Sheridan, M.N. (1965) The Fine Structure of the Electric Organ of
T. marmorata. J. Cell Biol. 24:129-141.
-14-
Figure Legends
Figure 1. This is a comparison the capacities of three different
size rays. Experiment 1 is the average peak voltage generated by
the ray, Experiment 2 is the average number of spikes per train
generated by the ray, Experiment 3 is the average time interval
between spikes in the ray.
Figure 2. This graph is designed to demonstrate the downward
trend of the top voltage of each string as compared to the
increase number of spikes in the associated string. Solid
box is th 19 cm ray, emptly box is the 17 cm ray, crossed box
is the 16 cm ray.
Figure 3. This graph demonstrates clearly the effect of
curare on the voltage amplitude of the electroplaque column
over time. The interval without data points reflects a
mechanical failure in the recording instrumentation.
Figure 4. A whole animal shock train generated by the 19 cm
ray.
Figure 5. A fluctuating whole animal shock train generated by the
16 cm ray.
-15-
Figure 6. One single spike blown up from a spike train generated
by the 19 cm ray.
Figure 7. One single neuromusculat junction potential from an
isolated column.
Figure 8. Facilitation recorded from a single isolated column,
stimulus:80 Hertz, 12 Volts.
Figure 9. Two consecutive neuromuscular junction potential trains
exhibiting facilitation and synaptic fatigue.
Figure 10a. Neuromuscular junction potential of one column in cube
of tissue immediately following curare application.
b. Same junction recorded 18 minutes after curare added.
c. Same junction recorded 39 minutes after curare added.
-16.
8

5 (

O
a
O
ONC
magnitude (Y. 4AS)
-
5
O
5O
5
I
Peck voltoge ()
aaaaaaaaaavaaa-





S



—
8
—
—
0
I

(

O
voltage (V)
-
o +
Figure 4
Figure
Figure 6
Figure
Figure 8
Figure 9
Figure
10