ABSTRACT

ELECTROPHYSIOLOGICAL RESPONSES OF THE LATERAL LINE
AND HEART TO STRESSES OF HYPOXIA, CYANIDE,
AND DDT IN RAINBOW TROUT
BY

Thomas Gordon Bahr

Studies of lateral-line nerve and heart reSponses in
rainbow trout, Salmo gairdnerii, to certain pollutional
stresses are presented. Spontaneous and evoked lateral-line
neural discharges and electrocardiograms were measured
ig_§i£3 from curarized preparations exposed to hypoxic and
aSphyxic conditions, and cyanide and DDT poisoning.

ASphyxiation and cyanide poisoning caused a reduction
in heart rate, changed wave forms of the electrocardiogram,
and depressed both Spontaneous and evoked activity from the
lateral—line nerve. Evoked reSponses persisted longer under
stress than spontaneous activity. Discontinuation of
stress was followed by recovery of both heart and lateral-
line activity. Normal function of the lateral line is de-
pendent on blood circulation and it is believed that
ischemic conditions in the lateral line were responsible for

the depressing effects of the stresses.

 

:trézt‘ns‘ngwmgit‘w :1!“ WW 7' r71 2:11-51

 

 

Thomas Gordon Bahr

Intravenous injection and water exposure to DDT caused
little if any change in lateral line or heart activity.
No change of these parameters was observed in fish suffer-
ing characteristic DDT poisoning tremors.

Hypoxic conditions resulted in a slight decrease of
the heart rate and changes in the electrocardiogram were
noted. Electrical activity recorded from the lateral—line
nerve remained unchanged.

The methodology developed for this study proved to be
a useful tool for analyzing neurotoxic effects of pollution-

related stresses.

 

 

ELECTROPHYSIOLOGICAL RESPONSES OF THE LATERAL LINE

AND HEART TO STRESSES OF HYPOXIA, CYANIDE,

AND DDT IN RAINBOW TROUT

BY

Thomas Gordon Bahr

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Fisheries and Wildlife

1968

 

 

 

 

454/02/
5-/d’—6‘7

 

 

ACKNOWLEDGMENTS

I wish to express my sincere gratitude to Dr. Robert C.
Ball for his helpful guidance and stimulation throughout my
graduate program. His unselfish willingness in allowing me
the freedom to pursue the topic of this thesis was also
most greatful.

I am indebted to Dr. Ralph A. Pax for his invaluable
assistance in many phases of this study. His generosity in
loaning me equipment is kindly appreciated.

My appreciation is also extended to Dr. Paul O. Fromm
and Dr. Niles R. Kevern for their helpful comments and con—
tinual efforts on my behalf.

I thank Dr. Harry K. Stevens for the many hours he
Spent critically reading the manuscript.

This study was financially supported by a Predoctoral
Research Fellowship (5-Fl-WP-26,004-05) Sponsored by the
Federal Water Pollution Control Administration of the

United States Department of the Interior.

ii

 

 

TABLE OF CONTENTS

Page

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . 1

LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . 6

METHODS. . . . . . . . . . . . . . . . . . . . . . . . 12

Fish Holding and Feeding. . . . . . . . . . . . . 12

Experimental Fish Chamber . . . . . . . . . . . . 15

Surgical Methods. . . . . . . . . . . . . . . . . 19

Immobilization . . . . . . . . . . . . 19

Exposure of the lateral— line nerve . . . . . 21

Canulation of the cardinal vein. . . . . . . 26

Electrophysiological Methods. . . . . . . . . . 28

Recording from the lateral— line nerve. . . . 28

Recording electrical activity from the heart 29

Instrumentation. . . . . . . . . . . . . . . 50

Stimulation of lateral—line receptors. . . . 54

Chemical Methods. . . . . . . . . . . . . . . . . 42

DDT. . . . . . . . . . . . . . . . . . . . . 42

Cyanide. . . . . . . . . . . . . . . . . . . 44

Tubocurarine . . . . . . . . . . . . . . . . 44

Saline . . . . . . . . . . . . . . . . . . . 44

RESULTS. . . . . . . . . . . . . . . . . . . . . . . . 46

Experiments on Control Fish . 46

Spontaneous activity from the lateral line . 46

General observations. . . . . . . . . . 46

Nerve cutting experiments . . . . . . 48

Evoked activity from the lateral line. . . . 54

Electrical activity from the heart . . . . . 60

Experiments on Stressed Fish. . . . . . . . . . . 64

DDT experiments. . . . . . . . . . 64
Spontaneous activity from the lateral

line . . . . . 64

Evoked activity from the lateral line . 69

Heart activity. . . . . . . . . . . . . 69

Hypoxia experiments. . . . . . . . . . . . . 7O

 

 

TABLE OF CONTENTS - Continued

Spontaneous activity from the lateral
line . . . .
Evoked activity from the lateral line
Heart activity. . . . . . . . . . . .
Asphyxia experiments . . . . . .
Spontaneous activity from the lateral
line . . .
Evoked activity from the lateral line
Heart activity. . . . . . . . . . . .
Cyanide experiments. . . . . .
Spontaneous activity from the lateral
line . . . .
Evoked activity from the lateral .line
Heart activity. . . . . . . . . . . .

DISCUSSION . . . . . . . . . . . . . . . . . . . . .
DDT Experiments . . . . . . . . . . . . . . . .
Hypoxia Experiments . . . . . . . . . . . . . .
Asphyxia Experiments. . . . . . . . . . . . . .
Cyanide Experiments . . . . . . . . . . . . . .

LITERATURE CITED . . . . . . . . . . . . . . . . . .

iv

Page

106
108
109
115

119

-- - 11-1.;—

 

 

 

 

FIGURE

1

N

.p.

5.

6.

oo

10.

11.
12.

LIST OF FIGURES

Diagram of experimental fish chamber, head tank,
oxygan metering equipment, and toxicant metering
system . . . . . . . . . . . . . . . . . . . . .

Electrode apparatus used for recording neural
activity from lateral—line nerve in rainbow
trout. . . . . . . . . . . . . . . . . . . . . .

Block diagram of instrumentation used for record-
ing electrical activity from the heart and
lateral-line nerve . . . . . . . . . . . . . . .

Secured fish showing position of ECG and lateral-
ling nerve electrodes. . . . . . . . . . . . . .

Entire experimental setup. . . . . . . . . . . .
Spontaneous lateral-line activity in control
fish demonstrating reduced frequency upon pro-

gressively cutting the nerve . . . . . . . . .

Examples of various evoked reSponses produced by
stimulating the lateral—line . . . . . . . . . .

Typical electrocardiogram (ECG) from rainbow
trout using a unipolar electrode . . . . . . . .

Effect of intravenous injection of DDT on Spon-
taneous activity recorded from the lateral-line
nerve. . . . . . . . . . . . . . . . . . . . . .

Effect of lowered oxygen tension in water on
heart rate . . . . . . . . . . . . . . . . . . .

Examples of ECG changes during hypoxia . . . . .

Effect of asphyxiation on Spontaneous and evoked
neural discharge from lateral—line nerve . . . .

Page

18

25

57

41

52

55

61

67

75
77

81

 

 

LIST OF FIGURES - Continued

FIGURE

15. Effect of 4 ppm KCN on spontaneous and
neural discharge from the lateral-line

14. Effect of 1 ppm KCN on Spontaneous and
neural discharge from the lateral-line

evoked
nerve.

evoked
nerve.

15. Effect of 0.5 ppm KCN on Spontaneous and evoked

neural discharge from the lateral-line

nerve .

16. Effect of 1 ppm KCN exposure on Spontaneous and
evoked neural discharge from the lateral-line

nerve.................

o a o

17. Relationship between Spontaneous and evoked
neural activity during asPhyxia and cyanide

exposure. . . . . . . . . . . . . . . .

18. Heart rate effects during exposure of fish

preparations to 4 ppm KCN . . . . . . .

19. Heart rate effects during exposure of fish

preparations to 1 ppm KCN . . . . . . .

20. Heart rate effects during exposure of fish

preparations to 0.5 ppm KCN . . . . . .

vi

0

Page

86

88

90

95

96

99

101

105

 

 

 

 

 

 

 

 

 

INTRODUCTION

Studies in the past which have been concerned with
effects of pollution stress upon physiological changes in
fish species have largely ignored effects of toxic pollutants
upon functional changes in the nervous system. This is un—
fortunate since a very large number of toxic substances are
recognized as Specific nervous system poisons. In addition
to Specific neuro-toxins there are a number of chemicals not
normally considered to be neurotoxic which can alter physio-
logical processes within the body in such a manner as to
have profound secondary effects upon the nervous system.
Because the nervous system is extremely sensitive to changes
in the internal environment of the body and because it is
unparalleled in its role of immediate usefulness to the
animal it is not surprising that a large variety of sub-
stances can cause severe disability or even death by
secondarily affecting the nervous system.

It is believed that a more detailed understanding of
neural mechanisms in fish, coupled with knowledge of effects
of various toxic substances associated with pollution are

necessary preliminaries for understanding the complexity of

-1:.-‘r —<!

 

is

tic

bet

PM
the
on
of
ar

t1

 

 

 

biological changes occurring in inhabitants of water receiv—

ing toxic wastes.

Most of the information regarding effects of toxicants
upon the nervous system has been obtained from experiments
on animals other than fish, in particular, mammals. There
is a growing awareness that applying this type of informa—
tion to fish can be wholly inadequate because of the numer—
ous biochemical and physiological differences existing
between the two classes of animals.

The objective of this study was to develOp electro-
physiological methodology for investigating one aSpect of
the nervous system in trout, the lateral-line system. In
order to describe the basic physiological characteristics
of the lateral—line system suitable procedures were tested
and adopted. In the second phase of the study I attempted
to demonstrate the effects of some stresses encountered in
a pollutional situation. In addition to furnishing knowledge
of immediate usefulness, this study may serve more impor-
tantly by providing much needed background information for
more intensive investigations of effects of pollution on
fish neurOphysiology.

This thesis will present data on the following electro-
physiological measurements obtained from rainbow trout:

1. Spontaneous neural discharge from the lateral-line

nerve.

2. Evoked neural discharge from the lateral-line nerve

following mechanical stimulation.

 

 

 

5. Electrocardiograms (ECG'S).

The above measurements were made on fish that had been
subjected to a number of stressful situations. Because of
the nature of this study the majority of the fish used for
experimentation served as their own controls. In other
words, a "base line" for each measurement to be made in a
particular fish was established from that same fish prior to
subjecting it to an experimental stress condition. This
thesis will present results from four groups of experiments
considering the following stresses:

1. Exposure to the insecticide, DDT, via the water

and also via the blood stream.

2. Exposure to reduced partial pressures of dissolved

oxygen in the water.

5. Exposure to conditions causing asphyxiation.

4. Exposure to the metabolic inhibitor, potassium

cyanide.

The insecticide, DDT, is reported to have a direct
action on the nervous system. Exposure to conditions of
reduced oxygen tension (hypoxia), and asphyxia are presumed
to have a more or less indirect action on the nervous system.
The poison, potassium cyanide, produces wideSpread metabolic
depression and may also have direct effects on the nervous
system.

.Rainbow trout, Salmo gairdnerii, were chosen as the

experimental fish for these reasons:

 

 

1. They were readily available from local hatcheries.

2. They are easy to maintain under laboratory condi-
tions.

5. They have a well-deve10ped lateral-line system.

4. Background information for physiological applica-
tion is documented.

5. They are considered an important Sport fish and are
empirically designated as a pollution intolerant
Species.

The choice of using the lateral—line system in these
experiments was made for a number of reasons. The first con-
cerns the possible role of the lateral line with the symptoms
commonly observed in fish suffering DDT poisoning. One of
the first symptoms of poisoning is characterized by hyper-
excitability. This is evidenced by violent locomotor activ—
ity on the part of the fish following slight mechanical
disturbances in the water. These characteristic motor
reSponses in poisoned fish might be partially explained by
malfunction in sensory inputs to the central nervous system.
The lateral-line system is one such input. It is conceivable
that excessive motor activity by the fish may be initiated
by a prolonged period of repetitive after-discharge from
lateral—line receptors following a single stimulus.

Secondly, the lateral-line nerve and associated receptor
organs are located just under the Skin in very close proxim-

ity to the water bathing the fish. This not only allows

 

 

easy
pose
may
thro
cana
woulc

linir

tinuo
activ
occur
affer:
physi<
activj
freque
ation

measu-

 

 

easy access to the nerve for dissection and recording pur—

poses but is also significant since waterborne toxicants
may possibly enter the endolymph of the lateral—line canal
through the small pores in the skin immediately above the
canal. If toxicants entered the canal in this manner they
would come in immediate contact with the sensory neuromasts
lining the base of the canal.

Thirdly, the lateral-line nerve is in a state of con-
tinuous Spontaneous neural discharge. The Spontaneous
activity presumably reflects physical and/or chemical events
occurring at the junction of the receptor cells and the
afferent lateral-line nerves. If a toxicant were to change
physiological conditions at these junctions the Spontaneous
activity might be expected to change. By recording the
frequency of neural discharge prior to a given stress situ-
ation one has a convenient base line against which to

measure effects of the Stress.

 

 

 

 

LITERATURE REVIEW

The lateral-line system in aquatic vertebrates has
been a subject of wideSpread research activity in recent
years. Interest in this subject was primarily aroused by
the development of sonar techniques during World War II.
Because of the intimate relationship between underwater
acoustics and function of the lateral-line system there has
been active support for physical and biological investiga-
tion of this sensory system.

Literature concerning the morphology and physiology of
the lateral-line system in trout is unfortunately scarce.
Only one publication was found that dealt Specifically with
the lateral—line system of trout (Hoagland, 1955b). Hoagland
reports that the trunk lateral—line nerve of brook trout,
.Salvelinus fontinalis (Mitchill), iS in a State Of continu—
ous Spontaneous activity and this activity can be reduced by
cutting the nerve or by cooling it. Hoagland was able to
obtain neural reSponses from a single fiber and record its
frequency. However, his work with trout was not extensive;
his research effort was concentrated on the catfish,
Ameiurus nebulOSus Les. (Hoagland, 1955a, 1955b, 1954a,

and 1954b).

 

 

 

Although not primarily interested in the function of

the lateral-line as a system, Tasaka (1959) and Cragg and
Thomas (1957) used the lateral line of trout in studies of
neural conduction velocity and strength—duration relation—
ships. They do, however, mention that the lateral—line
nerve is Spontaneously active and that the Spontaneous
activity will cease if the nerve is removed from the fish
or if the circulation to the nerve is stopped.

Although morphology of the lateral—line system has been
described in many fish by a large number of investigators,
little has been done on trout. Some similarity probably
exists between the lateral—line organs in trout and those
in other Species. Dijkgraaf (1962) presents a very compre-
hensive review of the literature concerning the morphology
and physiology of the lateral—line system. Dijkgraaf dis-
tinguishes between ampullary lateral-line organs, found in
elasmobranchs, and “ordinary" lateral-line organs, found in
most fish and amphibians. He subdivides the ordinary
lateral—line organs into free sensory neuromasts and canal
organs. The lateral-line system in trout is of the canal
organ type. Briefly, the canal organ of the type found in
trout consists of a longitudinal canal located just under
the skin which courses the entire length of the flank on
either side of the fish. Protruding into the fluid—filled
canal are numerous sensory endings arising from patches of

sensory neuromasts lining the length of the canal.

 

 

 

Endings within the canal itself consist of numerous sensory

hairs, embedded in a jelly-like substance called the cupula.
Shearing movement of the cupula with reSpect to the top of
the sensory cell is reSponsible for initiating the evoked
neural response in the afferent nerve. Sensory neuromasts
are innervated by fibers of the lateralis branch of the vagus
nerve. Cell bodies of these fibers are found in the acoustic
tubercle of the medulla oblongata.

A search of the literature revealed no studies concern-
ing effects of pollution stress on the neurophysiology of
the lateral line in any Species.

Neur0physiological studies of effects of the insecticide
DDT have been conducted almost exclusively on insects and
other invertebrates. Classical symptoms in DDT-poisoned
insects include hyperexcitability, convulsions, and eventual
paralysis (Welsh and Gordon, 1947; Roeder and Weiant, 1948;
Yamasaki and Ishii, 1952: Lalonde and Brown, 1954). Similar
symptoms are also common in fish, birds, and mammals. The
mode of action of this insecticide is still unclear although
there have been numerous revealing studies. Yamasaki and
Narahashi (1957a, 1957b) were first to indicate that DDT
specifically affected the nerve action potential. Their work
with axons of the American cockroach demonstrated that DDT
treatment caused an increase in the duration of the negative
after-potential of the nerve action potential. More re-

cently, Narahashi and Haas (1968) using voltage—clamp

 

 

 

techniques on lobster giant axons were able to clearly

demonstrate that DDT inhibits the turning-off process of the
peak transient current during an action potential and also
that the steady state current is inhibited.

Matsumura and O'Brian (1966a, 1966b) working with the
molecular structure of the site of action of DDT upon nerve
membranes in cockroaches have demonstrated that the insecti—
cide forms a charge-transfer complex with a component of the
membrane. It is possible that this interaction with the
membrane is responsible for the conductance changes observed
by Narahashi and Haas.

Effects of cyanide upon the nervous system were studied
by Schoepfle and Bloom (1959) and Schoepfle (1965). These
studies demonstrated that the Spike height of an action
potential becomes smaller even though there is no change in
the resting membrane potential. This finding indicates that
cyanide affects sodium conductance before cyanide has a
significant effect on the metabolic Na-K exchange mechanism.
A later study by Schoepfle and Eikman (1967) demonstrated
this more directly by measuring the efflux of radioactive
sodium from the sciatic nerve of the frog. Cyanide did not
affect the sodium extrusion rate for the first 20 to 40
minutes and there should have been no change in the intra-
cellular sodium concentration or sodium equilibrium potential
during this period. Nevertheless, reduction in the height of
the action potential was, in fact, observed within the first

20 minutes.

 

 

 

DI
CE
we

the

stn
that
pois

CODC

actix
(Gare
Shelt
creas<
envirc
neuron
was b1
used t
no det
Normal ‘
in the ‘
W‘

maY nc

 

10

The effects of cyanide on trout have largely been
bioassays, testing survival times of test fish after exposure
to various concentrations of the poison (Herbert, 1952;

Downing, 1954; Burdick et al., 1958). Effects of cyanide on

 

cardiac rhythm in embryos of Fundulug heteroclitus (killifish)
were determined by Brinley (1950). By visual observation of
the heart in the embryos he found that the heart rate would
decrease after exPosure to the poison and upon removing the
stress the heart rate would return to normal. He also found
that the auricles of the heart were more resistant to cyanide
poisoning than the ventricles. No literature could be found
concerning effects of cyanide on fish electrocardiograms.
rEffects of low dissolved oxygen conditions on heart
activity have been studied by a number of investigators
(Garey, 1962; Holeton and Randall, 1967a, 1967b; Randall and
Shelton, 19657 Randall and Smith, 1967). Compensatory de-
creases in heart rate occurred in trout when exposed to hypoxic
environmental conditions. Increased activity in inhibitory
neurons of the heart created the reduced rate. This effect
was blocked by the drug, atrOpine. Electrocardiograms were
used to monitor the heart rate in these studies, however,
no detailed description of changes in the ECG are given.
Normal ECG's in several Species of fish have been described
in the past (Oets, 1950; Kisch, 1948; Robertson gt_al., 1966).
When cardiac activity becomes severely reduced the heart

may no longer be able to perfuse body tissues. Lack of blood

 

 

 

In

sysl
isck
nerv
able

than

 

11

flow in a tissue is termed ischemia. Effects of ischemia

on peripheral nerves have been well documented in mammals

and to a lesser extent in amphibians, primarily frogs
(CooPer, 1925; Forbes and Ray, 1925; Heinbecker, 1929;
‘Lehmann, 1957; Magladery §t_§l., 1950; Fox and Kenmore, 1967).
No Similar studies were found relating to fish Species.

In short, studies have demonstrated that central nervous
system neurons could not withstand more than a few minutes of
ischemia before becoming permanently damaged. Peripheral
nerves, however, were more resistant to ischemia and were
able to recover even after blood flow had ceased for more

than an hour.

 

trout
They‘
Depart
Michig
circul
Units,
nated
plasti
Dechlo
a 5 cu

Water '

  
  

°XY9en

 

 

METHODS

Fish Holding and Feeding

Fish used in this study were two—year old rainbow
trout, Salmo gairdnerii,weighing from 170 to 540 grams each.
They were furnished by the State of Michigan Conservation
Department, Research and Development Laboratory, Grayling,
Michigan. About 50 fish were kept in each of two 150 gallon
circular Fiberglass holding tanks manufactured by Frigid
Units, Inc. of Toledo, Ohio. A continuous flow of dechlori-
nated tap water, about 5 l/min, was Sprayed through a small
plastic nozzel at an angle to the surface of the water.
Dechlorination was accomplished by passing tap water through
a 51cu ft activated charcoal filter. The fine Spray of
water into the tanks was sufficient to maintain the dissolved
oxygen level at near saturation and also had enough force to
set up a current of approximately 2 ft/sec within the tanks.
Fish maintained their position in the tanks by swimming
against the current and by doing so obtained a small amount
of exercise. The cold tap water maintained a temperature of
about 150 C which varied less than 10 C per day. The center
of each tank was equipped with two standpipes, concentrically

placed (outer standpipe Open at the bottom, inner standpipe

12

 

tot
Fish

noti

whit}

coinc

l
ings L
transf
Plexig
fish 11
fresh c
Place
of recc
cals.
allowed
1.
2.

 

 

15

Open at the top), allowing the feces which accumulated on
the bottom to flow through the standpipe system to the drain.
The fish were fed daily on a diet of dry pellet trout
food (Formula 65-W, Glencoe Mills, Glencoe, Minn. and Strike
Fish Feed, Country Best, Agway, Inc., Syracuse, N. Y.) fur—
nished by the Michigan Conservation Department. From time
to time this diet was supplemented with chOpped beef liver.
Fish were maintained in very good health on this diet and
noticeable growth was observed during the study period.
Holding tanks were illuminated with incandescent lamps
which were electrically switched on and off with a timer to

coincide with the natural photoperiod.
Experimental Fish Chamber

For purposes of obtaining electrOphysiological record—
ings under controlled conditions fish were individually
transferred from the holding tanks to a small rectangular
Plexiglas chamber. The chamber was slightly larger than the
fish itself and was provided with a continuous supply of
fresh oxygenated water. The chamber provided a convenient
place to conduct surgery, to confine fish after attachment
of recording electrodes, and to expose fish to toxic chemi—
cals. The design of the chamber and associated equipment
allowed the following conditions to be satisfied:

1. Fish could be completely covered with water.

2. The water temperature could be held constant

 

 

 

 

ch

anc
sec
wit
thrc
Sip}

in t

from
A com
lishe
about
head y
the ta
10Cate(
tank t]
Inc”
and or

thehu,

 

 

14

throughout an experiment.
5. A constant flow of water could be passed over the
gills of the fish.
4. Oxygen tension in the water could be carefully
controlled and monitored during an experiment.
Immobilized fish were placed on their side in the
chamber and the water inlet tube was inserted into the mouth
and secured by a steel pin which passed through the tube
and through the lower jaw of the fish. Fish were further
secured by fixing the dorsal fin to the side of the chamber
with the aid of a spring loaded clamp. Water left the chamber
through an adjustable siphon at the rear. By moving the
siphon either up or down one could adjust the water level
in the chamber.
Water to the chamber was at the same temperature and
from the same source as that supplying the holding tanks.
A constant flow of water into the chamber inlet was accomp-
lished by first passing it into a 2-liter head tank located
about 2 feet above the level of the chamber. The constant
head was maintained by passing an excess flow of water into
the tank and allowing it to overflow through a drain port
located near the t0p. Pure oxygen was bubbled into the head
tank through an air stone. Spherical float flowmeters (RGI,
Inc., Great Neck, N. Y.) were placed in line with the water
and oxygen inputs to the head tank. The oxygen tension in

the water could be quickly changed to any desired level by

 

adj
fou
the
tha
nece
SUPP
wate
tank
tion
oxygt
l/nix

in t}

place
the d
ment.
StOppe
throng
0fpla
in the
Chamber
”33 dir
the pro
Velocit
tion 01
to Prev

3“ the

 

 

15

adjusting the oxygen and water flows to predetermined values
found by prior calibration. .The dissolved oxygen level of
the dechlorinated water entering the head tank was less
than 1 ppm, thus to achieve low oxygen levels it was not
necessary to use oxygen Stripping techniques on the water
supply to the chamber. To achieve oxygen saturation in the
water to the fish chamber the flow of oxygen to the head
tank was increased. Flow rates required for water satura—
tion, for example, were 1.5 l/min of water and 0.4 l/min of
oxygen. Water flowed to the fish chamber at a rate of 0.67
l/min and the remaining 0.85 l/min left the overflow port
in the t0p of the head tank.

A polarographic oxygen sensor (Beckman Model 777) was
placed between the head tank and the fish chamber to monitor
the dissolved oxygen during the entire course of an experi-
ment. The sensor was passed through a hole cut in a rubber
stopper and the st0pper containing the sensor was inserted
through a hole cut into the side of a‘2 inch diameter piece
of plastic pipe. Using large stoppers this pipe was placed
in the water-flow line between the head tank and the fish
chamber (Figure 1). The water flow into the sensing chamber
was directed at the surface of the oxygen sensitive area of
the probe. This provided a continuous water flow, the
velocity exceeding the minimum required for accurate Opera-
tion of the sensing unit. The sensing chamber was tilted
to prevent entrapment of air bubbles within the chamber and

on the sensitive area of the probe.

 

 

an
1y
thE

 

16

The oxygen meter was calibrated prior to each experi-
ment by exposing the probe to the air then setting the meter
to read a partial pressure of oxygen of 160 mm Hg. Previous
experimentation proved that readings from the oxygen meter
calibrated in this fashion agreed very closely to dissolved
oxygen measurements done on the same water with chemical
oxygen determinations using the Winkler Method (Standard
Methods, 1960).

In one series of experiments fish were exposed to vari—
ous water concentrations of potassium cyanide. This was
accomplished by metering a known flow of the chemical from
a reservoir to the inflowing water supply to the fish
chamber. A Beckman Model 746 Solution Metering Pump was used
for this purpose. The flow rate from the pump could be ad—
justed to the nearest 0.01 ml/min from 0.00 to 20.00 ml/min
by turning a calibrated dial which in turn would increase or
decrease the length of travel of a Teflon piston within the
pump. Flow from the pump would pulsate in two second inter-
vals. To prevent pulses of concentrated cyanide solution
from passing over the gills of the fish as a result of the
pulsating nature of the pump, a mixing chamber was placed
between the inlet of the fish chamber and the juncture of the
pump inlet to the water line (Figure 1). Mixing occurred in
the turbulence created by water entry into the mixing chamber

and the volume of the mixing chamber (250 ml) was sufficient—
rly large to damp out the cyanide pulses. By pumping dye into

the system the mixing proved to be very good.

 

 

 

 

 

17

Figure 1.——Diagram of the experimental fish chamber, head
tank, oxygen metering equipment, and toxicant
metering system. Water flow through the system

is indicated by arrows.

 

18

 

OVERFLOW
PORT
a’

 

READ TANK

 

\ I C
' ”‘9 ‘ 0 '

O O
OXYGEN METER

 

 

 

 

 

 

{9:} 7‘

METERING
PUMP

 

 

 

 

FLOW METERS

°2
VALVES

TOXIN
RESERVO!R

 

 

 

 

'mxmo FISH cums“
CRIMIER

Figure 1

 

 

 

 

 

 

Tub.
hib:
of t
phys
immo

of t

0n he
used
Condu
later.
of the
the s;
needle
by run
length
latera

COMIC

 

19
Surgical Methods

Immobilization

Fish to be used in the chamber just described were
removed from the holding tanks with a net and anesthetized
in a solution of 100 ppm MS-222 (tricaine methane sulfonate,
Sandoz) for five minutes. The fish were then weighed and
given an intramuscular injection of d-tubocurarine chloride
(Sigma Chemical Co.) at a dosage of 20 mg/kg live weight.
Tubocurarine is a drug which causes muscle paralysis by in-
hibiting neural transmission at the motor end plate region
of the muscle. This drug enjoys wideSpread use in neuro~
physiological investigations because it achieves total
immobilization of the test animal without severe depression
of the central nervous system.

To determine whether or not tubocurarine had an effect
on neural discharge from lateral-line organs in the fish
used in this study, a number of separate eXperiments were
conducted where the animals were immobilized by pithing and
lateral—line discharges measured before and after injections
of the drug. Pithing was accomplished by either severing
the spinal cord in the anterior cervical region with a steel
needle or by completely destroying the brain and spinal cord
by running a c0pper wire through the brain and down the entire

length of the Spinal column. Electrodes were attached to the

lateral-line nerve (as described in the next section) and

control recordings were made of both Spontaneous and evoked

 

 

 

20

neural activity before administration of the drug.
Tubocurarine was then injected intramuscularly in doses two
to three times those normally given and the lateral line
recordings were closely observed. Results from these eXperi-
ments did not demonstrate any depressing effect of this drug
on the lateral-line nerve and also the mode of pithing

seemed to have little effect on the nerve.

Prior to placing the curarized fish in the experimental
chamber the gills were rinsed with fresh water to rid them
of excess mucus which accumulated during anesthetization.
Very frequently fish would vomit stomach contents shortly
after curarization. When this occurred fish were massaged
to remove remaining stomach contents then the mouth and gills
were rinsed again with fresh water. Since a curarized fish
is unable to move its mouth and operculum it is not only
unable to ventilate but also unable to conduct gill cleaning
reflexes. Failure to rinse the mouth and gill region prior
to placing the fish into the chamber did, on several oc—
casions, result in asphyxiation of the test animal.

The fish was then placed into the chamber on its side
and the water input tube inserted into its mouth. At this
time a unipolar electrocardiogram (ECG) electrode was in-
serted into the base of the left pectoral fin and pushed

down just under the skin to a midventral position about one

centimeter posterior to the heart. Construction of the ECG

electrode will be discussed later. The water level in the

 

 

 

 

 

 

21

chamber was then raised above the fish and the ECG trace
on the oscilloscope observed for about one hour. If the
heart continued to function properly during this period
the experiment was continued. If the heart showed signs of
failure at this time the fish was discarded and a new fish

used.

Exposure of the lateral-line nerve

Following a one hour period of normal heart function
the next Step was to isolate the lateral—line nerve and
attach recording electrodes. In most of the fish tested the
lateral-line nerve was found just under the skin lying one
or two millimeters dorsally to the lateral-line markings on
the surface of the fish. The lateral-line nerve (lateralis
branch of the vagus) leaves the cranium behind the gill
region and courses the entire length of the trunk to the
caudal peduncle. There is one nerve trunk on each side of
the fish.

To isolate the nerve a shallow longitudinal incision
was made with a sharp blade just through the skin about 5 mm
below the lateral line and at a point about one-third of the
way down the trunk toward the tail of the fish. The length
of the incision was about 1 cm. Using forceps and a blunt
probe the Skin overlying the musculature surrounding the
lateral-line nerve was separated by running the probe between
the skin and muscle. An area 2 cm in diameter directly over

the lateral-line nerve was freed in this manner and a

 

 

 

22

circular piece of skin, 1 cm in diameter, was cut free from
the fish with scissors. This exposed a circular area of
flesh containing a small longitudinal groove. This groove
was covered with a thin layer of membranous tissue or was
an Open canal depending on the depth Of the dissection.
This canal, running the entire length of the flank from
Operculum to caudal peduncle, is the lateral—line canal.
The sensory neuromasts of the lateral line are found along
the base of this canal. Afferent nerves from the neuromasts
join the main lateral-line nerve trunk immediately beneath
the canal. As the nerve trunk courses toward the head it
picks up more and more of these afferent fibers.

With the aid of a dissecting microscope the nerve was
separated from the surrounding musculature by Spreading the
muscle apart with fine forceps. At times it was necessary
to cut away strands of connective tissue with small scissors
in order to completely free the nerve. It is important not
to stretch the nerve which may cause permanent damage. If
the dissection was done correctly there was very little
bleeding.

A small glass cylinder, 1.5 cm in diameter and about
1 cm long, with a flanged lip at one end was then positioned
into the circular incision by forcing the lip under the
loosened skin surrounding the incision. This was done by
placing a portion of the lip under the skin at a convenient

starting point and stretching the Skin over the remaining lip

25

by inserting a blunt probe between the skin and the glass
lip, then working the probe around until the entire lip was
covered with skin. When completed the skin fit tightly
against the glass cylinder forming a water—tight seal.
The water level in the chamber could then be raised over the
entire fish and still provide a water free area of flesh
within the cylinder. A cross section of the glass cylinder
after being positioned in the fish is shown in Figure 2.
Two fine silver wire electrodes were attached to the
tOp of the glass cylinder with beeswax and soldered to two-
conductor shielded cable. The electrodes were constructed
from 26 gauge silver wire that had been electrolytically
etched to fine points in nitric acid. The tip of each
electrode was bent to form a hook and both were bent so that
the tips would extend into the interior of the glass cylinder.
In most experiments the nerve was cut centrally (between
the brain and the recording electrodes) and the distal portion
of the nerve lifted over the electrodes. The cut end of the
distal portion of the nerve was firmly secured to the anterior-
most electrode by pinching the wire hook tightly against the
nerve. The other electrode was simply IOOped under the nerve
a few millimeters posterior to the first. The nerve and

electrodes were then covered with mineral oil to prevent the

nerve from drying out.

-Many electrode preparations were tried during this study

but the one just described proved to be the best. Using this

 

 

 

 

 

 

 

 

 

24

Figure 2.--Electrode apparatus used for recording neural
activity from lateral-line nerve in rainbow
trout. Flanged glass cup inserted under skin
excludes water from electrode area. Top dia—
gram is an enlarged cross section of area

indicated in lower figure.

 

25

K To Preamp

. .-...—~ .. -—- Shielded Cable

- SEES-ix I k \‘b

an. fifiégg a...

e “a: \
Silver

Electrodes

  
 
   

Glass _.
Chamber

—_

 

 

........
n —

 

 

 

Figure 2

 

 

 

26

technique a fish could be completely covered with water,
the nerve was kept from drying out, and recordings could
be made from the lateral-line nerve for many hours with

no apparent ill effects to the fish nor to the lateral—line
reSponses. NO other method tested had all of the above

characteristics.

Canulation of the cardinal vein

In experiments requiring blood injection of chemicals
a canula was inserted into the posterior cardinal vein near
the tail. Because small amounts of heparin (an anticoagu—
lant) were introduced into the blood circulation during this
process the clotting ability of the blood was reduced.
Consequently, the lateral—line incision was made prior to
the canulation to prevent hemorrhage at the lateral—line
Site. Very little bleeding occurred at the lateral—line
incision if done first.

The posterior cardinal vein was exposed by making a 2-5
cm longitudinal incision just anterior to the caudal peduncle,
about 5 mm below the lateral line. The underlying muscle
was cut and separated down to the Spinal column and the in-
cision held Open with the aid of retractors. The flesh
adhering to the ventral Spines of the vertebrae was removed
with a blunt probe and forceps exposing the vein coursing
caudad between the bony Spines.

The canula was fashioned from a 20 cm length of Clay—

Adams PE 10 Intramedic Polyethylene tubing, inside diameter

 

 

 

 

 

27

0.011 inches, to which was fitted a 26 gauge needle. The
other end of the canula was attached to a 1 ml tuberculin
syringe. The canula and syringe were filled with Courtlands'
saline, pH 7.40 (Wolf, 1965), and contained 1 percent

heparin to prevent clotting. The needle was inserted into
the vein in the direction Of the blood flow (caudad) with

the aid of a dissecting micrOSCOpe. After successful inser-
tion the retractors were closed tightly against the canula.
This held the tubing firmly against the fish, hence, there
was no need to suture it in place. Best results were obtained
when the sharp beveled edge of the needle was dulled before
being placed into the vein. This prevented excessive cutting
of the vessel wall and resulted in very little hemorrhage.

On a few occasions, however, faulty placement of the canula
resulted in hemorrhage and when this happened bleeding was so
severe it completely obstructed the view of the canulation
site, making corrective measures very difficult. It was
necessary to discard these fish.

Injection of solution was accomplished by drawing blood
into the tubing up to a point just below the needle, removing
the needle from the tubing, then inserting a needle from a
second syringe (containing the solution to be injected) into
the blood-filled tube. The plunger of the second syringe was
depressed until all Of the blood in the canula was diSplaced
by the injection fluid. A measured amount of solution could

then be accurately injected into the vein. After the solution

 

 

 

28

had been injected the original syringe containing the
heparinized saline was replaced in the reverse order of the

procedure described above.
Electrophysiological Methods

Recording from the lateral-line nerve

Nerve impulses conducted toward the central nervous sys-
tem along the lateral-line nerve were recorded using the
extracellular silver wire electrodes described in the
previous section. As was stated, the distal portion of the
centrally cut nerve was attached to two electrodes as shown
in Figure 2. The electrode nearest the cut end of the nerve
was grounded at the preamplifier and the other electrode was
connected to the input of the preamplifier. With the
electrodes attached in this manner the nerve impulses, as
they passed the first electrode, would cause a negative de-
flection of the oscillOSCOpe beam after being amplified
through the preamplifier. Such a recording is termed mono-
phasic, that is, one nerve impulse causes one deflection of
the oscillosc0pe beam. Because the frequency Of neural
discharge was an important measurement in this study it was
necessary to display the nerve impulses in a manner which
allowed accurate counting. MonOphasic recording techniques

proved to be the best suited for this purpose.

 

29

Recording electrical activity from the heart

Electrocardiogram (ECG) measurements were routinely
taken during most of the experiments as a check on heart
function. The main purpose of conducting these measurements
was to use the ECG as a convenient cardiotachometer. After
examining many ECG records from fish subjected to various
stresses certain characteristic changes were observed in
the pattern of electrical activity from the heart. For this
reason it was decided to standardize the ECG recording method
so that comparisons could be made between different fish.
Construction and placement Of the ECG electrodes are
described below.

Because fish used in this study were covered with water
it was not possible to make good ECG recordings using common
surface electrodes. To avoid this difficulty electrodes had
to be placed under the skin and kept from making electrical
contact with the water. The electrodes were constructed by
soldering steel pins to insulated wire, then insulating the
pin and wire with several coats of insulating varnish. The
insulation at the sharp point of the pin was then scraped Off
leaving a few millimeters of electrically conductive metal
at the end of the pin. The pin was inserted at the base Of
the pectoral fin and then pushed just under the Skin until
the end Of the pin reached the ventral midline one centimeter
posterior to the heart. Insertion of this electrode was done

the same way in every fish. The wire from this electrode

 

 

 

50

was directly connected to the vertical amplifier of the lower
beam of the oscilloscope. Another electrode was grounded
to the water, completing the circuit. The voltages recorded

from the heart using this method were about one millivolt.

Instrumentation

Before weak electrical signals from the lateral-line
nerve could be diSplayed on an oscilloscope it was necessary
to amplify them. This was done in the majority of the experi—
ments with a Grass P8 A.C. Pre-amplifier (Grass Instrument
Co., Quincy, Mass.). The pre-amplifier could be Operated
either differentially (amplifying the voltage difference
between two electrodes) or single-ended (amplifying the volt-
age difference between one electrode and ground). The Open-
end mode was used most frequently because it would give a
larger signal to noise ratio than the differential mode.
The Open-ended mode Of Operation, however, was more prone to
pick up external 60 Hz (60 cycles per second) electrical
noise than was the differential mode. This was eliminated
to a large extent by proper grounding of the preparation.
The pre-amplifier was equipped with high and low frequency
noise filters which were set to exclude electrical signals
above 4 KHz (half amplitude frequency) and signals below 0.5
Hz. The high frequency filter did not appreciably attenuate
the wave form of a nerve impulse recording.

Leads from the nerve—electrode preparation to the pre-

amplifier and from the pre—amplifier to the oscilloscope

 

 

 

lulllllili l I I:

 

51

were shielded and the Shields grounded. The fish chamber
and pre-amplifier were enclosed in a common Shielded cage
which was also grounded. The peak-to-peak noise level of
the system as observed on the oscilloscope was less than
10 uV.

Electrical activity from both the lateral—line nerve
and the heart was diSplayed on the screen of a Tektronix
Type 502-A Dual Beam Oscilloscope (Tektronix, Inc., Portland,
Ore.), the lateral-line activity on the upper beam and the
ECG on the lower beam. Permanent records of oscillosc0pe
traces were made by photographing the screen with a C—27
Oscilloscope Camera (Tektronix Inc.).

Frequency of lateral-line impulses was measured either
directly from the photographs or was electrically determined
with the use Of a counter. Direct frequency measurements
from photographs involved counting the number of Spikes
(nerve impulses) appearing over a known time interval. For
rexample, if 10 Spikes were counted over a distance Of 10 cm
on the oscilloscope screen and the sweep rate of the beam
was 10 msec/cm the impulse frequency would be 10 impulses per
100 msec or 100 per sec.

Electrical frequency determinations were done with a
decade scaler, the same type of scaler as used in radioactive
isotope counting. Initial attempts to count nerve Spikes
by connecting the pre-amplifier directly to the scaler were

unsuccessful. This was because the input gate of the scaler

 

 

 

 

 

 

 

52

would exclude electrical pulses With rise times greater

than 10 usec. A nerve impulse from the lateral line has a
rise time of about 500 usec, much too slow to pass the input
gate of the scaler. For this reason it was necessary to
alter the wave form of a nerve impulse to decrease its rise
time. This was accomplished by placing an electrical trigger
circuit between the pre-aimplfier and the scaler. The circuit
is a bistable multivibrator, commonly termed a Schmitt Trigger,
which I constructed from a circuit described by Brophy (1966).
An amplified nerve impulse, a relatively slow rounded wave,

is changed by the trigger into a rectangular wave with a

rise time Of about 1 usec which could then pass the input

gate of the scaler and be counted.

Understanding the principle of Operation of the trigger
is a necessary prerequisite for interpreting data Obtained
from it. It Operates as follows: If an input voltage exceeds
a given threshold the resting voltage of the trigger output
rapidly switches to a higher value. When the input voltage
then falls below this threshold the output voltage returns
to its resting value. The threshold voltage required to
"fire" the trigger can be adjusted by changing the amount of
voltage bias at the input. This was done by turning a cali-
brated ten—turn.potentiometer which was part of a voltage
divider circuit placed across the input of the trigger.
Supplying positive voltage to the input with this voltage

divider lowers the threshold of the trigger enabling nerve

 

 

 

 

 

 

 

 

 

 

55

impulses of lower voltage to fire the trigger. The exact
amount of bias given to the input of the trigger was de-
termined by counting the number of turns which the potenti-
ometer was moved (the potentiometer was connected to a
mechanical revolution counter subdivided into 100 gradua—
tions). By diSplaying the output of the trigger on one beam
of the oscillOSCOpe next tO the input signal of the trigger
(amplified nerve impulses), diSplayed on the other beam of
the oscilloscope, it was easy to see which nerve Spikes had
the ability to fire the trigger. By careful visual observa-
tion of these two simultaneous events it was possible to
calibrate the potentiometer settings Of the bias circuit to
the size of the nerve spike (voltage) one desired to count.
It should be emphasized that this system did not have the
capability of selecting impulses of a certain voltage range
(a window), but rather could only select a threshold, all
nerve impulses of sufficient voltage to exceed this threshold
being counted and those below the threshold excluded.

There is one difficulty encountered with this counting
system that should be mentioned at this time. This occurs
when two or more nerve impulses reach the recording electrodes
very close together. The second impulse may reach the elec-
trodes before the first has totally passed. This event is
seen on the oscilloscope as a multiple peaked wave. The
difficulty arises at the input of the trigger circuit. If

the first nerve impulse does not return to below threshold

 

 

 

:54

voltage before the second reaches the input of the trigger,
the output of the trigger will remain at its higher voltage
until the input voltage does fall below the threshold. As
a result only one impulse will be recorded at the counter
when actually two or more nerve impulses actually occurred.
During the experiments, however, this effect was largely
eliminated by reducing the firing frequency of the lateral
line by cutting the nerve until neural activity from only a
few receptor units was being recorded. By reducing the fir—
ing frequency the occurrence of overlapping impulses was
greatly reduced.

To determine impulse frequency using the counter system
described above, one connects the pre-amplifier to the trigger-
counter circuit, selects the threshold, and counts the number
of nerve impulses occurring over an interval Of time, usually
one minute. The average impulse frequency was found by

dividing the number of impulses by the time interval.

Stimulation of lateral line receptors

Methodology described to this point has been involved
primarily with recording spontaneous neural activity from the
lateral-line nerve. In many experiments the receptors of the
lateral—line system were stimulated and the evoked neural
response was recorded. The standard stimulus used for this
purpose was to drop water from an Opening in a glass tube on
to the surface of the water directly over the lateral line Of

the fish. The water drOp apparatus was located in a fixed

 

 

 

 

CU.

Fig
lat.

91‘ a;

 

55

position over the fish chamber 50 that the drOp of water
would always hit the surface of the water at the same place
and fall from a given height. 80 that the evoked neural
reSponse resulting from this mechanical stimulation could
be easily seen on the oscilloscope it was necessary to Syn-
chronize the sweep of the oscilloscope beam with the falling
drop of water. This was done by placing two silver wires
immediately below the opening of the glass tube. As a water
drOp fell from the end of the tube it would Simultaneously
make contact with the end of both wires, completing a circuit
consisting of a 22.5 v battery placed in series with the
external D.C. synchronizing input of the oscilloscope.
When the circuit was closed by the falling water drop the
beam of the oscillOSCOpe would start to sweep across the
screen and the nerve impulse burst (resulting from mechanical
stimulation of the lateral line receptors) would always ap-
pear in the same position on the screen. If the sweep of
the beam was not synchronized to the falling of the water
drops the bursts of neural activity would appear in different
positions along the sweep of the beam making it very diffi-
cult to see, much less photograph.

A block diagram Of the instrumentation is shown in
Figure 5. Figure 4 is a photograph of a fish with ECG and
lateral-line nerve electrodes in place. Figure 5 is a photo-

graph Of the entire experimental setup.

 

 

 

 

 

 

 

56

Figure 5.--Block diagram of instrumentation used for record-
ing electrical activity from the heart and
lateral-line nerve. Nerve activity diSplayed
on upper beam of oscilloscope and heart activity
(ECG) on lower beam. In experiments of evoked
reSponses, water drop stimuli synchronize sweep
Of scope by completing the circuit shown just
above the fish chamber. In spontaneous activity
exPeriments nerve impulses are electronically

counted on a decade scaler for frequency analysis.

 

 

SHIELDED CABLE

TD EXT SYNC
0N
OSCILLDSCDPE

182 K

NERVE IMPULSE
22.5 V

“—"‘ LEAD

 

SHIELDED CAGE

Figure 5

 

 

 

 

 

58

Figure 4.--Secured fish showing position of electrocardio-
gram electrode (A) and lateral-line nerve

electrodes (B).

 

 

rocardio

rve

 

 

Figure 4

 

 

 

 

 

 

 

40

Figure 5.--Entire experimental setup.

 

 

 

 

 

 

wa.
wet
for
the

can

the

 

 

Chemical IdethOds

DDT

The DDT used in this study was a 995plus percent pure
p,p' isomer of 1,1,1-trichloro-2,2-bis(p-chlorophenyl)ethane,
M.W. 506.48. The chemical was Obtained from the City Chemi—
cal Corp. of New York, N. Y., and is considered a Pesticide
Reference Standard by the Entomological Society of America.

Stock solutions Of DDT were made up by dissolving the
chemical in absolute ethanol. The best method found to inject
the chemical into the blood Stream was to first make a saline-
DDT suSpension (10 percent ethanol) by rapidly mixing the
DDT-ethanol solution with the fish saline. This suSpension
was immediately injected. Waiting for only a few minutes
would cause the micro-crystals of DDT to precipitate in the
form of large particles. These particles would get caught in
the small Opening of the 26 gauge needle at the end Of the
canula preventing successful injection of the insecticide into
the vein.

Direct injection of the DDT—ethanol solution into the
vein proved to be very unsatisfactory. Upon contact with the
blood the solution would rapidly precipitate serum proteins
and the DDT would crystallize at the same time. Such in—
jections severely clogged the canula and if large enough
amounts of the precipitated material were allowed to enter

the vein heart failure soon followed.

 

 

 

lull-ll El. 51h

I: I...

 

Attempts were made to injeC3t DDT into
“9

soybean oil as a carrier. In 11355 than ten minutes fellow.
ing injections of 0.1 to 0.5 ml of pure Soybean oil the
heart would Show signs of failure. It was assumed that the
high viscosity of the oil was partially reSponsible for the
drastic effect on the heart. No other Oils were tested as
possible carriers for the DDT.

Early in the study, before the solution metering appara_
tus was used in conjunction with the fish chamber, a few ex—
periments were conducted whereby fish were exposed to very
high levels of DDT in water. In these experiments the purpose
was to expose fish to the chemical for a period of time long
enough to induce tremors and convulsions. To accomplish
this, fish were individually placed in 5 gallon aquaria which
contained a concentration of 15 ppm DDT. The DDT was added
to the water in an ethanol solution, then carefully mixed.
The aquaria were aerated, covered, and placed in a cold water
bath at 150C. Fish were exposed in this manner for about Six
hours, time enough for fish to Show advanced symptoms of DDT
toxicity.

Blood and nerve tissue from a few fish exposed to DDT
were analyzed for their DDT content. This was not a major
objective of this study and consequently will not be discussed
in detail. Briefly the chemical analysis consisted Of con-
verting all of the DDT to DDE by treatment with alcoholic KOH,

extracting the DDE in petroleum ether, drying the ether in

 

wa,

Sa]

‘

tot

7

 

 

 

44

sodium sulfate, and then analYZiJQg the final solut'
ion for
DDE in a gas chromatograph emPIIIYlng an electron captu
re
detector. Unknowns were compared Mdth standards and the

results were expressed as ppm p,p' DDT.

Cyanide
Stock solutions of potassium cyanide were made by dis-
solving the chemical in distilled water. Fresh solutions

were made every other day and stored in the refrigerator.

Tubocurarine

Stock solutions of d—tubocurarine chloride, 5 mg/ml,
were made by dissolving the chemical in distilled water.
At this concentration the use of a magnetic mixer is highly
recommended to aid in dissolving the chemical. The curare
solution was kept refrigerated at all times, and the powder

was kept in a dessicator and also refrigerated.

Saline

The trout saline solution used for tissue moistening and
as a DDT carrier for intravenous injection was made according
to the formula described by Wolf (1965). It consisted of

the following:

Glucose . . . . . . . . 1.00
Water . . . . . . . . . 1000

NaCl. . . . . . . . . . 7.25 g
CaC12-2H20. . . . . . . 0.25 9
KCl . . . . . . . . . . 0.58 g
NaH2P04-H20 . . . . . . 0.41 g
NaHC03. . . . . . . . . 1.00 g
MgSO4-7H20. . . . . . . 0.25 g

g

m

 

45

The pH of the saline was adjusted to 7.40 by adding either
HCl or NaOH. Very small amounts <3f acid or base were re_
quired for this adjustment, hence, the ionic strength of the
saline was changed very little.

To give the saline anticoagulant properties when used
to fill canulae, a small amount of heparinized-saline was
made by dissolving enough heparin in the saline to give a

one percent heparin solution. The heparin was a Grade I

sodium salt obtained from the Sigma Chemical Co.

 

 

dam
0ft4

act

dor

inn

del

dam

 

RESULTS

Experiments on Control Fish

Spontaneous activity from lateral line

General observations. A continuous discharge of high
frequency nerve impulse activity was typically observed in
unstimulated lateral-line nerves, neural activity seemingly
the result of independent Spontaneous firing of individual
neurons within the main nerve hundle. In recordings from
over 180 different fish preparations less than 10 were found
lacking Spontaneous activity. It is presumed that the lateral-
line nerve from the fish having no spontaneous activity was
damaged during dissection.

It was found that rough handling of the nerve would often
damage it. Gently pulling on the nerve with a forceps would
often cause severe reduction in the level of Spontaneous
activity and many times completely silenced the nerve. If a
large fish was forced into the fish holding chamber causing
dorso-ventral compression of the lateral-line, the fibers
innervating the compressed area would be permanently silenced.
It is possible that this type of compression damaged the
delicate fibers innervating the neuromasts in this area or

damaged the neuromasts themselves.

46

 
 

 

 

As was stated in the previCD

U3 S
e '
Ctlon the lateral line

recordings were made on preparaiiions in which the lateral-
line nerve was cut free from its cantral connection with the
brain. Results of preliminary experiments did not demon—
strate any obvious change in the Spontaneous activity upon
cutting the nerve centrally. It should be mentioned that
Spontaneous activity was observed from the centrally cut end
of the nerve on several occasions. This activity, however,
did not seem to arise from the brain, but rather from a loca-
tion within a few millimeters from the recording electrodes.
The Spontaneous activity may have originated from a neuromast
group which was still attached to the main nerve at the
recording electrode site or it may represent an injury dis-
charge of the type described by Hoagland (1954b). Similar
discharge was also seen from the distal end of the nerve,
arising from an area a few millimeters distal to the record—
ing electrodes. This type of discharge occurred only rarely
and was not present in preparations which were intended for
analysis of Spontaneous activity frequency. One could
quickly determine its presence by noting the lack of reSponse
to mechanical stimuli.

Spontaneous activity did not seem to be influenced by
the contralateral nerve (the lateral-line nerve on the other
side of the fish). Leaving it intact, cutting it free from
central connection with the brain, or completely removing it

from the fish had no effect on the level of Spontaneous

 

 

 

 

hypo
atta

 

48

activity in the opposite nerve. Similarly, transectiOn or
complete destruction of the Spinéil Cord aISO had little in—
fluence on lateral-line Spontaneous activity.

Nerve cutting experiments. Counting the number of
nerve impulse peaks (spikes) over a given time interval from
oscillosc0pe photographs of Spontaneous activity revealed
impulse frequencies exceeding 2000 spikes per second. This
figure represents a conservative estimate because many of the
larger Spikes which were counted as single impulses may
actually have been the summated effect of two or more smaller
Spikes. Summation will occur if two or more impulses, con—
ducted in separate fibers, reach the recording electrodes
simultaneously. One thus observes a single large Spike rather
than two or more smaller ones. To demonstrate this, the fre-
quency of Spontaneous discharge may be purposely reduced by
cutting through the nerve at points caudal to the recording
electrodes. This will eliminate Spontaneous activity arising
from neuromasts caudal to the cut. With the Spontaneous
activity reduced in this manner the probability of any two
(or more) impulses reaching the electrodes at the same time
is reduced. If summation is reSponsible for creating the
large Spikes one should observe a decrease in the relative
numbers of large Spikes after such an Operation.

A series of ten eXperiments were conducted to test this
hypothesis. In each experiment the recording electrodes were

attached to the centrally cut nerve (Figure 2) at a point

 

49

one-third of the way to the tai.1- (Figare 4). Th

e Spontaneous
activity from this preparation VW35 then photographed from
the diSplay on the screen of the OSCilloscope. Successive
transverse cuts were then made into the skin and through
the nerve starting near the tail and progressing toward the
recording electrodes. After each cut the Spontaneous activity
was photographed and the distance between the point where
the nerve enters the body at the recording electrodes to the
point where the nerve was cut was measured. The frequency
of Spontaneous activity was determined after each cut by count-
ing the number of distinct Spikes on the photographs over a
time interval of 200 msec.

A nerve length after each cut was expressed as a percent

of its original length, the distance between entry of the
nerve into the body at the recording electrodes to its termi-
nation in the caudal peduncle representing 100%. Similarly,
Spontaneous neural activity after each cut was expressed as
a percent of its original frequency. Expressing neural
activity and nerve lengths in terms of percentages enabled
fish of different lengths to be quantitatively compared and
compensated for unavoidable differences in placement of re-
cording electrodes along the side of the fish. Frequency
measurements in uncut nerves varied from fish to fish depend—
ing on where the recording electrodes were placed. If placed
closer to the head the frequency would be greater than if

placed closer to the tail. This was a result of greater or

 

fewer numbers of neuromaStS corltxrlbuting to the o
verall

frequency. By defining the initLial frequency as 100% these
differences were eliminated.

Results of the ten experiments demonstrated that as
more and more of the distal portion of the nerve was removed
a correSponding decrease in the number of large Spikes
(150-200 uv) always occurred. The decrease in numbers of
large Spikes continued until approximately 60% of the nerve
distal to the recording electrodes was removed. With addi-
tional cutting the decrease in number of large Spikes becomes
difficult to distinguish, however, there was a noticeable
reduction in the total number of Spikes. This decrease in
frequency continued until the nerve was cut at the point
where it leaves the body for attachment to recording elec-
trodes. When cut here all neural activity Stopped.

An example of the spontaneous activity patterns associ-
ated with progressive elimination of distal portions of the
lateral-line nerve is shown at the top of Figure 6. Each
photograph correSponds to the oscilloscope display of spon-
taneous activity immediately after each cut was made. The
percentages to the right of each trace represent the amount
of intact nerve distal (caudal) to the recording electrodes.
Notice the decreased number of large Spikes and eventually
the decreased frequency as the nerve is pared down.

At the bottom of Figure 6 is a plot of the frequency of

Spontaneous discharge (expressed as a percent of its initial

 

 

 

 

 

Figure 6.--Spontaneous lateral-line activity in control

fish demonstrating reduced frequency upon
progressively cutting the nerve. Top--Oscil-
loscope traces of neural activity correSponding
to percent of remaining original nerve length

distal to recording electrodes. Recordings

 

from one fish. Bottom-~Plot of Spontaneous
activity (percent of initial frequency) vs.
percent of intact nerve length distal to record-
ing electrodes. Results taken from ten different

fish.

 

 

 

 

 

52

spouuusous Acnvm

WMWmM‘WW “’0 % tuner

 

 

 

WWI/MM) 75% ”‘7‘”
WWW 50% INTACT
— .7. mm
L___J
10 mm [200!"
80' .
p- o
: . - -
' : 60- ° '
’- 3— ' 0
u
E 2 + ' .
§ = .
= E 40' . o O
E .
h
9 , o
,_ 20d -
I O
3 .
I
1"
l I I I I I I I 1
20 40 so so

PERCENT INTRCT NERVE DIST“. T0

RECORDING ELECTRODES

Figure 6

 

 

 

 

55

frequency) versus the percent of the intact nerve distal to
the recording electrodes. This plot was made from results
of the ten experiments just described. The frequency of
Spontaneous discharge does decrease with progressive elimi_
nation of distal portions of the nerve. The curve iS steep
to the left and becomes much flatter to the right. This ob-
servation supports the assumption that summation is occurring
at levels of high Spontaneous activity. With summation one
is unable to distinguish all of the impulse Spikes thus
counting efficiency will be low. This condition correSpondS
to points on the right hand side of the plot. Decreasing

the frequency of Spontaneous activity by cutting the nerve
will reduce the number of large Spikes but at the same time
allow the previously hidden smaller Spikes to now be seen.

As a result, the rate of decrease of activity will remain

low until summation has been significantly reduced. Summa—
tion was significantly reduced after about 60% of the nerve
was eliminated and as can be seen on the plot in Figure 6

the curve tends to become steeper to the left.

In fish preparations to be used for lateral—line fre-
quency measurements later in the study the lateral—line
nerve was pared down by cutting through the nerve about 5 cm
distal to the recording electrodes. The resulting Spontane-
ous activity frequency was about 500 spikes per second.

By paring the nerve down in this manner the problem of sum-

Ination was largely eliminated. This enabled the observer to

 

 

 

 

 

54

accurately distinguish individual impulses and detect subtle

frequency changes which might otherwise be hidden.

Evoked activity from the lateral line
After recording electrodes were attached to the Spon—
taneously firing lateral-line nerve it was found that a
number of mechanical disturbances were able to elicit bursts
of neural activity clearly seen above the background of
Spontaneous activity. This evoked activity was present in
intact nerve preparations as well as pared-down preparations.
Bursts of evoked neural activity could be induced by tapping
gently on the side of the fish holding chamber, by drOpping
water on to the surface of the water overlying the lateral
line, by rippling the water within the chamber, by gently
tapping the fish with a probe or by numerous other methods.
The lateral line was very sensitive to even the slightest
disturbances. The almost unnoticeable vibration set up by
a cooling fan in an amplifier on the same table as the fish
preparation would evoke very clear reSponses from the
lateral line. All of the evoked reSponses just mentioned
were not artifacts introduced into the electronics of the
system but were actually bursts of neural activity. This

was clearly demonstrated by the disappearance of the reSponse

after the nerve had been cut at the point where it leaves the

body for connection to the recording electrodes. Examples

of a few types of evoked reSponses are shown in Figure 7.

It is interesting to note that the stimulus need not be

 

 

 

 

 

 

55

Figure 7.--Examp1es of various evoked reSponses produced
by stimulating the lateral line. (A) Normal
background of Spontaneous activity; (B) Lateral
line stimulated by scraping probe on rough edge
of fish chamber; (C) Stroking finger across
skin over lateral line; (D) Gently tapping head
of fish with probe; (E) Directing a jet of water
directly on the lateral line: (F) Tapping side

of fish chamber with fingernail.

56

SPONTANEOUS ACTIVITY

‘K . . , . II ‘I A

     
  

|————c
0.2 ssc
I 200;”

. EVOKED ACTIVITY

v "I. '.“‘ “ I \
.Ivl’r'”. 'q'I-A" '
-a.'~.a-‘WII.I5WulstétaWW
l———I
: 0.2 SEC

 

I ZOOuV

Figure 7

 

 

57

aiIECtJQIapplied to the lateral line to evoke a reSponse,
For example, very sharp bursts of neural activity were
produced by gently tapping the head of the fish with a probe,
The purpose of investigating evoked responses was not
to determine what constitutes an adequate stimulus but rather
to select one type of stimulus and note its reSponse under
the Stress conditions described previously. The stimulus
selected for this purpose was the impact of a water drOp

hitting the surface of the water in the fish chamber directly

 

over the lateral line. With the aid of a mechanical manipu—
lator a pipet could be held at a fixed position above the
fish, directing water drOps to the preparation from the same
,height throughout an experiment. In every exPeriment the
drop fell a distance of 10 centimeters.

The top trace in Figure 12 shows a typical evoked re—
Sponse caused by the falling water drop. The burst can be
characterized by an immediate increase in the amplitude and
frequency of impulses, the base line of the oscilloscope
trace increasing approximately 200 uv. The first Spikes of

the burst often exceed 600 uV in amplitude and are usually

 

the largest Spikes in the entire burst. There is a gradual
decrease in Spike height as the burst continues and occa-
sionally a second series of large Spikes is seen near the end
of the evoked reSponse. The initial burst of activity lasts
approximately 35 msec and in most cases is followed by a

period of about 20 msec of reduced neural activity.

 

 

 

 

58

F01lowing the period of decreased activity there is a
secondary period of evoked neural activity which corresPohds
to reflected surface waves created by the falling water

drOp (surface waves were allowed to Stop before another drOp
of water fell to the preparation).

By synchronizing the falling of the water drop to the
sweep of the oscillosc0pe as described earlier it was pos-
sible to diSplay the evoked response in the same position
on the screen of the oscilloscope with each drOp. Because
each evoked burst appeared in the same place it was possible
to examine an "averaged" reSponse. This was done by photo-
graphing consecutive oscilloscope sweeps, superimposed upon
each other by keeping the shutter of the camera Open for 20
complete sweeps. Any Spike occurring in the same position
sweep after sweep could be seen on the photograph as a single,
well defined, Spike. The shape of this spike could be con—
sidered an “average" of 20 Spikes. Spikes due to Spontaneous
activity, having no fixed time relation to the falling of
the water, would appear randomly along the movement of the
beam and appear as a cloudy area above the base line when
observed on a photograph.

For purposes of this study it was desirable to ascribe
a quantitative measurement to the evoked reSponse in order
to evaluate changes in it induced by stressful conditions.

By employing the averaging method discussed above it was

possible to do so. For lack of a better method it was decided

 

 

 

 

 

1|

1

59

to use the amplitude of the first Spike in the evoked re-
sponse for this quantitative measurement. The large initial
Spike of the evoked reSponse actually consists of many
smaller, summated spikes. A common term for a spike of

this nature is "compound action potential." The fact that
this Spike is compound can be demonstrated by noting its
gradual decrease in amplitude as the nerve is progressively
pared down by partially cutting through the nerve a few
centimeters caudal to the recording electrodes. The more

the nerve was cut the greater was the reduction in the ampli-
tude of the first Spike. Had this Spike been the result of
an action potential conducted in a single fiber it would have
maintained its original size until the single fiber was cut.
At this time the Spike would have completely disappeared.

A measurement of the amplitude of the first Spike of
the evoked reSponse is a measure of the number of active
fibers contributing to it. The greater the number of fibers
contributing, the larger the spike (Katz, 1966).

The amplitude of the first Spike in an averaged evoked
reSponse was quite well defined and not difficult to measure.
An example of such a Spike can be seen at the tOp of Figure
16b.

The amplitude of the first Spike of the evoked reSponse
under pre-treatment conditions of an experiment was measured,
and subsequent changes in its amplitude were expressed as a

percent of the pre-treatment value. As with Spontaneous

 

    
   

.
n.
'1

6C)

activity, using percents wit?1 evoked responses allowed
comparison among different fiSh to be put 0n the same basis,
In experiments to be discussed later, the evoked activity

as well as the Spontaneous activity were measured in lateral-
line nerve preparations which had been pared down by cutting
through the nerve about three centimeters caudal to the

recording electrodes.

Electrical activity from the heart

The top oscilloscope trace in Figure 8 is an e1ectro-
cardiogram (ECG) from a curarized rainbow trout showing five
consecutive heart beats. An enlarged trace of the electrical
activity from one beat is shown in the second trace. On it
can be seen the three main waves of the ECG, the P—wave,
QRS complex, and the T-wave. The P—wave is the first wave
of the ECG and correSponds to electrical activity created by
myocardial depolarization of the atrium. The QRS complex is
the wave of electrical activity resulting from depolarization
of the ventricle and repolarization of the atrium. The begin-
ning of this wave complex started about 0.2 seconds after
the start of the P—wave. Repolarization of the ventricle
gives rise to the last wave of the ECG, the T—wave. The
waave began approximately 0.4 seconds after the start of the
QRS complex.

The P—wave was typically monophasic, rising in the
positive direction, then falling directly to the base line.

The QRS complex was also monophasic in the positive direction,

 

 

 

 

61

Figure 8.--Typical electrocardiogram (ECG) from rainbow
trout using a unipolar electrode. Top--
OscilloscoPe traces of ECG indicating positions
of the P, QRS, and T waves, and the P-R and Q-T
intervals. Bottom-~Histogram of average rate.
P-R interval, and Q-T interval. Data obtained
from 27 separate fish. Vertical bar indicates

one standard deviation.

 

 

 

-l
NERR'I’ RATE —IIN

 

7: 0.45

 

 

 

 

0.27 PO.5
IOO‘ i=79.8
"'27 -o.4
30'
_ -003
XIILZI
6¢)- n-27
-O.2
40‘
-O.'l
2°
RATE P-R O -T
INTERVAL INTERVAL

Figure 8

 

WAVE INTERVAL - SEC

 

 

 

p'.

V

h
i
I

§ 1

   
     

.
.§

65

lumnever it demonstrated a SlJLght amOunt of negative over-
Shoot at the completion of the WaVe. The T-wave was either
completely monOphasic in the negative direction or biphasic,
starting in the negative direction, overshooting the base
line in the positive direction, then finally returning
directly to the base line.

The exact placement of the ECG electrode within the
fish had a large influence on the amplitude (voltage) of
the ECG waves. If placed deeper into the surrounding muscle
ventral to the heart, the waves would be much larger than if
placed just under the Skin. Careful measurement of wave
voltages was not an important consideration of this study,
however.

The ECG parameters which were not dependent upon precise
placement of the electrode included the heart rate, and time
intervals between the various waves. The two intervals
measured from the ECG were the P-R interval and Q-T interval.
The P-R interval, time from the beginning of the P-wave to
the point of maximum voltage in the QRS complex, is the time
required for the myocardial depolarization to travel from
the pacemaker region in the atrium to the ventricle. The
Q-T interval, time from the start of the QRS complex to the
end of the T—wave, is the time required for complete depolar-
ization and repolarization of the ventricle. The P-R and Q-T
intervals are indicated just below the second trace in

Figure 8.

 

 

 

 

Ii

     

64

Measurements of heart rate’ P~R interval, and Q-T
interval were made<x127 curarized fish. Water temperature
within the fish holding chamber in these as well as all
other experiments was 15°C. The results of the 27 experi-
ments are Shown in the histogram at the bottom of Figure 8,
The mean heart rate was 79.8 beats per minute, the mean

P—R interval was 0.21 seconds, and the mean Q—T interval was

0.45 seconds.
Experiments on Stressed Fish

DDT experiments

Spontaneous activity from the lateral line. Saline
suSpensions of DDT were perfused into the posterior cardinal
vein of curarized trout preparations. The frequency of
Spontaneous neural discharge from the distal end of the
centrally cut lateral-line nerve was monitored using the
Schmitt trigger-decade scaler counting system. The Schmitt
trigger was adjusted to include nerve impulse Spikes of all
sizes.

In six experiments at six different injection levels of
DDT, each fish received a 0.4 ml injection of the suSpension.
The spontaneous activity from the lateral-line nerve was
recorded for one hour before and for two hours after injec-
tion of the insecticide. The dosage level in each injection
ranged from 0.01 mg to 4.0 mg. One fish was used at each

injection level. The blood concentration of DDT resulting

 

 

 

At ‘
5pc
81:

b0

 

 

 

 

. 1
from these injections was call-Cu ated to range from about

2 ppm to slightly over 750 pp“1 immediately after injection.
Calculations were based on an aSSUmed blood volume of 2.25%
of the live weight of the fish (Schiffman and Fromm, 1959).
In addition to the six experiments listed above, four fish
each received a control injection of 0.4 m1 saline (10 per-
cent ethanol). Frequency measurements of Spontaneous activ-
ity were done in a similar manner.

Figure 9 shows the results of these ten experiments.

At the top of this figure is a plot of the frequency of
Spontaneous activity (percent of initial frequency) versus
elapsed time following control injections of saline. The
bottom of Figure 9 shows a similar plot of experiments where
DDT was included in the injections. The six dosages used

in these experiments are indicated at the bottom of the plot.
None of the Spontaneous activity measurements in the six DDT
eXperiments appeared to differ enough from Similar measure—
ments in the four control experiments to warrant statistical
analysis of the data.

In a series of six similar experiments where fish were
immobilized by spinal transection rather than by curariza-
tion the same results were obtained. The Spontaneous discharge
frequency in four fish receiving 0.1 mg, 0.5 mg, 1.0 mg, and
4.0 mg of DDT respectively did not appear to differ from the
measurements recorded in two control fish. These experiments

were also continued for two hours after injection.

 

 

66

Figure 9.--Effect of intravenous injection of DDT on

 

Spontaneous activity recorded from the lateral-
line nerve. Top--Control injections (0.4 m1)

of saline. Four experiments. Bottom--Injections
of DDT-saline suSpensions (0.4 m1). Dosage of
DDT per injection is indicated at the bottom of

the plot. Six experiments, one at each dose.

 

SPONTANEOIS ACTIVITY - PERCENT OF INITIAL FREQUENCY

67

 

 

 

 

 

 

CONTROL
120 i o
no 0 11 O 0O o O o oo 0
UV U 0 O V U 0 o U --
° ° 0 o o
80 "
4O "
I I I I I I I I I I 1—1
20 4O 6O 30 IOO I20
OOT INJECTEO
120' A
I 1 .8. a o
I I." II n . III-alelll IIIIIIIIIIII II IIIII‘-Inil-IIIIIIIIIIIII'll-Ciel-
mo 6"?!“ U“ 69 .A. A
I I3 A
80"
.. III! 001'
o ..... 8'91
A---- _
‘0 n.....o_.2
. ..... OJ
1 e ..... 0.5
......4.0
I I I I I l l I I I I fi
20 4O 60 CO 100 I20
MINUTES AFTER INJECTION

Figure 9

 

 

58

An interesting observation vfiis made in this series of experi—
ments. At dosage levels of 0'5 m9 DDT and above, fish would
exhibit eye twitching movements and occasional trembling of
opercular and jaw muscles. These are symptoms commonly
observed in fish that have been poisoned by DDT. The control
fish showed no such symptoms. The muscular activity was
restricted to the head region and no movement was observed

below the level of the Spinal transection. It is believed

that the symptoms observed in these fish was induced by the

r..

action of DDT. The observations just mentioned demonstrate
that spontaneous activity from the lateral-line nerve remains
unchanged even in the presence of obvious symptoms of DDT
poisoning.

In four experiments individual fish were placed into
water containing a concentration of 15 ppm DDT. The purpose
of these experiments was to observe the spontaneous activity
from the lateral-line nerve and note any abnormal changes in
discharge frequency or discharge patterns after the fish
had reached the terminal stages of DDT poisoning. Shortly
after one hour of exposure to the insecticide fish became
hyperexcitable, demonstrated occasional periods of violent
swimming movements, and exhibited erratic eye and jaw move-
ments. These symptoms increased in severity for about three
hours. After about four hours of exposure fish became calmer
and laid almost motionless on the bottom of the aquaria.

The only movements that could be observed were erratic eye

 

69

movements, opercular trembling, and occasional trembling in
the trunk. Fish were kept for two more hours in this con-
dition (six hours total exposure) before being removed for
the preparation table. Fish were then pithed and connected
to the recording apparatus as before. None of the four
fish showed any signs of abnormality in the discharge of
Spontaneous activity from the lateral-line nerve.
On three occasions successful attempts were made at

injecting a saline suSpension of 100 ppm DDT directly into

 

the lateral-line canal. Recordings of neural activity were
made before and after the injection as in other experiments.
Here, as before, no change could be detected in the frequency
of lateral-line discharge.

Evoked activity from the lateral line. Measurements of
evoked neural reSponses from the lateral line were also
conducted in all of the DDT experiments. No changes could
be found in the amplitude of the response nor any other para-
meter associated with the evoked burst following any of the
DDT treatments.

Heart activity. The heart rate was the only ECG para-
meter measured in the DDT experiments. None of the DDT
treatments had a significant effect on the heart rate. This
was also true for fish that had been exposed for six hours
to 15 ppm DDT in the water. Casual observation of the ECG

waves in these fish also showed no apparent abnormalities.

 

7O

HYPOXia experiments

The oxygen tension in the water perfusing the gills of
the fish preparation was lowered by reducing the flow of
oxygen to the head tank (Figure 1). A total of nine fish
were used in these eXperiments, three fish at each of three
reduced oxygen levels. The three dissolved oxygen levels
used, expressed as oxygen partial pressures (p02), were 50,

40, and 50 mm Hg. On a part per million basis these values I
correSpond to 5.5, 2.8, and 2.1 ppm, respectively. I
Spontaneous activity fggmethe lateral line. Measure-

ments of Spontaneous activity were conducted for two hours
with the fish eXposed to oxygen saturated water, p02 of 160
mm Hg, and then for three hours after the oxygen tension had
been reduced. Frequency measurements of neural activity

were obtained from oscilloscope photographs rather than using
electrical counting methods because of an uneXpected mal-
function in the preamplifier. Although much more time consum-
ing than electrical counting this method proved to be entirely
adequate.

None of the reduced dissolved oxygen experiments demon-
strated Significant changes in the frequency of Spontaneous
neural activity from the lateral-line nerve.

Evoked activity frqmpghe lateral line. Evoked bursts
of neural activity also remained unchanged following the

stress of hypoxia.

 

 

  

7i

 

Heart activity. Electrocardiograms were recorded in
all of the eXperiments listed above. Figure 10 summarizes
the heart rate data from the above experiments. The three
plots in this figure correspond to heart rates obtained at
reduced p02 levels of: 50 mm Hg, (A); 40 mm Hg, (B); and
50 mm Hg, (C).

At a p02 of 50 mm Hg the heart rate decreased slightly
over the three hour period of exposure, drOpping only about
10% from its pre-treatment rate. At a p02 of 40 mm Hg the
heart rate decreased to approximately 80% of its initial
value over the first hour of exposure. During the second
hour of exposure the heart rate gradually increased by about
10% then decreased approximately 20% in the third hour.

A p02 of 50 mm Hg had the greatest effect on the heart rate.
At this level the heart rate decreased over 20% during the
first hour. Following this initial decrease the heart rate
stabilized at approximately the same level for the second
hour and then began to gradually decrease again, approaching
a rate of 60% of its pretreatment value after three hours.

Electrocardiograms taken during the course of these
eXperiments demonstrated some interesting changes. Within
the first 20 minutes of exposure to a p02 of 50 mm Hg the
P-R interval of the ECG increased from a normal value of
about 0.2 seconds to nearly 0.4 seconds. During the same
time period the wave form of the QRS complex changed from a

typically large monOphasic wave to a much smaller biphasic

 

 

 

 

 

 

 

72

Figure 10.--Effect of lowered oxygen
heart rate. T0p3-Oxygen
50 mm Hg; Center--Oxygen

40 mm Hg; Bottom--Oxygen

50 mm Hg. Three fish used in each experiment.

tension in water
partial pressure
partial pressure

partial pressure

on
of
of

of

Heart rate is expressed as a percent of the

initial rate in the unstressed condition.

HEART RATE- X or INITIAL RATE

 

 

 

100
I .

m E O A
“d

20-

 

 

‘
N
u-

100
80

 

40 "
2°-

 

 

l I
I 2 3
ELAPSED TIME - HOURS

Figure 10

—“‘_

 

 

 

74

wave. The T-wave also changed during this period, becoming
progressively larger both in amplitude and duration. The
Q-T interval was very difficult to measure because the end
of the T-wave was not well defined.

The delay of ventricular depolarization and repolariza_
tion created by the prolonged P—R interval caused the T—wave
to nearly coincide with the P-wave of the next heart beat.

This event, however, did not seem to change the basic heart

 

rate. For the duration of the first hour the changes just
described for the first 20 minutes persisted. The PdR
interval remained unchanged, the QRS complex continued to
demonstrate a lower-than-normal voltage, and the T-wave
continued to slightly increase in size. Occasionally the
QRS complex would regain its monophasic form, but only for
a few minutes at a time and at a subnormal voltage.

The decrease in heart rate during the first hour did
not appear to be caused primarily by a decreased pacemaker
rate because consecutive P-waves maintained normal intervals
on the ECG. However, an entire beat would occasionally be
missed, the ECG Showing no activity from the end of a T—wave
to the beginning of a later P-wave. This silent period would
last for the duration of one cardiac cycle and then the P-wave
would reappear exactly one beat later. Less frequently the
Silent period would last for a duration of two beats. The
heart rate was calculated by counting the number of complete

cardiac cycles over a period of one minute. The apparent

 

 

 

     

75

decrease in heart rate was primarily due to missed beats,
not to a reduced peacemaker rhythm. This was largely true
for the first hour. .During the second hour, however, the
pacemaker rhythm did decrease. This was evidenced by an
increase in the interval between consecutive P—waves. Also
during the second hour the heart became more regular, missing
very few beats. The regularity of the heart during the
second hour accounted for the apparent plateau in Figure 10c.
The decrease in rate during the third hour resulted from a
combination of reduced pacemaker rhythm and reappearance of
missed beats.

Although changes in the ECG patterns given above were
reSponses to a p02 of 50 mm Hg, similar changes were observed
in eXperiments conducted at oxygen partial pressures of 40
and 50 mm Hg. In these cases the exposure times required
to elicit similar changes were longer, but nonetheless present.

Figure 11 shows a series of ECG records from two fish
after various exposure times to a partial pressure of oxygen
of 30 mm Hg. The top six traces in the figure (a-f) demon—
strate most of the abnormalities discussed above. The first
trace (a) is a normal ECG from a fish exposed to oxygen-
saturated water (p02 of 160 mm Hg). The second trace (b) is
the ECG in the same fish after ten minutes of eXposure to a
partial pressure of 50 mm Hg. Note the increased P-R interval,
the biphasic QRS complex, and enlarged T-wave. In this trace,

as well as in those following, the T—wave partially overlaps

 

 

 

76

Figure 11.--Examples of ECG changes during hypoxia.

Oxygen partial pressure in the water was

reduced to 50 mm Hg. T0p--(a), ECG at 160

mm Hg p02; (b—f), ECG at times after the p02

in the water had been reduced. Bottom--Same

experimental conditions upon a different

fish. (a), ECG at 160 mm Hg p027 (h-j),

ECG at times after p02 was reduced.

 

POrST

I Cilia

I——-——'I
h - IOMin 'ISEC
IllmV

Figure 11

 

 

 

 

of

Af‘

 

78

with the P-wave of the next heart beat. The third trace (c)
shows a QRS complex after it had temporarily regained its
monophasic shape. Its amplitude, however, is much smaller
than normal. Trace f shows a missed ventricular beat
evidenced by two consecutive P-waves in the absence of a
QRS complex, a phenomenon occurring quite frequently in later
stages of hypoxic stress. The four traces (g-j) were ob-
tained from a different fish under the same experimental
conditions and are given for comparison. Note the very large
T-wave after the oxygen tension had been lowered for 17 and
25 minutes (i and j).

Two of the fish tested at an oxygen partial pressure
of 50 mm Hg were allowed to recover from the hypoxic stress.
After three hours of exposure to this reduced dissolved
oxygen level, the saturation level was re—established.
The ECG from both fish showed signs of recovery within minutes.
The biphasic QRS complex returned to its normal mon0phasic
Shape within eight minutes. Within 20 minutes the ECG ap-
peared normal except for a Slower rate. In both fish it
took approximately one hour for the rate to fully recover.
The ECG was carefully observed for about three hours after

complete recovery and no ill effects were noted in the ECG.

Asphyxia experiments
Spontaneous activity from the lateral line. Nine fish
were subjected to conditions of aSphyxia by discontinuing

the flow of oxygenated water over their gills. This was done

 

 

79

by Clamping the input and output tubes of the fish chamber,

The effect of aSphyxiation upon Spontaneous activity
from the lateral line was quite severe. Reduction of dis-
charge frequency occurred soon after the water flow had
ceased; six out of nine fish exhibited complete loss of Spon—
taneous activity in less than ten minutes. The other two
fish lost Spontaneous activity in about 20 minutes.

After the Spontaneous neural discharge had ceased the
clamps were removed from the input and output tubes and
oxygenated water was again allowed to perfuse the gills.

In every case the Spontaneous activity returned to normal
within 40 minutes, the majority attaining normal levels in
20 minutes.

Evoked activity from the lateral line. The discharge
of neural activity resulting from water drop stimuli was
also measured in these experiments. It was found that the
evoked reSponse was much more resistant to this Stress than
was the Spontaneous activity. At the time Spontaneous
activity had disappeared about 50% of the evoked reSponse
still remained. With the Spontaneous activity gone the
evoked reSponse could be clearly seen rising above the silent
base line. Figure 12 shows six oscilloscope traces of the
typical reSponse of Spontaneous and evoked activity to
conditions of aSphyxia. Neural activity to the left of the
arrow at the bottom of the figure is Spontaneous activity

and neural activity to the right is evoked. Note the rapid

 

 

 

80

Figure 12.--Effect of aSphyxiation on Spontaneous and

evoked neural discharge from lateral-line
nerve. Numbers to the right represent the
elapsed time in minutes after water flow
over the gills was discontinued. Arrow at
the bottom marks the time when the water
drop stimulus hit the water over the lateral
line. Spontaneous activity to the left of

the arrow and evoked activity to the right.

{u

5
n

,
. 'I

It”

 

svonnuzous "MED
IIISCNAIGE DISCHARGE

I ‘.
M" 5' 0 Min

v' v I. Mm . ,
MKW'WW WI I'Wq

7 Nil

I—-——"I
IO ISEC

  

Figure 12

 

I I
J

I _

‘ -I

U
Q.
A?

82

 

decrease in Spontaneous activity and the relative persistence
of the evoked activity.

The relationship between Spontaneous and evoked activity
can be best described by the lower plot in Figure 17. In
this plot each point represents the evoked activity which
correSponded to the Spontaneous activity at a given time dur-
ing aSphyxia. It should not be implied by the position of
the ordinate and abscissa that the evoked activity is neces-
sarily dependent on the Spontaneous activity. The graph was
constructed for the purpose of demonstrating a relationship
rather than dependence. Evoked activity is exPressed as a
percent of its initial amplitude (first Spike) and Spon—
taneous activity expressed as a percent of its initial
frequency. The curve through the points was fitted by eye.
When Spontaneous activity was reduced by half the evoked
activity was still at 80% of its initial level. When Spon-
taneous activity disappeared the evoked activity had de-
creased 50%.

Recovery of the evoked reSponse was also observed when
oxygenated water was again allowed to perfuse the gills.

Full recovery was observed in about 20 minutes.

Heart activity. Within four minutes after the onset of
aSphyxia the amplitude of the QRS complex decreased by more
than 70%° The P-R interval increased and the duration and

amplitude of the T-wave increased. The heart rate decreased

 

 

from a

than 40
fell tc
was aln
tmewa
followi
and tht
second
it was
W
being
recove
two mi

The w;

Cyani

three
1 Ppn
actl‘

hear

anal
Phot
Meas

to t

 

83

from a normal value of about 80 beats per minute to less
than 40 per minute. After seven minutes the heart rate
fell to about 20 beats per minute and the T-wave of the ECG
was almost gone° Electrical activity from the heart at this
time was very erratic and difficult to interpret. In the
following minutes the waves of electrical activity weakened
and the heart would beat only once every five to seven
seconds. Within ten minutes after the onset of aSphyxiation
it was judged that heart function, at best, was very poor.
When water flow across the gills was again resumed after
being off for a total of from 20 to 40 minutes the heart
recovered rapidly. Regular rhythm was restored in less than
two minutes and the normal rate in less than five minutes.

The wave form of the ECG appeared normal within two minutes.

Cyanide experiments

A total of nine fish were used in cyanide experiments,
three fish at each of three cyanide concentrations (0.5 ppm,
1 ppm, and 4 ppm KCN). In each experiment the Spontaneous
activity, evoked activity, and electrical activity from the
heart were measured.

Spontaneous activity from the lateral line. Frequency
analysis of Spontaneous neural discharge was conducted using
photographic rather than electrical counting techniques.
Measurements were made for a two-hour period before and up

to ten hours after exposure to the cyanide.

 

84

The bottom (solid) lines on the plots in Figures 15,

14, and 15 correSpond to the Spontaneous activity measure-
ments in the three series of experiments done at 4 ppm, 1 ppm,
and 0.3 ppm, reSpectively. Spontaneous activity is plotted
against accumulated dose per gram of fish rather than time
because of the size differences encountered between fish.
Small fish (less than 200 g) reacted much quicker to the
poison than the larger fish (over 500 9) because a constant
cyanide flow was being forced over the gills irreSpective of
the size of the fish. A small fish was thus forced to receive
a larger dose of cyanide per unit weight than a larger fish.
From a toxicological standpoint it was more meaningful to

take fish weight into consideration after studying the condi-
tions of these eXperiments.

As can be seen from Figures 15, 14, and 15 the Spontan-
eous activity from the lateral—line nerve decreased markedly
following exposure of fish preparations to the three levels
of cyanide. The most rapid decrease in Spontaneous activity
occurred at the higher exposure concentrations. At 4 ppm
(Figure 13) the Spontaneous activity decreased to half its
original level after about 150 ug/g of cyanide had passed over
the gills and approached a zero level of activity after about
700 ug/g. At exposure levels of 1 ppm (Figure 14) the Spon-
taneous activity approached half of its original value after
approximately 500 ug/g of cyanide had passed the gills and

after 700 ug/g the Spontaneous activity was about 20% of its

 

 

85

Figure 13.--Effect of 4 ppm KCN on spontaneous and evoked
neural discharge from the lateral-line nerve.
Abscissa represents the amount of KCN passed over
the gills per gram of fish. Ordinate represents
the lateral-line neural discharge expressed as
a percent of the initial (pretreatment) activity.
Top curve (dashed) is the evoked discharge and
the bottom curve (solid) is the Spontaneous
discharge. Symbols similarly shaped represent

the same fish.

 

86

ma mnzmum

mmHm ho w\ZUM ml II mAAHU mm>o QMmm<m mnHz<wU mo HZDOE<

09¢ eat
. . .

ecu
_

 

 

w. n. —
N; r S u. a
._ ... .

.. ”a a i. a a, a

.n ._ _: ... .2 .s v

2,. ...: ... ... I ... ...
. ._ :. ... ... ... . _ _
_. a.“ ... v. v. .. ..i. z. —:

 

HDHVHDSIG TVHHHN TVILINI J0 lNHOHHd

 

 

 

Figure 14.--Effect of 1 ppm KCN on spontaneous and evoked

neural discharge from the lateral-line nerve.
Abscicca represents the amount of KCN passed
over the gills per gram of fish. Ordinate
represents the lateral-line neural discharge
expressed as a percent of the initial (pretreat-
ment) activity. Top curve (dashed) is the
evoked discharge and the bottom curve (solid)

is the Spontaneous discharge. Symbols similarly

Shaped represent the same fish.

 

 

8&3

ecu
{ I

dd wusmflm

an: ..8 mass m: 3 made EB Emmi 2:36 so 2:92..

coo ecu
_ .

 

 

   

ESXVHDSIG 1VHRSN TVILINI £0 INHOHHJ

 

 

 

 

m7 ‘7" "

 

 

 

89

Figure 15.—-Effect of 0.5 ppm KCN on spontaneous and evoked

neural discharge from the lateral-line nerve.
Abscissa represents the amount of KCN passed
over the gills per gram of fish. Ordinate
represents the lateral-line neural discharge
expressed as a percent of the initial (pretreat-
ment) activity. Top curve (dashed) is the
evoked discharge and the bottom curve (solid)

is the Spontaneous discharge. Symbols similarly

shaped represent the same fish.

 

 

mfi

shaman

  

mmHh ho M\ZUM ml II mAAHO mu>o numm<m NGHZ<>U no 825024

 

 

0.” 06¢ 00¢ eon
- P b n - n -
ION
I
.av
.
‘
. I
‘ . . ‘
I :69
I C
I . ‘ I
0 .ea
COIOII'IIOIIQIIIOSII < 4 . .
a UIIIIQIIIIIIIOIIDIII . a u
D u IIIJIIOIDIIIIIIIIIIIIII o d
d 4 (I --. :—

 

HDHVHOSIG TVHQHN TVILINI d0 lNHDHQd

 

91

initial value. Fish exposed to 0-3 ppm of cyanide (Figure
115) éiixi not react as fast to the chemical as with levels
‘Df 3— Eppm and 4 ppm. Fish required nearly 400 ug/g of cyanide

before the spontaneous activity from the lateral line fell
to

50%, and after exposure to 700 ug/g of the poison the
a<=tlli—\Iity was still at 40%. Figure 16 (top) shows oscillo—
scz<>l;>ee traces of Spontaneous activity in one fish after
e3<l§’<>£5ure to 1 ppm KCN. For this particular fish approximately
1£3(:) LLg KCN per gram of fish passed over the gills per hour.

Evoked activity from the lateral line. The tOp lines
(dashed) in Figures 15, 14, and 15 correspond to the evoked

acz‘lzivity measurements after exposure of fish preparations to
tklfia same three concentrations of cyanide. In every case the
e“’<3}<ed activity decreased more slowly than the Spontaneous
a~<=tivity. In experiments with cyanide concentrations of

4 Iapm (Figure 15) the evoked activity fell to approximately

SCX% of its initial value after slightly over 300 ug of cyanide
EHEr gram of fish had passed the gills. The evoked activity

remnained at this level until about 700 ug/g passed the gills;
then the evoked reSponse fell off sharply.

After approxi-
Inately 900 ug/g had passed over the gills the evoked reSponse

nearly disappeared.

Experiments at 1 ppm cyanide demonstrated a more gradual

decrease in evoked activity (Figure 14). The evoked reSponse
gradually declined until the accumulated cyanide dose was
700 ug per gram of fish. The evoked activity then fell off

 

 

 

 

 

 

 

Figure 16.-—Effect of 1 ppm KCN exposure on Spontaneous

and evoked neural discharge from the lateral-
line nerve. Top--Spontaneous activity at
indicated elapsed times during KCN exposure.
Bottom--Evoked discharge from lateral line at
indicated times after exposure. Each trace
represents 20 superimposed sweeps of the

oscilloscope beam.

 

 

 

- 3|"

  
 

. 3hr 30 Min

3 use I zoo "'

Figure 16

 

 

 

  
  

, l
”V
L I

  
      
    

.u
I
I

94

more Sharply, reaching 50% Of its initial value after exposure

to 9 OO (lg/g.

Exposure to 0.5 ppm of cyanide (Figure 15) resulted in
a very mild decrease in evoked activity from the lateral line.
The decrease was nearly linear, the response falling only 20%

aft-ell: the fish was exposed to 700 ug of cyanide per gram of
fish -

The decrease in evoked reSponse during 1 ppm eXperiments
ls Shown in the bottom series of oscilloscope traces in

FiQUre 16. Traces at the top and bottom of the figure were

o13ta.ined from the same fish.
The relationship between Spontaneous and evoked activity
is; Shown at the top of Figure 17. This is the same type of

Plot that was described in the section on asphyxia. Each

Point represents the evoked reSponse observed at a given
1eVel of Spontaneous activity. Data from all three dosage
le\rels are included in the plot. AS can be seen there was
VGBry little difference in the spontaneous-evoked relationship
1Detzween cyanide experiments and aSphyxia exPeriments.

Recovery of the Spontaneous activity and evoked reSponse

was found to occur after the flow of cyanide to the prepara-

tion was discontinued. All three fish exposed to the 4 ppm

level recovered, both the Spontaneous and evoked neural
activity returning to normal levels after about one hour.

Recovery experiments were not done at the other two cyanide

levels.

 

7?

4!

 

 

 

 

Figure 17.-~Relationship between Spontaneous and evoked

 

 

neural activity during aSphyxia and cyanide
exposure. The abscissa represents the Spon-
taneous activity (percent of the initial
frequency) and the ordinate represents the
evoked activity (percent of the initial
amplitude). Plot at the tOp was constructed
from data obtained from cyanide experiments
and the bottom plot from data obtained in

aSphyxia experiments.

 

EVOKED ACTIVITY - PERCENT OF INITIAL DISCHARGE AMPLITUDE

 

96

1001

 

 

401 0
CYANIDE EXPERIMENTS
20-
l I I I I I U | 1 '
100]

 

ASPHYXIA EXPERIMENTS

 

 

l I I I I ' '

I I
20 40 60 80 100

SPONTANEOUS ACTIVITY -- PERCENT OF INITIAL FREQUENCY

Figure 17

 

 

...: I
A j " I. i '1.

u, *

97

Heart activity. Electrocardiograms were recorded
throughout the cyanide experiments. Figures 18, 19, and
20 show the effect of cyanide upon the heart rate. The
percent of the initial rate was plotted against the accumu-
lated dose of KCN which passed over the gills per unit
weight of the fish. Nearly every fish in the three series
of experiments demonstrated an initial rate increase of 5
to 10%. Following this initial increase the heart rate
dropped sharply to about 20% of the initial rate in the
experiments conducted at 4 ppm KCN, and to approximately 40%
in experiments at 1 ppm and 0.5 ppm. The low point of this
rapid rate decrease occurred at nearly the same point in
each series of experiments: after about 100 ug of cyanide
per gram of fish had passed the gills. After this period
of reduced heart activity the heart rate increased. The
heart rate in the 4 ppm experiments increased from 20% to
roughly 50% by the time 500 ug/g had passed the gills. From
this point on the rate gradually decreased, falling just
under 20% after an accumulated eXposure of 1000 ug/g. In
experiments conducted at a concentration of 1 ppm KCN the
heart rates demonstrated similar increases. The heart rates
rose from 40% to approximately 60% after the fish were ex-
posed to 500 ug/g and in two of the three fish the rate
remained near this level until exposed to 800 ug/g. The rate
in the third fish fell in a manner very similar to heart

rates in fish exposed to 4 ppm KCN. The heart rates in the

    
 

  

  

 

 

9?

 

 

 

 

 

 

Figure 18.-—Heart rate effects during exPosure of fish

preparations to 4 ppm KCN. Abscissa repre-
sents the amount of KCN passed over the gills
per gram of fish. The ordinate represents
the heart rate expressed as a percent of its
initial value. Different symbols represent

different fish.

 

 

 

 

99

 

ma muzmflm

mm: .8 323 w: I made ago 33% 2256 so 5592

...: 8. 2x. 2: ...:
P

 

 

 

31V! TVILINI J0 LNEOXSd -- ELVK IHVEH

 

 

 

Figure 19.--Heart rate effects during exposure of fish

preparations to 1 ppm KCN. Abscissa repre-
sents the amount of KCN passed over the gills
per gram of fish. The ordinate represents the
heart rate expressed as a percent of its
initial value. Different symbols represent

different fish.

 

 

 

101

I
ma wusmam
mmHm no m\zox m: In mngo mm>o nmmm<m mnHz<wo mo HZDOZ<
000— can 000 00¢ 00“
I) b» b n h b) p. n p -

 

 

 

HIVH TVIIINI do lNHDHHd -- BIVU IHVHH

 

 

 

 

 

Figure 20.—-Heart rate effects during eXposure of fish

preparations to 0.5 ppm KCN. Abscissa repre—
sents the amount of KCN passed over the gills
per gram of fish. The ordinate represents
the heart rate expressed as a percent of its
initial value. Different symbols represent

different fish.

 

 

 

105

ON musmflm
xmam ac w\zox ma ul msuau mm>o ammmam mon<wu mo ezpoz<
00» coo 00¢ DON
- p n - n p n -

 

 

 

-— SIVH JUVHH

HIVH TVIIINI JO lNHDHBd

 

104

other two fish decreased to 30% of original levels after
1000 ug/g. An increasing heart rate after an initial drOp
was also observed in fish that were exposed to 0.3 ppm KCN.
This increase was very gradual, reaching the 60% level of
heart activity after an accumulated eXposure of 700 ug/g.

The wave forms observed from the ECG recordings taken
during these eXperiments demonstrated changes that were
nearly identical to those observed in the hypoxia experiments.
These included: reduction in the amplitude of the QRS com-
plex; change in the QRS complex from a monOphasic wave form
to a biphasic wave form; increased length of the P—R interval;
and increased size of the T—wave. The slight increase in
heart rate at the beginning of the cyanide experiments re—
sulted from a slight increase in the rate of the pacemaker
as evidenced by the shorter intervals between successive
P-waves. None of the other measurements conducted on the ECG
demonstrated any changes during this period. The changes in
the ECG described above began during the rapidly falling
phase of the heart rate. The decrease in rate resulted mostly
from missed beats but the pacemaker discharge rate also de-
creased slightly. After the initial drop in rate the heart
became much more rhythmic, beating at regular intervals.

This accounted for the rate increase observed after the large
initial drop. The rate never reached its pretreatment value
because of the substantially decreased pacemaker rhythm.

The eventual decrease in heart rate resulted from a combi-

nation of reduced pacemaker rhythm and missed beats.

105

In the recovery experiments after eXposure of fish
preparations to 4 ppm KCN the heart rate in each fish
approached its pretreatment value within about 40 minutes.
Recovery of the ECG waves to their normal Shape occurred

within 20 minutes.

 

 

 

DISCUSSION

DDT Experiments

 

The finding that DDT exposure did not change any
electrOphysiological measurements conducted on the lateral
line was quite interesting. These were unexpected results
considering the profound effects of this chemical on the
nervous systems of other animals. The first explanation
which came to mind concerning these "negative results" was
that the insecticide was not actually getting to the lateral—
line tissue in large enough amounts to elicit changes. This
seemed to be a valid consideration until a few revealing
observations were made later in the study.

The most direct approach to the question of whether
or not DDT was able to get to the lateral-line nerve was to
expose fish to the chemical and then measure the amount of
DDT in the lateral-line nerve. This was done on three fish.
Two fish were given intravenous injections of 2 mg DDT and
the third fish was exposed to 15 ppm DDT in the water for
six hours. A fourth fish, not exposed to DDT, was used as
a control. The lateral-line nerve from the fish exposed to
15 ppm DDT in the water for six hours contained a DDT con—

centration of abouts ppm. The lateral-line nerves from the

106

 

 

 

 

107

two fish receiving the intravenous injection contained 2.2
and 2.5 ppm DDT two hours after injection. No DDT was
found in lateral-line nerves from the control fish.

The blood concentration of DDT was measured in one fish
at various times after injection. Samples of blood were
taken from a second canula placed caudal (upstream) to the
injection canula. The initial blood concentration of DDT
was calculated to be roughly 400 ppm. After 20 minutes the

measured blood concentration of DDT was only 15 ppm; after

 

one hour, 8 ppm; and after two hours, 5 ppm. This evidence,
although obtained from a Single fish, demonstrated that the
injection technique was successful in introducing the
insecticide into the systemic circulation. Disappearance
of the DDT from the blood indicates that it is presumably
entering other body tissues, as evidenced by the appearance
of DDT in the lateral-line nerve from the same fish.

Whether or not DDT was able to get to the lateral line
is probably irrelevant when one considers the effects of DDT
on the whole animal. AS was shown earlier, injections of DDT
did elicit the characteristic poisoning symptoms and at the
same time no changes could be identified in the discharge of
neural activity from the lateral line. This observation
Strongly indicates that the lateral line plays a minor role
in the symptoms characteristic of acute DDT poisoning.
Injections of very large amounts of DDT might well affect the

lateral-line system, however, from the evidence given above

 

 

 

 

 

108

above it is likely that if theSe effects did occur they
would be of minor importance to the fish. By this time the
fish would already have been suffering the terminal effects

of the insecticide by its action on another part of the body.
Hypoxia Experiments

Perhaps the most significant finding in hypoxia eXperi-
ments concerns heart function. Briefly restating the results
of these experiments I found that exposure of fish to oxygen
partial pressures of 40 and 50 mm Hg for three hours reduced
the heart rate by 20% and exposure to 30 mm Hg for the same
length of time reduced the rate by approximately 40%. After
two hours of exposure the heart rate was at the 80% level at
all three of the reduced oxygen levels.

Randall and Smith (1967) describe different results.
They demonstrate that when rainbow trout were exposed to
hypoxic conditions the heart rate rapidly decreased at oxygen
partial pressures below 80 mm Hg. For example at a p02 of
50 mm Hg the heart rate decreased 50% and at 40 mm Hg it de—
creased approximately 60%. These values are well below those
observed in my eXperiments.

The reason for this apparent discrepancy is probably
due to the fact that Randall and Smith did not immobilize
their fish with curare as I did. Curare is a powerful acetyl-
choline antagonist and can block impulse transmission across

synapses utilizing acetylcholine as a chemical mediator.

 

 

 

 

109

Randall and Smith have shown that the heart of the rainbow
trout is innervated by cholinergic inhibitory fibers from
the vagus nerve. A cholinergic nerve is one which releases
acetylcholine at a synapse or neuromuscular-junction. These
inhibitory fibers are responsible for causing reflex inhibi-
tion of heart activity during hypoxic conditions. Because
curare is an antagonist of acetylcholine it is suSpected
that the heart rate was partially prevented from Slowing
down during the hypoxic conditions of my eXperiments because
inhibitory impulses to the heart were blocked by the action
of curare. Randall and Smith have shown that atr0pine
(another acetylcholine antagonist) has a similar effect.
Lateral-line function did not appear to be affected by
the hypoxic conditions of the experiments. It appears as
though the lateral-line nerve is able to maintain normal

function as long as blood circulation continues.
ASPhyxiation Experiments

The rapid decline in normal neural activity of the
lateral-line system under conditions of aSphyxia appears to
be closely related to heart function. During aSphyxia the
heart rate rapidly decreased and in about ten minutes the
heart had nearly stOpped beating. It is probably safe to
assume that blood circulation was severely reduced and perhaps
even lacking in some tissues. Ischemic conditions (lack of

blood flow) not only result in oxygen lack at the tissue level

 

 

 

 

110

but also cause a build-up of metabolites because there is no
means for their removal.

The accumulation of metabolic products in the vicinity
of the nerve is probably responsible for decreased neural
activity in the lateral-line nerve during aSphyxia. Metabo-
lites which can occur include potassium, lactic acid, carbon
dioxide, and perhaps others. Hashimura and Wright (1958)
State that progressive anoxia in frog nerves results in an
accumulation of extracellular potassium. It has been indi-
cated by Adrian (1956) that excessive amounts of extracellular
potassium will cause hyperpolarization of neural membranes
and eventually render the nerves functionless. It is possible
that potassium will also accumulate extracellularly in trout
under the conditions of aSphyxia.

Holeton and Randall (1967b) found that lactic acid will
accumulate extracellularly in trout exposed to hypoxic con-
ditions, presumably due to increased anaerobic metabolism.
It is very likely that this occurred in my aSphyxiation ex-
periments. Carbon dioxide is also known to accumulate under
anoxic conditions. Both lactic acid and carbon dioxide will
depress nerve function. The resulting decrease in pH raises
the surface tension causing the nerve cell membrane to
increase its resistance to water soluble ions (Fox and
Kenmore, 1967). Sodium and potassium passage across the
membrane is hindered and the net effect is hyperpolarization

of the nerve membrane. Hyperpolarization, as was demonstrated

 

 

 

 

 

 

111

in the case of excessive amounts of extracellular potassium,
will depress nerve function. Nerve depression is brought
about by increasing its threshold of excitability.

ASphyxiation caused by discontinuing the flow of oxy-
genated water across the gills led to conditions which were
favorable Usaccumulation of metabolites in the vicinity of
the lateral-line nerve. It is the action of metabolites on
the nerve which is believed to have caused the depressed
activity in spontaneous and evoked neural activity from the
lateral—line nerve.

During hypoxia experiments the heart continued to func-
tion throughout the test period hence, a build-up of meta—
bolic products in the vicinity of the lateral-line nerve was
probably prevented. This would eXplain the inability of the
hypoxic stress to depress neural activity in these exPerimentS.

It was interesting to note that the evoked activity
from the lateral-line nerve was more resistant to conditions
of aSphyxiation than was the Spontaneous activity. It is
tempting to explain this phenomenon based on a theory that
different fibers are reSponSible for conducting evoked and
Spontaneous nerve impulses. If fibers reSponsible for carry-
ing the evoked nerve impulses were less susceptible to
ischemic conditions than were the fibers carrying Spontaneous
impulses it follows that the Spontaneous activity would be
first to disappear. I have no evidence to support this

however.

 

 

 

 

112

It is more likely that both Spontaneous and evoked
impulses are carried in the same fibers. This is evidenced
by the findings of Sand (1937) who demonstrated that the
frequency of Spontaneous discharge in single fibers increased
during stimulation. His work was conducted on one of the
lateral—line canal organs in the head of the skate. When one
records activity from a nerve trunk containing hundreds of
individual fibers a Simultaneous increase in firing frequency
of a number of these fibers will appear as a burst of evoked
activity. The evoked activity discussed throughout this
thesis is probably a summated burst of this type and is not
the result of newly recruited fibers.

Ischemic conditions resulting from aSphyxiation will
cause hyperpolarization of neural membranes. In the lateral-
line nerve the hyperpolarization presumably affects all of
the fibers equally. It is possible that events reSponsible
for initiating evoked and Spontaneous activity differ in
their ability to elicit a nerve impulse. If the threshold
of the nerve is raised by hyperpolarization the "stronger“
of the two events would be more able to initiate an impulse.
I maintain that the evoked activity is more resistant to
ischemic conditions than the Spontaneous activity for the
above reason.

.The biochemical and physiological basis for the events
which cause Spontaneous and evoked activity have been discussed

in detail by Harris and Flock (1967). Their model is based

 

 

 

 

 

113

on results of eXperimentS conducted on the Asian toad,
Xenopus 1aevis, and probably applies to fish aS well.

Their model is this: An afferent fiber of the lateral-line
nerve makes synaptic contact with the base of a sensory hair
cell. The t0p of the hair cell is equipped with the cupula
which moves during mechanical stimulation. Next to the area
of synaptic contact with the afferent fiber are numerous
synaptic visicles located within the hair cell. Each vesicle

contains a discrete quantum of chemical transmitter. When

 

the transmitter contacts the postsynaptic membrane (afferent
nerve membrane adjacent to the sensory hair cell) after being
released into the synaptic cleft (Space between hair cell
and nerve terminal) the postsynaptic membrane iS depolarized.
Many efferent terminals from adjacent hair cells unite to
form a single fiber. If enough of the terminals are depolar—
ized by simultaneous release of transmitter from the hair
cells the resulting waves of depolarization (excitatory post-
synaptic potentials, EPSP's) will summate and exceed a
threshold in the region where the fibers join. This is the
region where the regenerative action potential (nerve impulse)
begins. Depolarization of this region beyond a certain
threshold value will initiate events leading to an “all or
nothing" action potential which then is conducted to the
central nervous system.

Spontaneous neural activity is presumed to result from

random leakage of transmitter substance from the hair cell

 

 

 

 

 

114

into the synaptic cleft. When enough quanta are simultaneous-
ly released from different hair cells, the EPSP in the regen-
erative region will exceed threshold and a Spontaneous

impulse will ensue. An evoked action potential is not the
result of random leakage of the transmitter. Upon Stimulating

the sensory cells in a given region of the lateral line, the

 

cells will Simultaneously release quanta of transmitter sub-
stance giving rise to a large EPSP and eventually an action
potential in the nerve. These events do not depend on the
probability of several hair cells releasing transmitter quanta
at the same time as was true for Spontaneous impulses.

AS the nerve terminals encounter conditions of ischemia
the nerve membranes begin to hyperpolarize due to the action
of the accumulating metabolites discussed earlier. This pro-
gressive hyperpolarization increases the threshold of excit—
ability in the regenerative region of the efferent fiber.

AS a result, more and more hair cells must simultaneously
release transmitter substance in order to develop an EPSP
which is large enough to exceed the rising threshold. The
EPSP'S developed from random leakage of transmitter material
are generally much smaller than the EPSP'S created by the
Simultaneous evoked release of the transmitter after stimula-
tion. Thus, as the threshold rises, Spontaneous nerve im-
pulses will disappear before the evoked impulses. During
recovery from ischemic canditions, the evoked impulses will

be first to appear followed by the Spontaneous impulses.

 

 

 

115

The observations made concerning the disappearance and re-
appearance of evoked and spontaneous neural activity in my
aSphyxiation experiments agree very well with the model
described above.

The rapid decrease in heart rate which occurred during
the aSphyxiation experiments probably resulted from the
direct effects of anoxia upon myocardial tissue rather than
reflex inhibition mediated by inhibitory neurons to the
heart. This was evidenced by the observation that gradual
slowing of the basic heart rate did not occur. Rather, the
rhytum was progressively interrupted by periods of missed
beats and the wave patterns on the ECG became very erratic
and weak. Reflex neural inhibition would cause a rate

decrease in the absence of such drastic changes in the ECG.
Cyanide Experiments

All three concentrations of cyanide used in these experi-
ments (0.3, 1, and 4 ppm KCN) caused reduction in heart rate,
spontaneous activity, and evoked activity from the lateral-
line nerve.

Doudoroff et al. (1956), demonstrate that molecular HCN
is responsible for the toxic effects of cyanide compounds
and not the ionized CN— form. The cyanide ion is only found
in minute quantities after KCN is dissolved in acid, neutral,
or slightly alkaline water. At a pH 8.0, 93% of the KCN

exists in the undissociated HCN form (Wuhrmann and Woker,.

 

 

 

 

116

1948) and at lower pH's this percent increases. The pH of
the water used to perfuse my fish preparations was less than
8,0, thus it can be assumed that nearly all of the potassium
cyanide was in the form of the undissociated HCN molecule.

In considering both the lateral-line nerve and the heart
reSponses to cyanide it was evident that the higher concen—
trations of cyanide would elicit changes faster than the
lower concentrations. From this observation it was surmised
that the reSponse time was a function of the accumulated
amount of cyanide which had passed over the gills of the fish
preparation. It was also observed that smaller fish would
react quicker to the cyanide than would the larger fish at
a given concentration. This indicated that the weight of
the fish should be taken into consideration when comparing
cyanide effects between different fish. For these two reasons
the magnitude of the physiological reSponses observed in the
cyanide experiments was plotted against the amount of cyanide
which passed over the gills per unit live weight of fish.

When the data were plotted in this manner (Figures 13 through
18) the position of a major physiological change would usually
occur at the same position on the abscissa. This was largely
true for heart rate data. For example, the rapidly decreasing
phase of the heart rate occurred when approximately 50 ug of
KCN per gram of fish has passed over the gills (Figures 16, 17,
and 18). Lateral-line reSponses did not coincide as well,

however.

 

 

 

Decreases in Spontaneous and evoked activity from the

lateral line were found to be greater at the higher cyanide
concentrations, even though the same amount of cyanide had
passed over the gills per unit weight of fish. This might
be explained by an excretion or detoxifying process if the
following assumptions are made: (1) The detoxification or
excretion process proceeds at a relatively fixed rate below
that of the cyanide uptake rate; and (2) The rate of uptake
of cyanide across the gills is a relatively linear function
of the cyanide concentration in the water.

If these assumptions are true, at lower cyanide concen-
trations a greater fraction of the cyanide crossing the gills
would be detoxified or excreted than higher cyanide concen-
trations. The net effect of such a process is that higher
concentrations of cyanide will “load" the fish with the poison
faster than at lower concentrations. This could explain why
higher concentrations of the chemical have a greater physio—
logical effect than lower concentrations even though the
accumulated dose per gram of fish was the same.

Lateral-line responses in cyanide experiments were simi-
lar to those observed in aSphyxiation experiments. Evoked
reSponses persisted long after the'Spontaneous activity had
ceased, indicating that a hyperpolarization process in the
nerves may have been occurring. Heart function was severely
affected and ischemic conditions may well have been develop—

ing the lateral line. Anoxic conditions were also very likely.

 

118

Cyanide affects the electron tran5port system so that
oxygen cannot be biochemically utilized by body tissues.
This causes tissue anoxia even though the blood may be
saturated with oxygen. Cyanide may also directly affect
the nerve cell membrane (Schoepfle, 1963) but this was dif-
ficult to determine from these experiments.

An interesting finding was that fish preparations were
able to recover from cyanide poisoning. Recovery was noted
in the heart and also Spontaneous and evoked activity from
the lateral-line nerve. This suggests that a continuous
detoxification or excretion process is in effect.

The effect of cyanide on the heart was puzzling. The
rapid reduction in heart rate during the early exposure
periods was probably due to anoxia of the myocardial tissue.
The heart has a very high metabolic demand for oxygen and
cyanide could easily block its ability to use it. Because
the myocardium cannot tolerate an oxygen debt it is not
surprising that it reacted so quickly to cyanide poisoning.
What is puzzling is the temporary recovery of the rate follow-
ing the rapid decrease. This finding was reproducible in
nearly every fish tested. Perhaps this temporary rate re-
covery was not at the expense of an increased energy eXpendi-
ture. For instance, the rate could be higher but at the same
time the force of contraction could be less. This was perhaps
evidenced by the observation that the voltages recorded from

the ECG were lower during this period of temporary recovery.

 

 

LITERATURE CITED

Adrian, R. H. 1956. The effect of internal and external
potassium concentration on the membrane potential of
frog muscle. J. Physiol. London 133:631-658.

Brinley, F. J. 1930. The effect of cyanide on the cardiac
rhythm of embryos of Fundulus heteroclitus. Physiol.
Zool. 3:283-290.

BrOphy, J. J. 1966. Basic Electronics for Scientists.
McGraw-Hill Co. N. Y. 471 pages.

Burdick, G. E., H. J. Dean, and E. J. Harris. 1958.
Toxicity of cyanide to brown trout and smallmouth bass.
New York Fish and Game J. 5:133-163.

Cooper, S. 1923. The rate of recovery of nerves in
aSphyxia. J. Physiol. London 58:41—48.

Cragg, B. G. and P. K. Thomas. 1957. The relationships
between conduction velocity and the diameter and
internodal length of peripheral nerve fibers.

J. Physiol. London 136:606-614.

Dijkgraaf, S. 1962. The functioning and significance of
the lateral-line organs: Biol. Rev. 38251—105.

Doudoroff, P. 1956. Some exPeriments on the toxicity of
complex cyanides to fish. Sewage and Industr. Wastes
28:1020—1040.

Downing, K. M. 1954. The influence of dissolved oxygen
concentration on the toxicity of potassium cyanide to
rainbow trout. J. Exp. Biol. 31:161-164.

Forbes, A. and L. H. Ray. 1923. The condition of survival
of mammalian nerve trunks. Am. J. Physiol. 64:435-465.

Fox, J. L. and P. I. Kenmore. 1967. The effect of ischemia
on nerve conduction. Exptl. Neurol. 17:403-419.

119

 

 

 

Gar

Ha:

 

 

 

120

Garey, W. F. 1962. Cardiac reSponses of fishes in aSphyxic
environments. Biol. Bull. 122:362-368.

Harris, G. G. and R. Flock. 1967. .Spontaneous and evoked
activity from the Xenopus laevis lateral line, p. 135-
161. In Cahn, P., (ed.), Lateral Line Detectors.
Indiana Univ. Press, Bloomington.

Hashimura, S. and E. B. Wright. 1958. Effect of ionic
environment on excitability and electrical properties
of frog single nerve fiber. J. Neurophysiol. 21:24-44.

Heinbecker, P. 1929. Effect of anoxemia, carbon dioxide and
lactic acid on electrical phenomena of myelinated fibers
of the peripheral nervous system. Am. J. Physiol.
89:58—83.

Herbert, D. W. M. and J. C. Merkens. 1952. The toxicity of
potassium cyanide to trout. J. Exptl. Biol. 29:623-
649.

Hoagland, H. 1933 A. Electrical reSponses from the lateral-
line nerves of catfish. I. J. Gen. Physiol. 16:695-
714.

. 1933 B. Quantitative analysis of reSponses from
lateral-line nerves of fishes. II. J. Gen. Physiol.
16:715-732.

. 1934 A. Electrical reSponses from lateral-line
nerves of fishes. III. J. Gen. Physiol. 17:77-82.

. 1934 B. Electrical reSponses from the lateral-
line nerves of fishes. IV. The repetitive discharge.
J. Gen. Physiol. 17:195-209.

Holeton, G. F. and D. J. Randall. 1967 A. Changes in blood
pressure in the rainbow trout during hypoxia. J. Exptl.
Biol. 46:297-305.

Holeton, G. F. and D. J. Randall. 1967 B. The effect of
hypoxia upon partial pressure of gases in the blood
and water afferent and efferent to the gills of rainbow
trout. J. Exptl. Biol. 46:317-327.

Katz, B. 1966. Nerve, Muscle, and Synapse. McGraw-Hill Co.
N. Y. 193 pages.

Kisch, B. 1948. Electrographic investigations of the heart
of fish. Exptl. Med. Surg. 6:31-62.

 

 

 

 

121

Lalonde, D. I. V., and A. W. A. Brown. 1954. The effect of
insecticides on the action potentials of insect nerve.
Can. J. 2001. 32:74-83.

Lehmann, J. E. 1937. The effect of aSphyxia on mammalian
A nerve fibers. Am. J. Physiol. 119:111-120.

Magladery, J. W., D. B. McDougal, and J. Stoll. 1950.
Electrophysiological Studies of nerve and reflex activ-
ity in normal man. II. The effects of peripheral
ischemia. Bull. Johns Hopkins Hosp. 86:291—312.

Matsumara, F., and R. D. O'Brien. 1966 A. Absorption and
binding of DDT by the central nervous system of the
American cockroach. J. Agr. Food Chem. 14:36-38.

. 1966 B. Interactions of DDT with components of
American cockroach nerve. J. Agr. Food Chem. 14:39-41.

Narahashi, T., and H. G. Hass. 1968. Interaction of DDT
with components of lobster nerve membrane conductance.
J. Gen. Physiol. 51:177-198.

Oets, J. 1950. Electrocardiograms of fishes. Physiol.
Comp. Oecol. 2:181-186.

Randall, D. J., and G. Shelton. 1963. The effect of changes
in environmental gas concentrations on the breathing
and heart rate of a teleost fish. Comp. Biochem. Physiol.
9:229-239.

Randall, D. J., and J. C. Smith. 1967. The regulation of
cardiac activity in fish in a hypoxic environment.
Physiol. Zool. 40:104-113.

Robertson, 0. H., M. A. Krupp, N. Thompson, S. F. Thomas, and
S. Hane. 1966. Blood pressure and heart weight in
immature and Spawning Pacific Salmon. Am. J. Physiol.
210:957-964.

Roeder, K. D., and E. A. Weiant. 1948. The effect of DDT on
sensory and motor structures in the cockroach leg.
J. Cellular Comp. Physiol. 32:175—182.

Sand, A. 1937. The mechanism of the lateral line sense
organs of fishes. Proc. Roy. Soc. B, 123:472—495.

Schiffman, R. H., and P. O. Fromm. 1959. Measurement of
some physiological parameters in rainbow trout (Salmo
gairdnerii). Can. J. Zool. 37:25-32.

 

 

122

Schoepfle, G. M. 1963. Effects of cyanide at different
concentrations on single nerve fibers. Am. J. Physiol.
204:77-80.

Schoepfle, G. M., and F. E. Bloom. 1959. Effects of cyanide
and dinitrophenol on membrane properties of single
nerve fibers. Am. J. Physiol. 197:113-115.

Schoepfle, G. M., and E. A. Eikman. 1967. Cyanide depression
of sodium efflux in desheathed frog sciatic nerve. Am. J.
Physiol. 212:1273-1278.

Standard Methods for the Examination of Water and Wastewater.
1960. 11th Ed. Published jointly by APHA, AWWA, and
WPCF. N. Y. 626 pages.

Tasaka, I. 1939. Strength duration in normal and narcotized
nerve fibers. Am. J. Physiol. 125:367—371.

Welch, J. H., and H. T. Gordon. 1947. The mode of action
of certain insecticides on the arthropod nerve axon.
J. Cellular Comp. Physiol. 30:147-154.

Wolf, K. 1963. Physiological salines for freshwater teleosts.
Prog. Fish Cult. 25:135-140.

Wuhrmann, K., and H. Woker. 1948. Beitrage zur Toxikologie
der Fische. II. Experimentelle Untersuchungen fiber
die Ammoniak— und Blausaurevergiftung. Schweiz.

Z. Hydrol. 11, 210—227.

Yamasaki, T., and T. Ishii. 1952. Studies on the mechaniSm
of action of insecticides. IV. The effects of insecti—
cides on the nerve conduction of insects. Oyo Kontyu.
7:157-170.

Yamasaki, T., and T. Narahashi. 1957 A. Increase in the
negative after-potential of insect nerve by DDT. Studies
on the mechanism of action of insecticides. XIII.
Botyu—Kagaku (Scientific Insect Control). 22:296-304.

. 1957 B. Intracellular microelectrode recordings
of resting and action potentials. Studies on the mechan—
ism of action of insecticides. XIV. Botyu-Kagaku
(Scientific Insect Control). 22:305-311.

 

 

 

 

 

 

 

 

"‘lilillllilllllifs