EPA-600/3-77-133
December 1977
Ecological Research Series
                PROCEDURES  FOR  MEASURING COUGH
                          (GILL  PURGE) RATES  OF  FISH
                                         Environmental Research Laboratory
                                        Office of Research and Development
                                        U.S. Environmental Protection Agency
                                               Duluth, Minnesota  55804

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology.  Elimination  of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental  Health Effects Research
      2.  Environmental  Protection Technology
      3.  Ecological Research
      4.  Environmental  Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency  Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on  the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems  are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
 This document is available to the public through the National Technical Informa-
 tion Service, Springfield, Virginia 22161.

-------
                                                  EPA-600/3-77-133
                                                  December 1977
PROCEDURES FOR MEASURING COUGH (GILL PURGE) RATES OF FISH
                           by
                   Robert A. Drummond
                   Richard W. Carlson
        Environmental Research Laboratory-Duluth
                 Duluth, Minnesota 55804
            ENVIRONMENTAL RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S.  ENVIRONMENTAL PROTECTION AGENCY
                 DULUTH,  MINNESOTA 55804

-------
                                 DISCLAIMER
    This report has been reviewed by the Environmental Research Laboratory-
Duluth, U.S. Environmental Protection Agency, and approved for publication.
Mention of trade names or commercial products does not constitute endorsement
or recommendation for use.
                                     ii

-------
                                   FOREWORD
     When this manuscript appears in print,  twelve months will have elapsed
since the passage of the Toxic Substances Control Act by the United States
Congress.  Never before in the history of any large industrial country has
there been a requirement to so extensively test and evaluate all chemicals
before they are produced and marketed.  This is a truly monumental task,
particularly as we must evaluate not only the thousands of chemicals currently
in use, but new ones produced annually.

     This report outlines results of nearly  ten years of research at the
Environmental Research Laboratory-Duluth. For years, toxicologists have
sought short term methods to predict long term effects of chemicals.  The gill
purge response described in this report  has  a better correlation between the
endpoints observed and the no-effect concentrations measured for a variety of
toxic chemicals than any other approach  currently available.  Furthermore, it
is quick and relatively inexpensive to perform.  The opinion of our laboratory
is that this test has had adequate testing and that there is a sufficient data
base upon which to conclude that it is one of the best measurements to predict
the long term effects of toxic materials yet devised.  We further believe that
this test has the best biomonitoring potential for use directly on waste
streams of any biomonitoring technique yet proposed.  We think that this
technique is ready to be used on a wide  scale both in screening toxic materials
for the Toxic Substances Control Act,  and for monitoring complex and simple
effluents from current discharges.

     From this publication regulatory agencies ought to be able to develop
precise test protocol for unique and specific application.


                                      Donald I.  Mount, Ph.D.
                                      Director
                                      Environmental Research Laboratory
                                      Duluth, Minnesota
                                     iii

-------
                                  ABSTRACT
     The cough (gill purge) is an interruption in the normal ventilatory cycle
of fish that serves to clean the gills of accumulated particulate matter.  A
review of the literature shows that the cough occurs in a variety of fresh-
water and marine fish; that both mechanical and chemical stimulation apparently
can cause fish to increase their cough rates; and that an increase in coughing
is a rapid and sensitive endpoint for studying chemicals and effluents.  In
reviewing the test methods and apparatus for measuring cough rates of fish,
we conclude the electrode chamber method offers more potential as a bioassay
tool for assessing the respiratory responses of fish due to toxicant exposure.
Recommended test procedures, based on our experience, for using the electrode
chamber method are given.
                                       iv

-------
                                   CONTENTS
Foreword	
Abstract	    iv
Figures	   vii
Tables	   vii
Acknowledgments	viii

     1.  Introduction  	     1
     2.  Conclusions 	     2
     3.  Recommendations 	     3
     4.  Review of Literature	     4
             Physiological studies 	     4
             Methods for detecting coughs  	     4
             Response to physical and chemical substances  .......     5
                 Metals  	     5
                 Pesticides  	     6
                 Effluents 	     6
                 Miscellaneous 	     7
     5.  Recommended Test Method 	     8
     6.  Recommended Test Apparatus	    11
             Chambers	    11
                 Basic design	    11
                 Multiple chamber units  	    11
                 Flow-through chambers 	    13
                 Immersed chambers 	    13
                 Placement of electrodes 	    13
             Recording system  	    14
                 Amplifiers	    16
                 Recording pens	    16
                 Recording paper 	    16
                 Automatic data processing 	    16
     7.  Recommended Test Procedures 	    18
             Physical and chemical procedures  	    18
                 Lighting	    18
                 Disturbances  	    18
                 Temperature, oxygen, pH 	    19
                 Toxicant-delivery system  	    19
             Biological procedures 	    19
                 Selection of fish	    19
                 Transfer and acclimation to chambers	    22
                 Preliminary observations  	    23
                 Number of test fish	    23
                 Monitoring and count intervals  	    23
                 Length of exposure	    24

-------
     8.   Interpretation and Application of Cough-Frequency Data 	    25
             Statistical considerations 	    25
             Pattern of response	    26
             Utility of the response	    28

References	    30
Appendices

     A.   An air-lift chamber system designed for effluent testing ....    36
     B.   A four-plex, immersed chamber system designed for effluent
           testing	    38
     C.   Examples of fish coughs (arrows) recorded by a rectilinear
           polygraph	    40
                                      vi

-------
                               LIST OF  FIGURES


Number                                                                    Page

   1   Respiratory patterns  of  bluegill sunfish  recorded  simultaneously
         by the pressure-transducer and electrode  chamber method 	    9

   2   Schematic of the  electrode  chamber  system 	   12

   3   Wiring schematic  for  multiple chamber  units 	   15

   4   Dilution-water and  toxicant-delivery systems   	   21

   5   Cough-response patterns  of  brook trout exposed to  copper,
         methylmercuric  chloride,  and chlorinated  hydrocarbons  	   27
                                LIST  OF  TABLES
   1   Lowest  concentrations  of  ten  compounds  causing  significant
         increases  in brook trout  cough  frequency  and  MATC  ranges
         for this species	29
                                     vii

-------
                               ACKNOWLEDGMENTS
     Figures and related art-work in this report were prepared by Ms.  Barbara
Halligan.  All drafts of this paper were typed by Ms. Terry Highland.  We
thank both for their cheerful help.
                                    viii

-------
                                  SECTION 1

                                INTRODUCTION


     Various behavioral and physiological aberrations of fish have been used
over the years to indicate stress due to the presence of toxic materials.
Changes in cough frequency, opercular rate, respiration rate (oxygen consump-
tion), feeding patterns, locomotor activity, fin movements, eye movements,
reproductive displays, swimming ability, avoidance,  and the ability to respond
to a stimulus (conditioned response) all have been used as a measure of such
stress.  Of these, respiratory impairment in fish is a sensitive indicator of
toxicant stress, but coughing rate shows more promise as a significant
response to sublethal concentrations than opercular rate (Sprague 1971).  We
believe that change in cough frequency has wide potential application as a
bioassay tool because (1) the response occurs in a variety of fish, (2) many
chemical substances are known to elicit the response, (3) the endpoint is
rapid and sensitive, (4) the degree of response usually occurs in direct
proportion to the toxicant concentration, (5) the cough is predictive of
long-term adverse effects at levels near the maximum acceptable toxicant
concentration (MATC) for most of the single toxicants investigated to date,
and (6) an organism can act as an integrating sensor of potentially toxic
conditions when exposed to complex effluents.

     Although fish coughs have been recognized for many years and a consider-
able amount of effort has been expended on examining this phenomenon, no
comprehensive review of the fish-cough literature is presently available.
One purpose of this paper is to fulfill this need.  Also, we have found that
some procedures are better than others for detecting and recording the cough
and other movements of fish.  The main purpose of this paper is to describe
these procedures and discuss the state-of-the-art as we see it.  This discussion
will include observations we have made that remain unpublished as well as
those we and others have published.  Although this report is written primarily
for those who do not know how such tests can best be conducted and how the
data can be applied, we hope that it will also contain useful information for
those presently using increases in cough rate as a test endpoint.

-------
                                  SECTION 2

                                 CONCLUSIONS
1.  The cough is an interruption in the normal ventilatory cycle of fish, which
    in ordinary water serves to clean the gills of accumulated particulate
    matter.  When particulate content of the water is increased, cough rates
    increase accordingly.  Increases in cough rate also can be elicited by
    exposing fish to chemicals and effluents devoid of particulate matter.
    Thus, both mechanical and chemical stimulation can cause fish to increase
    their cough rates.

2.  A review of the literature shows that the cough occurs in a variety of
    freshwater and marine fish; that, as a test endpoint,  it is rapid and
    sensitive; that the response occurs at sublethal levels of toxicants, and
    usually in direct proportion to toxicant concentration; and that it is
    predictive of long-term adverse effects on fish at levels near the
    maximum acceptable toxicant concentration (MATC) for most of those single
    toxicants tested to date.

3.  Respiratory movements of fish, including coughs, can be detected and
    recorded by use of pressure transducers that measure water-pressure
    changes within the buccal or opercular cavities of a cannulated fish, or
    by electrode chambers that use external, unimplanted electrodes to detect
    the action potentials associated with the respiratory  movements of a free-
    swimming fish.  Although the pressure transducer method has yielded
    considerable information about the respiratory events  that occur in fish,
    the electrode chamber method offers more potential as  a bioassay tool for
    assessing the respiratory responses of fish due to toxicant exposure.

-------
                                  SECTION 3

                               RECOMMENDATIONS
1.  We recommend that studies be initiated to  determine why fish respond to
    various chemical substances by increasing  the rate  at  which they cough.
    The mode of action needs to be established according to the type of
    chemical to which the fish are exposed.

2.  The influence of physical factors such as  salinity, conductivity,
    temperature, and pH on cough rates needs further study.

3.  More comparative studies are needed to determine if concentrations known
    to cause long-term adverse effects also cause significant increases in
    cough rates of fish tested under similar experimental  conditions.   Correla-
    tions of this type are required to assess  the utility  of using increases
    in coughing as a predictor of long-term (chronic) effects.

4.  Computer programs should be developed that are capable of distinguishing
    coughs from other biological activities, tabulating these data,  and
    statistically analyzing the results.

5.  Methods for statistically analyzing and graphically interpreting cough-rate
    data should be standardized.

6.  We recommend that in the future fish "coughs" be referred to as  "gill
    purges" and that the purge rate be expressed as gill purges per  minute.
    Gill purge describes the physiological event more accurately than terms
    such as cough, cough reflex, gill clearing or cleaning, etc.

-------
                                  SECTION 4

                            REVIEW OF LITERATURE
PHYSIOLOGICAL STUDIES

     In ordinary water fish periodically rid the gill surfaces of accumulated
foreign material by a gill-clearing action referred to as a "cough" or "cough
reflex".  Fish coughs have been observed and studied for many years beginning
with Bert (1870), Francois-Franck (1906), Kuiper (1907), and Bijtel (1949).
Coughs have been observed in freshwater and marine teleosts (Hughes and Shelton
1958; Hughes 1960), elasmobranchs (Satchell and Maddalena 1972) and lampreys
(Rovainen 1977).  Neurological events associated with coughing have also been
studied (Shelton 1959; Young 1970, 1972; Ballintijn and Roberts 1976).

     So far as we know, there are two basic types of coughs.  A small, or weak,
cough consists of a greater than normal expansion and contraction of the buccal
and opercular cavities, particularly the buccal.  This action produces a
higher than normal positive pressure in the buccal cavity and results in a
forceful movement of water across the gills in the normal direction.  This
pulse may be preceded by a momentary reversal of water flow over the gills
(Holeton and Jones 1975).  A large, or strong, cough occurs whenever the
pressure gradient between the buccal and opercular cavities is reversed.  In
this case the buccal cavity develops a high, uncharacteristic negative pressure
relative to the opercular cavity, and as a result water flow reverses across
the gills.  During the reversal debris or mucus is ejected rapidly through
the mouth.  The reversal phase is quickly followed by a prolonged, high positive
pressure in the buccal cavity, after which normal pressure cycles are resumed.
The physiological basis of the cough is well documented, and important findings
have been summarized by Hughes (1973), Hughes and Morgan (1973), Holeton and
Jones (1975), Hughes (1975), and Hughes and Adeney (1977).


METHODS FOR DETECTING COUGHS

     Over the years a number of researchers have counted fish coughs by
observing the fish directly.  This method is still used by a few investigators
(McPherson 1973; Bull and Mclnerney 1974; Hargis 1976), but most researchers
have found it advantageous to record coughs at an instrument station distant
from the test area.  One approach is to record the water-pressure changes that
take place in the buccal and opercular cavities of the fish.  A tube is placed
adjacent to these cavities (Hughes and Shelton 1958) or a cannula is inserted
directly into the fish (Saunders 1961), and changes in pressure are measured
by a pressure transducer.  The resulting signals are then displayed on an
oscilloscope or recorded on a polygraph strip-chart recorder.  Several

-------
researchers have used this method with slight modifications (Ballintijn and
Hughes 1965; Schaumburg, Howard, and Walden 1967; Hughes and Saunders 1970;
Hughes and Roberts 1970; Young 1972; Sparks, Cairns, McNabb, and Sutter
1972; Davis 1973; Sellers, Heath, and Bass 1975; Bimber, Boenig, and Sharma
1976).

     Another approach differs from the above in that action potentials (bio-
electric potentials) associated with the animal's movements (Camougis 1960;
Goodman and Weinberger 1971; Heath 1972) are measured instead of pressure
changes.  Electrodes, usually stainless steel wire, are placed at each end of
an electrode chamber containing one free-swimming fish.  Electrical signals in
the microvolt range emanating from the fish are picked up by the electrodes and
transmitted via shielded cable to high-gain preamplifiers.  The amplified
signals are then displayed on a strip-chart recorder (Spoor, Neihelsel, and
Drummond 1971; Drummond, Spoor, and Olson 1973; Drummond, Olson, and Batterman
1974).
RESPONSE TO PHYSICAL AND CHEMICAL SUBSTANCES

     In addition to the debris and other foreign materials found in ordinary
water (Heimstra, Damkot, and Benson 1969), fish respond to other physical and
chemical substances by changing the rate at which they cough.  In general,
cough frequency increases as the concentration of these substances increases,
but the mode of action is not well understood.  Schaumburg et al. (1967)
suggested that increases in coughing might be due to the lowering of the
dissolved oxygen at the gill surface.  Hughes (1975) felt that coughing is
mainly a response to some form of mechanical stimulation, the precise nature of
which is unknown, but Davis (1973) concluded that the response was due to
chemical as well as mechanical stimulation.  Substances known to affect the
cough rates of fish are grouped in four categories:  metals, pesticides,
effluents, and miscellaneous.

Metals

     Drummond et al. (1973) reported that cough frequency in brook trout
(Salvelinus fontinalis) increased significantly at a copper concentration as
low as 0.009 mg/1 and that cough rate increased proportionately as the
concentration was increased.  Because 0.009 mg/1 was just below the maximum
acceptable toxicant concentration (MATC, described by Mount and Stephan 1967)
determined by McKim and Benoit (1971), the response was considered predictive
of the concentration range of copper likely to have no long-term adverse
effects on the species.  Cough frequencies of sockeye salmon (Onchorhynchus
nerka) increased significantly at all copper concentrations above 0.024 mg/1
(Martens, Gordon, and Servizi 1970).  Increases in coughing were noted among
rainbow trout (Salmo gairdneri) and bluegill sunfish (Lepomis macrochirus)
exposed to zinc (Skidmore 1970; Sparks e± al. 1972; Hughes 1975).  Sprague
(1964) observed that young Atlantic salmon (Salmo salar) coughed whenever
they swam into water containing a high concentration of copper or zinc.  The
effect of mixtures of copper and zinc on the cough rate of rainbow trout has
also been studied (Sellers jit _al. 1975).  Bluegill sunfish showed significant
increases in cough frequency at cadmium concentrations of 0.050, 0.084, and

-------
0.115 mg/1, and the increases were proportional to the concentrations
(Mclntosh and Bishop 1976).

     McPherson (1973) found a linear relationship between the concentration
of methylmercuric chloride (MMC) and the number of cough reflexes per 100
opercular movements when testing rainbow trout and coho salmon (Oncorhynchus
kisutch).  The lowest concentration of MMC and mercuric chloride to cause a
significant increase in cough frequency in brook trout was reported by Drummond
et^ a±. (1974) to be 0.003 mg/1.  Because there were no discernible adverse
effects to the first generation of brook trout at concentrations below 0.0029
mg/1 MMC during a three-generation exposure  (McKim, Olson, Holcombe, and Hunt
1976), the cough response was determined to be a short-term indicator of the
long-term effects of MMC to that species.  Cough frequency also increased in
direct proportion to MMC concentrations tested above 0.003 mg/1.

Pesticides

     A direct relationship was demonstrated between the concentration of DDT
and the cough rate of coho salmon (Schaumburg jt^ jal^ 1967); a significant
increase in cough rate occurred at as low a concentration as 0.005 mg/1.  Bull
and Mclnerney (1974) examined several behavioral endpoints of coho salmon and
found that only coughing increased in response to increasing concentrations
of the insecticide Fenitrothion.  Rainbow trout exposed to DDT, dieldrin, and
carbaryl (a methyl carbamate insecticide) showed increases in cough frequency,
and, except for carbaryl, the increases were proportional to toxicant
concentration (Lunn, Toews, and Free 1976).  Concentrations of 1.0 and 5.0 mg/1
diaquat caused yellow perch (Perca flavescens) to increase their cough frequency
significantly; the response was greater at the higher concentration (Bimber et
al. 1976).

Effluents

     With the exception of kraft pulp-mill effluents (KPME) little work has
been done in which fish coughs are used as an indicator of effluent toxicity.
Walden, Howard, and Froud (1970) conducted several experiments with KPME
collected from two different pulp mills.  They found dramatic increases in
cough frequency of rainbow trout exposed to high concentrations of effluent.
As effluent concentration approached lethal levels, the cough rate rose
steadily, became erratic, and could not be correlated with the concentration.
But at lesser effluent concentrations cough frequency increased, reached a
maximum, and then declined to the normal pre-exposure rate.  The maximum
increases in cough frequency at these lesser concentrations were directly
proportional to the effluent concentration.

     Davis (1973) exposed sockeye salmon to KPME and found that increases in
cough rate began at 20 percent of the 96-hr LC50.  He noted that other
workers reported reduced specific growth rates for this species at the same
concentration.  Howard and Walden (1974) stated that the cough reflex is
one of several biological indicators that show promise for monitoring the
effectiveness of treatment systems designed to upgrade the quality of pulp-
mill effluents.  They showed that rainbow trout exposed to an untreated KPME
increased their cough frequency significantly, but when exposed to the same

-------
effluent after proper treatment, cough frequencies were not significantly
different from those of control fish.   In a more recent study Carlson and
Drummond (1977) demonstrated that increases in bluegill sunfish cough rate
were commensurate with effluent concentration when the fish were exposed to
a variety of treated industrial and municipal effluents.  We concluded that
cough frequency is a rapid and sensitive physiological characteristic for
evaluating the relative quality of treated effluents and variations in quality
of same-source effluents, before discharge into receiving waters.

Miscellaneous

     Brook trout increased their cough frequency when exposed to chlorine
and sulfides (Dandy 1967, 1972), and rainbow trout exposed to chlorine showed
a similar response (Bass 1975).  Servizi, Gordon, and Martens (1969)  noted that
sockeye salmon smolts showed increases in coughing at initial hydrogen sulfide
concentrations between 0.3 and 1.1 mg/1.   Ammonia, alkylbenzene sulfonate, and
coal dust were reported to cause increases in fish-cough frequency (Hughes
1975).  MacLeod and Smith (1966) found that fathead minnows (Pimephales
promelas) increased their cough rate relative to fiber concentration  when
exposed to suspension of pulpwood fibers.  Water-soluble fractions of oils
caused pink salmon fry (Oncorjvynchus gorbuscha)  to increase their cough rate
in proportion to concentrations > 30%  of the 96 hr median tolerance limit
(Rice, Thomas, and Short 1977).

     Water temperature and pH reportedly influence the rate at which  fish
cough.  While studying the respiratory pumps of rainbow trout, Hughes and
Roberts (1970) noted an increase in the number of reversals and double rever-
sals in respiratory pressure (presumably coughs) as the water temperature was
raised.   Rainbow trout tested at pH 6  showed significantly fewer coughs than
those treated at pH 9 (Hargis 1976),  although cough rates of brook trout tested
at pH 6 and 9 were not significantly different according to Drummond  e_t al.
(1974).

-------
                                  SECTION 5

                           RECOMMENDED TEST METHOD
     In comparing the methods for measuring fish respiratory movements Heath
(1972) concluded that observing fish directly had several disadvantages as a
test method.  Although the procedure does have some disadvantages, it can be
a useful technique to provide first-hand observations that could not be obtained
by other means.  Heath (1972) further concluded that recording respiratory
pressures with cannulated fish is a method that yields the most information
with the least effect on the fish.  Detection or measurement of coughing is
mentioned only in conjuction with indwelling cannulae, even though he noted
a one-to-one correspondence between opercular movements recorded simultaneously
by using cannulae and external electrodes (electrode chambers).  Heath's comments
seem to imply that the electrode chamber method is suitable for recording opercu-
lar movements, but that the cannulae/pressure-transducer method should be used
to record cough, opercular and other respiratory movements.  We compared the
two methods and believe that the electrode chamber method, when used with the
proper instrumentation, is the better technique of the two for a number of
reasons.

     We simultaneously recorded coughs and opercular movements of bluegill
sunfish by using the cannulation (Saunders 1961) and the electrode chamber
(Spoor et_ jal_. 1971)  methods.  We found that coughs can be distinguished from
the more rhythmic opercular waveforms with either method and that a one-to-one
correspondence exists between coughs and opercular movements recorded
simultaneously as action potentials and water-pressure changes (Figure 1).
The recordings of the coughs and opercular movements obtained simultaneously by
the two methods also were verified by observing some of the fish and correlating
our observations with what appeared on the strip-chart records.  In addition to
coughs and opercular movements, both methods enabled us to detect and record
changes in opercular amplitude, yawns (McCutcheon 1970), and body movements
(Appendix C).  The electrode chamber method also enabled us to detect and record
fin movements, muscle spasms, and, in some instances, the heart beat of trout.
Although these signals are usually masked by the stronger action potentials
associated with cough, opercular,  and body movements, some of the weaker signals
can be separated from the other signals by electronic filters (Venables and
Smith 1972; Rommel 1973;  Strange,  Dean, Dorman, and Fletcher 1975).

     The electrode chamber method has a big advantage over the cannulae/pressure-
transducer method in that it permits study of free-swimming fish that have
not suffered the trauma of anesthetization and operative procedures.  We have
observed that cannulated fish never fully recover (i.e., cough and opercular
rates remain higher than normal) from the effects of surgery and having a

-------
     Action
     potential
     Buccal
     pressure
                          Cough
                                                                 1.0 sec
    b.
     Action
     potential
      Buccal
      pressure
                                             Cough
                                                                 1.0 sec
Figure 1.   Respiratory patterns of bluegill sunfish recorded simultaneously by
           the pressure-transducer and electrode chamber method while a) in
           control water, and b) undergoing toxicant stress.  Note presence of
           other coughs  (unmarked).

-------
cannula imbedded in the oral cavity.   The longer the cannula remains in the
fish, the more stressed they become.   In addition,  the erosion of tissue
around the cannula may cause it to come loose from the fish.  For these reasons
we recommend that the electrode chamber method be used for measuring the cough
and opercular rates of fish exposed to chemical substances or to complex
mixtures such as effluents.
                                      10

-------
                                  SECTION 6

                         RECOMMENDED TEST APPARATUS
     The basic system consists of electrode chambers,  amplifiers,  and servo-
pens to record the various action potentials produced  by the fish on
strip-chart records (Figure 2).   The easiest way to  set  up this system is to
build the chambers yourself and purchase a commercial  polygraph (2-12 channels
are available) consisting of preamplifiers,  amplifiers,  servo-pens,  and a
strip-chart drive.  A commercial polygraph costs about one thousand  dollars
per channel, but cost of the instrumentation can be  reduced by using a
switching device so that one channel can be used to  monitor more than one
fish (Figure 2).  In the following detailed description  of the system we will
emphasize those considerations that we  believe  are important.
CHAMBERS

Basic Design

     The basic design of the electrode chamber has been given by Spoor et al.
(1971).  Although the chambers can be modified to fit a particular need,  the
concept remains unchanged.   The chambers should be rectangular and narrow
(roughly, not more than 3x  the width of a fish) so that the fish is forced to
assume a position perpendicular to the ends of the chamber, i.e.,  facing one
of the two electrodes.   If  the chambers are too wide, the fish can assume a
position off to one side of the chamber or parallel to the electrodes; this
results in considerable loss of signal amplitude.  For toxicity tests the
chambers should be made of  glass cemented together with a clear silicone
adhesive.  A horizontal, cylindrical chamber may be required for studying
some fish species such as fathead minnows (Drummond 1977).

Multiple Chamber Units

     Chambers can be put together either individually or in banks of two, four,
six, etc.  Each fish is still monitored individually whether the chambers are
constructed as single or multiple units.  It is convenient to have the same
number of chambers and recording channels in each bank; for example, a bank
of four chambers is best used with a four-channel polygraph.  Also, working
with and cleaning four units of four chambers is easier than working with
16 individual chambers.  We have described individual chambers,  but these
designs can be expanded readily into multiple units.
                                     11

-------
                                         Strip-chart Record
                                      Rectilinear Pen Recording
                                               System
                                              Amplifier
                                            Preamplifiers
                                                INPUT
                                           I   I
          | Switching Device (optional)
          —n—TV
Electrode
Chambers.
                 . .j
        Wire Electrodes.
  Figure 2.  Schematic of  the electrode chamber system.
                              12

-------
Flow-through Chambers

     Incoming water normally enters one end of the chamber,  passes through the
chamber, and exits through a drain at the other end.   Constant level headboxes,
fitted with capillary tubes or syringe needles, are the most convenient and
reliable method of water delivery (Figure 4,  Section 7).  Chambers can be
drained by a glass standpipe or by a hole, fitted with a stopper and drain
tube, in the endplate.

     Water can also be delivered to a chamber by air-lifting it up from a tray
placed immediately below the chamber.  The air-lifted water  passes through the
chamber, returning to the tray via a standpipe drain (Appendix A).  Although
this system can be used for flow-through testing, it was designed to be used
as a recirculating test system and is particularly useful for testing imported
samples of effluents.  A detailed description of this design, based on an
actual system used for testing effluent samples, is given in Appendix A.

Immersed Chambers

     Chambers designed to be immersed in a tank (aquarium) of water are
suitable for both flow-through and recirculating tests (Appendix B).  In the
flow-through mode this design is especially useful for on-site monitoring of
effluents because separate flows to each chamber (multiple chamber design)
are not required.  As with the airlift chambers, they can be used in the
flow-through mode during the acclimation period and,  when the test begins, can
be converted to the recirculating mode.  Immersed chambers require wire-mesh
electrodes at each end to keep the fish inside the chamber and permit
circulation of water through the chamber.  Recirculation is  very important
because without it, fish quickly deplete the dissolved oxygen content of the
water inside the chamber.  Aeration is the easiest way to promote necessary
circulation.  Details of a four-plex, immersed chamber design used in some
of our effluent tests are given in Appendix B.

Placement of Electrodes

     Stainless steel wire or mesh electrodes are glued with  silicone to the
endplates of the chamber.  The distance between the electrodes should not
exceed two times the length of the fish.  For versality, one electrode can be
glued to the endplate and the other to a moveable glass baffle (Appendix A).
This arrangement permits increasing or decreasing the interior length of the
chamber to accomodate fish of different sizes.  Unless mesh  electrodes are
required to keep the fish inside immersed chambers, wire electrodes (1.5 mm
diameter) work just as well as mesh electrodes.

     Each pair of chamber electrodes consists of a positive  and a negative
electrode.  Although electrodes are usually arranged in pairs (Figure 2),
this arrangement need not be used when several chambers are  put together as
a unit.  In this case the positive electrodes are placed at  the inlets and
a common negative electrode is placed at the outlet of the chambers.  Two
examples of how the negative electrode can be attached to the chambers
follow:
                                     13

-------
                NEGATIVE  ELECTRODES
    N-notch filled with
       siliconeglue
GLASS
WIRE MESH
One word of caution.   This arrangement  (Figure 3) worked well with a Narco
physiograph, but did  not work with a  four-channel, homemade amplifier based on
the design given by Drummond and Dawson (1974).  The problem seemed to be in
the circuitry in that a single input  was recorded on all four channels.  The
choice of using pairs of independent  electrodes  (Figure 2) or individual
positive electrodes and a common negative electrode (Figure 3) may depend on
the type of polygraph used.   We suggest testing both types of electrode
arrangements to determine which one works best.

     Electrodes should be connected to  the amplifiers with shielded leads.
In addition to shielded leads,  the water supply  (headboxes) and ungrounded
equipment in the vicinity of the test area should be grounded.  All ground
wires should be connected to the same ground potential.  Larger microphone
cables work well with the pin jack/electrode connection described by Spoor
et^ a±. (1971).  Smaller diameter leads, such as Belden shielded instrumenta-
tion cable, are convenient to work with because of their size, but tend to
break after being repeatedly connected  and disconnected when pin jack
connectors are used.   To avoid this problem we permanently attached the
shielded leads to male pin terminals, in a junction box, and connected the
electrodes to these terminals with patch cords (Figure 3).  Patch cords, having
female terminals at each end, can be  made up from a kit (Waldom/Molex Designer
Kit No. WM-72); they  are easy to connect and disconnect from the electrodes,
and they are durable.
RECORDING SYSTEM

     A number of commercial  polygraphs are suitable for detecting and recording
coughs and other movements of  fish.  Although we cannot recommend one brand
over another, we can outline those considerations we believe to be important.
                                    14

-------
            4-CHANNEL  POLYGRAPH
               SWITCHING DEVICE


,
1

v^
,
1

1
1
1
1
» 1
1

V
1
1
J 2-wire leads
K (shielded)
i
v^
1
!"L 	 junction box
K
\ \\ 	 male pin terminals
i i i
i i '
I i
' (etc.)


I
8
I
11
I
8
k
1
                          cords

                     wire electrodes
                                          	 positive
                                          	negative
Figure 3.  Wiring schematic for multiple chamber units.
                        15

-------
Amplifiers

     Input signals in the low microvolt range require two stages of
amplification (preamplifiers and amplifiers) to amplify them into the millivolt
range.  Polygraphs capable of amplifying human electroencephalogram input
signals are suitable for amplifying the action potentials of fish studied in
electrode chambers.  Solid-state amplifiers are recommended.  Built-in 60-Hz
filtration is highly desirable, but can be added if necessary by using some
type of filtering device (Ferris 1974; Ewing and Ashworth 1974).  Elaborate
filtration may not be necessary in a well-grounded system.  For example, a
5-mfd non-polarized capacitor placed across the output of a Narco physiograph
amplifier will effectively eliminate any remaining 60-Hz noise.  If funds are
not available to purchase a commercial unit, amplifiers can be constructed
(Drummond and Dawson 1974) and the output signals recorded on any rectilinear,
strip-chart recorder.  Commercial polygraphs are recommended in that all
components are integrated into a single, compact unit.

Recording Pens

     Use ja rectilinear, servo-pen recording system rather than a curvilinear
pen system.  It is very difficult to distinguish coughs from opercular and
other movements on curvilinear tracings because the shape of the waveform is
distorted.  Curvilinear tracings are suitable for recording opercular
movements, but do not provide a clear-cut recording of the cough reflex.

     Many rectilinear pen systems are available.  Some have a capillary or
pressurized inking system; others are equipped with a thermal or pressure
writing system.  Inking systems are quite reliable even though messy to work
with at times.  Heated pens also work well if used with a high quality, heat-
sensitive paper coated with a plastic material.

Recording Paper

     Cost of the recording paper must be considered.  Ewing and Ashworth
(1974) stated that electrical recording paper is more expensive than thermal
paper, and both types are more costly than paper suitable for ink writing.
Both sides of inking paper, at least the Z-fold type, can be used to further
reduce the cost.   Since monitoring is usually done quite frequently, paper
should be long enough to permit extended monitoring without changing paper.
Z-fold packets or rolls 152 m (500 ft) long are recommended.  For economy
the paper should be run at the slowest possible speed, yet fast enough so
that the coughs can be seen clearly and counted.  We found 0.1 cm/sec to be
an optimum paper speed for monitoring most fish species, but faster speeds
(0.25-2.5 cm/sec) should be available for use as the need arises.

Automatic Data Processing

     Although strip-charts provide a permanent record of the fish's movements,
counts and other measurements must be made manually.  This can become a
tedious and time-consuming task in a large-scale testing program.  An
alternative is to feed the amplified signals into some type of automatic
data-processing unit (Cairns et_ a±. 1973); Morgan and Kuhn 1974; Dessy 1976;

                                     16

-------
Morgan 1976, 1977).  As far as we know,  coughs  have  not  been separated from
the other signals with this type of instrumentation.   The development  of a
software program should not be too difficult, however, that  will  separate
coughs from other fish movements and automatically tabulate  cough rates
before and after fish are exposed to chemical substances or  effluents.
                                    17

-------
                                  SECTION 7

                         RECOMMENDED TEST PROCEDURES
     Since many procedures apply to toxicity testing in general, e.g.,
treatments for diseased fish, we will describe only those procedures closely
allied to the electrode chamber method.  Other appropriate references
should be consulted before cough-response tests are started (U.S. Environmental
Protection Agency 1975; American Public Health Association 1976).
PHYSICAL AND CHEMICAL PROCEDURES

Lighting

     If a diel lighting regimen is followed, we suggest a 16-hr photoperiod
and a lighting system similar to that described by Drummond and Dawson (1970).
The system will eliminate sharp transitions between light and dark, more
nearly approximating natural sunrise and sunset.  Nevertheless, fish still
react to the light-on and light-off periods with an increase in locomotor
activity and opercular rate, and their opercular rate also tends to be higher
during the day than it is at night, especially during the early morning hours.
If desired, diel respiratory patterns (Sparks et al.  1972) can be nearly
eliminated by using a continuous photoperiod.  Another advantage of using
continuous lighting is that a test can be started at  any time of the day.
For instance, imported effluent samples often arrive  during the evening hours,
and tests can be started immediately rather than allowing the effluent sample
to age further.

Disturbances

     If two or more chambers are constructed as a unit, an opaque shield may
be needed between the chambers to prevent test fish from seeing one another.
Although the bluegill sunfish in our cough tests were not influenced by the
activities of their counterparts in adjacent chambers, other fish species
may react differently.  Also, all experimental fish must be well-shielded
from visual and vibratory disturbances external to the test setup.  Outside
visual stimuli, such as a person walking by an unshielded test area, can
greatly influence the fish's behavior.  Typically, the fish will cease all
movements, including respiratory, for a short time and then go through a
period of increased locomotor activity and opercular  rate.  Opaque plastic
curtains or styrofoam shields will effectively reduce or eliminate disturbances
altogether.
                                      18

-------
Temperature, Oxygen. pH

     Cairns, Morgan, and Sparks (1974) showed that bluegill sunfish breathing
rates fluctuated with diurnal temperature changes, but they did not determine
if cough rates also fluctuated with temperature change.  Drummond (unpublished
data) found that fluctuations in dissolved oxygen concentrations have a
greater effect on opercular rate than on cough frequency.  If the dissolved
oxygen concentration is lowered, coughs will increase at about the same rate
as opercular movements, i.e., if the opercular rate increases twofold, the
cough rate will also.  In contrast, fish exposed to a sublethal toxicant
concentration in water with adequate dissolved oxygen may show a several-fold
increase in cough rate without a corresponding increase in opercular rate.
At lethal concentrations cough rate will increase about twice as fast as the
opercular rate.  Cough and opercular rates of some species may also be
affected by extremes in pH (Hargis 1976).   These variables should be monitored
and controlled, but if they cannot be closely regulated then the degree of
fluctuation should be specified and accounted for when analyzing test results.

Toxicant-Delivery System

     Two methods of toxicant delivery gave consistently good results.  Methods
A and ]} shown in Figure 4 both deliver the toxicant solution to a cell where
it is mixed with incoming dilution water.   The desired mixture is then fed to
a constant level headbox for further mixing and distribution to the electrode
chambers.  Method _A requires a variable speed, multi-channel peristaltic pump,
but is a versatile method because toxicant flow rates can be changed by simply
altering pump speed.

     Method li is a straightforward, inexpensive technique that works well
when properly installed.  The stock solution is maintained at a fairly constant
level by periodically raising the screw-jack beneath the stock bottle.  To
maintain a steady toxicant flow rate, a large diameter bottle is required since
the vertical loss of solution (and corresponding decrease in headpressure) is
less in a wide bottle than it is in a narrow bottle over the same time span.
The flow rate through the toxicant-delivery lines should be slow enough (0.1-
2.0 ml/min) so that it will not decrease significantly overnight due to the
decrease in headpressure.  The rate at which the toxicant enters the mixing
cell is therefore regulated by (1) the height of the stock bottle, (2) the
level of the solution in the bottle, (3) the inside diameter of the Teflon
tubing, and (4) an adjustable clamp attached to the Teflon delivery line.
A mariotte bottle could be used instead, but we found flow rates to vary with
fluctuations in room temperature, especially when solvent-toxicant stock
solutions were required.


BIOLOGICAL PROCEDURES

Selection of Fish

     In addition to those fish species mentioned in the review of literature,
we have observed coughs in green sunfish (Lepomis cyanellus), largemouth bass
                                     19

-------
Figure 4.   Dilution-water and toxicant-delivery systems


EW   -   incoming dilution water

     -   main headbox, large well-aerated reservoir
HB-1 -   constant level distribution headbox (aeration optional); can provide
         flows to one or more mixing cells

M]   -   dilution water-toxicant mixing cell

HB-2 -   constant level distribution headbox

SP   -   standpipe overflow drain

DT   -   hypodermic or capillary delivery tubes; provide equal flows to
         chambers

_A    -   Method ^A:  Toxicant delivery

         SB  -  stock bottle

         TG  -  Tygon tubing

         PP  -  peristaltic pump

JB    -   Method ]i:  Toxicant delivery

         SB  -  stock bottle

         TF  -  Teflon tubing

         AC  -  adjustable clamp (optional)

         SJ  -  screw jack
                                      20

-------
Figure 4.  Dilution-water and toxicant-delivery systems,
                            21

-------
(Micropterus salmoides), black crappie (Pomoxis nigromaculatus), northern pike
(Esox lucius), and channel catfish  (Ictalurus punctatus).  Other species
probably possess a cough mechanism  also.  We do not  recommend a standard test
fish, but suggest that priority be  given to salmonids, bluegills, or green
sunfish and, if test procedures can be worked out, the fathead  minnow.
Salmonids are perhaps the most amenable species to work  with in cough tests
because their coughing movements are so readily distinguishable from locomotor
and other respiratory movements on  strip-chart records.  Most of these species
are obtainable from government or private hatcheries or  can be  cultured and
reared in the laboratory.  Imported fish should be held  in holding  tanks or
aquaria for several weeks to allow  recovery from handling and transportation-
induced stress.  The fish should be treated prophylactically against disease
only when necessary.

Transfer and Acclimation to Chambers

     Fish can be transferred to the electrode chambers with a dip net.  Since
fish are stressed for some time after initial transfer,  acclimation under
flow-through conditions will help eliminate any metabolic waste products or
fear substances that might inhibit  their recovery.   Further transfer is not
required, except when the immersed  chamber design is used (Appendix B).  In
this case the dilution water must be siphoned out of the acclimation tank,
leaving the fish out of water temporarily, and replaced  with an effluent or
toxicant mixture to obtain the desired concentration.  Alternatively, water
can be siphoned out of a spare tank, replaced with the desired  mixture, and
the electrode chamber containing the test fish transferred from their
acclimation tank to the prepared tank.  Either method is preferable to another
transfer with a dip net because handling-induced stress  is reduced  and
recovery is quicker.

     When airlift chambers are used for recirculating tests, stess  due to
transfer is eliminated by allowing  the fish to remain in their  chambers while
the test is being set up.  The fish are acclimated under flow-through conditions,
the incoming water is stopped, and  a toxicant or effluent is added  after an
appropriate amount of water is removed from the tray located below  the chambers
(Appendix A).  Since water is removed only from the  tray, the volume of water
remaining in the electrode chambers must be included in  calculations of test
concentrations.  Hence, effluent tests cannot be conducted in 100 percent
effluent (by volume) unless the fish are transferred.  This problem is
apparently inconsequential, however, because test concentrations greater than
50 percent may not be necessary to  assess the quality of an effluent (Carlson
and Drummond 1977).

     Acclimation time required for  the fish to adjust to the electrode chambers
is dependent to a certain extent on the fish species selected.  Brook trout
require 4 days, bluegill sunfish 3  days, and green sunfish 2 days to recover
from the effects of transfer.  The  criterion we have used to determine proper
acclimation is that time period required for cough and opercular rates and
locomotor activity to decline to a  base level, i.e., that point at  which these
activities do not vary significantly (P=0.05) from one day to the next.  In
no case has this period been less than 48 h.
                                       22

-------
     Feeding is not recommended even though the acclimation and exposure
period may last 7 days.   If the fish are maintained in the chambers for
longer periods, they should be fed once daily with a pelletized fish food.
When fed, the chambers must be routinely cleaned to prevent buildup of feces
and unused food, or the fish must be periodically transferred to clean chambers.

Preliminary Observations

     Before starting cough tests, several fish should be individually observed
to correlate visual observations of the different locomotor and respiratory
activities with the strip-chart recordings.   Indirect observation through
one-way glass is preferable since fish often alter their behavioral patterns
when observed directly.   For instance,  bluegills under moderate toxicant stress
may stop coughing while an observer is present.   Fish should be observed under
control conditions as well as during exposure to a selected toxicant.  Observe
as many fish as possible at both sublethal and lethal toxicant concentrations
while looking for characteristic cough patterns.   When counting coughs, count
all movements regardless of whether they appear to be weak or strong coughs
(see review of literature).  Some examples of cough movements recorded by a
rectilinear polygraph are given in Appendix C.   When coughs and other fish
movements can be identified with confidence on the record sheets, visual
observations can be discontinued.

Number of Test Fish

     At sublethal concentrations intolerant fish may increase their cough rate
greatly, whereas the more tolerant individuals may show little, if any, increase.
If this variation in individual tolerance is not taken into account, the data
may be biased and the test results invalidated.   This variation may be minimized
by using larger numbers of fish, but the number of fish tested at each
concentration is largely dependent on the purpose of the test.  If the purpose
of the test is to detect spills of toxic materials in an on-site monitoring
situation or to do laboratory screening of chemical substances or effluents
(Carlson and Drummond 1977), then two fish per concentration may be all that is
necessary.  But if the intent is to rank chemicals according to their toxicity,
as evidenced by increases in cough frequency, then at least four fish per
concentration is a desirable number.  A minimum of eight fish per concentration
should be tested if the results are used to predict the concentration range at
which long-term adverse effects are expected to occur (Drummond et al. 1973,
1974).  Doubling the above numbers is biologically desirable, but will add
considerably to the cost of the test.

Monitoring and Count Intervals

     Frequency of the monitoring periods is dependent on the type of test
conducted and the length of time required for cough frequency to reach a
maximum.  As a general rule, the more observations made, the better, since  a
large number (N) is desirable from the standpoint of statistical analyses.

     Unless fish are being used to detect spills of toxic materials, each fish
need not be monitored continuously.  We suggest monitoring each fish at least
once every 2 h during the 24-h period preceding the start of a test and for


                                      23

-------
the duration of the test.   If previous experience shows that cough frequencies
increase gradually over a period of days,  the fish can be monitored less
frequently, perhaps every 4-6 h.  Each fish should be monitored for at  least
15 min during each scheduled monitoring period.

     A representative 5- or 10-min count interval should be selected from
each 15-min monitoring period and the total number of coughs counted.   A count
interval should be selected that is smaller than the monitoring period  for  two
reasons:  first, it is too laborious to count all coughs over a 15-min  period;
and second, portions of the total record may be  unreadable because swimming
activity obscures cough and opercular movements.  However, a 5- or 10-min
count interval is necessary rather than a 1-min  count interval because  fish
cough rates vary somewhat from minute to minute.  These data are usually
expressed as coughs per minute for use in statistically analyzing test
results.

Length of Exposure

     Generally, if fish do not show significant  increases in cough rate within
the first 96 h of exposure, they will probably not respond to prolonged
exposure.  An exception to this generality is fish exposed to the chlorinated
hydrocarbons DDT and endrin (Schaumburg et^ al. 1967; Carlson, unpublished
data).  Fish exposed to these compounds show increases in cough rate that are
time and concentration dependent.  For instance, bluegills exposed to 6.0 yg/1
endrin responded within 24 h with an increase in cough rate, but at 0.4 vg/1
an increase was noted only after 240 h.  Therefore, the exposure period may
have to be extended for as long as 10 days to evaluate the effects on cough
rates of certain chemical substances.
                                     24

-------
                                 SECTION  8

          INTERPRETATION AND APPLICATION  OF  COUGH-FREQUENCY  DATA
STATISTICAL CONSIDERATIONS

     Percentage change in cough frequency  seems  adequate  to  reflect  stress  in
screening tests, but other tests require more  elaborate treatment.   Although
there is no "standard" method for statistically  analyzing cough-frequency
data, and none is proposed at this time, we will outline  those  methods  that
have been used to judge the significance of an effect.

     In using the electrode chamber method one can  statistically compare a
control group with an experimental group,  or compare  individual fish (or the
same group of fish) before and after exposure.   In  tests  with copper (Drummond
et^ al. 1973) and mercury (Drummond et al.  1974),  we compared the pretreatment
cough frequencies to the frequencies on that day of exposure during  which a
group of fish, as a whole, showed a maximum increase.  A  paired t-test  was
used to test for significance of differences between  cough frequencies  before
and during exposure (period of maximum increase).   Walden et^ al. (1970) also
used peak cough-frequency periods to evaluate  the effect.

     The selection of peak cough-frequency periods  to evaluate  an effect has
the disadvantage of having to wait for the response to reach a  maximum. To
avoid this delay, we also have used Dunnett's  procedure  (Steel  and Torrie
1960) for making a one-sided comparison between  a treatment  and a control mean.
The effective concentrations were judged to be those  at which at least  one-half
of the eight test fish showed significant  (P,  95%)  increases in cough frequency
(Drummond, unpublished data).  Other workers  (Mclntosh and Bishop 1976; Rice
e^ _§!_. 1977; Maki, personal communication) report using analysis of  variance
for analyzing cough and opercular data. Differences  between means or log means
were used in making these analyses.

     Linear interpolation or extrapolation from  a plotted line  may be a better
method for pinpointing no-effect concentrations. Walden  et  al. (1970)  plotted
the maximum and 5-h mean cough rates against effluent concentrations and
extrapolated (best visual fit) to what they termed  a  "minimum response
concentration."  They indicated that linear plotting  is a suitable technique
for extrapolating to a minimum response concentration, provided the  test
concentrations are marginally above the level  that  initiates an increase in
coughing.  We agree that linear plots should be  based on  responses at test
concentrations reasonably close to the no-effect level in that  cough rates  may
become erratic at high concentrations.
                                     25

-------
PATTERN OF RESPONSE

     The time at which peak cough frequencies occur for a group of fish and
the pattern of decline after reaching a maximum may be important factors.
Cough frequency in brook trout exposed to copper (Drummond ej^ al. 1973)
increased to a peak by the 20th hour of exposure.  The pattern was the same
for all test concentrations as schematically illustrated in Figure 5-A.  When
exposed to methylmercuric chloride (Drummond et^ al. 1974) the peak response
was reached in 3 days (Figure 5-B).  Fish exposed to chlorinated hydrocarbons
(Schaumburg e^ _al_. 1967; Carlson, unpublished data) showed an entirely
different pattern of response, as shown in Figure 5-C.  The cough rate of
trout exposed to neutralized kraft pulping effluents peaked during the second
hour of exposure (Walden ^t al. 1970).  These results suggested that the
period of peak response may be diagnostic of the type of chemical substance
to which the fish are exposed.

     The response may, after reaching a peak, decline rapidly or slowly toward
pretreatment levels under conditions of continuous exposure.  The speed at
which cough frequencies return to normal varies similarly when exposures are
stopped at time of maximum response.  Thus, the rate of decline may also be
an important factor in classifying the type of chemical substance.

     Although we have emphasized cough frequency because it is a more definitive
response to sublethal concentrations than opercular rate and amplitude, various
combinations of these respiratory responses, differences in their intensities,
and differences in the rates at which they change may reflect the degree of
effect that would be imposed upon the fish should toxicant exposure be
prolonged.  These responses can be categorized as follows:

     Level I response:  Cough frequency, opercular rate, and amplitude vary,
but do not increase significantly from one day to the next.  Response
indicates normal expected variation for fish in ordinary water.

     Level II response;  Cough frequencies increase significantly without a
corresponding increase in opercular rate or amplitude.  Response indicates
adverse long-term (chronic) effects.

     Level III response;  Large increases in cough frequency accompanied by
moderate increases in opercular rate and amplitude.  Response indicates
sublethal to lethal effects.

     Level IV response;  Large and rapid increases in cough frequency,
opercular rate,  and amplitude.  Response indicates lethal (acute) effects.

     Level V response;  Cough frequency, opercular rate, and amplitude greatly
reduced or erratic after a level IV response.  Response indicates that death
is imminent.

Because these categories are founded on limited data, we have not attempted
to quantify the break-points between them.
                                     26

-------
         o
         UJ
         a
         a:
         LI.
         o
         o
         o
         UJ
B
         UJ
         cr
                 r(death)
                          I            2
                                    DAYS
Figure 5.   Cough-response patterns of brook trout exposed to copper  (A),
           methylmercuric chloride (B), and chlorinated hydrocarbons (C).
           H,  I, and L  denote high, intermediate, and low concentrations.
                                     27

-------
UTILITY OF THE RESPONSE

     The fish-cough test appears to have wide potential application as an
endpoint for screening new chemicals for relative toxicity, predicting long-
term effect concentrations, evaluating the quality of drinking water (Jung
1973) and surface water, and as an indicator of effluent toxicity.  The latter
is especially significant because the organism integrates the complex
interrelationships of several possible toxicants present in a waste into a
single, well-defined response.  Sublethal biological effects cannot be predicted
from physical-chemical measurements, nor can they be shown from 96-h acute
mortality tests  that only consider survival.

     Of these possible applications, the cough response appears most useful as
an endpoint for  predicting long-term effect concentrations.  In addition to
copper  (Drummond et^ _al_. 1973) and mercury (Drummond ejt al. 1974), we have
found that other materials (Table 1) also cause significant increases in brook
trout cough frequency near those MATC levels known to cause long-term effects
such as increased mortality, reduction in growth rate, poor reproduction, and
physical deformities.  Although the present data base is small, information
is sufficient to warrant further studies in which predicted effect concentrations
can be  verified  by long-term exposures.  If the cough test is consistently
demonstrated to  be a good indicator of MATC concentrations, it will also provide
a short-term technique for evaluating the chronic toxicity of mixtures of
toxicants.
                                      28

-------
       TABLE 1.   LOWEST CONCENTRATIONS OF TEN COMPOUNDS CAUSING

     SIGNIFICANT INCREASES IN BROOK TROUT COUGH FREQUENCY AND MATC

                       RANGES FOR THIS SPECIES.
Effective concentration (yg/1)*
Compound
Cadmium chloride
Sodium dichromate
Lead nitrate
Zinc sulfate
Copper sulfate
Methylmercuric chloride
Diazinon
Cough test
5
860
80
1,392
10
3
25
MATC range
2 - 32
200 - 3502
58 - 1192
532 - 1,3682
9 - 172
0.3 - 0.92
< 0.82
  (Tech. grade 93.6%)

Malathion
  (Tech. grade 95.0%)

Lindane
  (ESA Pest. Ref.
   Standard 100%)

DDT
  (ppf isomer 77.2%)
7


6
 < 163


  9 - 173



unavailable
Measured concentration rounded to nearest whole number.

2McKim (1977).

3Andrew, R. W., et^ aL.  Evaluation of an application factor hypothesis.
 Envir. Res. Lab.-Duluth, unpublished manuscript.
                                   29

-------
                                 REFERENCES
American Public Health Association.  1976.  Standard methods for the
     examination of water and wastewater.  14th ed.  1193 p.  New York.

Ballintijn, C. M., and G. M. Hughes.  1965.  The muscular basis of the
     respiratory pumps in the trout.  J. Exp. Biol.  43:349-362.

Ballintijn, C. M., and J. L. Roberts.  1976.  Neural control and proprioceptive
     load matching in reflex respiratory movements of fish.  Fed. Proc.
     35:1983-1991.

Bass, M. L.  1975.  A study of lethality and toxic mechanisms of intermittent
     chlorination to freshwater fish.  Ph.D. Thesis.  Virginia Polytechnic
     Institute and State University, Blacksburg, VA.  70 p.

Bert, P.  1870.  Lecons sur la physiologie compare'e de la respiration.  J. -B.
     Ballie"re, Paris.

Bijtel, J. H.  1949.  The structure and the mechanism of movement of the gill-
     filaments in teleostei.  Arch. Ne"erl. Zool.  8:1-22.

Bimber, D. L., R. W. Boenig, and M. L. Sharma.  1976.  Respiratory stress in
     yellow perch induced by subtoxic concentrations of diquat.  Ohio J. Sci.
     76:87-90.

Bull, C. J., and J. E. Mclnerney.  1974.  Behavior of juvenile coho salmon
     (Oncorhynchus kisutch) exposed to sumithion (fenitrothion) an organo-
     phosphate insecticide.  J. Fish. Res. Board Can.  31:1867-1872.

Cairns, J., Jr., J. W. Hall, E. L. Morgan, R. E, Sparks, W, T. Waller, and G. F.
     Westlake.  1973.  The development of an automated biological monitoring
     system for water quality.  Virginia Wat. Res. Center, Blacksburg, VA.
     Bull. 59.  50 p.

Cairns, J., Jr., E. L. Morgan, and R. E. Sparks,  1974.  The response of
     bluegills (Lepomis macrochirus Rafinesque) in a pollution monitoring
     system to a diurnal temperature change.  Trans. Am. Fish. Soc.
     103:138-140.

Camougis, G.  1960.  Recording bioelectric potentials from aquatic animals.
     Turtox News  38:156-157.

Carlson, R. W., and R. A. Drummond.  1977.  Fish cough response - a method for
     evaluating quality of treated complex effluents.  Water Res. (in press).

                                     30

-------
Dandy, J. W. T.  1967.  Effects of chemical characteristics of the environment
     on the activity of an aquatic organism.   Ph.D. Thesis.  Univ. Toronto,
     Toronto, Ont.  263 p.

Dandy, J. W. T.  1972.  Activity response to chlorine in brook trout, Salvelinus
     fontinalis (Mitchill).   Can. J.  Zool.   50:405-410.

Davis, J. C.  1973.  Sublethal effects of bleached kraft pulp mill effluent on
     respiration and circulation in sockeye salmon (Oncorhynchus nerka).  J.
     Fish. Res. Board Can.  30:369-377.

Davis, J. C.  1976.  Progress in sublethal  effect studies with kraft pulpmill
     effluent and salmonids.   J. Fish. Res. Board Can.   33:2031-2035.

Dessy, R. E.  1976.  Microprocessors  - an end user's view.  Science.  192:511-518.

Drummond, R. A.  1977.  Disposable electrode chamber for measuring opercular
     movements of fathead minnows. Prog.  Fish-Cult.  39:94.

Drummond, R. A., and W. F. Dawson. 1970.   An inexpensive method for simulating
     diel patterns or lighting in the laboratory.  Trans. Am. Fish. Soc.
     99:434-435.

Drummond, R. A., and W. F. Dawson. 1974.   An inexpensive solidstate amplifier
     for detecting movements  and electrical potentials  of fish.  Trans. Am.
     Fish. Soc.  103:391-392.

Drummond, R. A., G. F. Olson, and A.  R. Batterman.  1974.  Cough response and
     uptake of mercury by brook trout (Salvelinus fontinalis) exposed to
     mercuric compounds at different  hydrogen-ion concentrations.  Trans. Am.
     Fish. Soc.  103:244-249.

Drummond, R. A., W. A. Spoor, and G.  F. Olson.  1973.  Some short-term indicators
     of sublethal effects of  copper on brook trout, Salvelinus fontinalis.  J.
     Fish. Res. Board Can.  30:698-701.

Ewing, G. W., and H. A. Ashworth.  1974.  The laboratory recorder.  Plenum
     Press, New York.  129 p.

Ferris, C. D.  1974.  Introduction to bioelectrodes.  Plenum Press, New York.
     243 p.

Francois-Franck, C. A.  1906.  Analyse graphique de movements respiratoires des
     poissons teleosteens.  Compt. Rend. Seances Soc. Biol. Filiales  60:799-802.

Goodman, D. A., and N. M. Weinberger.  1971.   Submerged electrodes in an
     aquarium:  validation of a technique for remote sensing of behavior.
     Behav. Res. Meth. Instr.  3:281-286.

Hargis, J. R.  1976.  Ventilation and metabolic rate of young rainbow trout
     (Salmo gairdneri) exposed to sublethal environmental pH.  J. Exp. Zool.
     196:39-44.


                                       31

-------
Heath, A. G.  1972.  Critical comparison of methods for measuring fish
     respiratory movements.  Water Res.  6:1-7.

Heimstra, N. W., D. K. Damkot, and N. G. Benson.  1969.  Some effects of silt
     turbidity on behavior of juvenile largemouth bass and green sunfish.  U.S.
     Fish Wildl. Serv., Bur. Sport Fish. Wildl. Tech. Pap. 20.

Holeton, G. F., and D. R. Jones.  1975.  Water flow dynamics in the respiratory
     tract of the carp (Cyprinus carpio L.).  J. Exp. Biol.  63:537-549.

Howard, T. E., and C. C. Walden.  1974.  Measuring stress in fish exposed to
     pulp-mill effluents.  Tappi.  57:133-135.

Hughes, G. M.  1960.  A comparative study of gill ventilation in marine teleosts.
     J. Exp. Biol.  37:28-45.

Hughes, G. M.  1973.  Respiratory responses to hypoxia in fish.  Am. Zool.
     13:475-489.

Hughes, G. M.  1975.  Coughing in the rainbow trout (Salmo gairdneri) and the
     influence of pollutants.  Rev. Suisse Zool.  82:47-64.

Hughes, G. M., and R. J. Adeney.  1977.  Variations in the pattern of coughing
     in rainbow trout.  J. Exp. Biol.  68:109-122.

Hughes, G. M., and M. Morgan.  1973.  The structure of fish gills in relation
     to their respiratory function.  Biol Rev.  48:419-475.

Hughes, G. M., and J. L. Roberts.  1970.  A study of the effect of temperature
     changes on the respiratory pumps of the rainbow trout.  J. Exp. Biol.
     52-177-192.

Hughes, G. M., and R. L. Saunders.  1970.  Response of the respiratory pumps
     to hypoxia in rainbow trout CSalmo gairdneri).  J. Exp. Biol.  53:529-545.

Hughes, G. M., and G. Shelton.  1958.  The mechanism of gill ventilation in
     three freshwater teletostean fishes.  J. Exp. Biol.  35:807-823.

Jung, K. D.  1973.  Substances extremely toxic to fish and their importance for
     a fish test warning system.  GWF-Wasser/Abwasser.  114:232-234.

Kuiper, T.  1907.  Untersuchungen Uber die Atmung der Teleostei.  PflUgers.
     Arch. Ges. Physiol.  177:1-107.

Lunn, C. R., D. P. Toews, and D. J. R. Pree.  1976.  Effects of three pesticides
     on respiration, coughing, and heart rates of rainbow trout (Salmo gairdneri
     Richardson).  Can. J. Zool.  54:214-219.

MacLeod, J. C., and L. L. Smith, Jr.  1966.  Effect of pulpwood fiber on oxygen
     consumption and swimming endurance of the fathead minnow, Pimephales
     promelas.  Trans. Am. Fish. Soc.  95:71-84.
                                      32

-------
Martens, D.  W.,  R.  W.  Gordon,  and J.  A.  Servizi.   1970.   Toxicity of copper to
     sockeye and pink  salmon during their freshwater life.   Int.  Pac. Salmon
     Fish.  Comm., Manuscript Report,  27  p.

McCutcheon,  F.  H.  1970.   Stimulation, control  and phylogenetic projection of
     the teleostean yawn  reflex.   Comp.  Biochem.  Physiol.   34:399-344.

Mclntosh, A., and W. Bishop.  1976.  Distribution and effects of heavy metals
     in a contaminated lake.  Water Resour.  Res.  Center,  Purdue Univ.,  Tech.
     Rep. 85.  69 p.

McKim,  J. M.  1977. Evaluation of tests with early life  stages of fish for
     predicting long-term toxicity.  J.  Fish. Res. Board  Can.  34:1148-1154.

MzKim,  J. M., and D. A. Benoit.  1971.   Effects of long-term exposures to
     copper on survival,  growth,  and reproduction of brook  trout (Salvelinus
     fontinalis).  J.  Fish.  Res.  Board Can.  28:655-662.

McKim,  J. M., G. F. Olson, G.  W.  Holcombe,  and  E.  P.  Hunt.   1976.   Long-term
     effects of methylmercuric chloride  on three  generations of brook trout
     (Salvelinus fontinalis):   toxicity, accumulation,  distribution, and
     elimination.  J.  Fish.  Res.  Board Can.  33:2726-2739.

IfcPherson,  B. P. 1973.   Effects  of methylmercury exposure  on sea water
     adaptation of  juvenile  salmonids.   M.S. Thesis.   Oregon State Univ.,
     Corvallis,  Ore.  89  p.

Morgan, W.  S. G. 1976.   Fishing  for toxicity:  biological  auto-monitor for
     continuous water  quality  control.   Effl. Water Treat.   16:471-475.

Morgan, W.  S. G. 1977.   Biomonitoring with fish:   an aid to industrial
     effluent and surface water quality  control.   Prog. Water Tech.  (Great
     Britain)  9:703-711.

Morgan, W.  S. G., and  P.  C.  Kuhn.  1974.  A method to monitor the effects of
     toxicants upon breathing  rate of largemouth  bass (Micropterus salmoides
     Lacepede).   Water Res.  8:67-77.

Mount,  D. I., and C. E. Stephan.   1967.   A method for establishing acceptable
     toxicant limits for  fish—malathion and the  butoxyethanol ester of 2,4-D.
     Trans.  Am.  Fish.  Soc.   96:185-193.

Rice, S. D., R.  E.  Thomas, and J. W.  Short.  1977.  Effect  of petroleum
     hydrocarbons on breathing and cough rates  and hydrocarbon uptake-depuration
     in pink salmon fry.   In:   Physiological Responses of Marine Biota to
     Pollutants. Academic Press, Inc.,  New York.   p. 259-277.

Rommel, S.  A.,  Jr.   1973. A simple method for  recording  fish heart and
     operculum beats without the  use of  implanted electrodes.  J.  Fish. Res.
     Board  Can.   30:693-694.
                                      33

-------
Rovainen, C. M.  1977.  Neural control of ventilation in the lamprey.  Federation
     Proc.  36:2386-2389.

Satchell, G. H., and D. J. Maddalena.  1972.  The cough or expulsion reflex in
     the Port Jackson shark, Heterodontus portus jacksoni.  Comp. Biochem.
     Phys.  41A:42-62.

Saunders, R. L.  1961.  The irrigation of the gills in fishes.   I.  Studies of
     the mechanism of branchial irrigation.  Can. J.  Zool. 39:637-653.

Schaumburg, F. D., T. E. Howard, and C. C. Walden.  1967.  A method to evaluate
     the effects of water pollutants on fish respiration.  Water Res.  1:731-737.

Sellers, C. M., Jr., A. G. Heath, and M.  L. Bass.  1975.  The effect of sublethal
     concentrations of copper and zinc on ventilatory activity, blood oxygen and
     pH in rainbow trout (Salmo gairdneri).  Water Res.  9:401-408.

Servizi, J. A., R. W. Gordon, and D. W. Martens.  1969.  Marine disposal of
     sediments from Bellingham Harbour as related to  sockeye and pink salmon
     fisheries.  Int. Pac. Salmon Fish. Comm. Prog. Rep. 23.  38 p.

Shelton, G.  1959.  The respiratory center in the tench (Tinea tinea).  I.
     The effects of brain transection on respiration.  J. Exp.  Biol.  36:191-202.

Skidmore, J. F.  1970.  Respiration and osmoregulation in rainbow trout with
     gills damaged by zinc sulphate.  J.  Exp. Biol.  52:481-494.

Sparks, R. E., J.  Cairns, Jr., and A. G.  Heath.   197.2.  The use of bluegill
     breathing rates to detect zinc.  Water. Res.  6:895-911.

Sparks, R. E., J.  Cairns, Jr., R. A. McNabb, and G. Sutter, II.  1972.
     Monitoring zinc concentrations in water using the respiratory response
     of bluegills (Lepomis macrochirus Rafinesque).  Hydrobiologia  40:361-369.

Spoor,  W. A.,  T.  W.  Neiheisel, and R. A.  Drummond.  1971.  An electrode chamber
     for recording respiratory and other movements of free-swimming animals.
     Trans. Am. Fish. Soc.  100:22-28.

Sprague, J. B.  1964.  Avoidance of copper-zinc  solutions by young salmon in
     the laboratory.  J. Water Poll. Cont. Fed.   36:990-1004.

Sprague, J. B.  1971.  Measurement of pollutant  toxicity to fish.  III.
     Sublethal effects and "safe" concentrations.  Water Res.  5:245-266.

Steel,  R. G. D.,  and J. H. Torrie.  1960.   Principles and procedures of
     statistics.   McGraw-Hill Co., Inc.  New York.  481 p.

Strange, J. R., J.  W. Dean, G. K. Dorman,  and D. J. Fletcher.  1975.  New
     method for recording heart and respiratory  rates in catfish.  Sci. Biol.
     1:17-18.
                                      34

-------
U. S. Environmental Protection Agency.   Committee on Methods for Toxicity Tests
     with Aquatic Organisms.   1975.   Methods for acute toxicity tests with
     fish, macroinvertebrates, and amphibians.   Rep. EPA-660/3-75-009.  Nat.
     Tech. Inf. Ser., U.S. Dep. Comm.,  Springfield,  VA.   61 p.

Venables, G.,  and E. N. Smith.  1972.  A simple method for recording opercular
     rate and heart rate of unrestrained fish.   Prog.  Fish-Cult.  34:233-234.

Walden, C. C., T. E. Howard,  and G.  C.  Froud.   1970.  A quantitative assay of
     the minimum concentration of kraft mill effluents which affect fish
     respiration.  Water Res.  4:61-68.

Young, S.  1970.  EMG activity in tench (Tinea  tinea)  gill lamellae and its
     association with coughing.  J.  Physiol. (London)   215:37-38.

Young, S.  1972.  Electromyographic activity during  respiration and coughing
     in tench (Tinea tinea L.).  J.  Physiol. (London)   227:18-19.
                                      35

-------
                                 APPENDICES


Appendix A.  An air-lift chamber system designed for effluent testing.   System
             also can be used for testing single toxicants.


     Each chamber (15 cm long x 4 cm wide x 10.5 cm high)  exposes fish  to
effluent or control water air-lifted up glass tubes from an  effluent tray into
the chamber (Figure A-l).  One electrode is glued to the drain end,  and the
other electrode is mounted on a 3-cm x 10-cra moveable glass  baffle.   The baffle
is suspended in the chamber by exiting the electrode on one  side and a  short
stainless steel wire hanger on the other, both resting in notches cut in the
top of each side.  Three sets of notches, at 6 cm,  9 cm, and 12 cm from the
drain end, permit chamber size to be adjusted to accommodate any fish 3-10 cm
long.  A 2-cm square of plexiglas glued to the rear upper part of the moveable
baffle prevents the fish from pushing the electrode toward the water inlet when
a glass lid is in place and weighted.  The 4-cm x 13-cm lid  covers the  chamber
except for the water inlet.  A sliding 6-mro-diameter glass standpipe adjusts
chamber volume to 250 for a centrarchid or 125 ml for small  salmonids.

     Four individual chambers are placed in a three-sided box constructed so
that the drain end of each chamber extends beyond the bottom on the  open side
(Figure A-l).   The box is made of 1-cm plywood coated with fiberglass and rests
on the sides of a water bath.  Each water bath consists of a 5 1/2-gal  aquarium
containing an effluent tray in place of the four-plex, Immersed chamber described
in Appendix B.  The effluent tray (23 cm long x 18  cm wide x 13 cm high)  is
divided into two equal compartments, and the bottom edge of  a drain  hole drilled
in one end for each half of the tray is exactly 10  cm off the tray bottom.
Volume of each compartment is 2,000 + 50 ml when filled to the drain hole and
supplies water or effluent to two individual electrode chambers.

     Each lift tube is a 23-cm length of 8-mm glass tubing connected with
rubber tubing to one side of a 1/4-in. Kimax glass  connecting u-tube.  The
constriction is cut off the u-tube to make it flush with the lift tube, and the
unused common opening of the u-tube is plugged with glass glue.  The other
side of the u-tube extends over the front of the plywood box into the electrode
chamber water inlet (Figure A-l).  Air is supplied  through 3-mm glass tubing
bent- 180 degrees at one end to fit into the bottom  of the lift tube.  Air flow
is regulated with three-way valves, and when properly adjusted a maximum lift
of 65 ml/min is possible.  Flows are maintained at  approximately 50  ml/min
through each chamber during testing, and the chambers drain  directly back into
the effluent tray.  During cecirculating tests the  drain hole is stoppered, and
the water-delivery tubes are flipped back out of the effluent trays.  The
dilution water then flows into the surrounding water baths to maintain  effluents
and fish at test temperature.

                                     36

-------
      Electrode.
Incoming
Acclimation
Water	
Tygon.
                                    \
                    Electrode/baffle
1-
                                   Plywood Box
                          Effluent Tray
Pi
 5

I*
fa
y,\
'/•
                                                         Air  line
                                                 U
      i
                            Water Bath
                                                            .Drain Hole
                 Figure A-i.  Airlift chamber  system.
                                  37

-------
Appendix B.  A four-plex, Immersed chamber system designed for effluent testing.
             System also can be used for testing single toxicants.


     Each four-plex consists of a series of glass partitions 18 cm long x 10
cm high and sides mounted on a base plate 18 cm long x 15 cm wide (Figure B-l).
End pieces (3 cm wide x 7 cm high) glued 3 cm off the base separate the
partitions and sides and are mounted flush with one end, but recess 7 cm from
the other end.  Electrodes are constructed of 18-cm-long stainless steel wires
spot-welded to five-mesh-per-inch stainless steel screens.  The common
(negative) electrode is a single 15-cm-wide x 3-cm-high screen glued across
the openings of the non-recessed end, and individual (positive) electrodes are
3-cm x 4-cm screens glued across each recessed chamber opening.  Recessing the
positive electrodes reduces signal pick-up from adjacent fish chambers.

     The four-plex is designed to immerse fish in a tank of effluent or control
water (Figure b-1).  Each effluent tank is a 5-1/2-gal all-glass aquarium (40
cm long x 20 cm wide x 25 cm high) with glass brackets glued to the interior
walls to hold a glass plate 11.5 cm off the bottom.  A drain hole is drilled
in one end of the aquarium so that its bottom edge is exactly 18 cm from the
bottom.  Final volume of the effluent tank is 14 1 + 100 ml when filled to
drain level, and a four-plex resting on the glass plate will be covered to a
depth of 7 cm.  Glass aeration tubes are placed in each effluent tank to
promote water circulation through the four-plex.  Each effluent tank containing
a four-plex is placed in a large water bath, and water flow during acclimation
or flow-through testing is through the drain hole into the water bath and then
to drain.  During recirculating tests the drain hole is stoppered, and the
water-delivery tubes are flipped back out of the effluent tanks.  The dilution
water then flows into the surrounding water baths to maintain effluents and
fish at test temperatures.
                                      38

-------
       Common
       Electrode.
 oxl
                                                   Individual
                                                  .Electrode
                    Top  View (chamber  only)
Incoming
Acclimation
Water	
Tygon
     Water
      Bath
                    Common
                    Electrode
              Individual
              .Electrode
                                                      u
                               Drain Hole
Effluent Tank
                           Side View
                                                    Drain

                Figure B-l.  Innnersed chamber system.
                                39

-------
Appendix C.  Examples of fish coughs (arrows) recorded by a rectilinear
             polygraph.  Note;  Arrows point to representative coughs.
             others are present but are not marked.
                                      40

-------
     a.
     b.
     c.
    d.
                                                              Locomotor activity
Ventilating movement
Figure C-l.  Respiratory patterns of four bluegill sunfish (Lepomis macrochirus)  in control  water'.
             Chart speed:  0.1 cm/sec.

-------
to
                                  ventilating movement
        Figure C-2.  Respiratory patterns of four bluegill  sunfish  (Lepomis macrochirus) exposed  to a
                     sublethal level of a toxicant.  Chart  speed:   0.1 cm/sec.

-------
  Recording
  Speed:

  O.I cm/sec
  0.25 cm/sec
  0.5 cm/sec
  1.0 cm/sec
                                    Ventilating movement
  2.5 cm/sec
                      \
                                                         2.5 sec
                                              Cough
      j\
A; i
                                                                 i
                                                          1.0 sec
Figure C-3.  Respiratory pattern of bluegill, sunfish (Lepomis macrochirus)
            at 72-hr exposure to a sublethal level of zinc.  Strip-chart
            records are of the same fish, but at different paper speeds.
                                 43

-------
    Recording
    Speed-

    CX I cm/sec
    0.25 cm/sec
    O.5 cm/sec
     1.0 cm/sec
                          Ventilating movement
                                                              2.5 sec
                         Cough
    2.5 cm/sec
                                                              1.0 sec
Figure  C-4.
Respiratory pattern of bluegill sunfish (Lepomls macrochirus)
at 24-hr exposure  to a lethal level of zinc.   Strip-chart
records are of the same fish, but at different paper speeds.
                                     44

-------
    Recording
    Speed:

    O.I cm/sec
    0.25 cm/sec
    0.5 cm/sec
    1.0 cm/sec
                                                              10 sec
                                   Normal ventilating movement
                                                              2.5 sec
    2.5 cm/sec
                               .Cough
                                                               1.0 sec
Figure C-5.   Respiratory pattern of bluegill sunfish  (Lepomis macrochirus)
             at  30-hr exposure to a lethal level  of pentachlorophenol.
             Strip-chart recordings are of the same fish, but at different
             paper speeds.
                                    45

-------
Figure C-6.
Respiratory patterns of brook trout (Salvelinus fontinalis)
while a,b) in control water, and c,d) undergoing toxicant
stress.  Chart speed was 0.1 cm/sec.
                                    46

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
   EPA-600/3-77-133
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE

  PROCEDURES  FOR MEASURING COUGH  (GILL PURGE) RATES OF
  FISH
                                5. REPORT DATE
                                  December  1977 issuing  date
                                6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

  Robert  A.  Drummond and Richard W.  Carlson
                                8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Environmental  Research Lab. - Duluth,  MN
  Office of Research and Development
  U.S. Environmental Protection Agency
  Duluth, Minnesota 55804
                                                            10. PROGRAM ELEMENT NO.
                                  1BA608	
                                11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
               SAME AS ABOVE
                                13. TYPE OF REPORT AND PERIOD COVERED
                                  Jlnal
                                                            14. SPONSORING AGENCY CODE
                                                               EPA/600/03
15. SUPPLEMENTARY NOTES
16. ABSTRACT
       The cough  (gill  purge) is an interruption in the normal ventilatory cycle  of fish
  that serves to  clean  the gills of accumulated particulate matter.   A review of  the
  literature shows  that the cough occurs  in  a  variety of freshwater  and marine  fish;
  that both mechanical  and chemical stimulation apparently can cause fish to increase
  their cough rates;  and that an increase  in coughing is a rapid  and sensitive  endpoint
  for studying chemicals and effluents.   In  reviewing the test methods and apparatus for
  measuring cough rates of fish, we conclude the electrode chamber method offers  more
  potential as a  bioassay tool for assessing the respiratory responses of fish  due  to
  toxicant exposure.  Recommended test procedures,  based on our experience, for using
  the electrode chamber method are given.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                  b.lDENTIFIERS/OPEN ENDED TERMS  C. COS AT I Field/Group
  Fresh water  fishes
  Cough
  Respiration
  Ventilation
  Toxicity
  Effluents
  Methodology
Reviews-
Pollution
Tests
Electrode  chambers
Gill purge
Opercular  movements
Action potentials
Test procedures
06C
06F
06S
06T
14B
14C
18. DISTRIBUTION STATEMENT
  RELEASE TO PUBLIC
                                               19. SECURITY CLASS (This Report)
                                                UNCLASSIFIED
                                              21. NO. OF PAGES

                                                     55
                  20. SECURITY CLASS (Thispage)

                    UNCLASSIFIED
                                                                          22. PRICE
EPA Form 2220-1 (R«v. 4-77)   PREVIOUS EDITION is OBSOLETE
                                             47
                                    OUSGPO: 1978 - 757-140/6634 Region 5-11

-------