Protection
Ageney
WmtfngtonDC 20460
EPA 560/6-82402
PB82-232992
Auguit 1982
Tox ic Substances
SEPA
ental Effects
-------
GUIDELINES AND SUPPORT DOCUMENTS
FOR
ENVIRONMENTAL EFFECTS TESTING
Part One
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
PREAMBLE
The following guidelines describe methods for performing testing
of chemical substances under the Toxic Substances Control Act
(TSCA). These methods include the state-of-the-art for
evaluating certain properties, processes and effects of
chemical substances. They are intended to provide guidance
to test sponsors in developing tes!t protocols for compliance
with test rules issued under Section 4 of the TSCA. They
may also provide guidance for testing which is unrelated
to regulatory requirements. Support documentation is
included for some of these guidelines. It is expected that
additional guidelines and support documentation will be
incorporated later as the state-of-the-art evolves or the
need for them warrants.
Since these guidelines are divided into three sections which
cover the diverse areas of health effects, environmental
effects and chemical fate testing, there are some differences
in the ways they are presented. These differences are
explained in an introduction prepared for each section.
-------
INTRODUCTION -TO ENVIRONMENTAL EFFECTS .TESTING GUIDELINES
In order to assess the environmental risk associated with a
chemical s'ubstance, a-determination of ecotoxicity
potential, along with information on the transport and fate
of the substance in the environment, is needed. The extent
and degree to which a ch.emical may pose a potential hazard
can be characterized in part by its ecotoxicity to plant,
animal and microbial species which are valued for economic
or ecological importance. Ecotoxicity can be evaluated on
the basis of-those acute, subacute or chronic effects which
result in death, inhibition of reproduction, or an
impairment of growth and development of an organism.
Ecotoxicity also can occur as•a result of the presence or
accumulation of a chemical in or on one organism which is
not affected by the chemical but serves as basic food source
for another organism which is affected.
Whether a chemical substance will cause ecotoxic effects is
greatly dependent upon the organism and the stage in the
life cycle in which exposure occurs and the conditions under
which exposure occurs. The toxicity of a chemical may not
be the same to all organisms or to all levels of biological
organization. Ideally, such testing should employ test
organisms or systems which provide for the broadest range of
taxonomic representation and biological processes within the
constraints of the costs and resources available. The Test
Guidelines for-Environmental Effects Testing have been
selected with these constraints in mind.
The Guidelines use single species of plants and animals and
-------
incorporate the best state-of-the-art methodologies for
testing purposes. It is anticipated that the species and
methodologies used in the Guidelines will be reviewed and
revised as the state-of-the-art changes. Such changes may-
include other single species tests, multi-species tests or
microcosm tests.
The Test Guidelines and Support Documents are identified by
the prefixes EG and ES, respectively, and are numbered
sequentially. Where applicable, each Test Guideline is
supported by a document which provides the scientific
background and rationale used in the development of the Test
Guideline. In some cases, a Support Document provides
support for two Test Guidelines.
-------
INDEX TO ENVIRONMENTAL EFFECTS TEST GUIDELINES AND SUPPORT DOCUMENTS
Document Guideline
Daphnid Acute Toxicity Test
Daphnid Chronic Toxicity Test
Mys id Shrimp Acute Toxicity Test
Mys id Shrimp Chronic Toxicity Test
Oyster Acute Toxicity Test
Oyster Bioconcentration Test
Penaeid Shrimp Acute Toxicity Test
Algal Acute Toxicity Test
Fish Acute Toxicity Test
Fish Bioconcentration Test
Fish Early Life Stage Toxicity Test
Seed Germination/Root Elongation
Toxicity Test
Early Seedling Growth Toxicity Test
Plant Uptake and Translocation Test
Avian Dietary Test
Bobwh ite. Reproduction Test
Mallard Reproduction Test
Daphnid Chronic Toxicity Test (OECD)
Algal Acute Toxicity Test (OECD)
Fish Acute Toxicity Test (OECD)
Fish Bioconcentration Test (OECD)
Support
EG-1
SG-2
EG-3
EG-4
EG-5
EG-6
T— v"~i "7
CJo — /
EG-8
BG-9
EG-10
EG-11
EG-1 2
EG-1 3
EG-1 4
EG-15
EG-1 6
EG-17
EG-18
EG-1 9
EG-20
BG-21
Document
ES-1
ES-1
ES-2
ES-2
ES-3
ES-3
ES-4
ES-5
ES-6
ES-7
ES-8
ES-9
ES-10
ES-11
ES-12
ES-1 3
ES-14
-------
EG-1
August, 1982
DAPHNID ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-1
Guideline for Testing of Chemicals August, 1982
DAPHNID ACUTE TOXICITY TEST
(a) Purpose. This guideline is intended for use in
developing data on the acute toxicity of chemical substances and
mixtures ("chemicals" 'subject to environmental effects test
regulations under the Toxic Substances Control Act (TSCA) (Pub.L.
94-469, 90 Stat. 2003, 15 U.S.C. 2601 et seg.). This guideline
prescribes an acute toxicity test in which daphnids (Daphnia
magna or JD. pulex) are exposed to a chemical in static and flow-
through systems. The United States Environmental Protection
Agency will use data Erom this test in assessing the hazard a
chemical may present in the aquatic environment.
(b) Def initions. The definitions in Section 3 of the Toxic
Substances Control Act (TSCA) and Part -792--Good Laboratory
Practice Sta idards apply to this test guideline. In addition,
the following definitions apply to this guideline:
(1) "Brood stock" means the animals which are cultured to
produce test organisms through reproduction.
(2) "EC50" means that experimentally derived concentration of
test substance in dilution water that is calculated to affect 50
percent of a test population during continuous exposure over a
specified period of time. In this guideline, the effect
measured is immobilization
-------
EG-1
August, 1982
measured is immobilization.
(3) "Ephippium" means a resting egg which develops under
the carapace in response to stress conditions in daphnids.
(4) "Flow-through" means a continuous or an intermittent
passage of test solution or dilution water through a test chamber
or culture tank with no recycling.
(5) "Immobilization" means the lack of movement by the test
organisms except for minor activity of the appendages.
(6) "Loading" means the ratio of daphhid biomass (grams,
wet weight) to the volume (liters) of test solution in a test
chamber at a point in time, or passing through the test chamber
during a specific interval.
(7) "Static system" means a test system in which the test
solution and test organisms are placed in the test chamber and
kept there for the duration of the test without renewal of the
test solution.
(c) Test procedures — (1) Summary of the test. (i) Test
chambers are filled with appropriate volumes of dilution water.
In the flow-through test, the flow of dilution water through each
chamber is adjusted to the rate desired. The test chemical is
introduced into each treatment chamber. The addition of test
chemical in the flow-through system is conducted at a rate which
is sufficient to establish and maintain the desired
-------
EG-1
August, 1982
concentration in the test chamber. The test is started within 30
minutes aftert'ne test chemical has been added and uniformly
distributed in static test chambers or after the concentration of
test chemical in each flow-through ttl-. t chamber reaches the
prescribed level and remains stable. At the initiation of the
test, daphnids which have been cultured and acclimated in
accordance with the test design are randomly placed into the test
chambers. Daphnids in the test chambers are observed
periodically duringthe test, the immobile daphnids removed, and
the findings recorded.
(ii) Dissolved oxygen concentration, pH, temperature, the
concentration of test chemical and other water quality parameters
are measured at specified intervals in selected test chambers.
Data are collected during t .e test to develop concentration-
response curves and determine EC50 values for the test chemical.
(2) [Reserved]
(3) Range-finding test. (i) A range-f inding test should
be conducted to establish test solution concentrations for the
definitive test.
(ii) The daphnids should be exposed to a series of
widelyspaced concentrations of the test chemical (e.g., 1, 10,
100 mg/1, etc), usually under static conditions.
(iii) A minimum of five daphnids should be exposed to each
-------
US-1
August, i982
concentration of test chemical for a period of 48 hours. The
exposure period may be shortened if data suitable for the purpose
of the range-finding test can be obtained in less time. No
replicates are required and nominal concentrations of the
chemical are acceptable.
(4) Definitive test. (i) The purpose of the definitive
test is to determine the concentration-response curves and the
24- and 48- hour EC50 values with the minimum amount of testing
beyond the range-finding test.
(ii) A minimum of 20 daphnids per concentration should be
exposed to five or more concentrations of the chemical chosen in
a geometric series in which the ratio is between 1.5 and 2.0
(e.g., 2, 4, 3, 16, 32 and 64 mg/1). An equal number of daphnids
should be placed in two or more replicates. If solvents,
solubilizing agents or eraulsifiers have to be used, they should
be commonly used carriers and should not possess a synergistic or
antagonistic effect on the toxicity of the test chemical. The
concentration of solvent should not exceed 0.1 ml/1. The
concentration ranges should be selected to determine the
concentration-response curves and EC50 values at 24 and 48
hours. Concentration of test chemical in test solutions should
be analyzed prior to use.
(iii) Every test should include controls consisting of the
same dilution water, conditions, procedures and daphnids from the
-------
EG-1
August, 1982
same population (cultare container), except that none of the
chemical is added.
(iv) The dissolved oxygen concentration, temperature and pH
should be measured at the beginning of the test and I-it 24 and 48
hours in each chamber.
(v) The test duration is 48 hours. The test is
unacceptable if more than 10 percent of the control organisms
appear to be immobilized, stressed or diseased during the 48 hour
test period. Each test chamber should be checked for immobilized
daphnids at 3, 6,, 12, 24 and 48 hours after the beginning of the
test. Concentration-response curves and 24-hour and 48-hour EC50
values for immobilization should be determined along with their
95 percent confidence limits.
(vi) In addition to immobility, any ;bnormal behavior or
appearance should also be reported.
(vii) Distribution of daphnids among test chambers should
be randomized. In addition, test chambers within the testing
area should be positioned in a random manner or in a way in which
appropriate statistical analyses can be used to determine the
variation due to placement.
(viii) The concentration of dissolved test chemical (that
which passes through a 0.45 micron filter) in the chambers should
be measured as often as is feasible during the test. In the
-------
Augus t, -1-93 2
static test the concentration of test chemical should be
measured, at a minimum, at the beginning of the test and at the
end of the test in each test chamber. In the flow-through test
the concentration of test chemical should be measured at a
minimum; (A) in each chamber at the beginning of the test and at
24 and 43 hours after the start of the test; (B) in at least one
appropriate chamber whenever a malfunction is detected in any
part of • the test substance delivery system. Among replicate test
chambers of a treatment concentration, the measured concentration
of the test chemical should not vary more than 20 percent
(+ or -).
(5) [Reserved]
(6) Analytical measurements--( i ) Tes t chemical . De ionized
water should be used in making stock solutions of the test
chemical. Standard analytical methods should be used whenever
available in performing the analyses. The analytical method used
to measure the amount of test chemical in a sample should be
validated before beginning the test by appropriate laboratory
practices. An analytical method is not acceptable if likely
degradation products of the test chemical, such as hydrolysis and
oxidation products, give positive or negative interferences which
cannot be systematically identified and corrected mathematically.
(ii) Numerical. The number of immobilized daphnids should
-------
EG-1
August, 1982
be counted during each definitive test. Appropriate statistical
analyses should provide a goodness-of-f it determination for the
concentration-response curves. A 24- and 48- hour EC50 and
corresponding 95 percent interval should be calculated.
(d) Test conditions — (1) Test species--(i) Selection.
(A) The cladocerans, Daphnia magna or D. pulex , are the test
species to be used in this test. Either species may be used for
testing of a particular chemical. The species identity of the
test organisms should be verified using appropriate systematic
keys. First ins tar daphnids, _<_ 2 4 hours old, are to be used to
start the test.
(3) Daphnids to be used in acute toxicity tests should be
cultured at the test facility. Records should be kept reaarding
the source of the initial stock and culturing techniques. All
organisms used for a particular test should have originated from
the same source and be from the same population (culture
container).
(C) Daphnids should not be used for a test (_!_) if cultures
contain ephippia; (_2_) if adults in the cultures do not produce
young before day 12; (_3_) if more than 20 percent of the culture
stock die during the two days oreceeding the test; (_4_) if adults
in the culture do not produce an average of at least three young
per adult per day over the seven day period prior to the test
7
-------
Augus '• 1982
and (J5_) if daphnids have been used in any portion of a previous
test, either in a treatment or in a control.
(ii) Acclimation. (A) Daphnids should be maintained in
100 percent dilution water at the test temperature for at least
48 hours prior to the start of the test. This is easily
accomplished by culturing them in the dilution water at the test
temperature. Daphnids should be fed prior to the test.
(3) During culturing and acclimation to the dilution water,
daphnids should be maintained in facilities with background
colors and light intensities similar to those of the testing
area.
(iii) Care and handling. (A) Daphnids should be cultured
in dilution water under similar environmental conditions to those
used in the test. Organisms should be handled as little as
possible. When handling is necessary it should be done as
gently, carefully and quickly as possible. During culturing and
acclimation, daphnids should be observed carefully for ephippia
and other signs of stress, physical damage and mortality. Dead
and abnormal individuals should be discarded. Organisms that
touch dry surfaces or are dropped or injured in handling should
be discarded.
(B) Smooth glass tubes (I.D. greater than 5 mm) equipped
with rubber bulb should be us eel for transferring daphnids
-------
EG-1
August, 1982
with minimal culture media carry-over. Care should be exercised
to introduce the daphnids below the surface of any solution to
avoid trapping air under the carapace.
(iv) Feeding. A variety of foods (e.g., unicellular green
algae) have been demonstrated to be adequate for daphnid culture.
Daphnids should not be fed during testing.
(2) Facilities--(i) Apparatus. (A) Facilities needed to
perform this test include: (1) containers for culturing and
acclimating daphnids; (2) a mech'anism for controlling a^nd
maintaining the water temperature during the culturing,
acclimation, and test periods; (3) apparatus for straining
particulate matter, removing gas bubbles, or aerating the water
as necessary; and (4) an apparatus for providing a 16-hour light
and 8-hour dark oho tope riod with a 15 - 30 minute transition
period. In addition, the flow-through system should contain
appropriate test chambers in which to expose daphnids to the test
chemical and an appropriate test substance delivery system.
(B) Facilities should be well ventilated and free of fumes
and disturbances that may affect the test organises.
(C) Test chambers should be loosely covered to reduce the
loss of test solution or dilution water due to evaporation and to
minimize the entry of dust or other particulates into the
solutions.
-------
August, 1982
(ii) Construction materials. (A) Materials and equipment
that contact test solutions should be chosen to minimize sorption
of test chemicals from the dilution water and should not contain
substances that can be leached into aqueous solution in
quantities that can affect the test results.
(B) For static tests, daphnids can be conveniently exposed
to the test chemical in 250 ml beakers or other suitable
containers.
(C) For flow-through tests, daphnids can be exposed in
glass or stainless steel containers with stainless steel or nylon
screen bottoms. The containers should be suspended in the test
chamber in such a manner to insure that the test solution flows
regularly into and out of the container and that the daphnids are
always submerged in at least five centimeters of test solution.
Test chambers can be constructed using 250 ml beakers or other
suitable containers equipped with screened overflow holes,
standpipes or V-shaped notches.
(iii) Dilution water. (A) Surface or ground water,
reconstituted water or dechlorinated tap water are acceptable as
dilution water if daphnids will survive in it for the duration of
the culturing, acclimation and testing periods without showing
signs of stress. The quality of the dilution water should be
constant and should meet the following specifications:
10
-------
EG-1
August, 1982
SUBSTANCE
MAXIMUM CONCENTRATION
20. mg/liter
2 mg/liter
5 mg/liter
1 ug/liter
<3 ug/liter
50 ng/liter
50 ng/liter
25 ng/liter
Particulate matter
Total organic carbon or
:hemical oxygen demand
Un-ionized ammonia
Residual chlorine
Total organophosphorus pesticides
Total organochlorine pesticides plus
polychlorinated biphenyls (PCBs) or
organic chlorine
(B) The above water quality parameters under paragraph
(d)(2)(iii)(A) of this section should be measured at least twice
a year or whenever it is suspected that these characteristics may
have changed significantly. If dechlorinated tap water is used,
daily chlorine analysis should be performed.
(C) If the diluent water is from a ground or surface water
source, conductivity and total organic carbon (TOG) or chemical
oxygen demand (COD) should be measured. Reconstituted water can
be made by adding specific amounts of reagent-grade chemicals to
deionized or distilled water. Glass distilled or carbon-filtered
deionized water with a conductivity less than 1 u ohm/cm is
acceptable as the diluent for making reconstituted water.
11
-------
EG-1
August, 1932
(iv) Cleaning. All test equipment and test chambers should
ba cleaned before each test using standard laboratory procedures.
(v) Test substance delivery system. In flow-through tests,
proportional diluters, metering pump systems or other suitable
devices should be used to deliver test chemical to the test
chambers. The system should be calibrated before each test.
Calibration includes determining the flow rate through each
chamber and the concentration of the test chemical in each
chamber. The general operation of the test substance delivery
system should be checked twice daily during a test. , The 24-hour
flow through a test chamber shouljd be equal to at least five
times the volume of the test chamber. During a test, the flow
rates should not vary more than 10 percent from any one test
chamber to another or from one time to any other.
(3) Test parameters. Environmental parameters of the water
contained in test chambers should be maintained as specified
below:
(i) Temperature of 20 ± 1°C.
(ii) Dissolved oxygen concentration between 60 and 105
percent saturation. Aeration, if needed to achieve this level,
should be done before the addition of the test chemical. All
treatment and control chambers should be given the same aeration
treatment.
12
-------
EG-1
August, 1982
(iii) The number of daphnids pla-ced'in a test chamber should
not affect test results. Loading should not exceed forty
daphnids per liter test solution in the static system. In the
flow-through test, loading limits will vary depending on the flow
r^te of dilution water. Loading should not cause the dissolved
oxygen concentration to fall below the recommended levels.
(iv) Photoperiod of 16 hours light and 8 hours darkness,
with a 15-30 minute transition period.
(e) Reporting. The sponsor should submit to the US EPA all
data'developed .by the t0s t that are suggestive or predictive of
acute toxicity and all concomitant gross toxicological manifes-
tations. In addition to the reporting requirements prescribed in
Part 792--Good Laboratory Practice Standards, the reporting of
test data should include the following:
(1) The name of the test, sponsor, testing laboratory, study
director, principal investigator and dates of testing.
(2) A detailed description of the test chemical including
its source, lot number, composition (identity and con-centration
or major ingredients and major impurities), known physical and
chemical properties and any carriers or other additives used and
their concentrations.
(3) The source of the dilution water, its chemical
13
-------
EG-1
August, 1982
characteristics (e.g., conductivity, hardness, pH, etc.) and a
description of any pretreatment.
(4) Detailed information about the daphnids used as brood
stock, including the scientific name and method of verification,
age, source, treatments, feeding history, acclimation procedures
and culture method. The age (in hours) of the daphnids used in
the test is also reported.
(5) A description of the test chambers, the volume of
solution in the chambers, the way the test was begun (e.g.,
conditioning, test chemical additions), the number of test
organisms per test chamber, the number of replicates per
treatment, the lighting, the method of test chemical introduction
or the test substance delivery system and the flow rate (in flow-
through test) expressed as volume additions per 24 hours.
(6) The concentration of the test chemical in each test
chamber at times designated for static and flow-through tests.
(7) The number and percentage of organisms that were
immobilized or showed any adverse effects in each test chamber at
each observation period.
(8) Utilizing the average measured test chemical
concentration, concentration-response curves should be fitted to
immobilization data at 24 and 48 hours. A statistical test of
goodness-of-fit should be performed and the results reported.
14
-------
EG-1
August, 1932
(9) The 24- and 48- hour EC50 values and their respective 95
percent confidence limits using the mean measured test chemical
concentration and the methods used to calculate botn the EC50
values and their ;conf idence limits.
(10) All chemical analyses of water quality and test
chemical concentrations, including methods, method validations
and reagent blanks.
(11) The data records of the culture, acclimation and test
temperatures.
(12) Any deviation from this test guideline and anything
unusual about the test, e.g., diluter failure, temperature
fluctuations, etc..
15
-------
EG-2
August, 1982
DAPHNID CHRONIC TOXICITY TEST'
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
• U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances BG-2
Guideline for Testing of Chemicals August, 1982
DAPHNID CHRONIC TOXICITY TEST
(a) Purpose. This guideline is intended for use in
developing data on the chronic toxicity of chemical substances
and mixtures ("chemicals") subject to environmental effects test
regulations under the Toxic Substances Control Act (TSCA) (P.L.
94-469, 90 Stat. 2003, 15 U.S.C. 2601 et seq. ) . This guideline
prescribes a chronic toxicity test in which daphnids are exposed
to a chemical in a renewal or a flow-through system. The United
States Environmental Protection Agency will use data from this
test in assessing the hazard a chemical may present to the
aquatic environment.
(b) Def i nit ions . The definitions in Section 3 of the Toxic
Substances Control Act (TSCA), and the definitions in Part 792
Good Laboratory Practice Standards apoly to this test
guideline. In addition, the following definitions apply to this
guideline:
(1) "Brood stock" means the animals which are cultured to
produce test organisms through reproduction.
(2) "Chronic toxicity test" means a method used to determine
the concentration of a substance in water that produces an
adverse effect on a test organisms over an extended period of
time. In this test guideline, mortality and reproduction (and
optionally, growth) are the criteria of toxicity.
-------
EG-2
August, 1982
(3) "EC50" means that experimentally derived concentration
of test substance in dilution water that is calculated to affect
50 percent of a test population during continuous exposure over a
specified period of time. In this guideline, the effect measured
is immobilization.
(4) "Ephippium" means a resting egg which develops under the
carapace in response to stress conditions in daphnids.
(5) "Flow-through" means a continuous or intermittent
pas's age !of tels't solution or dilution water through a test chamber
or culture tank with no recycling.
(6) "Immobilization" means the lack of movement by daphnids
except for minor activity of the appendages.
(7) "Loading" means the ratio of daphnid biomass (grams, wet
weight) to the volume (liters) of test solution in a test chamber
at a point in time or passing through the test chamber during a
specific interval.
(8) "MATC (Maximum Acceptable Toxicant Concentration)" means
the maximum concentration at which a chemical can be present and
not be toxic to the test organism.
(9) "Renewal system" means the technique in which test
organisms are periodically transferred to fresh test solution of
the same composition.
(c) Test procedare--( 1) Summary of the test. (i) Test
-------
EG-2
August, .1982
chambers are filled with appropriate volumes of dilution water.
In the flow-through test the flow of dilution water through each
chamber is then adjusted to the rate desired. The test substance
is introduced into each test chamber. The addition of test
substance in the flow-through system is done at a rate which is
sufficient to establish and maintain the desired concentration of
test substance in the test chamber.
(ii) The test is started within 30 minutes after the test
substance! has been added and uniformly distributed in the test
chambers in the renewal test or after the concentration of test
substance in each test chamber of the flow-through test system
reaches the prescribed level and remains stable. At the
initiation of the test, daphnids which have been cultured or
acclimated in accordance with the test design, are T ^ndomly
placed into the test chambers. Daphnids in the test chambers are
observed periodically during the test, immobile adults and
offspring produced are counted and removed, and the findings are
recorded. Dissolved oxygen concentration, pH, temperature, the
•concentration of test substance, and other water quality
parameters are measured at specified intervals in selected test
chambers. Data are collected during the test to determine any
significant differences (P < 0.05) in immobilization and
reproduction as compared to the control.
-------
EG-2
August, 1982
(2) [Reserved]
(3) Range-find ing test. (i) A range-finding test should be
conducted to establish test solution concentrations for the
definitive test.
(ii) The daphnids should be exposed to a series of widely
spaced concentrations of the test substance (e.g., 1, 10, 100
mg/1), usually under static conditions.
(iii) A minimum of five daphnids should be exposed to each
concentration of test substance fo.r a period of time which allows
estimation of appropriate chronic test concentrations. No
replicates are required and nominal concentrations of the
chemical are acceptable.
(4) Definitive test. (i) The purpose of the definitive
test is to determine concentration-response curves, EC50 values
and effects of a chemical on immobilization and reproduction
during chronic exposure.
(ii) A minumum of 20 daphnids per concentration should be
exposed to five or more concentrations of the chemical .chosen in
a geometric series in which the ratio is between 1.5 and 2.0
(e.g., 2, 4, 8, 16, 32, 64 mg/1). An equ.al number of daphnids
should be placed in two or more replicates. The concentration
ranges should be selected to determine the concentration-response
curves, EC50 values and MATC. Solutions should be analyzed for
chemical concentration
-------
EG-2
August, 1982
prior to use and at designated times during the test.
(iii) Every test should include controls consisting of the
same dilution water, conditions, procedures and daphnids from the
same population (culture container), except that none of the
chemical is added.
(iv) The test duration is 21 days. The test is unacceptable
if:
(A) more than 20 percent of the control organisms appear to
be immobilized, stressed or diseased during the tes, t;
(B) each control daphnid living the full 21 days produces an
average of less than 60 young;
(C) any ephippia are produced by control animals.
(v) The number of immobilized daphnids in each chamber
should be recorded on days 7, 14, 21 of the test. After
offspring are produced, they should be removed from the test
chambers every two or three days. Counts of the cumulative
number of offspring per adult (number of young divided by the
number of adults in each chamber) and the cumulative number of
immobilized offspring per adult should be recorded on days 14,
and 21 of the test. Concentration-response curves, EC50 values
and associated 95 percent confidence limits for adult
immobilization should be determined for days 7, 14 and 21. A
MATC should be determined for the most sensitive test criteria
-------
EG-2
August, 1982
measured (number of adult animals immobilized, number of young
per adult and number of immobilized young per adult).
(vi) In addition to immobility, any abnormal behavior or
appearance should also be reported .
(vii) Distribution of daphnids among the test chambers
should be randomized. In addition, test chambers within the
testing area should be positioned in a random manner or in a way
in which appropriate statistical analyses can be used to
determine the1 variation due to placement.
(5) [Reserved]
(6) Analytical measurements--( i ) Tes t chemical . Deionized
water should be used in making stock solutions of the test
substance. Standard analytical methods should be used whenever
available in performing the analyses. The analytical method used
to measure the amount of test substance in a sample should be
validated before beginning the test by appropriate laboratory
practices. An analytical method is not acceptable if likely
degradation products of the test substance, such as hydrolysis
and oxidation products, give positive or negative interferences
which cannot be systematically identified and corrected
ma themat ic ally.
(ii) Numerical. The number of immobilized adults, total
offsring per adult and immobilized offspring per adult should be
-------
Augus t, 1982
counted during each definitive test. Appropriate statistical
analyses should provide a goodness-of-fit determination for the
adult immobilization concentration-response curves calculated on
days 7, 14 and 21. A 7-, 14- and 21-day EC50, based on adult
immobilization and corresponding 95 percent confidence intervals,
should be calculated. Appropriate statistical tests (e.g.,
analysis of variance, mean separation test) should be used to
test for significant chemical effects on chronic test criteria
(cumulative !nu!mber of immobilized adults', cumulative number of
offspring per adult and cumulative number of immobilized
offspring per adult) on days 7, 14 and 21. An MATC shoul|d be
calculated using these chronic test criteria.
(d) Test conditions — (1) Test species . (i) Selection.
(A) The cladocerans, Daphnia magna or D. pulex , are the species
to be used in this test. Either species can be utilized for
testing of a particular chemical. The species identity of the
test organisms should be verified using appropriate systematic
keys .
(B) First ins tar daphnids, _<_ 24 hours old, are to be used to
start the test.
(ii) Acquis ition. (A) Daphnids to be used in chronic
toxicity tests should be cultured at the test facility. Records
should be kept regarding the source of the initial stock and
7
-------
EG-2
August, 1982
culturing techniques. All organisms used for a particular test
should have originated from the same population (culture
container).
(B) Daphnids should not be used for a test if:
(JJ cultures contain ephippia;
(2_) adults in the cultures do not produce young before day
12;
(_3_) more than 20 percent of the culture stock die in the two
Jdays preceding the test;
(_4_) adults in the culture do not produce an average of, at
least three young per adult per day over the seven day period
prior to the test;
(_5_) daphnids have been used in any portion of a previous test
either in a treatment or in a control.
(iii) Feeding . (A) During the test the daphnids should be
fed the same diet and with the same frequency as that used for
cu_lturing and acclimation. All treatments and control(s) should
receive, as near as reasonably possible, the same ration of food
on a oer-animal basis.
(B) The food concentration depends on the type used. Food
concentrations should be sufficient to support normal growth and
development and to allow for asexual (parthenogenic)
reproduction. For automatic feeding devices, a suggested rate is
-------
EG-2
August, 1982
5-7 mg food (either solids or algal cells, dry weight) per liter
dilution water or test solution. For manual once-a-day feeding,
a suggested rate is 15 mg food (dry weight) per liter dilution
water or test solution.
( iv) Load ing. The number of test organisms placed in a test
chamber s lould not affect test results. Loading should not
exceed forty daphnids per liter in the renewal system. In the
flow-through test, loading limits will vary depending on the flow
rate o]f tine > dilution ;water. loading should not cause (:hd
dissolved oxygen concentration to fall below the recommended
level.
(v) Care and handling of test organisms. (A) Daphnids
should be cultured in dilution water under similar environmental
conditions to those used in the test. A variety of foods have
been demonstrated to be adequate for daphnid culture. They
include algae, yeasts and a variety of mixtures.
(B) Organisms should be handled as little as possible. When
handling is necessary it should be done as gently, carefully and
quickly as possible. During culturing and acclimation, daphnids
should be observed carefully for ephippia and other signs of
stress, physical damage and mortality. Dead and abnormal
individuals should be discarded. Organisms that touch dry
surfaces or are dropped or injured during handling should be
-------
EG-2
August, 1982
discarded.
(C) Smooth glass tubes (I.D. greater than 5ram) equipped with
a rubber bulb can be used for transferring daphnids with minimal
culture media carry-over.
(D) Care should be exercised to introduce the daphnids
below the surface of any solution so as not to trap air under the
carapace.
(vi) Acclimation. (A) Daphnids should be maintained in 100
percent dilution' water at the test temperatjure for at least 48
hours prior to the start of the test. This is easily
accomplished by culturing them in the dilution water at the test
temperature. Daphnids should be fed the same food during the
test as is used for culturing and acclimation.
(3) During culturing and acclimation to the dilution water,
daphnids should be maintained in facilities with background
colors and light intensities similar to those of the testing
area.
(2) Facilities — (i) General . (A) Facilities needed to
perform this test include:
(_1_) containers for culturing and acclimating daphnids;
(_2_) a mechanism for controlling and maintaining the water
temperature during the culturing, acclimation and test periods;
(_3_) apparatus for straining particulate matter, removing gas
10
-------
EG-2
Augus t, 1932
babbles, or aerating the water when water supplies contain
particulate matter, gas bubbles, or insufficient dissolved
oxygen, respectively;
(_4_) an apparatus for providing a 16-hour light and 8-hour
dark photoperiod with a 5- to 30-minute transition period;
(_5_) an apparatus to .ntroduce food if continuous or
intermittent feeding is used;
(_o_) in addition, the flow-through test should contain
appropriate test chambers in which to expose daphnids to the test
substance and an appropriate test substance delivery system.
(B) Facilities should be well ventilated and free of fumes
and other disturbances that may affect the test organisms.
(ii) Tea t chambers . (A) Materials and equip.'lent that
contact test solutions should be chosen to minimize sorption of
test chemicals from the dilution water and should not contain
substances that can be leached into aqueous solution in
quantities that can affect test results.
(3) For renewal tests, daphnids can be conveniently exposed
to the test solution in 250 ml beakers or other suitable
containers .
(C) For flow-through tests daphnids can be exposed in glass
or stainless steel containers with stainless steel or nylon
screen bottoms. Such containers should be suspended in trie test
11
-------
EG-2
August, 1982
chamber in such a manner to ensure that the test solution flows
regularly into and out of the container and that the daphnids are
always submerged in at least 5 centimeters of test solution.
Test chambers can be constructed using 250 ml beakers or other
suitable containers equipped with screened overflow holes,
standpipes or V-shaped notches.
(D) Test chambers should be loosely covered to reduce the
loss of test solution or dilution water due to evaporation and to
minimize the entry of dijast or lother pa'rtlculates into the
solutions.
(iii) Test substance delivery system. (A) In the flow-
through test, proportional diluters , metering pump systems or
other suitable -systems should be used to deliver the test
substance to the test chambers.
(3) The test substance delivery system used should be
calibrated before and after each test. Calibration includes
determining the flow rate through each chamber and the
concentration of the test substance in each chamber. The general
operation of the test substance delivery system should be checked
twice daily during a test. The 24-hour flow rate through a test
chamber should be equal to at least five times the volume of the
test chamber. During a test, the flow rates should not vary more
than 10 percent from any one test chamber to another or from one
12
-------
EG-2
August, 1982
time to any other. For the renewal test, test substance dilution
water should be completely replaced at least once every three
days .
(iv) Dilution water. (A) Surface or ground water,
reconstituted water, or dechlorinated ta, water are acceptable as
dilution water if daphnids will survive in it for the duration of
the culturing, acclimation, and testing periods without showing
signs of stress. The quality of the dilution water should be
constant and should meet the following specifications:
Substance Maximum Concentration
Particulate matter 20 mg/1
Total organic carbon or 2 mg/1
chemical oxygen demand 5 mg/1
Un-ionized ammonia 20 ug/1
Residual chlorine <3 ug/1
Total organophosphorus pesticides 50 ng/1
Total organochlorine pesticides
plus polychlorinated biphenyls (PCBs) 50 ng/1
or organic chlorine 25 ng/1
13
-------
EG-2
August, 1982
(B) The water quality characteristics listed above should be
measured at least twice a year or when it is suspected that these
characteristics may have changed significantly. If dechlorinated
tap water is used, daily chlorine analysis should be performed.
(C) If the diluent water is from a ground or surface water
source, conductivity and total organic carbon (TOG) or chemical
oxygen demand (COD) should be measured. Reconstituted water can
be made by adding specific amounts of reagent-grade chemicals to
deio;nized or distilled water. Glass dist illed ' or1 carbon filtered
deionized water with a conductivity of less than 1 microohm/cm is
acceptable as the diluent for making reconstituted water.
(D) If the test substance is not soluble in water an
appropriate carrier should be used.
(v) Cleaning of test system. All test equipment and test
chambers should be cleaned before each test following standard
laboratory procedures. Cleaning of test chambers may be
necessary during the testing period.
(3) Test parameters. (i) Environmental conditions of the
water contained in test chambers should be maintained as
specified below:
(A) Temperature of 20 ± 1°C.
(B) Dissoved oxygen concentration between 60 and 105 percent
saturation. Aeration, if needed to achieve this level, should be
14
-------
EG-2
August, 1932
done before the addition of the test substance. All treatment
and control chambers should be given the same aeration treatment.
(C) Photoperiod of 16-hours light and 8-hours darkness, with
a 15-30 minute transition period.
(ii) Additional measurements include:
(A) The concentration of dissolved test substance (that
which passes through a 0.45 micron filter) in the chambers should
be measured during the test.
(B) At a minimum,, the concentration of test substance should
be measured as follows:
(_1_) in each chamber before the test;
_(J2) in each chamber on days 7, 14 and 21 of the test;
(3_) in at least one appropriate chamber whenever a
malfunction is detected in any part of the test substance
delivery sys .em. Among replicate test chambers of a treatment
concentration, the measured concentration of the test substance
should not vary more than 20 percent.
(C) The dissolved oxygen concentration, temperature and pH
should be measured at the beginning of the test and on days 7, 14
and 21 in each chamber.
(e) Reporting . The sponsor should submit to the US EPA all
data developed by the test that are suggestive or predictive of
chronic toxicity and all associated toxicologic manifestations.
15
-------
EG-2
August, 1982
In addition to the reporting requirements prescribed in the Part
792--Good Laboratory Practice Standards the reporting of test
data should include the following:
(1) The name of the test, sponsor, testing laboratory, study
director, principal investigator, and dates of testing.
(2) A detailed description of the test substance including
its source, lot number, composition (identity and concentration
of major ingredients and major impurities), known physical and
chemical properties', and any carriers' or other additives used 'and
their concentrations.
' ' The source of the dilution water, its chemical
characteristics (e.g., conductivity, hardness, pH), and a
description of any pretreatment.
(4) Detailed information about the daphnids used as brood
stock, including the scientific name and method of verification,
age, source, treatments, feeding history, acclimation procedures,
and culture methods. The age (in hours) of the daphnids used in
the test should be reported.
(5) A description of the test chambers, the volume of
solution in the chambers, the way the test was begun (e.g.
conditioning, test substance additions), the number of test
organisms per test chamber, the number of replicates per
16
-------
EG-2
August, 1992
treatment, the lighting, the renewal process and schedule for the
renewal chronic test, the test substance delivery system and flow
rate expressed as volume additions per 24 hours for the flow-
through chronic test, and the method-of feeding (manual or
continuous) and type of food.
(6) The concentration of the test substance in test chambers
at times designated for renewal and flow-through tests.
(7) The number and percentage of organisms that show any
adverse effect in each test chamber ^t each observation period.
(8) The cumulative adult and offspring immobilization values
and the progeny produced at designated observation times, the
time (days) to first brood and the number of offspring per adult
in the control replicates and in each treatment replicate.
(9) All chemical analyses of water quality and test
substance concentrations, i eluding methods, method validations
and reagent blanks.
(10) The data records of the culture, acclimation, and test
temperatures.
(11) Any deviation from this test guideline, and anything
unusual about the test, (e.g., dilution failure, temperature
fluctuations ) .
(12) The MATC to be reported is calculated as the geometric
mean between the lowest measured test substance concentration
17
-------
EG-2
August, 1982
that had a significant (P<0.05) effect and the highest measured
test substance concentration that had no significant (P>0.05)
effect on day 7, 14 or 21 of the test. The most sensitive of the
test criteria (number of adult animals immobilized, the number of
young per female and the number of immobilized young per female)
is used to calculate the MATC. The criterion selected for MATC
computation is the one which exhibits an effect (a statistically
significant difference between treatment and control groups;
P<0.05) at the lowest test substance concentration for the
shortest period of exposure. Appropriate statistical tests
(analysis of variance, mean separation test) should be used to
test for significant test substance effects. The statistical
tests employed and the results of these tests should be reported.
(13) Concentration-response curves utilizing the average
measured test substance concentration should be fitted to
cumulative adult immobilization data at 7, 14, and 21 days. A
statistical test of goodness-of-fit should be performed and the
results reported.
(14) An EC50 value based on adult immobilization with
corresponding 95 percent confidence limits when sufficient data
are present for days 7, 14, and 21. These calculations should be
made using the average measured concentration of the test
subs tance.
18
-------
ES-1
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
DAPHNID ACUTE AND CHRONIC TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
TABLE OF CONTENTS
Subject Page
I. Purpose 1
II. Scientific Aspects 1
Test Procedures 1
General 1
Range-Finding Test 7
Definitive Test 7
Test Conditions 8
Test Species 8
Selection 8
Sources 11
Maintenance of Test Species 11
Handling and Acclimation 11
Feeding 15
Facilities 13
Construction Materials 13
Test Substance Delivery Syst.en l.o
Cleaning of Test System 20
Dilution Water 21
Lo ad i ng 12
Controls 23
Carriers 25
Randomization 26
Environmental Conditions 26
Dissolved Oxygen 26
Light 31
Temperature 31
Reporting 34
III. Economic Aspects 35
IV. References 37
-------
Office of Toxic Substances ES-1
August, 1982
Technical Support Document for Daphnid Acute and Chronic Tests
I. Purpose
The purpose of this document is to provide the
scientific background and rationale used in the .levelopment
of Test Guideline EG-1 wh ich uses Daphnia species to
evaluate the acute and chronic toxicity of c'neraical
substances. The Document provides an account of the
scientific evidence and an explanation of the logic used in
the selection of the test methodology, procedures and
conditions prescribed in the Test Guideline. Technical
issues and practical considerations relevant to the Test
Guideline are discussed. In addition, estimates of the cost
of conducting the test are. provided .
II. Scientific Aspects
A. Test Procedures
1. General. Relatively few industrial chemicals,
compared to the vast number produced, have been previously
tested by standard aquatic bioassay methods. As a result,
many cannot be classified as to their toxicological
properties. The acute and chronic toxicity tests will
provide some of the information needed to evaluate the
hazard posed to aquatic organisms from a chemical
substance. Although assessment of effects on higher levels
of biological organization is desirable, it is necessary to
begin with effects on individuals and small test
populations. Acute effects can be considered as those which
cause rapid damage to the organism by the fastest acting
mechanism of poisoning which can prove detrimental unless
the animal escapes the toxic environment at an early time.
The static acute toxicity test provides the most easily
-------
ES-1
August, 1982
obtainable measure of toxicity. This test provides
information on the immediate effects of short term exposure
to potential toxicants under carefully controlled
conditions. A single exposure to a potential toxicant in a
confined system may represent the worst possible test case
for weak or non-motile organisms wh ich may be
incapable of avoiding the chemical under natural conditions
(Curtis et al. 1979) .
The acute flow-through toxicity test is especially
applicable for those test substances which display high
oxygen demand, are highly volatile, are unstable in aqueous
solution, are biodegradable or are readily assimilated by
the test organism. By constantly maintaining test solution
concentrations and environmental factors such as dissolved
oxygen and pH within narrow limits, a more representative
indication of the potential toxic effects of the test
substance/is obtained as opposed to static test conditions
where degradation products and metabolic wastes may affect
the test organism as well as the test substance.
The proposed Daphnia chronic toxicity test guidelines
are designed to assess the effects of tes i; substances on the
survival and reproduction of Daphnia as a representative
macroinverteb cate. The duration of the test permits the
organism to be exposed.,to a chemical from shortly after
birth until well into adulthood. The organisms are exposed-
long enough to allow the adults to produce several broods of
second generation progeny. Initiating exposure shortly
after birth allows an assessment of the possible effects of
the test substance on such metabolic processes as
reproductive system, maturation, fecundity, and growth.
-------
ES-1
August, 1982
Exposure of the test organisms for more than "a complete
generation cycle (approximately ten days at the test
temperature, Beisinger et al. 1974) allows the testing
facility to assess and predict the potential effect OL the
test compound on a representative Daphnia population. The
number of -first generation test organisms unaffected at the
termination of the test provide an indication of
survivorship. Mathematica] treatment of the fecundity and
survival data provides an index of the inherent ability of a
population of test organisms to increase under similar
environmental conditions (Boughey,1973). Althcm3h the
1
chronic assays require more time and expense, they pro-
vide much more accurate predictive information concerning
the effects of the test substance (Liptak 1974). The
potential disadvantages of the static assays, which exclude
feeding of test organisms, include physico-chemical
modifications of the test substance and stress on the
organisms from their own metabolic wastes, dissolved oxygen
depletion, and starvation. These disadvantages are greatly
reduced by renewing the test solution-dilution water medium
at various rates while providing sufficient food.
The replacement of the medium can be continuous or at
fairly frequent intervals. Flow-through tests ace more
expensive because they require more dilution water, more
test substance, and more expensive apparatus. The setup,
breakdown, and maintenance time for flow-through systems is
greater and requires more experienced personnel.
Intermediate results can be obtained between flow-
through and static bioassays by renewing the medium at
periodic intervals. This system may be satisfactory if
-------
ES-1
August, 1982
water is limited, if labor costs are not too high and if
bioassays are conducted relatively infrequently.
Laboratories that perform bioassays on a regular basis may
find flow-through tests to be cheaper and more effective
than renewal tests (Liptak 1974).
The difference in;results between renewal and flow-
through chronic assays may be significant, depending in part
on the renewal frequency and the physical characteristics of
the test substance. Nebeker and Puglisi (1974) investigated
the effects of several PCB's on Daphnia magna using weekly
renewal and flow-through techniques. The LC50 values for
Arochlor 1254 for three-week assays were 31 ppb in the
renewal test and 1.3 ppb in the flow-through test. The
authors attribute the thirty-fold difference in toxicity to
cumulative toxicity, the volatility of the PCB's and
sorption to bacteria, algal waste materials, test container
and food surfaces.
The frequency of test chamber renewal may also affect
test results. Banner and Halcrow (1977) observed a
significant difference in ephippia production and mortality
with different dilution water replacement frequencies.
Daphn ia magna maintained under the same photoperiod, feeding
rate, temperature, and density, but with renewal on an
alternate day basis, exhibited 1.2 percent ephippial
production and 3.8 percent mortality. Daphnids maintained
under a weekly replacement scheme exhibited 8.7 percent
ephippial production and 35 percent mortality at the end of
the test period. The renewal system test guideline
recommends alternate day renewal in order to maintain the
best practicable conditions for the test organisms.
-------
ES-1
August, 1982
The choice of an acute test duration of 48-hours was
chosen for the acute tests because of certain biological
characteristics of daphnids. Adema (1978) compared the
mortality of one-day old daphnids which were fed an algal
diet to those which were not fed, under identical
conditions. Adema demonstrated that 43 hours is the maximum
time which ins tars can be" deprived of food without suffering
increased mortality. The 48 hour test duration also allows
sufficient time for the instars to molt at least once. Lee
and Buikema (1979) report there is a molt-enhanced
sensitivity to certain compounds.
The events in the daphnid life cycle are dependent on a
number of factors including temperature, food concentration
and type. Four distinct periods may be recognized in the
daphnid life cycle (Pennak 1973): egg, juvenile, adolescent,
and adult. There are few juvenile instars. Generally, D.
pulex has 3-4 instars, and _D. magna 3-5. The adolescent
period is a single ins tar prior to the first adult ins tar
during which the eggs reach full development in the ovary.
The number of adult instars varies. D. pulex generally has
13 - 25 instars, while D. magna from 6 - 22. Molting occurs
at an approximate rate of once a day for the juveniles and
at a rate of every two days for the adults under favorable
condi tions .
Richman (1958) observed that D. pulex released the
first brood at age eight days at 20 °C. Anderson and Jenkins
(1942) noted that D. magna released its first brood at the
age of seven days at 25 °C. Anderson (1932) recorded sexual
maturity in _D. magna from six to ten days when raised in a
temperature range of 18 to 23°C.
-------
ES-1
August, 1982
The test duration of 21 days, starting with first
ins tar dap hn ids, will expose the daphnids to the test
substance for approximately 11 total ins tars, including
approximately seven adult ins tars (Anderson and Jenkins
1942, Richman 1958). The total number of instars in the
treatments will be a function of the effects of the test
substance on juvenile growth and survival, reproductive
maturation, and adult growth and survival.
Anderson et al. (1937), using D. pulex, and Anderson
and Jenkins (1942) using _D. magna, observed a peak in the
number of living young produced in both species at the 10th
to llth ins tar which corresponds to days 20 to 23 after
experimental initiation. Instars subsequent to the 10th and
llth iastar generally produced less living young per molt.
Richman (1958) using D. pulex did not observe a decrease in
the number of young produced until approximately days 24 and
28. The three studies indicate that the test period will
provide time for several adult ins tars 'and should encompass
or closely approach the life stages of maximal
reproduction. Winner and Farrell (1976) assessed the
reproductive sensitivity of D. pulex to copper and concluded
that the experiment could have terminated anytime after the
third brood (approximately day 13) without altering
conclusions as to the effect of cooper on reproduction.
It is important that treatment and controls be
monitored closely, especially after approximately day 5 to
determine the release time of the first brood, frequency of
subsequent broods and the number of living young produced
per brood.
-------
ES-1
August, 1932
1 . Range-Finding .Test
The concentration range for the definitive tests is
normally chosen based on the results of a range-finding
test. Range-finding tests with daphnids are usually short-
term bioassays which use fewer organisms per test substance
concentration than required- for the definitive test. In all
cases, the range-finding test is conducted to reduce the
expense involved without having to repeat a definitive test
due to inappropriate test substance concentrations.
2. Definitive test
The results of the definitive test will be used to
establish any statist ically significant (PL 0.05)
differences between, the treatments and the control(s)
pertaining to survivorship and reproductive capabilities.
These parameters should provide an indication of the
potential effects of the test substance on representative
populations of the test organism.
Certain test parameters as promulgated in the Test
Guidelines are more or less inflexible or have a narrow
range of acceptable values. Slight variations of these
parameters have been demonstrated to have a significant
effrect on the test organisms. Small variations of
temperature, for example, can produce changes in the
metabolic rate of daphnids (Kaestner 1970, Bunting 1974), as
well as on their response to toxicants (Bunting and
Robertson, 1975). Other test parameters establish maximum-
minimum criteria or a broad range of acceptable values.
ASTM (1980) concluded that, at present, all conditions
do not warrant very precise control without invalidating
test results. However, some experimental conditions should
-------
ES-1
August, 1982
be more narrowly controlled. Therefore, recommendations for
test procedures are made which, if followed, will yield
results that are scientifically sound without causing
unnecessary testing costs.
B. Test conditions
1. Test species
a. Selection. Daphnids are of ubiquitous
occurrence and are an important link in many aquatic food
chains (Kring and O'Brien 1976, Gulati 1978, Makarewicz and
Likens 1979). Species of the genus Daphnia are major
components of the freshwater zooplankton throughout the
world (Hebert 1978). Because of their predominantly
herbivorous nature, the daphnids represent an intermediate
trophic linkage between the primary producers and the
carnivores and predators of higher trophic levels (Gulati
1978) .
Daphnia are found in a variety of aquatic environments,
except for rapid streams, brooks and grossly polluted 'water
(Pennak 1978). Two of the most common species found in
ponds, lakes and permanent pools are D. magna and D.
pulex . _D. magna, the largest of the daphnids, is generally
found in the northern and western parts of North America
while D. pulex is distributed over the entire North American
continent (Pennak 1978, Suikema et al. 1976). Both D. magna
and D. pulex have been used in toxicity tests. D. magna has
been used more extensively due to its large size, ease of
culture, short generation time, and sensitivity to toxic
compounds (Dewey and Parker 1964, Frear and Boyd 1967,
Builkema et al . 1976). Buikema, _e_t _al_. (1976) maintain that
the cosmopolitan distribution of D. pulex, as well as its
-------
ES-1
August, 1982
adaptability to a wider range of habitats, warrants its use
as a test organism.
Daphnia have been used to assess the toxicological
properties of a number of different types of compounds for
both survival and the possible effects on reproduction:
insecticides (Adema 1978); herbicides (3chober and Lampert
1977); organic compounds (Adema 1978; Canton et al 1975);
metals (Winner and Farrell 1976; Bertram and Hart 1979,
Biesinger and Christens en 1972); PCB's (Nebeker and Puglisi
1974); nitrilotriacetate (NTA) (Biesinger et al. 1974);
polyethyleneimine (PEI) (Stroganov et al. 1977).
Daphnids have been shown to be very sensitive organisms
for assessing the possible deleterious effects of chemicals
on other aquatic forms. D. magna was more susceptible to
several xenobiotics tested than other invertebrates and fish
tested (Canton eet al. 1975, Leeuwangh 1973). Kenaga (1978)
used 75 insecticides and herbicides to assess the
comparative toxicology of several different species
including birds, rats, fish, shrimp, honeybees, and daphnids
as indicators in toxicity screening tests. Comparisons
between the test organisms and test substances indicated
that the daphnids were extremely sensitive to a number of
compounds. Comparisons of the test species, within a
chemical class, indicated that the invertebrates (daphnids
and shrimp) were the most sensitive test form for the entire
spectrum of chemicals tested (Kenaga 1978).
In a subsequent review of the toxicity of more than
30,000 test compounds, Kenaga and Moqlenar (1979)
demonstrated the enhanced sensitivity of D. magna compared
to four species of fish, five species of aquatic vascular
-------
ES-1
August, 1982
plants, and the alga Chlorella.
The sensitivity of D. magna was compared with that of
D. pulex in two sets of experiments (Winner and Farrell
T9~76, Canton and Adema 1978). The conclusions indicated
that the two species do not significantly differ in their
sensitivities to the compounds tested.
Winner and Farrell (1976) studied the acute and chronic
toxicity of copper (as copper sulfate, pentahydrate) to four
species of daphn.ids. The daphnids used were divided into
two groups, the larger species, D. magna and D. pulex, and
the smaller species, D. parvula and D. ambigua. It was
observed that the acute toxitiity of D. magna and d. pulex
for copper did not differ significantly (72-hour LC5Q: D.
magna, 86.5 ug/1; D. pulex, 86.0 ug/1; P > 0.05). However,
two smaller species did display a slightly enhanced
sensitivity (72-hour LC50; p. parvula, 72.0 ug/1; D^_
ambigua, 67.7 ug/1. Comparison of chronic exposure data
demonstrated consistent similarities in sensitivity with
regard to survival. Survivorship curves for the two lowest
test conditions (20 ug/1 and 40 ug/1) were never
significantly different (P > 0.05) from those of the
controls for the four species of Daphnia. Enhanced
mortality was observed at the next highest concentration (60
ug/1) for all test species. The authors attribute this
increase in toxicity between 40 ug/1 and 60 ug/1 to a
saturation of the complexing capacity of the natural pond
water used as dilution water.
Canton and Adema (1978) investigated the sensitivity of
D. magna, D. pulex and D. cucullata to 15 different
compounds (13 organic, 2 inorganic). Comparison of acute
10
-------
ES-1
August, 1982
toxicity test data pooled from two different laboratories
using the three test species indicated that with only two
exceptions (aniline and pentachlorophenol), there were no
significant differences (P > 0.05) between the species.
Because of its size (manageability), as well as its
comparable sensitivity, _D_. magna was designated the daphnid
of choice for Dutch laboratory studies.
As data are generated using both species under
carefully defined conditins, the intercomparability of the
two test species for chronic testing can be assessed in
future work.
b. Sources'. Daphriids as" a group display
taxonomically troublesome variations in details of setation
and in carapace, head and postabdomen morpiu logy.
Test species may be obtained from field collections,
supply houses or established cultures and should be
identified and documented. Verification of .either species
used should be performed using the systematic keys of Brooks
(1959) or Pennak (1978) for D. magna, and Brandlova et al.
(1974) for D. pulex.
2. Maintenance of Test Species
a. Handling and Acclimation. All organisms
used in a test should be of the same species and from the
same source to reduce variability of the test results.
Laboratory culture of daphnids from a single innoculum
provides test organisms of similar history. Reproduction
can be restricted to the parthenogenetic production of only
females when suitable culture conditions are maintained.
This insures a supply of experimental animals with genetic
variability limited to the heterozygos ity of the parent
11
-------
ES-1
August, 1982
(APHA 1975, Kaestner 1970, Pennak 1978). Under proper
conditions, such as sufficient food, favorable temperature
and uncrowded conditions, D. magna cultures have been
maintained for many years (Kaestner 1970). Daphnia from
cultures in which ephippia (resting eggs) are being produced
should not be used for testing, as the production of
ephippia indicates unfavorable culture conditions and
production of males.
Several culture methods have been described, with no
one method universally accepted. Needham et al. (1959)
describes several historical methods of culture, some of
which ari still utilized. Hutchinson (1967)' reviews several
cultural, schemes using primarily algae as food.
The following'references are not meant to be
exhaustive, but to provide the testing facilities with some
recent information of various culture methods: D'Agastino
and Provasoli (1970), algae; Murphy (1970), algae; Burns,
(1969), mixed algae; Kring and O'Brien (1976), algae; Berge
(1978), algae; Canton et al. (1975), algae; Lee and Buikema
(1979), mixed algae; Winner and Farrell (1976), algae and
vitamins; Schultz and Kennedy (1976), algae and yeast; Dewey
and Parker (1964), algae and yeast; Buikema et al. (1976),
trout chow pellets; Beisinger and Christensen (1972), trout
fry food granules and grass; Fear and Boyd (1967), manure-
soil; and Whitten et al. (1976), hard-boiled egg yolk.
It is left to the experience and discretion of the
testing facilities to decide which method(s) prove to be the
most reliable.
Acclimation to new environmental condition(s) is
accomplished by various biochemical and biophysical
processes. Capacitive adaptations are those which permit
12
-------
ES-1
August, 1982
relative constancy of biological activity over a normal
range of environmental parameters. To accommodate the
necessary biochemical/physiological changes required for the
test organism to adapt to new environmental conditions, the
rate of chan-ge of the factors should be such as to avoid
additional stress.
Once the desired conditions are established, the
animals should be held at these conditions for a period of
time to determine that no delayed symptoms of stress appear
which could bias test results.
The recommended rates of temperature acclimation
(l°C/dayi), plus culturilng in dilution water for
approximately 21 days, are designed to allow the animals to
make the necessary physiological adjustments prior to
exposure to the treatments.
Food type, concentration, and feeding rate should
approximate test conditions as closely as possible. This
allows the animals to adjust to test conditions and yields
to the investigator a preliminary assessment of the
effectiveness of the feeding regime to be used during the
extended test period.
A culture should not be used as a source of test
organism if (a) the individuals appear stressed or diseased;
(b) it possesses adults that do not produce young by day 12,
which would indicate delayed maturation or infertility; (c)
it has adults that do not produce an average of at least
three young per adult per day over a seven-day period, which
would indicate reproductive impairment due to genetic or
culture conditions such as crowding, inadequate diet or some
pathogen; (d) ephippia are being produced, which would
indicate stressful conditions; or (e) mortality exceeds 3
13
-------
ES-1
August, 1982
percent during the 48 hours immediately preceding the test.
These criteria are designed to prevent bias in the test
results caused by the use of reproduct ively inferior
daphnids, which could result from heredity or a stressed
parent. Stressed culture conditions causing metabolic
dysfunctions in the parent may be reflected in the tey t
organisms. Such dysfunctions as reduced yolk synthesis
could result in an inferior test daphnid.
First ins tar D. magna or D. pulex are the -initial tes t
stage. Animals 0-24 hours old are to be used. This age
class can be collected by an overnight .separation of gravid
females, and insures that all the test organisms will be
f irs t-ins tar, pre-molt.
Dewy and Parker (1964) described a separation chamber
consisting of funnels with screen openings. The instars
passed through the screen a nil were collected in receiving
jars while the adults remained in the funnel. This iaethod
resulted in the production of animals of known age with a
minimum of labor and time.
Static acute assays indicate an enhanced sensitivity of
first ins tar daphnids as compared to the later juvenile or
early adult stages (Sanders and Cope 1956).
Schultz and Kennedy (1976) and Lee and Buikema (1979),
again using static acute assays, demonstrated enhanced
sensitivity of Daphnia spp. at molting. Such mechanisms
as changes in permeability of the body surface and the
incorporation of large volumes of water were postulated to
explain the enhanced toxic effect during the molt period.
Exposure of the daphnids shortly after brood release
insures exposure to the test substance prior to the first
molt. Exposure of the test organism at an early age is
14
-------
ES-1
August, 1982
designed to assess any possible effects on the organism
which may not manifest themselves until later in the life
cycle.
Subjecting the first ins tar to the test substance
allows an evaluation of effects on such physiological
processes as ovary maturation (reflected in fertility),
fecundity, and the time from brood release to the first
post-release molt. The effects on the parent generation can
also be reflected in growth rate and survivorship.
b. Feeding . Daphnids should not be fed
during the -acute tests. The presence of food in the test-
medium may have several;effects: (1) The test subs-tance 'may
be absorbed onto the food particles and either increase or
decrease its toxic effects (Adema, 1978). (2) It may alter
the dissolved oxygen content by increasing BOD (biological
o.xygen demand), increase dissolved oxygen by photosyn the tic
activity, or reduce dissolved oxygen by respiratory
demands. (3) Feeding may alter the physiology of the
ins tars and change the uptake and metabolism of the test
substance: (4) It may introduce more variability into the
test.
Feeding is required during culture of daohnids and in
renewal and flow-through chronic tests. Food concentration
and type is extremely important because it
can affect: (1) the concentration of test substance needed
to elicit a response, (2) diurnal dissolved oxygen levels,
and (3) the physiological state of the test organism.
A large number of variables concerning feeding are
evident from the literature. These include food type(s),
feeding rates, use of supplements, frequency of renewal of
15
-------
ES-1
August, 1982
test medium and dilution water and any food included in
natural dilution waters. Recommended feeding regimes for
daphnid testing may be based on the literature compilation
in Table 1. Final feeding conditions are left to the
discretion of the individual laboratory based on experience
and the satisfaction of the test guidelines pertaining to
control mortality and minimum number of control orogeny.
These two- criteria are design.ed to insure, in part, proper
feeding techniques during test conditions.
The food used should be sufficient to maintain the test
organism in a nutritional state which will support normal
metabolic activity and reproductive capabilities! • This is
advisable in order to avoid introducing starvation as a
variable into test results.
Adema (1978) states that the feeding of 4.0 x 107 to
6.0 x 107 Chlorella pyrenoidosa cells per liter per adult D.
magna per day is the optimum amount of food for reproduction
and was the same concentration used in cultures.
Overfeeding may compromise test results through (1)
<-3.
-------
S3—1
August, 1932
Table 1: Several feeding regimes used in chronic toxicity
us ing Daphnia.
Calculated Feeding Rate
Food Type ( amount/I/Maphnid/day) Reference
Chlorel la pyr enoidos a
Chiorel La pyrenoidos a
Yeast
Chlorella pyrenoidos a
Yeast Extract
Grass and Trout
Pellets
Yeast
Yeast
Seenedesmus obliquus
Yeast and
Scenedesrnus acutus
Chlaymdomonas
reinhardti
2.5 - 6.0 x 107
eel Is
1.3 x 107 cells
6.5 x 109 cells
4 x 107 cells
7 Tig solids
2 mg Yeast
1.2 x 105 cells
4 x 105 cells
4 mg
7.5 mg
Adema (1978)
Bertram and
Hart (1979)
Be rye (1978)
Bies inger and
Chris tens en
(1972)
Bunner and
Hal crow
(1977)
Dewey and
Parker
( 1964)
Schober and
Lampert
(1977)
Winner and
Parrel1
(1976)
pulex and Chlamydomonas spp. , observed a three-fold change
in the cumulative number of young produced with
Chlamydomonas concentrations ranging from 25 x 103 cells per
ml (30 young) to 100 x 103 cells per ml (92 young).
Bunner and Halcrow (1977), using D. magna fed on yeast
17
-------
E3-1
August, 1982
and maintained at the same photoperiod and density,
observed an increase in ephipoia production of 18.6 percent
on a starvation diet (0.025 mg yeast per animal per day) as
opposed to 9.6 percent ephippia on a diet of 0.05 mg yeast
per animal per day.
3 . Facilities
a. Construction Materials. Construction
materials and any equipment that may contact, s tock
solutions, test solutions or any water into which the test
organisms will be placed should not contain any substances
that can be leached into the aqueous medium, Such
substances could introduce an error into the test results or
stress the test organisms by direct or indirect toxic
effects.
Materials and equipment should be chosen to minimize or
eliminate the occurrance of sorption and leaching, which may
reduce the effective concentration of the test substances
and introduce a potential error in test results, or which
may introduce contaminants into the system.
b. Test Substance Delivery System. In flow-
through tests, the delivery of constant concentrations of
test substances is required to reduce variability in test
results. Large fluctuations in test substance concentration
will give abnormally high or low responses, depending upon
the mechanism of toxic actions. Proportional diluters with
metering pumps or continuous-flow infusion pumps have been
used extensively to maintain constant test substance
concentration. For the flow-through acute and life-cycle
test guideline, all tests should be conducted in
intermittent flows from a diluter or in continuous flow with
18
-------
E3—1
August, 1982
the test substance added by an infusion pump. The
procedures of Mount and Brungs (1967) and Hansen et al.
(1974) are recommended if the test substance can be added
without a carrier; the device described by Hansen et al.
(1974), if a carrier is necessary; or procedure of Banner et
al. (1975), if pumps are required for continuous flow.
Proportional diluters operate on a sequential filling
and emptying of water chambers. The water chambers are
calibrated to contain a measured amount of water. Separate
water chambers can be provided for test substance and
diluent v/ater. Diluent and test substance waters are mixed
and delivered to the test aquaria. 'The cyclic action of the
diluent is regulated by a solenoid valve connected to the
inflow dilutioa water. The system is subject to electrical
power failure, so an alternate emergency power source is
recommended .
The proportional diluter is probably the best system
Cor routine use; it is accurate over extended periods of
time, is nearly trouble free, and has fail-safe provisions
(Lemke et al. 1978). A small chamber to promote mixing of
test substance-containing water and dilution water may be
used between the diluter and the test aqaaria for each
concentration. If replicate chambers are used in this test,
separate delivery tubes should be run fro;a this mixing
chamber to the appropriate replicate chambers. If an
infusion pump is used, a glass baffle should be employed to
insure mixing of the test substance and dilution water.
Calibration of the test substance delivery system should be
checked carefully before and during each test. This should
include determining the flow rate through each test aquarium
and measuring the concentration of test substance in each
19
-------
ES-1
August, 1982
test aquarium. The general operation of the system should
be checked twice daily.
The use of municipal water supplies is not
recommended. Municipal waters often contain high
concentrations of potentially harmful components such as
chlorine, chloramines, copper, fluoride, lead, zinc
and iron. A carbon filtered dechlorinated water may be
acceptable if Daphnia can be cultured in it. Caution should
be exercised since municipal water may vary considerably in
quality and chemical characteristics associated with
seasonal changes (e.g., extensive chlorination following
heavy storm activity), different sources and modifications
or repairs to the distribution system.
c. Cleaning of Test System. Standard
laboratory practices (e.g., USEPA 1974) are recommended to
remove dust, dirt, other debris, and residues from test
facilities. At the end of a test, test systems should be
washed in•preparation for the next test. This will prevent
chemical residues and organic matter from becoming embedded
or absorbed into the equipment. I-t is also recommended tht
any silicon cement which has been exposed to a test
substance is replaced prior to future tests to avoid
contamination due to sorption properties.
Rinsing and priming the system with dilution water
before use (conditioning) allows equilibrium to be reached
between the chemicals in the water and the materials of the
test system. The test system may sorb or react with
substances in the dilution water. Allowing this equilibrium
to take place before use lessens the chances of water
chemistry changes during a test.
20
-------
F]S-1
August, 1932
d. Dilution Water. An adequate supply of
dilution water is required in which the daphnids will
survive, grow, and reproduce. This is necessary to insure
that the test organ is as will not be stressed or adversely
affected by the dilution water.
Dilution water quality should be maintained within a
certain range to allow for standardization and comparability
of test data. Changes outside the acceptable range
recommended may cause undue stress to the test organisms,
thus biasing test results. Variations in water chemistry
from the 'recommended range may also interfere chemically
with the test compound, either enhancing or diminishing the
toxicological properties. Criteria have been established
for several heavy metals and pesticides which have been
known to produce adverse effects on aquatic organisms (ASTM
1930).
The dilution water should be vigorously aerated prior
to use for culturing and testing. The recommended
saturation value of 90 to 100 percent should provide
sufficient oxygen under most conditions for daphnid
metabolic demands, as well as any chemical oxygen demand of
the test substance.
Test chambers should not be aerated after the test
organisms are introduced to prevent entrapment of air
bubbles under the daphnids' carapace.
Natural dilution water should be obtained from an
uncontaminated well, spring, or surface water source. Wells
and springs generally provide water of fairly constant
quality. Surface water sources are more likely to be
subjected to point or non-point source loadings. Any
peculiarities in local ground and surface water due to
21
-------
ES-1
August, 1982
geologic conditions that are not listed in the
specifications of the test guideline should be investigated
in terms of their effects on the test organisms.
Natural dilution water is usually the most cost-
efficient type of water, especially for extensive testing
and flow-through systems. Reconstituted water may be
prepared using ground or surface water which, in itself,
will not maintain daphnids. Reconstituted water,has the
advantage of having well defined chemical characteristics,
due to the specific chemical components defined in its
manufacture. Reconstituted water is however, less cost-
efficient than natural dilution water for large scale
renewal or flow-through testing due to the requirement of a
distilling apparatus and labor required to measure and
•nix the necessary chemical components.
e. Load ing . The use of 10 ins tars per 200 ml
test solution is recommended for static acute tests. This
loading should insure aderjuate dissolved oxygen for the
duration of the test period. Adema (1978) recommended a
loading of 10 ml test solution per instar based on oxygen
consumption data for the instar at 20°C. Adema suggested
this loading would result in a final dissolved oxygen
concentration of 80 percent saturation dilution water. The
recommended loading provides twice the amount of test
solution suggested by Adema and should provide a margin of
safety for the dissolved oxygen of both the organism and
chemical.
A recommended loading of one daphnid per test chamber
for chronic, reproductive studies is designed to meet the
dissolved oxygen requirements of the organism and to allow
22
-------
ES-1
1932
the observation of critical stages in the life cycle such as
initial release of ins tars and the subsequent brood
frequency and size.
Survivorship studies using five parent daphnids per 200
ml test solution may present some problems with maintaining
dissolved oxygen levels above 60 percent saturation,
depending on the food type and feeding rate. Should
dissolved oxygen values be observed below 60 percent
saturation, the use of a larger volume of test solution or ri
different food type and feeding rate is recommended.
For acute and chronic flow-through assays the near
constant changing of the test solution should prove
sufficient to maintain all environmental condicions within
the criteria for a definitive test.
f. Controls. Controls are required for every
test to insure that the observed effects are due to the test
substance and not to other factors.
In acute toxicity tests, a maximum of 10 percent
immobility is permissible in dilution water control daphnids
due to inherent biological factors and possible handling-
induced stress. Higher immobilization negates the test
results and indicates the need to determine the cause of the
increased immobilization in the control. Possible sources
of increased control immobility include culturing
techniques, acclimation procedures, handling techniques, or
testing facilities or procedures.
In chronic tests, to insure that the reproductive
capabilities of the test population are not impaired, a
lower limit of at least 60 young produced per control animal
(cumulative) has been established for the test duration.
23
-------
ES-1
August, 1982
Criteria proposed in the test guideline for culture
conditions are designed to reduce possible stress conditions
which could be reflected in reproductive impairment of the
organi^as . Similar criteria for the experimental procedures
are designed .to insure that the observed effects are due to
the test compound and not to crowding or feeding. Table 2
presents a summary of observed 21 day cumulative production
of young daphnids. The range in values can be attributed to
such factors as loading, temperature, food type and
concentration, dilution water characteristics, and in some
cases, the need to extract 21 day data from the results of
experiments-of longer duration.
Table 2: Cumulative production of young _D. magna during
a period of 21 days after birth.
Average Cumulative T^s h Temperature
Production of Young °C Reference
Per Daphnid
67-122 20 Berge
(1978)
73 25 Anderson
and Jenkins
(1942)
38-43 18 Nebecker
and Puglisi
(1974)
63 20 Schober and
Lampert
(1977)
30-92 20 Richman
(1953)
83 - Canton
et al .
(1975)
24
-------
ES-1
Augus t, 1.98 2
g. Carriers. Carriers should only be used when
they are necessary to solubilize hydrophobia test compounds.
Dimethylforraamide and triethylene glycol are the carriers of
choice due to their low volatility and toxicity, as well as
their ability to dissolve many organic compounds (ASTM
1980). Sax (1979) suggests that dime thylf ormamide not come
in contact with halogenated hydrocarbons and inorganic
nitrates due to the reactivity of the compounds. Schober
and Lampert (1977) observed a significant effect of the use
of ethanol as a carrier for Atrazin (chlorinated triazine
herbicide) using D. pulex for a 28-day exposure. Although
no effect was observed with Oil percent ethanol carrier, the
use of 0.5 percent carrier with the herbicide produced
effects greatter than the sum of the individual effects.
Significant differences (P < 0.05) were observed on such
parameters as number of young per animal and mean length.
Comparison of dilution water controls and 0.5 percent
ethanol, cphtrols indicated that use of the ethanol control
resulted in approximately a 40 percent reduction in the
number of young per animal over the experimental period.
Mean length data also indicated a carrier effect.
The investigation of Schober and Lampert (1977)
demonstrates possible errors associated v/ith the use of a
carrier and reinforces the recommenda tion that a carrier be
used only when necessary. The investigation also emphasizes
the need for a carrier control and the investigation of the
effects of two different concentrations of the same carrier
on the test organism.
Krugel et al. (1978) describe an apparatus for the
continuous dissolution of poorly soluble compounds for
b io as s ays .
25
-------
ES-1
August, 1982
h. Randomization. Randomization is required to
prevent conscious or unconscious biases from being
introduced. These biases can be in environmental conditions
such as temperature and lighting, daphnid selection and
?,is tr ibution.
4 . Environmental Conditions
a. Dissolved Oxygen. Daphni_a respond to
partial anoxia by synthesizing hemoglobin (Hoar 196 6,
Lockwood 1967, Kaestner 1970). This adaptation has
significant.survival value (increased life span and
increased egg production) when compared to those organisms
thatj lack or pcbssess reduced concentrations of hemoglobin.
In mature females, considerable hemoglobin enters the eggs,
accelerating embryonic development. After egg laying, the
level of hemoglobin in the adult's blood is .approximately
two-thirds of its normal value. The concentration in the
1'iloov, increases during the time the young are developing in
the brood pouch. The cycle is then repeated with the
release of the young and production of a new batch of
parthenogenic eggs.
Considering its molecular size, incorporation into the
eggs and possible replacement rates, hemoglobin synthesis
may represent a considerable energy demand on the
organism. Such demands can be inferred from the increase
rate of feeding at partial anoxic conditions. No chronic
toxicity test data could be located comparing Daphnia
response to varying concentrations of dissolved oxygen.
Adema (1978) determined the oxygen consumption of 25
egg-bearing adult daphnids at 20°C to be about 850
ug/02/day. The ins tars released from adults (25) in 24
26
-------
ES-1
August, 1982
hours consume an additional 600 ug/02/day. Adema suggests a
minimum of 2.5 liters of oxygen-saturated medium for 25
daphnid on an alternate day renewal scheme will satisfy
daphnid oxygen demands, as well as any other oxygen sinks
such as food remnants, excreta and test compound oxygen
demands .
The solubility of oxygen in freshwater at 20°C is
approximately 9 mg/1. The test guidelines recommend that
dissolved oxygen not fall below 60 percent saturation or
approximately 5.5 mg/1 at any time during the test. Kring
and O'Brien (1976), using _D.. pulex at 22°C, observed, that
when tphe oxygen concentration dropped below 3 mg/1 the
filtering .rate of D. pulex decreased drastically. The same
authors cite unpublished data of a critical oxygen
concentration of about 3 mg/1 for D. mag n a. The recommended
minimum concentration of 60 percent saturation is well above
the critical oxygen concentrations observed.
Kring and 0 ' Brien (19'75 } observed that exposure for
less than one hour to oxygen concentrations of 1 mg/1 caused
a negligible depression in the filtering rate of JD. pulex.
Longer exposures to dissolved oxygen values less than the 3
mg/1 critical concentration resulted in depressed filtering
rates (60 percent reduction) for an eight-hour period.
After 24 hours of exposure, the filtration rate increased to
near normal values. Continued exposure to dissolved oxygen
values less than the critical concentration resulted in the
ability of the animals to resume and surpass the initial
high filtering rates, presumably because the low oxygen
concentration stimulated the production of hemoglobin. The
daphnids synthesized increasing amounts of hemoglobin
27
-------
as-i
August, 1982
associated with the time they were exposed to low dissolved
oxygen values increased. Consideration of the increase in
hemoglobin synthesis and the high filtering rates associated
with long-term exposure to low dissolved oxygen suggests
that hemoglobin synthesis is energetically inefficient.
Maintenance of test dissolved oxygen values above 60 percent
saturation should prevent any biochemical stress on the test
organism associated with hemoglobin synthesis which could
decrease the energy available for other metabolic processes.
The presence or abs'ence of dissolved molecular oxygen
in the test solution may also affect the form of the metals
and ions in the medjium. Severjal general types oE ; redox \
reactions of ionic species of metals have been demonstrated,
depending in part on pH, the presence of organic complex ing
agents, the presence of other ionic species such as the
carbonate ion and the presence of molecular oxygen (Faust
an:l Hunter 1967). Several of the reactions are reversible;
thus various equilibria can be established, depending on the
chemical composition of the test medium. Maintenance of
dissolved oxygen levels in excess of 60 percent saturation
should reduce the variability of the ionic shifting for
metallic test substances. No data could be located to allow
comparison of the toxicity of the different ionic forms of a
given metal. Due to the chemical complexity of the test
medium, such as food type and concentration, metabolic
products and test system-test substance interaction,
maintaining dissolved oxygen values at a given ninimum
should help to reduce variability in test results.
The concentration of dissolved oxygen should be
monitored closely in static acute test chambers to insure
28
-------
ES-1
August, 1982
that the oxygen level is above the required minimum. During
a test, the chambers are not to be aerated under any
circumstances. This is specified to prevent air entrapment
under the ins tar carapace as well as not to enhance
volatilization of the test substance. If levels cannot be
maintained above 60 percent, the static flow-through method
should be used.
The potential for oxygen depletion in the renewal test
does exist, and depends on loading, food type and
concentration, feeding frequency, and renewal frequency.
/
Several renewal studies listed in Table 3 show a range of
values for these parameters. Only .one study, Winner and
Farrell (1976), presents any dissolved oxygen data. These
authors state that observe'"! dissolved oxygen values were
always in excess of 95 percent saturation.,
Table 3. Summary of Renewal Chronic Assay Techniques.
Initial loading Replacement
(daphn ids/ volume) Frequency Food
Frequency
Reference
1/40 ml
1/40 ml
1/100 ml
1/50 ml
1/40 ml
4 Days C. pyrenoidosa 1 day Bertram and
and Yeast Hart (1979)
1 Week Trout Pellets 1 Week Biesinger and
Chris tens en
(1972)
2 Days Scenedesnus
and Yeast
2-3 Days Chlorella
3 Days hi amyd arenas
2 Days Schober and
Lampert
(1977)
2-3 Days Stroganov
et al.
(1977)
1 Day Winner and
Farrell
(1976)
29
-------
ES-1
August, 1932
The use of photosynthe'tlcal ly active algae can
provide significant oxygen to the test medium during the
light photoperiod. During the dark photoperiod, algal
respiration competes with the other oxygen sinks for
dissolved oxygen. Again, depending on a number oE
variables, significant oxygen depletion is a possibility.
Animals maintained on non-photosynthetic food with extended
replacement schemes, such as in Biesinger and Christensen
(1972), may be exposed to extended periods of low dissolved
oxygen. No'data on the effects of low dissolved oxygen on
reproduction and survivorship of the test organisms could be
loca ted .
The suggested scheme of alternate day renewal in
chronic testing should aid in maintaining dissolved oxygen
levels above the recommended minimum.' Dissolved oxygen
readings should be taken several t L^s during the first two
days of testing to determine that the proper levels are
maintained. Tt is especially important to determine
dissolved oxyjen values at the end of the dark photoperiod
when using algal food supplies.
The inability to maintain sufficient dissolved
oxygen values in a renewal system indicates the need for a
flow-through test or larger volume test chambers.
The flow-through techniques should not present any
significant dissolved oxygen problems due to (1) the
constant replacement of the test medium and (2) the volume
of the test chamber and loading ratio, which should provide
sufficient oxygen according to the data of Adema (1978).
This is not to imply that dissolved oxygen need not be
monitored.
30
-------
ES-1
1982
b_. Light. A 16-hour light, 8-hour dark
photoperiod with a 15 to 30-minute transition has been
suggested to meet biological requirements of daphnids and to
increase 'test standardization (ASTM, 1980). The recommended
photoperiod approximates the temperate summer light regime
which would support parthenogenic reproduction. The
transition period is recommended i:o simulate natural
conditions. A device for maintaining the photoperiod has
been described by Drummond and Dawson (1970).
Stross and Hill (1968), using D. pulex at five
individuals per 50 ml density at 19 °C for 30 days duration>
observed decreasing sexual reproduction with changing
photoperiod. At 12L:12D, approximately 90 percent sexual
reproduction was recorded, while increasing the light period
to 14L:10D resulted in 0 percent sexual reproduction.
Extrapolation to the recommended photoperiod at 16L:8D
should insure parthenogenic reproduction.
Wide spectrum fluorescent bulbs (Color Rendering, Index
greater than 90) and a light intensity at the surface of the
test chambers not exceeding 800 lux (74 ft. candles) is
essentially equivalent to the average cabletop conditions.
These facilities would allow the use of room lighting for
experimental conditions, provided it was controlled to the
recommended photoperiod and transition period.
Temperature. The selected test temperature
(20 ± 1°C) approximates room temperature, thus minimizing
the requirement for extensive temperature controlling
devices in culturing, acclimating and testing facilities.
31
-------
ES-1
August, 198-2
An accurate device controlling room temperature should
maintain the dap hn ids within the proper range of
temperatures.
The selected temperature also approximates summer water
temperatures for temperate lakes. The test temperature is
within the range in which _D. magna reproduces
parthenogenically. Kaestner (1970) reports asexaal
reproduction for D. magna in the temperature range 11°C to
27°C. It was also reported that temperatures below 15°C,
.together with stress conditions, resulted : in sexual
reproduction. Sexual reproduction was also reported as
temperatures approached 30°C.
Variations in test temperatures beyond thos.e suggested
could bias test results. Bunting (1974) observed a 50
percent reduction in the growth rate of juvenile D. magna at
15°C as compared to 20°C. This rate reduction manifested
itself in increased time periods between molts. At 25°C an
increased growth rate was observed as compared to 20°C,-'out
beyond 25°C a reduction in t a growth rate was observed,
indicating a potential thermal stress. Bunting and Robertson
(1975) observed a significant difference in the acute
effects of two herbicides, Aininotr izole and Amitrole, on
juvenile D. pulex at two temperatures. It required
approximately twice, the herbicide concentration at the lower
temperature, 15°C, to produce the same effect as at 20°C.
The narrow range of specified temperatures for these test
guidelines should reduce significant differences in reported
test results.
The information presented in Table 4 was derived from
32
-------
53—1
August, 1982
a search of articles concerning control mortality during
chronic toxicity tests. In some cases, the 21-day control
mortality was interpolated from long-term survivorship
curves. Based on the studies the recommendation of a 20
percent maximum control mortality criterion for 21 day
chronic tests should provide a reasonable testing
requirement.
Table 4: Percent Mortality in Controls for 21-
day experiments using daphnids.
Percent
Mortality
12
8-20
0
15
0
11
Test
Temperature1 Species
20 D . mag n a
18 D. magna
19 D. pulex
20 D. magna
20 D. pulex
20 D. pulex
Reference
Berge
(1978)
Nebecker
Puglis i
( 1974)
Bertram
and Hart
(1976)
Winner
and
Farrell
(1976)
Schober
and
Lamoert
(1977)
Winner
and
Fa ere]. 1
( 1976)
33
-------
ES-1
August, 1982
C. Reporting
1. Acute Tes ts. For each set of data, with a
ninimura of the 24- and 48- hoar observations, the EC50 and
95 percent confidence limits should be calculated based on
the mean measured concentration of the toxicant. A
concentration-response curve Cor each observation period
shall also be constructed. It is strongly suggested that a
statistician be consulted before the test is initiated to
insure that the specific test procedures used will satisfy
the statistical, requirements of the methods of data
analys is.
Acute toxicity tests usually produce quantal data, that
is, counts of the number of organisms in two mutually-
exclusive categories - alive/dead; affected/not affected. A
variety of methods can be used to calculate an EC50 and 95
percent confidence limits from quantal data containing two
or more concentrations at which the percent affected is
between zero and one hundred. The most widely used are the
probit moving average, and Litchfield-Wilcox >n methods
(Finney, 1964 and 1971, Steph.an 1977, Litchf ield and
Wilcoxon 1949). The method of Litchfield and Wilcoxon
(1949) produces a slope function which together with EC50
value allows reconstruction of the probit Line, The slope
of the straight line can be useful for interpolation of the
potential effects of concentrations other than those near
the EC50. The slope may also provide indications as to the
mode of toxicity or any change in the toxic effects over the
experimental period. Sprague (1969) discusses the
construction and interpretation of 'concentration-response
curves using several methods of data analysis. Thus, both
the EC50 values and the dose-response curves are necessary
34
-------
ES-1
August, 1982
for evaluating the hazard potential of a given test
subs tance.
2. Chronic Tests. The statistical methods used
to evaluate the effects of a test compound on survival and
reproduction should be described in full. The choice of;
methods is left to the testing facilities, but it is
suggested that a statistician be consulted prior to the
initiation of the test program.
Suggested methods include analysis of variance (ANOVA)
and appropriate mean separation tests (Sokal and Rahlf 1969,
Steele and Torrie 1960).
III. Econlomic Aspects .
The Agency awarded a contract to Enviro Control, Inc.
to provide us with an estimate of the cost for performing
static and flow-through acute toxicity tests and renewal and
flow-through chronic toxicity tests. Enviro Control
supplied us with two estimates; a protocol estimate and a
laboratory survey estimate.
Protocol Estimates
range mean
Acute (static and flow-through) $322-$965 $643
Chronic (renewal and flow-through) $2021-$6064 $4043
These estimates were prepared by separating the
guidelines into individual tasks and estimating the hours
used to accomplish each task. Hourly rates were then
applied to yield a total direct labor charge. An overhead
rate of 115 percent, other direct costs ($50 - acute, $450 -
chronic), a general and administrative rate of 10 percent
and a fee of 20 percent were then added to the direct labor
charge to yield the final estimate.
35
-------
ES-1
1982
jaboratory Survey Estimates
range
Acute (static and flow-through) $340-$1250 $743
Chronic (renewal and flow-through) $750-$10,000 $4178
The laboratory survey estimates were compiled from
three laboratories for the acute guideline and five
laboratories for the chronic guideline.
36
-------
ES-1
August, 1982
IV. REFERENCES
Adema DMM. 1978. Daphnia magna as a test animal in
acute and chronic toxicity tests. Hydrobiologia
59:125-134.
Anderson BG . 1932. The number of pre-adult ins tars ,
growth, relative growth and variation in Daphnia
magna. Biol. Bull. 63:81-98.
Anderson RG and Jenkins JC. 1942. A time study of
events in the life span of D. magna. Biol. Bull.
83:260-272.
Anderson BG, Lumer H and Zupancic LJ. 1937. Growth
and variability in Daphnia pulex. Biol. Bull. 73:444-
463.
APHA. 1975. American Public Health Association.
Standard Methods for the Examination of Water and
Wastewater. 14th ed . New York, N.Y. 1193 p.
ASTM. 1980. American Society for Testing and Material
Standard Practice for Conducting Acute Toxicity Test
with Fishes, Macroinverte'orates and Amphibians.
Philadelphia, PA.
Gerge WF. 1978. Breeding Daphnia magna.
Hydrobiologia. 59:121-123.
Bertram PE and Hart BA. 1979. Longevity and
reproduction of Daphnia pul^x exposed to cadmium
contaminated food or water. Environ. Pollut. 19:295-
305.
Biesinger XE and Christensen GM. 1972. Effects of
various metals on survival, growth, reproduction, and
metabolism of Daphnia magna. J. Fish. Res. Brd.
Canada. 29:1961^700.
Brandlova J, Bramdl A and Fernando CH. 1974. The
cladocera of Ontario with remarks on some species and
distribution. Canadian J. of Zoology. 50:1373-1403.
Brooks JL. 1959. Cladocera. In Freshwater Biology.
W.T. Edmonson ed . 2nd ed . New York: Wiley, po. 537-
656.
37
-------
SS-1
August, 1982
Buikema AL, Lee DR and Cairns J. 1976, Screening
bioassay using Daphnia pulex for refinery waste
discharged into freshwater, J. Test. Eval. 4:119-125.
Bunner HC and Halcrow K. 1977. Experimental induction
of the production of ephippia by Daphnia magna.
Ccustaceana. 32:77-36.
Bunting DL. 1974. Zooplankton: Thermal regulation and
stress. In Energy Production and Thermal Effects.
3.J. Gallager, ed. Ann Arbor, MI: Ana Arbor Science
Publ., pp. 50-55.
Bunting DL and Robertson EB . 1975. Lethal and
sub lethal effects of herbicides on zooplankton
species. Research Report No. 43. Water -Research
Center. Univ. of Tennessee.
Bujrns CW. 1969. Relation between filtering rate,
temperature and body s ize in four species of
Daphnia] . Limnol. Oceanogr. 14:693-700.
Canton JH, Greve PA, Sloof W and Esch GJ. 1975.
Toxicity, accumulation and elimination, studies of alpha
hexachlorocyclohexane with freshwater organisms of
different trophic levels. Water Research 9:1163-1169.
Canton JH and Adema DMM. 1978. Reproducibility of
short-term and reproduction toxicity experiments with
Daphnia magna and comparison of the sensitivity of
Daphnfa" magna with Daphnia pulex and Daphnia cucullata
in short-term experiments . Hydrobiologia 5~9 :13 5 -14 0 .
Curtis MW, Copland TL and Ward CH. 1979. Acute
toxicity of 12 industrial chemicals to freshwater and
saltwater organisms. Water Research 13:137-141.
D'Agostino AS and Provasoli L. 1970. Dixenic culture
of Daphnia magna, Straus. 3iol. Bull. 139-435-494.
Dewey TE and Parker BL. 1964. Mass rearing of Daphnia
magna for insecticide bioassays. J. Econ. Entomol.
57:821-825.
Doudoroff P. 1942. The resistance and Acclimation of
marine fishes to temperature changes. I. Experiments
with Girella nigrans (Ayers). Biol. Bull. 83:219-244.
33
-------
ES-1
August, 1932
Doudoroff P. 1945. The resistance and acclimation of
marine fishes to temperature changes. II, Experiments
with Fundulus and Atherinops Biol. Bull. 88:194-206.
Drummond RA and Daws on WF. 1970. An inexpensive
method for simulating diel patterns of lighting in the
laboratory. Trans. Amer. Fish. Soc. 99:434-435.
Faust SD -and Hunter JV. 1967. Principles and
Applications of Water Chemistry. Wiley, New York, N.Y.
643 pp.
Finney D J. 1964. Statistical Methods in Biological
Assay. 2nd ed. Hafner Publishing Co. New York, N.Y.
668 p.
Finney DJ. 1971. Probit Analysis. 3rd ed . Cambridge
Univ. Press. London. 333 p.
Frear DE and Boyd JE. 1967. Use oC Oaohnia magna for
the macrobioass ay of pesticides. I. Development of
standardized techniques for rearing Daohnia and
preparation of dose-mortality curves for pesticides.
J. Econ. Entomol. 60:1228-1236.
Gulati RD. 1978. The ecology of the common olanktonic
Crustacea of the freshwaters- in the Netherlands.
Hydrobiologia 59:101-112.
Hebert PDN. 1978. , The population biology of Daphnia.
Biol. Rev. 53:387-426.
Hoar WS. 1966. General and Comparative Physiology.
Prentice Hall. New York, N.Y. 815 p.
Hutchinson GE. 1967. A Treatise on Limnology. Volume
II. Introduction to Lake Biology and Limnoplankton.
Wiley, New York, N.Y. 1115 p.
Kaestner A. 1970. Invertebrate Zoology. Vol. III.
Intersc ience. New York, N.Y. 523 p.
Kenaga EE. 1978. Test organisms and methods useful
for early assessment of acute toxicity of chemicals.
Snv. Sci. Tech. 12:1322-1329.
39
-------
ES-1
August, 1982
Kenaga EE and Moolenar RJ. 1979. Fish and Daphnia
toxicity as surrogates for aquatic plants and algae.
Env. Sci. Tech. 13:1479-1430.
Kring RL a;-rl O'Brien W J. 1976. Effects of varying
oxygen concentrations on the filtering rate of Daphnia
oalex. Ecology 57:303-314.
Xrugel S, Jenkins D and Klein S. 1978. Apparatus for
the continuous dissolution of poorly water-soluble
compounds for bioassays. Water Research 12:259-272.
Lee DR and BuikeTia AL. 1979. Molt-related sensitivity
of Daphnia oulex in toxicity testing. J. Fish. Res.
Bd . Canada. 3~67l 129-1133,
Leeuwangh P. 1973. Toxicity tests with Daphnids: Its
application in the 'management of water quality.
Hydrobiologia 59:145-148.
Liptak BG. 1974. Environmental Engineers Handbook;.
Vol. I. 'Water Pollution. C'nilton Book Co., Radnor,
PA. 2013 p.
Litchfield JT and Wilcoxon F. 1949. A simplified
method for evaluation dose-effect experiments. J,
Pharmacology Exper. Therapeutics 96:99-113.
Lockwood APM. 1967- Physiology oE Crustacea.
Freeman. San Francisco, CA. 328 p.
Makarewicz JD and Likens GE. 1979. Structure and
function of the zooplankton community of Mirror Lake,
New Hampshire. Ecol. Monog. 49:109-127.
McMahon JW. 1965. Some physical factors influencing
the feeding behavior of Daphnia magna. Can. J. Zool.
43:603-611.
Murphy JS. 1970. A general method for the monaxenic
cultivation of the Daphindae. Biol. Bull. 139:312-332.
Nebeker AV and Puglisi FA. 1974. Effects of
polychlorinated biphenyls (PCB's) on survival and
reproduction of Daphnia, Gammarus and Tanytarsus .
Trans. Amer. Fish. Soc. 103:722-728.
40
-------
ES-1
August,- 1982
Needham JG, Galtsoff PS, Lutz FE and Welch PS. 1959.
Culture Methods foe Invertebrate Animals. Dover Press,
New York, N.Y. 590 p.
P'ennak RW. 1973, Freshwater Invertebrates of the
United States. John Wiley. New York, N.Y. 303 o.
Richman S. 1958. The trans formatioa of energy by
Daphnia pulex. Eco. Monog. 28:273-291.
Sanders HO. 1970. Toxicities of some herbicides to
six species of freshwater crustaceans. J. Wat. Po.lJ.ut.
Contr. Fed. 42:1544-1550.
Sax NL. 1979. Dangerous Properties of Industrial
Material. Reinhold, New York, N.Y. 1343 p.
Schober U and Lampert W. 1977. Effects of sublethal
concentrations of the herbicide Atraziu on growth and
reproduction of Daphnia pulex. Bull. Environ. Contam.
Tox. 17:259-277.
Schultz TW and Kennedy JR. 1976. Cytotoxic efEects of
the herbicide 3-amino-l, 2, 4-triazole on Daphnia
pulex . Biol. Bull. 151:370-385.
Sokal RR-and Rohlf F J. 1969. Biometry. W-.H. Freeman
Co., San Francisco, CA. 776 p.
Sprague JB. 1969. Measurement of pollution toxicity
to fish. I. Bioassay methods for acute toxicity.
Water Research 3:793-821.
Steele RGD and Torrie JH. 1960. Principles and
Procedures of Statistics. McGraw Hill, Inc. New York,
N.Y. .
Stephan CE. 1977- Methods For calculating and L>CCQ.
In Aquatic Toxicology and Hazard Evaluation. ASTM STP
~6~34. F.L. Mayer and J.L. Hamelink, eds Philadelphia,
PA. pp. 65-84.
Stroganov NS, Maksimova NN and Isakova YI. 1977.
Long-term residual effects of polyethyleneamine on
Daphnia. Hydrobiological J. 13:74-83.
41
-------
Augus t;
ES-1
1982
Stross RG and Hill JC.
winter diapause in the
Daphnia. Biol. Bull.
1968. Photoperiod control of
freshwater crustacean,
134:176-198.
Winner RW and Parrel 1 MP. 1976. Acute and chronic
toxicity of copper to four species of Daphnia. J.
Fish. Res. 3d. Canada. 33:1585-1691.
Whitten RH, ^energrass
Living Inver ebrates.
Burlington, C. 25 p.
WR and Best RL. 1976. Care of
Carolina Biologicdl Supply Co.
42
-------
BG-3
August, 1982
MYSID SHRIMP ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-3
Guideline for Testing of Chemicals August, 1982
MYSID SHRIMP ACUTE TOXICITY TEST
(a) Purpos e. This guideline is intended for use in
developing data on the acute toxicity of chemical substances and
mixtures ("chemicals") subject to environmental effects test-
regulations under the Toxic Substances Control Act (TSCA) (Pub.L.
94-469, 90 Stat. 2003, 15 U.S.C. 2601 e_t. seq. ) . This guideline
prescribes a test using mys id shrimp as test organisms to develop
data on the acute toxicity of chemicals. The United States
Environmental Protection Agency ( EJPA) will use data from these
tests in assessing the hazard of a chemical to the aquatic
environment.
(b) Def initions . The definitions in Section 3 of the Toxic
Substances Control Act (TSCA) and in Part 792—Good Laboratory
Practice Standards apply to this test guideline. The following
definitions also apply to this guideline.
(1) "Death" means the lack of reaction of a test organism to
gentle prodding.
(2) "Flow-through" means a continuous or an intermittent
passage of test solution or dilution water through a test chamber
or a holding or acclimation tank, with no recycling.
(3) "LC50" means that experimentally derived concentration
of test substance that is calculated to kill 50 percent of a test
population during continuous exposure over a specified period of
-------
EG-3
August, 1982
t ime.
(4) "Loading" means the ratio of test organisms biomass
(grams, wet weight) to the volume (liters) of test solution in a
test chamber.
(5) "Retention chamber" means a structure ithin a flow-
through test chamber which confines the test ore .nisms,
facilitating observation of test organisms and eliminating loss
of organisms in outflow water.
(6) ."Static system" mealns a test chamber in which the test
solution is not renewed during the period of the test.
(c) Test procedures — (1) Summary of the test. In
preparation for the test, test chambers are filled with
appropriate volumes of dilution water. If a flow-through test is
performed, the flow of dilution water through each chamber is
adjusted to the rate desired. The test substance is introduced
into each test chamber. In a flow-through test, the rate at
which the test substance is added is adjusted to establish and
maintain the desired concentration of test substance in each test
chamber. The test is started by randomly introducing my s ids
acclimated in accordance with the test design into the test
chambers. Mys ids in the test chambers are observed periodically
during the test, the dead mys ids removed and the findings
recorded. Dissolved oxygen concentration, pH, temperature,
-------
EG-3
August, 1982
salinity, the concentration of test substance, and other water
quality characteristics are measured at specified intervals in
test chambers. Data collected during the test are used to
develop concentration-response curves and LC50 values for the
test substance.
(2) [Reserved]
(3) Range-finding test. (i) A range-finding test should be
conducted to determine:
(A) whi[ch life stage (juvenile or young adult;) is to bel
utilized in the definitive test.
(B) the test solution concentrations for. the definitive
test.
(ii) The mys ids should be exposed to a series of widely-
spaced concentrations of test substance (e.g., 1, 10, 100 mg/1,
etc.), usually under static conditions.
(iii) This test should be conducted with both newly-hatched
juvenile (< 24 hours old) and young adult (5-6 days old)
mysids. For each age class (juvenile or young adult), a minimum
of ten mysids should be exposed to each concentration of test
substance for up to 96 hours. The exposure period may be
shortened if data suitable for the purpose of the range-finding
test can be obtained in less time. The age class which is most
sensitive to the test substance in the range-finding test should
-------
EG-3
August, 1982
be utilized in the definitive test. When no apparent difference
in sensitivity of the two life stages is found, juveniles should
be utilized in the definitive test. No replicates are required
and nominal concentrations of the chemical are acceptable.
(4) Definitive test. (i) The purpose of the definitive
test is to determine the concentration-response curves and the
48- and 96- hour LC50 values with the minimum amount of testing
beyond the range-finding test.
( i i-) The definitive1 test alhould be conducted on. the mys id
life stage (juveniles or young adults) which is most sensitive to
the test substance being evaluated.
(iii) A minimum of 20 mysids per concentration should be
exposed to five or more concentrations of the chemical chosen in
a geometric series in which the ratio is between 1.5 and 2.0
(e.g., 2, 4, 8, 16, 3 and 64 mg/1) . An equal number of mys ids
should be placed in two or more replicates. If solvents,
solubilizing agents or emulsifiers have to be used, they should
be commonly used carriers and should not possess a synergistic or
antagonistic effect on the toxicity of the test substance. The
concentration of solvent should not exceed 0.1 ml/1. The
concentration ranges should be selected to determine the
concentration-response curves and LC50 values at 48 and 96
hours. The concentration of test substance in test solutions
should be analyzed prior to use.
-------
EG-3
August, 1982
( iv) Every test should include controls consisting of the
same dilution water, conditions, procedures, and mys ids from the
same population or culture container, except that none of the
chemical is added.
(v) The dissolved oxygen concentration, temperature,
salinity, and oH should be measured at the beginning of the test
and at 24, 48, 72 and 96 hours in each chamber.
(vi) The test duration is 96 hours. The test is
unacceptable if mote than 10 percent of the control organisms die'
or exhibit abnormal behavior during the 96 hour test period.
Each test chamber should be checked for dead mys ids at 3, 6, .12,
24, 48, 72 and 96 hours after the beginning of the test.
Concentration-response curves and 43- and 96- hour LC50 values
should be determined along with their 95 percent confidence
1 imits.
(vii) In addition to death, any abnormal behavior or
appearance should also be reported.
(viii) Distribution of mys ids among test chambers should be
randomized. In addition, test chambers within the testing area
should be positioned in a random manner or in a way in which
appropriate statistical analyses can be used to determine the
variation due to placement.
( ix) The concentration of dissolved test substance (that
-------
EG-3
August, 1982
which passes through a 0.45 micron filter) in the chambers should
be measured as often as is feasible during the test. At a
minimum, during static tests, the concentration of test substance
should be measured in each chamber at the begi-nning and at the
end of the test. During the flow-through test, the concentration
of test substance should be measured (A) in each chamber at the
beginning of the test and at 48 and 96 hours after the start of
the test; (B) in at least one chamber containing the next to the
lowest test substance concentration at least 'ojice every 24 hburs
during the test; and (C) in at least one appropriate chamber
whenever a malfunction is detected in any part of the test
substance delivery system. Among replicate test chambers of a
treatment concentration, the measured concentration of the test
substance should not vary more than 20 percent.
(5) [Reserved]
(6) Analytical measur ements--( i ) Tes t chemical . Deionized
water should be used in making stock solutions of the test
substance. Standard analytical methods should be used whenever
available in performing the analyses. The analytical method used
to measure the amount of test substance in a sample should be
validated before beginning the test by appropriate laboratory
practices. An analytical method is not acceptable if likely
degradation products of the test substance, such as hydrolysis
-------
EG-3
August, 1982
and oxidation products, give positive or negative interferences
which cannot be systematically identified and corrected
ma thema t i c ally.
(ii) 'Numerical . The number of, dead mys ids should be counted
during each definitive test. Appropriate statistical analyses
should provide a goodness-of-f it determination for the
concentration-response curves. A 48- and 96- hour LC50 and
corresponding 95 percent interval should be calculated.
(d) 'Test conditions--(1) Tes!t species--(i) Selection.
(A) The mys id shrimp, Mysidopsis bahia, is the organism
specified for these tests. Either juvenile (< 24 hours old) or
young adult (5-6 days old) mys ids are to be used to start the
test.
(3) Mys ids to be used in acute toxicity tests should
originate from laboratory cultures in order to assure that the
individuals are of similar age and experiential history. Mys ids
used for establishing laboratory cultures may be purchased
commercially or collected from appropriate natural areas.
Because of similarities with other mysid species, taxonomic
verification should be obtained from the commercial supplier or
through an appropriate systematic key.
(C) Mys ids used in a particular test should be of similar
age and be of normal size and appearance for their age. Mys ids
-------
EG-3
August, 1982
should not be used for a test if they exhibit abnormal behavior
or if they have been used in a previous test, either in a
treatment or in a control group.
(ii) Acclimation. (A) Any change in the temperature and
chemistry of the dilution water used for holding or culturing the
test organisms to those of the test should be gradual. Within a
24-hour period, changes in water temperature should not exceed
1°C, while salinity changes should not exceed 5 Percent.
(B) During acclimation 'mys ids should be maintained1 in
facilities with background colors and light intensities similar
to those of the testing areas.
(iii) Care and handling. Methods for the care and handling
of mys ids such as those described in US EPA (1978) can be used
during holding, culturing and testing periods.
( iv) Feeding. Mys ids should be fed during test ng. Any
food utilized should support survival, growth and reproduction of
the mysids. A recommended food is live Artemia spp. (48-hour-old
naupli i) .
(2) Facilities — (i) Apparatus. (A) Facilities which may
be needed to perform this test include: (I) flow-through or
recirculating tanks for holding and acclimating mys ids ; (2) a
mechanism for controlling and maintaining the water temperature
during the holding, acclimation and test periods; (3) apparatus
-------
EG-3
August, 1982
for straining particulate matter, removing gas bubbles, or
aerating the water, as necessary; and (_4_) an apparatus for
providing a 14-hour light and 10-hour dark photoperiod with a 15
to 30 minute transition period. In addition, for flow-through
tests, flow-through chambers and a test substance delivery system
are required. Furthermore, it is recommended that mys ids be held
in retention chambers within test chambers to facilitate
observations and eliminate loss of test organisms through outflow
water. For static tests, suitable chambers for exposing test
mys ids to the test substance are required. Facilities should be
well ventilated and free of fumes and disturbances that may
affect the test organisms.
(B) Test chambers should be loosely covered to reduce the
loss of test solution or dilution water due to evaporation and to
minimize the entry of dust or other particulates into the
solutions.
(ii) Cleaning . Test substance delivery systems and test
chambers should be cleaned before each test following standard
laboratory practices,
(iii) Construction materials. (A) Materials and equipment
that contact test solutions should be chosen to minimize sorption
of test chemicals from dilution water and should not contain
substances that can be leached into aqueous solution in
-------
EG-3
August,_1982
quantities that can affect test results.
(3) For use in the flow-through test, retention chambers
utilized for confinement of test organisms can be constructed
with netting material of appropriate mesh size.
( iv) Dilution water. (A) Natural or artificial seawater is
acceptable as dilution water if mys ids will survive and
successfully reproduce in it for the duration of the holding,
acclimating and testing periods without showing signs of stress,
such as reduced growth and ' f ecundity . Mys ids should be cultured
and tested in dilution water from the same origin.
(B) Natural seawater should be filtered through a filter
with a pore size of < 2 0 microns prior to use in a test.
(C) Artificial seawater can be prepared by adding
commercially available formulations or by adding specific amounts
of reagent-grade chemicals to deionized water. Deionized water
with a conductivity less than 1 u ohm/cm at 12°C is acceptable
for making artificial seawater. When deionized water is prepared
from a ground or surface water source, conductivity and total
organic carbon (or chemical oxygen demand) should be measured on
each batch.
(v) Test substance delivery system. In flow- through tests,
proportional diluters, metering pumps or other suitable systems
should be used to deliver test substance to the test chambers.
10
-------
EG-3
August, 1932
The system used should be calibrated before each test.
Calibration includes determining the flow rate through each
chamber and the concentration of the test s.u'ostance in each
chamir^r. The general operation of the test substance delivery
system should be checked twice daily during a test. The 24-hour
flow through a test chamber should be equal to at least five
times the volume of the test chamber. During a test, the flow
rates should not vary more than 10 percent among test chambers or
across time.
(3) Test parameters. Environmental parameters of the-water
contained in test chambers should be maintained as specified
below:
(i) Temperature of 25 ± 2°C.
(ii) Dissolved oxygen concentration between 60 and 105
percent saturation. Aeration, if needed to achieve this level,
should be done before the addition of the test substance. All
treatment and control chambers should be given the same aeration
treatment.
(iii) The number of rays ids placed in a test solution should
not be so great as to affect results of the test. Thirty mys ids
per liter is the recommended level of loading for a static
test. Loading requirements for the flow-through test will vary
depending on the flow rate of dilution water. The loading should
11
-------
EG -3
August, 1982
not cause the dissolved oxygen concentration to fall below the
recommended levels.
( iv) Photoperiod of 14 hours light and 10 hours darkness,
with a 15-30 minute transition period.
(v) Salinity of 20 ± 3 o/oo.
(e) Reporting. The sponsor should submit to the EPA all
data developed during the test that are suggestive or predictive
of acute toxicity and all concomitant toxicologic
manifestations.. In addition to the general reporting
requirements prescribed in Part 792—Good Laboratory Practice
Standards, the reporting of test data should include the
following:
(1) The source of the dilution water, its chemical
characteristics (e.g., salinity- pH, etc.) and a description, of
any pretreatment.
(2) Detailed information about the test organisms, including
the scientific name and method of verification, age, source,
history, abnormal behavior, acclimation procedures and food used.
(3) A description of the test chambers, the depth and volume
of solution in the chamber, the way the test was begun (e.g.,
conditioning, test substance additions, etc.), the number of
organisms per treatment, the number of replicates-, the loading,
the lighting, the test substance delivery system and the flow
12
-------
EG-3
August, 1982
rate expressed as volume additions per 24 hours.
(4) The measured concentration of test substance in test
chambers at the times designated.
(5) The number end percentage of organisms that died or
showed any other adverse effects in the control and in each
treatment at each observation period.
(6) Concentration-response curves should be fitted to
mortality data collected at 24, 48, 72 and 96 hours. A
statistical test of goodness-of -f it should i be performed and the
results reported.
(7) The 43-hour and 96-hour LC50, and when sufficient data
have been generated, the 24-hour and 72-hour LCSO's and the
corresponding 95 percent confidence limits and the methods used
to calcula 3 the values. These calculations should be made using
the average measured concentration of the test substance.
(8) Methods and data records of all chemical analyses o£
water quality and test substance concentrations, including method
validations and reagent blanks.
(9) The data records of the holding, acclimation and test
temperature and salinity.
(f) References. U.S. Environmental Protection Agency,
1978. Bioassay Procedures for the Ocean Disposal Permit
Program. Environmental Research Laboratory, Office of Research
and Development. Gulf Breeze, Fl. EPA-600-9-78-010.
13
-------
EG-4
August, 1982
MYSID SHRIMP CHRONIC TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-4
Guideline for Testing Chemicals August, 1982
Mys id shrimp chronic toxicity test
(a) Purpose. This guideline is intended for use in
developing data on the, chronic toxicity of chemical substances
and mixtures ("chemicals") subject to environmental effects test
regulations under the Toxic Substances Control Act (TSCA) (Pub.L.
94-469, 90. Stat. 2003, 15 U.S.C. 2601 et seg. ) . This guideline
prescribes tests using mys ids as test organisms to develop data
on the chronic toxicity of chemicals. The United States
Environmental Protection Agency (EPA) will use data from these
tests in assessing the hazard of a chemical to the aquatic
environment.
(b) Def initions . The definitions in section 3 of the Toxic
Substances Control Act (TSCA) and in Part 7 9 2--Good Laboratory
Practice Standards apply to this test guideline. The following
definitions also apply to this guideline:
(1) "Chronic toxicity test" means a method used to determine
the concentration of a substance that produces an adverse effect
from prolonged exposure of an organism to that substance. In
this test, mortality, number of young per female and growth are
used as measures of chronic toxicity.
(2) "Death" means the lack of reaction of a test organism to
-------
EG-4
August, 1982
gentle prodding.
(3) "Flow-through" means a continuous or an intermittent
passage of test solution or dilution water through a test chamber
or a holding or acclimation tank, with no recycling.
(4) "Gl (Generation 1)" means those mysids which are used to
begin the test, also referred to as adults; G2 (Generation 2) are
the young produced by Gl.
(5)- "LC50" means that experimentally derived concentration
of test substance that is calculated to kill 50 percent of a test
population during continuous exposure over a specified period of
time.
(6) "Loading" means the ratio of test organism biomass
(gram, wet weight) to the volume (liters) of test solution in a
test chamber.
(7) "MATC" (Maximum Acceptable Toxicant Concentration) means
the maximum concentration at which a chemical can be present and
not be toxic to the test organism.
(8) "Retention chamber" means a structure within a flow-
through test chamber which confines the test organisms,
facilitating observation of test organisms and eliminating
washout from test chambers.
(c) Test procedures — (1) Summary of the test. (i) In
preparation for the test, the flow of test solution through each
-------
EG-4
August, 1982
chamber is adjusted to the rate desired. The test substance is
introduced into each test chamber. The rate at which the test
substance is added is adjusted to establish and maintain the
desired concentration of test substance in each test chamber.
The test is started by randomly introducing mysids acclimated in
accordance with the test design into retention chambers within
the test and the control chambers. Mys ids in the test and
control chambers are observed periodically during the test, the
dead mysids removed and the findings reported.
(ii) Dissolved oxygen concentration, pH, temperature,
salinity, the concentration of best substance and other water
quality characteristics are measured at specified intervals in
selected test chambers.
(iii) Data collected during the test are used to develop a
MATC (Maximum Acceptable Toxicant Concentration) and quantify
effects on specific chronic parameters.
(2) [Reserved]
(3) Range-finding test. (i) A range-finding test should be
conducted to establish test solution concentrations for the
definitive test.
(ii) The mys ids should be exposed to a series of widely
spaced concentrations of the test substance (e.g., 1, 10, 100
mg/1), usually under static conditions.
-------
EG-4
August, 1982
(iii) A minimum-of 10 mysids should be exposed to each con-
.centration of test substance for a period of time which allows
estimation of appropriate chronic test concentrations. No
replicates are required and nominal concentrations of the
chemical are acceptable.
(4) Definitive test. (i) The purpose of the definitive
test is to determine concentration-response curves, LC50 values,
and effects of a chemical on growth and reproduction during
chronic exposure.
(ii) A minimum of 40 mys ids per concentration should be
exposed to four or more concentrations of the chemical chosen in
a geometric series in which the ratio is between 1.5 and 2.0
(e.g., 2, 4, 8, 16, 32 and 64 mg/1). An equal number of mysids
should be placed in two or more replicates. If solvents,
solubilizing agents or emulsifiers have to be used, they should
be commonly used carriers and should not possess a synergistic or
antagonistic effect on the toxicity of the test substance. The
concentration of solvent should not exceed 0.1 ml/1. The
concentration ranges should be selected to determine the
concentration-response curves, LC50 values and MATC.
Concentration of test substance in test solutions should be
analyzed prior to use.
(iii) Every test should include controls consisting of the
same dilution water, conditions, procedures and mysids from the
-------
EG-4
August, 1982
sane population or culture container, except that none of the
chemical is added.
(iv) The dissolved oxygen concentration, temperature,
salinity and pH should be measured at the beginning of the test
and on days 7, 14, 21 and 28 in each chamber.
(v) The. test duration is 28 days. The test is unacceptable
if more than 20 percent of the control organisms die, appear
stressed or are diseased during the test. The number of dead
mysids in each chamber should 'be recorded on days 7, 14, 21 and
28 of the test. At the time when sexual characteristics are
discernable in the mys ids (approximately 10-12 days in controls;
I
possible delays may occur in mys ids exposed to test substances),
the number of males and females (identified by ventral brood
pouch) in each chamber should be recorded. Body length (as
.measured by total midline body length, from the anterior tip of
the carapace to the posterior margin of the uropod) should be
recorded for males and females at the time when sex can be
determined simultaneously for all mys ids in control and treatment
groups. This time cannot be specified because of possible delays
in sexual maturation of mys ids exposed to test substances. A
second observation of male and female body lengths should be
conducted on day 28 of the test. To reduce stress on the mysids,
body lengths can be recorded by photography through a stereo-
-------
EG-4
August, 1982
microscope with appropriate scaling information. As offspring
are produced by the Gl mys ids (approximately 13-16 days in
controls), the young should be counted and separated into
retention chambers at the same test substance concentration as
the chambers where they originated. If available prior to
termination of the test, observations on the mortality, number of
males and females and male and female body length should be
recorded for the G2 mys ids . Concentra-tion-response curves, LC50
values and associated 95 percent confidence limits for the number
of dead mys ids (Gl) should be determined for days 7, 14, 21 and
28. An MATC should be determined for the most sensitive test
criteria measured (cumulative mortality of adult mys ids, number
of young per female, and body lengths of adult males and
females).
(vi) In addition to death, any abnormal behavior or
appearance should also be reported.
(vii) Distribution of mys ids among test chambers should be
randomized. In addition, test chambers within the testing area
should be positioned in a random manner or in a way in which
appropriate statistical analyses, can be used to determined the
variation due to placement.
(viii) The concentration of dissolved test substance (that
which passes through a 0.45 micron filter) in the chambers should
-------
EG-4
August, 1982
be measured as often as -is feasible during the test. The
coacentration of test substance shoul.d be measured: (a) in each
chamber at the beginning of the test and on days 7, 14, 21 and
28; and (b) in at least one appropriate chamber whenever a
malfunction is detected in any part of the test substance
delivery system. Among replicate test chambers of a treatment
concentration, the measured concentration of the test substance
should not vary more than -20 percent.
(5) [Reserved"]
(6) Analytical measurements — (i) Tes t chemical. Deionized
water should be used in making stock S9lutions of the test
substance. Standard analytical methods should be employed
whenever available in performing the analyses. The analytical
method used to measure the amount of test substance in a sample
should be validated before beginning the test by appropriate
laboratory practies. An analytical method is not acceptable if
likely degradation products of the test substance, such as
hydrolysis and oxidation products, give positive or negative
interferences which cannot be systematically identified and
corrected mathematically.
(ii) Numerical . (A) The number of dead rays ids , cumulative
young per female and body lengths of male and female rays ids
should be recorded during each definitive test. Appropriate
7
-------
EG-4
August, 1982
statistical analyses should provide a goodness-of-fit
determination for the day 7, 14, 21 and 28 adult (Gl) death
concentration-response curves.
(B) A 7-, 14-, 21- and 28- day LC50, based on adult (Gl)
death, and corresponding 95 percent confidence intervals should
be calculated. Appropriate statistical tests (e.g., analysis of
variance, mean separation test) should be used to test for
significant chemical effects on chronic test criteria (cumulative
mortality of adults, cumulative number of young per female and
body lengths of .adult male and females) on designated' days. An
MATC should be calculated using these chronic test criteria.
(d) Test conditions — (1) Test species — ( i ), Selection.
(A) The mys id shrimp, Mysidopsis bahia, is the organism
specified for these tests. Juvenile mys ids , _<_ 24 hours old, are
to be used to start the test.
(3) Mys ids to be used in chronic toxicity tests should
originate from laboratory cultures in order to ensure the
individuals are of similar age and experiential history. Mys ids
used for establishing laboratory cultures may be purchased
commercially or collected from appropriate natural areas.
Because of similarities with other mys id species, taxonomic
determinations should be verified by the commercial supplier or
by an appropriate individual.
-------
EG-4
August, 1982
(C) Mys ids used in a particular • tes t should be of similar
age and be of normal size and appearance for their age.
(D) Mys ids should not be used for a test if they exhibit
abnormal behavior, or if they have been used in a previous test,
either in a treatment or in a control group.
(ii) Acclimation. (A) Any change in the temperature and
chemistry of the water used for holding or culturing the test
organisms to those of the test should be gradual. Within a 24-
hour period, changes in water temperature should not exceed 1°C,
while salinity changes should not exceed 5 Percent.
(B) During acclimation mysids should be maintained in
facilities with background colors and light intensities similar
to those of the testing areas.
(iii) Care and handling. Methods for the care and handling
of mys ids such as those described in US EPA (1978) can be used
during holding, culturing and testing periods.
(iv) Feeding. Mysids should be fed during testing. Any
food utilized should support survival, growth and reproduction of
the mysids. A recommended food is live Artemia spp. nauplii
(approximately 48 hours old).
(2) Facilities--(i) Apparatus. (A) Facilities which may
be needed to perform this test include: (_1_) flow-through or
recirculating tanks for holding and acclimating mys ids ; (_2_) a
-------
EG-4
August, 1982
mechanism for controlling and maintaining the water temperature
during the holding, acclimation and test periods; (_3_) apparatus
for straining particulate matter, removing gas bubbles, or
aerating the water, as necessary; and (_4_) an apparatus for
providing a 14-hour light and 10-hour dark photoperiod with a 15-
to 30-mihute transition period. In addition, flow-through
chambers and a test substance delivery system are required. It
is recommended that mys ids be held in retention chambers within
test chambers to facilitate observations and eliminate1 loss
through outflow water.
(B) Facilities should be well ventilated and free of fumes
and disturbances that may affect the test organisms.
(C) Test chambers should be loosely covered to reduce the
loss of test solution or dilution water due to evaporation and to
minimize the entry of dust or other particulates into the
solutions .
(ii) Cleaning . Test substance delivery systems and test
chambers should be cleaned before each test following standard
laboratory practices.
(iii) Construction materials. (A) Materials and equipment
that contact test solutions should be chosen to minimize sorption
of test chemicals from the dilution water and should not contain
substances that can be leached into aqueous solution in
10
-------
EG-4
August, 1982
quantities that can affect the test results.
(B) Retention chambers utilized for confinement of test
organisms can be constructed with netting material of appropriate
mesh size.
(iv) Dilution water. (A) Natural or artificial seawater is
acceptable as dilution water if mysids will survive and
successfully reproduce in it for the duration of the holding,
acclimating and testing periods without showing signs of stress,
s uch as reduced growth and fecundity. Mys ids should be cultur'ed
and tested in dilution water from the same origin.
(B) Natural seawater should be filtered through a filter
with a pore size of < 20 microns prior to use in a test.
(C) Artificial seawater can be prepared by adding
commercially available formulations or by adding specific amounts
of reagent-grade chemicals to deionized or glass-distilled
water. Deionized water with a conductivity less than 1 u ohm/cm
at 12°C is acceptable as the diluent for making artificial
seawater. When deionized water is prepared from a ground or
surface water source, conductivity and total organic carbon (or
chemical oxygen demand) should be measured on each batch.
(v) Test substance delivery system. Proportional diluters,
metering pumps or other suitable systems should be used to
deliver test substance to the test chambers. The system used
11
-------
EG-4
August, 1982
should be calibrated before each test. Calibration includes
determining the flow rate and the concentration of the test
substance in each chamber. The general operation of. the test
substance delivery system should be checked twice daily during a
test. The 24-hour flow rate through a chamber should be equal to
a't least five times the volume of the chamber. The flow rates
should not vary more than 10 percent among chambers or across
t ime.
(3) Test parameters. Environmental parameters of the water
contained in test chambers should be maintained as specified
below:
(i) 'Temperature of 25 _+ 2°C.
(ii) Dissolved oxygen concentration between 60 and 105
percent saturation. Aeration, if needed to achieve this level,
should be done before the addition of the test substance. All
treatment and control chambers should be given the same aeration
treatment.
(iii) The number of mys ids placed in. a test solution should
not be so great as to affect results of the test. Loading
requirements for the test will vary depending on the flow rate of
dilution water. The loading should not cause the dissolved
oxygen concentration to fall below the recommended levels.
( iv) Photoperiod of 14 hours light and 10 hours darkness,
12
-------
EG-4
August, 1982
with a 15-30 minute transition period.
(v) Salinity of 20 _+_ 3 °/oo.
(e) Reporting. The sponsor should submit to the EPA all
data developed by the test that are suggestive or predictive of
chronic toxicity and all concomitant toxicologic
manifestations. In addition to the general reporting
requirements prescribed in Part 792--Good Laboratory Practice
Standards, the reporting of test data should include the
following:
(1) The source of the dilution water, its chemical
characteristics (e.g., salinity, pH, etc.) and a description of
any pre treatment.
(2) Detailed information about the test organisms, including
the scientific name and method of verification, average length,
age, source, history, observed diseases, treatments acclimation
procedures and food used.
(3) A descrip'tion of the test chambers, the depth and volume
of solution in the chamber, the way the test was begun (e.g.,
conditioning, test substance additions, etc.), the number of
organisms per treatment, the number of replicates, the loading,
the lighting, the test substance delivery system, and the flow
rate expressed as volume additions per 24 hours.
13
-------
EG-4
August, 1982
(4) The measured concentration of test substance in test
chambers at the times designated.
(5) The first time (day) that sexual characteristics can be
observed in controls and in each test substance concentration.
(6) The length of time for the appearance of the;first brood
for each concentration.
(7) The means (average of replicates) and respective 95
percent confidence intervals for:
(A) Body length of males and females at the first
observation day (depending on time of sexual maturation) and on
day 28.
(B) Cumulative number of young produced per female on day
28.
(C) Cumulative number of dead adults on day 7, 14, 21 and
28.
(D) If available prior to test termination (day 28), effects
on G2 mys ids. (number of males and females, body length of males
and females and cumulative mortality).
(8) The MATC is calculated as the geometric mean between the
lowest measured test substance concentration that had a
significant (P<0.05) effect and the highest measured test
substance concentration that had no significant (P>0.05) effect
in the chronic test. The most sensitive of the test criteria for
14
-------
EG-4
August, 1982
adult (Gl) mysids (cumulative number of dead mysids, body lengths
of males and females or the number of young per female) is used
to calculate the MATC. The criterion selected for MATC
computation is the one which exhibits an effect (a statistically
significant difference between treatment and control groups;
P<0.05) at the lowest test substance concentration for the
shortest period of exposure. Appropriate statistical tests
(analysis of variance, mean separation test) should be used to
test for significant chemical- effects. The statistical tests
employed and the results of these tests should be reported.
(9) Concentration-response curves should be fitted to the
cumulative number of adult dead for days 7, 14, 21 and 28. A
statistical test of goodness-of-fit should be performed and the
results reported.
(10) An LC50 value based on the number of dead adults with
corresponding 95 percent confidence intervals for days 7, 14, 21
and 28. These calculations should be made using the average
measured concentration of the test substance.
(11) Methods and data records of all chemical analyses of
water quality and test substance concentrations, including method
validations and reagent blanks.
(12) The data records of the holding, acclimation and test
temperature and salinity.
15
-------
EG-4
August, 1982
(f) References. U.S. Environmental Protection Agency,
1978. Bioassay Procedures for the Ocean Disposal Permit
Program. Environmental Research Laboratory, Office of Research
and Development Gulf Breeze, FL: EPA-6 00/9-78-010.
16
-------
ES-2
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
MYSID SHRIMP ACUTE AND CHRONIC TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Table of Contents
Subject Page
I. Purpose 1
II. Scientific Aspects 1
Test Procedures 1
General 1
Range-Finding Test 3
Definitive Test 4
Test Conditions 6
Test Species 6
Selection 6
Sources 7
Maintenace of Test Species 8
Handling and Acclimation 8
Feeding 8
Facilities 10
General 10
Construction Materials 11
Test Substance Delivery System 13
Tes t Ch ambe rs 14
Cleaning of Test System 15
Dilution Water 16
Controls 18
Carriers 18
Randomization 19
Environmental Conditions 19
Dissolved Oxygen 19
Light 20
Temperature and Salinity 20
Reporting 21
III. Economic Aspects 24
IV. References 26
-------
Office of Toxic Substances ES-2
August, 1932
I. Purpose
The purpose of this document is to provide the
scientific background and rationale used in the development
of Test Guideline EG-2 which uses Mys id shrimp to evaluate
the toxicity of chemical substances. The Document provides
an account of the scientific evidence and an explanation of
the logic used in the selection of the test methodology,
procedures and conditions prescribed in the Test
Guideline. Technical issues and practical considerations
relevant to the Test Guideline are discussed. In addition,
estimates of the cost of conducting the test are provided.
II. Scientific Aspects
A. Test Procedures
1. .General
The choice of mys id toxicity tests (96-hour static, 96-
hour flow-through or 28-day chronic is based on several
considerations. A static test requires less equipment,
fewer chemical analyses and disposal of smaller quantities
of contaminated wastewater than flow-through systems.
Static tests are a relatively easy means to evaluate and
compare the acute effects of a test substance.
The flow-through system more closely simulates the
natural exposure process, eliminating problems associated
with accumulation of organic material and toxic metabolic
products. Test substances are more thoroughly mixed in a
flow-through system and problems of sorption to suspended
sediments, feces and uneaten food are reduced. In order to
produce valid toxicity test results, the flow-through test
should be used with test substances which have a high oxygen
demand, are highly volatile, are unstable, biodegradable or
-------
ES-2
August, 1982
are removed in significant amounts by the test organisms.
A more comprehensive evaluation of the potential
environmental hazard of a test substance is available from
chronic toxicity testing. The chronic test using mys ids
provides two important advantages over other toxicity
testing regimes. First, it permits an evaluation of
response to chronic exposure to a test substance. Second,
it allows a determination of effects of the test substance
throughout sequential life s,tages of the organism (juvenile,
adult, egg). These data can be used to estimate potential
adverse population and community changes associated with
shifts in growth and reproductive potential.
For the acute tests, 96 hours is a convenient interval
of time for starting and completing a .test in a normal five-
day work week, and is better than shorter periods for
estimating accumulative and other chronic effects. Because
set-up is the most expensive portion of a test, a 96-hour
test is only slightly more expensive than 24 or 48 hour
tests. Yet additional data on the LC50's over time and the
observations of other abnormal effects that do not appear in
shorter tests are gained for this slight increase in cost.
Although the 43 hour test can reduce costs, eliminate the
necessity of feeding of the mys ids during the tes't, and make
the test more comparable to the 48 hour Daphnia acute
toxicity tests, the 96-hour toxicity test was selected for
the mys id test guidelines because of greater potential for
determining the incipient LC50 (threshold limit for acute
toxicity) through extension of the toxicity curve. In
situations where the 96 hour LC50 does not permit estimation
of an incipient LC50 (lethal threshold concentration),
-------
ES-2
August, 1982
continuation of the test may allow better estimation of the
toxic effects of the test substance. In a review of 375
toxicity tests, Sprague (1969) found that a lethal threshold
clearly had not been reached in 42 tests, while in 122 other
tests the threshold was reached in four days or longer.
Because of the importance of the incipient LC50 in hazard
assessment, continuation of the acute toxicity tests past
the 96 hours is recommended for those test substances which
do not elicit a threshold concentration within the four day
test period.
For the life cycle test, , a 28-day experimental period is
used to permit testing through at least one complete life
cycle in Mysidopsis bahia at 25°C. Juveniles utilized for
tests reach sexual maturity within 12-14 days under normal
conditions at this temperature (US EPA 1978). However, test
substances may delay sexual maturity several days (Nimmo et
al. in press c). Tests longer than 28 days are not
recommended because of possible fouling of retention
chambers with subsequent decreases in the efficiency of the
flow-through system.
2. Range-Finding Test
The concentration range for the definitive test is
normally chosen based on the results of a range-finding
test. In the acute static and flow-through test, the range-
finding test also'serves as a means of determining which
life stage (juvenile or young adult) is most sensitive to
the test substance and should be used in the definitive
test. For the acute tests, range-finding tests are normally
short-term (24-96 hour), static or flow-through bioassays,
which utilize fewer organisms per test substance
-------
ES-2
August, 1982
concentration than the definitive test. For the life cycle
test, range-finding tests may take the form of acute, flow-
through tests using different life stages or determination
of incipient LCSO's to allow selection of appropriate
definitive test concentrations. In all cases, the range-
finding test is conducted to reduce the expense involved
with having to repeat a definitive test due to inappropriate
test substance concentrations.
3. Definitive Test
The concentration range for the definitive test is
chosen based on the results of preliminary range-finding
i ! ' '.
tests. By testing a minimum of five concentrations in the
acute bioassays, partial kills both above' and below the
median 50 percent mortality level are probable and will help
define the concentration-response relationship. The more
partial kills, the better the definition of the
concentration-response curve. The slope and shape of the
concentration-response curve may give insight into possible
mechanisms of action of a chemical and will allow estimation
of the effects of lower concentrations upon test
organisms. In addition, by having partial kill data, a
greater array of statistical methods can be used to
determine the LC50.
The utilization of the most sensitive of the two life
stages (juvenile or young adult), as required for the
definitive acute mys id tests, is based on evidence that
these two life stages exhibited differential mortality to
eleven pesticides (Nimmo et al . in press b) . The procedure
of testing both stages in the range-finding test and using
the most sensitive life stage in the definitive test permits
-------
ES-2
August, 1982
determination of toxicity data at less expense than
conducting two complete definitive tests.
Because of the increased expense of the 28-day mysid
life cycle test, the nature of the recorded data and
equipment limitations, testing of a minimum of only four
concentrations is required. These observations will allow
determination of the MATC (maximum acceptable toxicant
concentration) limits for the most sensitive life cycle
criterion recorded (mortality, body lengths of males and
females and numbers of young per female).
The jfujimber of mys ids exposed to each test substance
concentration (e.g. 20 in acute tests and 40 in chronic
test) is designed to allow adequate numbers for statistical
evaluation even with the presence of partial mortality
(Nimmo et al. 1977, 1978, US EPA 1978).
Measurement of test substance concentrations at
designated periods during static and flow-through tests
allows documentation of real test concentrations at
appropriate periods under acute and chronic conditions.
Chemical and physical parameters (temperature, pH,
dissolved oxygen and salinity) are recorded at specified
times to permit evaluation of the biological conditions
present for mysid survival in test solutions.
Specified observations on mortality and life cycle
characteristics are designed to allow an adequate evaluation
of concentration-response effects in both acute and chronic
mysid tests. In addition, these defined observation times
allow greater comparability of dose-response data between
different chemicals and laboratories.
-------
ES-2
August, 1982
B. Test Conditions
1. Test Species
a. Selection
The primary considerations in the selection of this
organism for toxicity testing were: (a) sensitivity to a
variety of chemical substances; (b) geographical
distribution and abundance, (c) ecological importance and
(d) existence of established culture methods for laboratory
rearing.
The test species, Mysidopsis bahia, is a member of the
family Mys(idae, which is found in most of the neritic zones
of the world's oceans. Mysidopsis bahia inhabits shallow
water grass flats along the eastern and western Gulf of
Mexico. They are particularly abundant from the Galveston
Bay system to southern Florida.
Mys ids occupy an important position in near shore food
webs. They constitute a major source of food for many fish
species, including catfish, flounder, anchovy, silvers ide,
sunfish and seatrout (Darnell 1958, Schuster 1959, Odum and
Herald 1972, Powell and Schwartz 1979). In addition to
their role in food chains of fish, mysids are important in
the conversion of organic detritus to living tissue in
estuarine environments (Hopkins, 1965).
To date, Mysidopsis bahia has been the most extensively
tested mys id shrimp (Nimmo et al. 1977, 1978, In press a, in
press b, in press c, USEPA 1978, Gripe et al. in press).
During its development as a test species (since 1977)
methods of culturing, holding and testing have been
established, and the methodologies developed at the EPA Gulf
Breeze Environmental Research Laboratory were considered
-------
ES-2
• Augus t, 1982
heavily in the design of this guideline. This laboratory,
and others, have established Mysidopsis bahia as a test
organism and have stressed the importance of the following
qualities: (1) small size; (2) short (18 day) life cycle;
(3) small brood size; (4) readily reared through all life
stages in culture; (5) commercially available (6)
representative of an ecologically important family and (7)
extremely sensitive to a variety of test substances. In
addition, this species has been selected as a test organism
for a variety of assessment programs in EPA and other
government agencies, as well as private laboratory
testing. The sensitivity characteristic of mys ids was
dramatically reported by Bionomics (EPA Contract No. 68-01-
4646). In testing of the acute toxicity (no effects
concentrations) of 35 priority pollutants, mysids were found
to be on the average more sensitive than any of the other
species tested (i.e., Selenestrum capricornutum, Skeletonema
cos tatum, Daphnia mag n a, Cypr i hod on variega tus and Lepomis
machrochirus ) .
b. Sources
It is recommended that the mys ids used for laboratory
testing as specified in this test guideline be obtained
commercially from a supplier willing to certify proper
taxonomic identification. Although field collection is
acceptable, it is highly recommended that test organisms
originate from culture stock. Definitive identification of
Mysidopsis bahia (Molenock, 1969), is difficult without
expertise, and it occurs sympatrically with two other
species of Mys id ops is . Reliable use of Mysidopsis bahia is
required for this testing procedure, and it is therefore
-------
ES-2
August, 1982
suggested that the mysids be cultured in the laboratory to
meet all testing needs. The ease with which this species
can be cultured -in the laboratory has been demonstrated
(Nimmo et al .' 1977, 1978).
Furthermore, laboratory culturirig of mys ids permits the
isolation of newly-hatched juveniles. This allows control
of the size, age and experiential history of mys ids used for
acute and chronic testing.
2. Maintenance of Test Species
a. Handling and Acclimation
The bay mys id, _M_. bahia, may be cultured in aquaria
supplied with either filtered flowing or recirculating
seawater. Details of _M. bahia culture can be obtained from
Nimmo et al. (1977) and USEPA (1978). A salinity of 20 °/oo
for mys id culture allows optimal reproductive conditions
(Nimmo et al. 1977, USEPA 1978) and reduces acclimation
problems related to transfer of animals from culture to test
water.
b. Feeding
Artemia spp. nauplii suggested for mys id feeding, can be
reared in the laboratory from commercially available eggs.
These eggs or any other appropriate food used for mys ids,
should not be used if the total organochlorine pesticide
plus oolychlorinated biphenyls exceeds 0.3 ug/g (wet
weight), or if organic chlorine exceeds 0.15 ug/g (v?et
weight). A recent study by Johns and Walten (1979) reported
that adult Mys id ops is bah i a fed Artemia spp. from San Pablo
Bay, California exhibited increased mortality, did not
reproduce and showed reduced growth rates. In contrast,
both juvenile and adult mys ids fed Artemia spp. strains
-------
ES-2
August, 1982
collected from Brazil, Australia, Italy and Utah maintained
high survival and growth rates. Additional studies have
indicated major differences in the nutritional value of
Artemia to brachyuran crustaceans (Johns et al. in press).
These results strongly imply that nutritional quality of
Artemia, possibly associated with pesticide or heavy metal
contamination, can significantly influence test results and,
therefore, should be considered.
In order to separate Artemia sop, nauplii (used for
mys id feeding) from their egg cases ,and other debris, a
light box may be employed. This system makes use of the
positive phototropism of Artemia to separate nauplii from
unwanted materials. It is important to isolate the nauplii
l
from the 'egg cases and to deliver only nauplii to the test
chambers in order to minimize build-up of organic debris
within the chambers. The decomposition of the entrapped egg
cases may directly or indirectly enhance or reduce the
i i
toxici'tiy of the test substance.
To isolate juvenile mys ids, ovigerous females may be
placed within a retention chamber, which is then submerged
into a five-liter glass battery jar or other suitable vessel
(USEPA 1978). The retention chamber should be slightly
smaller than the battery jar and should extend above the
water level of the battery jar. A slow flow of salt water
(approximately 4 drops/second) should be dripped into the
jar to sustain proper dissolved oxygen levels and prevent
stagnation. Juveniles pass through the cylinder mesh (one
millimeter mesh opening) at birth and attach to the walls of
the battery jar; thereby minimizing cannibalism by adult
females and facilitating capture. During this isolation
-------
ES-2
August, 1982
procedure, mysids should be given a supply of 24-hr old
Artemia spp. (nauplii). Juvenile mys ids may be removed from
the sides of the jar every 4-12 hours during this period
using a glass tube.
3. Facilities
a. General
In flow-through systems it may be necessary to have the
capability to vary and maintain the water temperature.
Since the temperature of the dilution water can be expected
to vary both daily and seasonally, facilities for adjusting
temperature may be needed to .maintain ; the desired culturijng,
holding, and testing temperatures. Filters are needed to
remove particulates and biological material from the
dilution water so the diluter system and retention chambers
will not become clogged, cause a change in test substance
concentrations, or lead to stagnation and oxygen depletion
within the retention chambers. The primary concera is to
minimize the confounding of results associated with the
differential sorption of the test substance on cell walls,
clay particles, etc. which in turn may enhance or reduce the
availability of the test substance to the mys ids.
To minimize these problems, the dilution water should be
filtered through a 20 micrometer or smaller pore-size filter
to sufficiently reduce the amount of suspended sediments,
organic material and biological organisms (phytoplankton,
zooplankton, fungi, bacteria, etc.).
Requiring filtration through a 20 micrometer or smaller
filter is based on recent modifications to the testing
procedure developed at the EPA Gulf Breeze Environmental
Research Laboratory (USEPA 1978). In addition, filtration
10
-------
ES-2
August, 1982
to five micrometers more effectively controls fouling of the
retention chamber mesh walls. Minimizing this problem also
prevents dissolved oxygen depletion and stagnation within
the retention chambers. Filtration through five micrometer
filters is attainable for the recommended flow rate and
quantity of water heeded to conduct the test.
Gas accumulation may also cause adverse effects and
therefore, a device which removes air bubbles may be
necessary. A suitable device is described by Penrose and
Squires (1976). When the dissolved oxygen in the. dilution
water is less that 60 percent, a device is needed to aerate
the water. Culture techniques recommend 70-100 percent
saturation for other marine crustaceans (Forster and Beard
of the test substance during aeration through
volatilization, aeration should be, conducted prior to
introduction of the test substance.
In order to attain optimal test results, it is necessary
to culture and test1 the organisms in an environment
considerate of both their behavioral and physiological
needs. Mys id shrimp are extremely sensitive to fluctuations
in these parameters which may be reflected in a number of
ways, all of which can affect test validity (Bahner et al.
1975) .
b. Construction Materials
All pipes, tanks, holding chambers, mixing chambers,
metering devices, and test chambers should be made of
materials that minimize the release of chemical contaminants
into the dilution water or the adsorption of the test
substances. Chemicals that may leach from construction
materials can stress test organisms, or possibly act
11
-------
ES-2
August, 1932
synergistically or antagonistically with test substances to
give inaccurate results. Generally, undesirable substances
are not leached from borosilicate glass, titanium, and
perfluorocarbon plastic. In addition, the tendency of these
muterials to absorb substances is minimal. Rubber,
polyvinyl chloride, copper, brass, galvanized metal, lead
and epoxy resins should not come in contact with dilution
water, stock solution, or test solutions because of the
toxic substances they contain. Cast iron should not be used
in water systems since rust may develop and result in
fouling. Teflon (Algoflon), Perspex, Polyethylene, Tygon,
Polypropylene, Polycarbonates (Makrolor) and Polyester
(Gabraster) have been shown to be non-toxic and suitable for
experiments with marine organisms (APHA 1975, USEPA 1978).
Retention chambers, aquaria delivery systems, pipes and
any tank exposed to solutions that may come in contact with
the organism should not be soldered or brazed, since lead,
tin, copper or zinc may be leached. Silicone adhesive is
the preferred bonding agent for all construction
materials. It is inert, and the solvent it generally
contains (acetic acid) is easily washed away or volatilized
from the system. A minimum amount, of the adhesive should
contact the test solution because it may absorb test
materials. If large amounts of the adhesive are needed for
strength, it should be applied to the outs ides of chambers
and apparatus to minimize contact.
In static testing, borosilicate glass, crystallizing
dishes, or similar containers may be used as test chambers.
Use of these dishes will minimize sorption of the test
substance into the chamber walls and minimize residues of
12
-------
E3-2
August, 1982
test substances or metabolites remaining after cleaning.
Glass plates can be placed over the test chambers to allow
for stacking and to minimize space requirements and
evaporative loss of test solution. Chamber size should be
adequate to insure proper loading and to minimize
cannibalism.
c. Test Substance Delivery System
In flow-through tests, the delivery of constant
concentratio-ns of test substances is required to reduce
variability in test results. Large, fluctua tions in test
substance concentration will! give abnormally high or j low
responses, depending upon the mechanism of toxic actions.
Proportional diluters with metering pumps or continuous flow
infusion pumps have been used extensively to maintain
constant test substance concentration. For the flow-through
acute and life cycle test guideline, all tests should be
conducted in intermittent flows from a diluter or in
continuous flow with the test substance added by an infusion
pump. The procedures of Mount and Brungs (1967) and Hansen
et al. (1974) are recommended if the test substance can be
added without a carrier; the device described by Hansen et
al. '(1974) if a carrier is necessary; or procedure of Banner
et al. (1975) if pumps are required for continuous flow.
Proportional diluters operate on a sequential filling
and emptying of water chambers. The water chambers are
calibrated to contain a measured amount of water. Separate
water chambers can be provided for test substance and
diluent water. Diluent and test substance waters are mixed
and delivered to the test aquaria. The cyclic action of the
diluent is regulated by a solenoid valve connected to the
13
-------
ES-2
August, 1982
inflow dilution water. The system is subject to electrical
power failure, so an alternate emergency power source is
recommended. In addition, the mysids should to be shielded
from the clicking sound of solenoid valves. If unshielded
from this disturbance, mys ids may jump out of the test
solution and stick to the sides of the retention chambers
(US EPA 1978) .
The proportional diluter is probably the best system for
routine use; it is accurate over extended periods of time,
is nearly trouble free, and has fail-safe provisions (Lemke
et al. 1978). ,A small chamber to promote mixing ,of test
substance-containing water and dilution water may be used
between the diluter and the test aquaria for each
concentration. If replicate , chambers are used in this test,
separate delivery tubes should be run from this mixing
chamber to the appropriate replicate chambers. If an
infusion pump is used, a glass baffle should be employed to
insur e mixing of the test substance and dilution water.
Calibration of the test substance delivery system should be
checked carefully before and during each test. This should
include determining the flow rate through each test aquarium
and measuring the concentration of test substance in each
test aquarium. The general operation of the system should
be checked twice daily.
d. Test Chambers
Retention chambers are suggested to prevent escape of
mys ids from the test system, reduce cannibalism and
facilitate counting and observation. Overcrowding enhances
cannibalism and assignment of five mys ids per retention
chamber is recommended to minimize this oroblem. The mesh
14
-------
SS-2
August, 1982
size (315 micrometer mesh opening) of the screen used in
construction of the retention chambers at the EPA laboratory
in Gulf Breeze minimizes the problems associated with
fouling by fungal, bacterial and algal growth, yet is still
small enough to retain the rays ids and food organisms. This
mesh size is slightly more porous than the 200 micrometer
opening mesh recommended by Nirnmo et al. (1977). Use of the
larger mesh size combined with 20 micrometer filtration
obviates the need for continuous lighting, which was
employed in early testing protocols (US EPA 1978) ,to minimize
fouling of the retention chambers.
e. Cleaning of Test System
Standard laboratory Practices (e.g, US EPA 19/4) are
recommended to remove dust, dirt, other debris, 'and residues
from test facilities. At the end of a test, test systems
should be washed in preparation for the next test. This
will prevent chemical residues and organic matter from
becoming embedded or absorbed into the equipment. It is
also recommended that any silicon cement which has been
exposed to a test substance is replaced prior to future
tests to avoid contamination due to sorption properties.
Rinsing and priming the system with dilution water
before use (conditioning) allows equilibrium to be reached
between the chemicals in the water and the materials of the
test system. The test system may sorb or react with
substances in the dilution water. Allowing this equilibrium
to take place before use lessens the chances of water
chemistry changes during a test.
Even after extensive washing, new facilities may still
contain toxic residues. The best way to determine whether
15
-------
ES-2
August, 1982
toxic residues remain is to rear mys ids through at least one
complete life cycle. If the mys ids survive and successfully
reproduce the test facililties can be considered to be free
of toxic residues.
f. Dilution L'ater
A constant supply of dilution water is required to
maintain constant experimental conditions. An interruption
in flow or change in water quality can change the chemistry
of the test system and possibly the response of the test
popalation. Therefore, the results of a test with variable
dilution water quality are n,ot ,cpmparable to tests run under
constant conditions and they are more difficult to
interpret.
For acute and chronic toxicity tests, .a minimum
criterion for acceptable dilution water is that healthy
mys ids will survive and reproduce in it without showing
signs of stress such as abnormal behavior ,(erratic swimming,
loss of equilibria i or lack of feeding activity).
Natural seawater, obtained from a source with similar
characteristics as those designated for the test species or
water from an area where the test organisms were obtained,
is preferable to artificial seawater. Dilution water should
be of constant quality and should be uncontaminated.
Contaminated water can affect results both directly and
indirectly. If natural seawater is used, it should meet the
following specifications for contaminant levels (APHA 1975).
16
-------
ES-2
August, 1982
Suspended solids < 20 mg/1
Total organic carbon < 10 rag/1
Un-ionized ammonia < 20 mg/1
Residual chlorine < 3 ug/1
Total Organophosphorous pesticides < 50 ug/1
Pesticides plus PCB's < 50 ug/1
In addition, water used to make reconstituted seawater
should meet or exceed the same water quality criteria.
Maintenance of desired salinities in the test aquaria
throughout the duration of the test may j pos e a problem.
When possible, water from an area of high salinity shoulds
be used; low salinities can then be obtained by adding
distilled or deionized water as needed. To increase
salinity, a strong, natural brine, which can be obtained by
freezing and then partially thawing seawater, can be used.
This procedure is suitable if limited amounts of seawater
are needed; however, it is recommended that artificial
seawater salts be used when large increases in salinities
are required (APHA 1975).
Due to the volume of water necessary to conduct a
chronic test and technical problems associated with
conditioning of the dilution water, the use of reconstituted
seawater for these tests may not be currently feasible
because of high costs and lack of information on proper
aging processes. Research is needed to determine methods in
which reconstituted water can be conditioned and aged with
the use of appropriate storage and selective filtration
before use of reconstituted seawater can become a viable
alternative.
17
-------
ES-2
August, 1982
g. Controls
Controls are required for every test to assure that any
effects which are observed are due to the test substance and
not to other factors. These may include effects from
construction materials, environmental fad'cors, nutritional
quality of food, vapors, stressed test organisms, etc.
Within the acute, 96-hour tests ten percent control
mortality may be present due to inherent biological
factors. Any increas'e above this is considered to be due- to
A \
conditiions of the test or the test organisms. The ten
percent mortality figure is representative for a wide
variety of organisms, including both fish and invertebrates
(ASTM, 1979) and is generally utilized for 96-hour testing.
Some ot this mortality in invertebrates may be associated
with injury during handling.
In an analysis of thirteen life cycle tests which
utilized _M_. bahia (Nimmo, unpublished laboratory data; Gulf
Breeze EPA Laboratory), control mortality over the testing
period (20-28 days) ranged from 0-31 percent. The mean
control mortality of these studies was 11 percent, with 34
percent of the test results greater than the 11 percent.
Control mortality of 20 percent in the chronic test will be
considered to be due to conditions of the test or the test
organisms .
h. Carriers
Carriers can effect test organisms and can possibly
alter the form of the test substance in water. For these
reasons it is preferable to avoid the use of carriers in
toxicity tests unless absolutely necessary to dissolve the
test substance. Since carriers can stress or adversely
18
-------
E3-2
August, 1982
effect test organisms, the amount of carrier should be kept
to a minimum. Recommended maxima are 0.5 ml/1 in static
tests and 0.1 ml/1 in flow-through tests (APHA 1975).
Triethlyene glycol has been found to exert the least
influence on mys id response to test s ubs-tances of several
carriers that have been used. Acetone and ethanol have a
stronger tendency to reduce the surface tension of the
water, and therefore, decrease "oxygen saturation (Veith and
Comstock 1975, Xrugel et al. 1978, APHA 1975).
i. Randomization
The test chamber position in the testing area andi
assignment of mys ids to test chambers are randomized to
prevent conscious or unconscious biases from being
introduced. These biases can be in environmental conditions
and distribution, dilutor system function, etc.
4. Environmental Conditions
a. Dissolved Oxygen
In flow-through testing, large variations in flow rates
to aquaria will result in undesirable differences in
exposure and test conditions between aquaria. Parameters
such as dissolved oxygen and test substance concentration
can decrease more rapidly in aquaria with low flow rates;
similarly metabolic products can build-up under these
conditions. Proper dilution water filtration and mesh size
of retention chambers are of utmost importance in
maintaining flow rate of test solutions and exchange within
retention chambers. Previous studies (Nimmo et al. 1977,
USEPA 1978) have found that a flow rate which allows five
test chamber volume changes per 24 hours is adequate to
obtain necessary conditions for mys id testing.
19
-------
ES-2
August, 1982
b. Light
A 14-hour light and 10-hour dark photoperiod is
recommended in an effort to approximate representative
natural conditions under which _M. bahia is found and thus
reduce test-related stress. If used, this photoperiod
should remain constant throughout culturing, holding and
testing, as any deviations could effect test results.
Mys ids appear very sensitive to light changes and
intensity- At very high intensities swimming is inhibited,
with mys ids sinking to the tank bottom. In very dim light,
shoaling behavior is disrupted (Steven 1961). With use, of, ;a
light-dark regime a transition period is recommended.
Mys ids are known to react to sudden light changes by jumping
out of the water. Gradual transition will avoid ' mortal ity
caused by mys ids "sticking" to test chamber walls.
c . Temperature and Salinity
Test temperature and salinity choices ( 25°± 2°C and 20
± 3 °/oo, respectively) were made after revie / of reports on
the habitat characteristics of _M. bahia (Nimmo et al . 1977,
Price 1978). These temperature and salinity ranges were
also found to produce the greatest reproductive success in
laboratory cultures of M_. bahia (Nimmo et al. 1977, US EPA
1973). Furthermore, minimizing variability in testing
conditions through specific temperature and salinity values
allows greater comparability of interlaboratory test results
and the development of a comparative toxicology data base.
An acceptable method for maintaining desired temperature and
salinity ranges in flow-through bioassays with marine
organisms is described in Bahner and Nimmo (1975).
20
-------
ES-2
August, 1982
C. Reporting
A coherent theory of the dose-response relationship, on
which the acute toxicity tests are based, was introduced by
Bliss (1935), and is widely accepted today. This theory is
based on four assumptions:
(1) Response is a positive function of dosage, i.e. it
is expected that increasing treatment rates should
produce increasing responses.
(2) Randomly selected animals are normally distributed
with respect to their response to a toxicant.
(3) Due to hq>meos tas is , response magnitudes are
proportional to the logarithm of the dosage, i.e. it
takes geometrically increasing dosages (stresses) to
produce arithmetically increasing responses (strains) in
test animal populations.
(4) In the case of a direct dosage of animals, their
resistance to effects is proportional to body mass.
Stated another way, the treatment needed to produce a
given response is proportional to the size of the
animals treated.
The concentration-response curve, where percent
mortality is plotted as a function of the logarithm of test
solution concentration, can be interpreted as a cumulative
distribution of tolerance within the population (Hewlett and
Plackett 1979). Experiments designed to measure tolerance
directly (Bliss 1944) have shown that tolerance is
lognc-rmally distributed within an experimental population in
most cases. Departures from the lognormal pattern of
distribution are generally associated with mixtures of very
susceptible and very resistant individuals within a
21
-------
ES-2
August, 19 8 2
population (Hewlett and Plackett 1979). In addition,
mixtures of toxicants can produce tolerance curves which
deviate significantly from the lognormal pattern (Finney
1971) .
If tolerances are lognormally distributed within the
experimental population, the resulting concentration-
response curve will be sigmoidal in shape, resembling a
logistic population curve (Hewlett and Plackett 1979).
While estimates for the mean lethal dose can be'made
directly from the dose response curve, a linear
transformation often is possible, using pro'oit (Bliss 1934,
Finney 1971) or logit (Hewlett and Plackett 1979)
trans formations .
Once the mortality data have |been transformed, a
straight line can be fitted to the points by a least
5 qua res
linear regression equation and confidence limits can be
determined for predicted mortality values. An additional
advantage is that the significance of the slope of the
regression line can be determined (Draper and Smith 1976).
While the mean lethal dose (LC50) can be estimated
graphically from the linearized dose-response curve (APHA
1975), other techniques are preferable since the graphical
method does not permit the calculation of confidence limits.
The probit method (Finney 1971), which is recommended in
the acute toxicity test guideline, uses the probit
transformation and the maximum likelihood curve-fitting
technique. Other appropriate tests used in data reduction
include the modified pro'oit method of Litchfield and
Wilcoxon (1949), the logit method (Ashton 1972) and the
22
-------
ES-2
August, 1982
moving average method (Thompson 1947).
If there are no partial kills in an experiment,
determination of the concentration-response curve is not
possible. In situations where there are no partial kills,
the binominal test (Siegel 1956) can be used to estimate the
LC50 and confidence limits around the LC50 value (Stephan
1977) .
If concentration-response data are plotted for each 24-
hour interval throughout the test, the LC50 determined from
each curve can be plotted as a function of time, yielding a
time-acute toxicity curjVe (APHA 1975). This curve
approaches a line parallel to the time axis asymptotically,
indicating a constant or threshold value for LC50. The
absence of a threshold LC50 may indicate the need for an
acute test of longer duration.
The statistical tests recommended for analyses of mys id
life-cycle data (mortality, body lengths of males and
females and young per female) were chosen to permit as
complete an interpretation of the quantifiable data as
possible. Under most conditions, the analysis of variance
(ANOVA) is a powerful statistical test allowing the
determination of significant differences between treatment
means through incorporation of data variability. This
statistical examination is especially important in
biological experimentation due to the presence of many
sources of inherent variability. In previous chronic mys id
testing, Nimmo et al. (In press b) employed the analysis of
variance with subsequent comparisons between means utilizing
Student-New-Kuels, Duncan's, Dunnett's, or Bonferron's
tests. Futhermore, the use of analysis of variance (ANOVA)
23
-------
ES-2
August, 1982
and mean separation tests have been employed in mysid
testing at the EPA Environmental Research Laboratory at Gulf
Breeze, Florida, and have given consistent results under the
experimental conditions stated .in the test guideline
document.
III. Economic Aspects
The Agency awarded a contract to Enviro Control, Inc. to
provide us with an estimate of the cost for performing
static and flow-through acute toxicity tests and flow-
through chronic toxicity tests. Enviro Control supplied us
with two estimates; a protocol! estimate and a ^Laboratory,
survey estimate.
Protocol Estimates
range mean
Acute (static and flow through) $ 322-$ 965 $ 643
Chronic $1653-$4959 $3306
These estimates were prepared by separating the
guidelines into individual tasks and estimating 'the hours
used to accomplish each task. Hourly rates were then
applied to yield a total direct labor charge. An overhead
rate of 115 percent, other direct costs ($50-acute, $415-
chronic), a general and administrative rate of 10 percent
and a fee of 20 percent were then added to the direct labor
charge to yield the final estimate.
24
-------
ES-2
August, 1982
Laboratory Survey Estimates
range mean
Acute (static and flow through) $ 340-$ 1250 $ 743
Chronic - $3000
The laboratory survey estimates were compiled from three
laboratories for the acute guideline and one laboratory for
the chronic guideline.
25
-------
Augus t,
ES-2
1982
IV. References
APHA. 1975. American Public Health Association.
American Water Association and Water Pollution Control
Federation. Standard methods for examination of water
and waste water, 14th ed. New York: American Public
Health Association.
Ash ton WD. 1972.
Publishing Co. New
The log it transformation.
York:
Hafner
ASTM. 1979. American Society for. Testing and
Materials. New standard practice for conducting basic
acute toxicity tests with fishes, macroinvertebrates and
amphibians. Philadelphia, PA: American Society for
Testing and Materials.
Banner LH and Nimmo DW. 1975. Methods to assess effects
of combinations of toxicant, salinity and temperature on
estuarine animals. Proc. Univ. Mo.'Ann. Conf. Trace
Subst. Environ. Health pp: 167-177.
Bahner LH, Wilson AJ, Shepppard JM, Patrick JM, Goodman
LR, and Walsh GE. 1977. Kepone bioconcentration,
accumulation, loss and transfer through estuarine food
chains. Chesapeake Sci. 13:299-308.
1975. A saltweater
Bahner LH, Craft CD, Nimmo DR. 1975. A saltweater
flow-through bioassay method with controlled temperature
and salinity. Prog. Fish. Cult. 37 ( 3 ) : 1 26-1 29 .
Berkson J. 1949. The minimum chi-square and maximum
likelihood solution in terms of a linear transform, with
particular reference to bioassay. J. Amer. Stat.
Assoc. 44:273-278.
Bionomics Inc. 1978. In-depth studies on health and
environmental impacts of selected water pollutants. EPA
Control No. 68-01-4646, 1977-1978.
Bliss
39.
CI. 1934. The method of probits. Science 79:38-
Bliss CI. 1934.
mortality curve.
The calculation of
Ann. Appl. Biol.
the dosage-
22:134-307.
26
-------
ES-2
August, 1982
Bliss CI. 1944. The U.S.P. collaborative rat assay
£or digitalis. J. Amer. Pharra. Ass. 33:225-245.
Gripe GM, Nimrao DR, Hamaker TL. In press. Effects of
two organophosphate pesticides on swimming stamina of
the mysid, Mysidopsis bahia. In: Vernberg FJ, Calabrese
A, eds . Pollution and physiology of marine organisms.
New York: Academic Press.
Darnell RM. 1958. Food habits of fishes and large
invertebrates of Lake Ponchatrain, Louisiana, an
estuarine community. Publ. Inst. Marine Sci. Univ.
Texas, 5:353-416.
Draper NR. and Smith H. 1966. Applied regression
analysis. New York: John Wiley and Sons.
Drumrnond RA and Dawson WF. 1970. An inexpensive method
for simulating diel pattern of lighting in the
laboratory. Trans. Amer. Fish. Soc. 99:434-435.
Finney AJ. 1971. Probit analysis. London, England:
Cambridge University Press.
Foster JRM and Beard TW. 1974. Experiments to assess
the suitability of nine species of prawns for intensive
cultivation. Aquaculture. 3:355-368.
Hansen DJ, Schimmel SE, Matthews E. 1974. Avoidance
of Aroclor 1254 by shrimp and fishes. Bull. Environ.
Contam. Tox . 12( 2 ) : 243-2 56.
Hewlett PS. and Plackett RL. 1979. The interpretation
of quantal responses in biology. University Park Press,
Baltimore, MD: 82 pp.
Hopkins TL. 1965. Mysid shrimp abundance in surface
waters of Indian River Inlet, Delaware. Chesapeake Sci.
6:36-91.
Johns DM and Walton W. 1979. International Study on
Artemia: X. Effects of food source on survival, growth,
and reproduction in the mysid, Mysidopsis bahia.
Abstract cited in Amer. Zoolog. 19(3):906.
27
-------
ES-2
August, 1982
John DM, Peters ME, Beck AD. In press. International
Study on Artemia (1): Nutritional value of geographical
and temporal strains of Artemia. Effects on survival
and growth of two species of brachyuran larvae. In:
Personne, Sorgeloos, Roels, Jaspers, eds. The brine
shrimp Artemia Vol. 3. Wetterner, Belgium: Universa
Press .
Klein-' ac Phee C, Howell WH, Beck AD. In press.
International study on Artemia. Nutritional value of
five geographical strains of Artemia to winter flounder
( Psuedopleuronectes Americanus") larvae . In: Personne,
Sorgeloos , Roels , Jaspers , eds. The Brine shrimp
Artemia Vol. 3. Wetterner, Belgium: Universa Press.
Krugel S,< Jenkins D, Kleen SA. 1978. Apparatus for the
continuous dissolution of poorly water-soluble compounds
for bioassays. Water Res. 12:269-272.
Lemke SE, Brungs WA, Hilligan B J. 1978. Manual for
construction and operation of toxici ty-tes ting
proportional diluters. EPA Report No. 600/3-78-072.
Litchfield JT, Jr. and Wilcoxon F. 1949. A simplified
method of evaluating dose-effect experiments. J. Pharm.
Exp. Ther. 96:99-113.
Molenock'J. 1969. Mysidopsis bahia, a new species of
mys id (Crustacea: Mysidacea) from Galveston Bay,
Texas. Tulane Studies in Zool. 15: 113-116.
Mount DI. and Brungs WA. 1967. A simplified dosing
apparatus for fish toxicology studies. Water Res. 1:21-
29.
Nimmo DR, Bahner LH, Rigby RA, Sheppard JM, Wilson A J,
Jr. 1977. Mys id ops is bah i a; an estuarine species
suitable for life-cycle toxicity tests to determine the
effects of a pollutant. In-: Aquatic toxicology and
hazard evaluation, ASTM STP 634. FL Mayer and Hamelink
JL, eds. Philadelphia, PA: American Society for Testing
and Materials, pps. 109-116.
28
-------
ES-2
August, 1982
Nimmo DR, Rigby RA, Bahner LH, Sheppard JM. 1978. The
acute and chronic effects of cadmium on the estuarine
mysid, Mys id ops is bah i a. Bull. Environ. Contam. Tox . 19:
80-85.
Nimmo DR, Hamaker TL, Matthews E, Young WT. In press
a. The long-term effects of suspended sediment on
survival and reproduction of the rays id shrimp,
Mys id ops is bahia in the laboratory. Marine Ecosystems
Analysis. National Oceanograon ic and Atmospheric
Administration, Department of-Commerce. Boulder,- CO.
Nimmo DR, Hamaker TL, Matthews E, Moore JC. In press
b. Effects of. eleven pesticides on Mys id ops is bahia
throughout its life cycle. In: Vernberg FJ and
Calabrese A, eds. Pollution and physiology of marine
organisms. New York: Academic Press.
Nimmo DR, Hamaker TL, Moor JC, Wood RA. In press c.
Acute and chronic effects of Dimilin on survival and
reproduction of Mys id ops is bahia. Philadelphia, PA:
American Society for Testing and Materials.
Odum WE and Heald EJ. 1972. Trophic analysis of an
estuarine mangrove community. Bull. Marine Sci. 22:
671-738.
Olney CE, Schauer PS, Simpson KL. In press..
International study on Artemia. Comparison of the
chlorinated hydrocarbons and heavy metals in five
dif-ferent strains of newly hatched Artemia In:
Personne, Sorgeloss, Roels, Jaspers, eds. The
brineshrimp Artemia. Wetterner, Belgium: Universa
Press.
Penrose WR and Spuires WR. 1976. Two devices for
removing supersaturated gases in aquarium systems.
Trans. Am. Fish Soc. 105: 116-118.
Powell AB and Schwartz FJ. 1979. Food of Paralichthys
dentatus and _P. 1 ethos tig ma (Pices: Bothidae) in North
Carolina estuaries. Estuaries 2:276-279.
Price WW- 1978. Occurence of Mys id ops is alrnyra Bowman
(Crustacea, Mysidacea) from the eastern coast of
Mexico. Gulf Research Reports 6:173-175.
29
-------
ES-2
August, 1982
Schuster CN. 1959. A biological evaluation of the
Delaware River Estuary. Info. Series Public No. 3.
Univ. Delaware Marine Laboratories.
Siegel, 1956. Nonparametric statistics for the
behavioral sciences. McGraw - Hill. Publ. Co. New York:
Sokal RR and Rohlf FT. 1969. Biometry. San Francisco,
CA: W.H. Freeman anc Co.
Sprague J3. 1969. Measurement of pollutant toxicity to
fish. I. Bioassay methods for acute toxicity. Water
Res. 3.:793-821.
Stephan CF. 1977., Methods for calculating an LC50.
In: Mayer FL and Hamelink, eds. Aquatic toxicology and
hazard evaluation. ASTM STP 634. Ph rlade'lphia, PA:
American Society for Testing and Materials. pp. 65-84.
Supplee VC and Lightner DV. 1976. Gas-bubble disease
due to oxygen supersaturation in raceway-reared
California brown shrimp. Prog. Fish Cult. 38:159.
Thompson WR. 1947- Use of moving averages and
interpolation to estimate median effective dose. I.
Fundamental formulae, estimation and error, and relation
to other methods. Bacterial. Rev. 11:115-145.
USEPA. 1974. U.S. Environmental Protection Agency.
Manual of analytical methods for the analysis of
pesticide residues in human and environmental samples.
Research Triangle Park, NC : U.S. Environmental
Protection Agency.
USEPA. 1978. U.S. Environmental Protection Agency.
Bioassay Procedures for the Ocean Disposal Permit
Program. Environmental Research Laboratory. Gulf
Breeze, Fl: U.S. Environmental Protection Agency. EPA-
600/9-78-010.
Veith GD and Corns tock VM. 1975. Apparatus for
continuously saturating water with hydrophobic organic
chemicals. J. Res. Fish. Bd. Canada 32: 1849-1851.
30
-------
EG-5
August, 1982
OYSTER ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EE-5
Guideline for Testing Chemicals August, 1982
OYSTER ACUTE TOXICITY TEST
(a) Purpose. This guideline will be used in developing data
on the acute toxicity of chemical substances and mixtures
("chemicals") subject to environmental effects test regulations
under the Toxic Substances Control Act (TSCA)(Pub.L. 94-469, 90
Stat. 2003, 15 U.S.C. 2601 et. seq. ). This guideline prescribes
tests to be used to develop data on the acute toxicity of
chemicals to Eastern oysters, Crassostrea virginica (Gmelin).
The United States Environmental Prote'dtion Agency (USEPA) will
use data from these tests in assessing the hazard of a chemical
to the environment.
(b) Def initions . The definitions in section 3 of the Toxic
Substances Control Act (TSCA) and in Part 792--Good Laboratory
Practice Standards are applicable to this test guideline. The
following definitions also apply:
(1) "Acute toxicity" is the discernible adverse effects
induced in an organism within a short period of time (days) of
exposure to a chemical. For aquatic animals this usually refers
to continuous exposure to the chemical in water for a period of
up to four days. The effects (lethal or sublethal) occurring may
usually be observed within the period of exposure with aquatic
organisms. In this test guideline, shell deposition is used as
the measure of toxicity.
-------
EG-5
August, 1982
(2) "EC 50" is that experimentally derived concentration of
a chemical in water that is calculated to induce shell deposition
50 percent less than that of the controls in a test batch of
organisms during continuous exposure within a particular exposure
period which should be stated.
(3) "Shell deposition" is the measured length ,f shell
growth that occurs between the- time the shell is ground at test
initiation and test termination 96 hours later.
(4) "Umbo" means the narrow end (apex) of the oyster shell.
(5) "Valve height" means the greatest linear dimension of
the oyster as measured from the umbo to the ventral edge of the
valves (the farthest distance from the umbo).
(c) Test procedures--(1) Summary of the test. (i) The
water solubility and the vapor pressure of the test chemical
should be known. Prior to testing, the structural formula of the
test chemical, its purity, stability in water and light, n-
octanol/water partition coefficient, and pKa values should be
known prior to testing. The results of a biodegradabil i ty test
and the method of analysis for the quantification of the chemical
in water should also be known.
(ii) For chemicals with limited solubility under the test
conditions, it may not be possible to determine an EC 50. If it
is observed that the stability or homogeneity of the test
-------
EG-5
August, 1982
chemical cannot be maintained, then care should be taken in the
interpretation of the results and a note made that these results
may not be reproducible.
(iii) Test chambers are filled with appropriate volumes of
dilution water. The flow of dilution water through each chamber
is adjusted to the rate desired. The test chemical is introduced
into each test chamber and the flow-rate adjusted to establish
and maintain the desired concentration in each test chamber.
Test Oysters which have been acclimated and prepared by grinding
away a portion of the shell periphery are randomly introduced
into the test and control chambers. Oysters in-the test and
control chambers are observed daily during the test for evidence
of feeding or unusual conditions, such as shell gaping, excessive
mucus production or formation of fungal growths in the test
chambers. The observations are recorded and dead oysters
removed. At the end of 96 hours the increments of new shell
growth are measured in all oysters. The concentration-response
curve and EC 50 value for the test chemical are developed from
these data.
(2) [Reserved]
(3) Range-f inding test. A range-finding test should be
conducted to establish test chemical concentrations for the
definitive test. The test is conducted in the same way as the
-------
EG-5
August, 1982
definitive test except a widely spaced chemical concentration
series (i.e. log-interval) is used.
(4) Definitive test. (i) Oysters which meet condition
criteria (age, size, reproductive status, health) and which have
been acclimated to test conditions should have approximately 3 to
5 mm of the shell periphery, at the rounded (ventral) end, grounc
away with a small electric disc grinder or other appropriate
device, taking care to uniformly remove the shell rim to produce
a smooth, rounded blunt profile. The oyster's valves should be
held together tightly during grinding to avoid vibrating the
shell and injuring the adductor muscle. Oysters of which so much
of the shell rim has been removed that an opening Into the shell
cavity is visible should not be used.
(ii) It is desirable to have shell growth values for the low
and high concentrations relatively close to, but different from,
0 and 100 percent. Therefore,, the range of concentrations to
which the oysters are exposed should be such that in 96 hours
relative to the controls, very little shell growth occurs in
oysters exposed to the highest concentration and shell growth is
slightly less than controls at the lowest concentration. Oysters
in the remaining concentrations should have increments of shell
growth, such that ideally, the concentration producing 50 percent
shell growth relative to the controls is bracketed with at least
-------
EG-5
August, 1982
one concentration above and one below it.
(iii) The test should be carried out without adjustment of
pH unless there is evidence of marked change in the p.H of the
solution. Then it i's advised that the test be repeated with pH
adjustment to that of the dilution water and the results
reported .
(iv) The test begins when at least 20 prepared oysters are
placed in each of the test chambers containing the appropriate
Concentrations of test substance and controls. The steady-state
flows and tes't chemical concentrations should be documented. At
least 5 test chemical concentrations should be used. The
dilution factor between concentrations should not exceed 1.8.
(v) The distribution of individual oysters among the test
chambers should be randomized. The oysters should be spread out
equidistantly from one another so that the entire test chamber is
used. The oysters should also be placed with the left (cupped')
valve down and the open, unhinged ends all oriented in the same
direction facing the incoming flow of test solution.
(vi) The oysters are inspected at least after 24, 48, 72 and
96 hours. Oysters are considered dead if touching of the gaping
shell produces no reaction. Dead oysters are removed when
observed and mortalities are recorded. Observations at three
hours and six hours are also desirable.
-------
EG-5
August, 1982
(vii) Shell growth is the primary criterion used in this
test guideline to evaluate the toxicity of the test chemical.
Shell growth increments in all oysters should be measured after
96 hours exposure. Record the length of the longest "finger" of
new shell growth to the nearest 0.5 mm. Oysters should be
handled very gently at this stage to prevent damage to the new
shell growth.
(viii) Records should be kept of visible abnormalities such
as loss of feeding activity (failure to deposit f eces ) , excessive
mucus production (stringy material floating suspended from
oysters), spawning or appearance of shell (closure or gaping).
( ix) The criteria for a valid definitive test are:
(A) The mortality in the controls should not exceed 10
percent at the end of the test.
(B) The dissolved oxygen concentration should be at least 60
percent of air saturation throughout the test.
(C) Oysters should not spawn during test. If they do the
test should be repeated with prespawn oysters.
(D) There should be evidence that the concentration of the
substance being tested has been satisfactorily maintained (e.g.,
within 80 percent of the nominal concentration) over the test
period. The total concentration of test substance (i.e. both
dissolved and suspended undissolved particulates) should be
-------
EG-5
August, 1982
measured; (JJ in each chamber at 0-hour, (2) in each chamber at
96-hours and (_3_) in at least one appropriate chamber whenever a
malfunction is detected in any part of the test chemical delivery
sys tern.
(E) Dissolved oxygen, temperature, salinity and pH
measurements should be made at the beginning of the test, at 48
hours, and at the end of the test in the control chambers and in
those test chambers containing the highest, lowest and a middle
concentration of the test substance.
(5) Test res ults . (i) At the end of the test, a one-way
analysis of variance followed with an appropriate ad hoc test
(the studentized Neuman-Keul's or Duncan's multiple range tests;
or Dunnetts' or Williams' pairwise comparison tests) should be
conducted on the oyster shell deposition test data. The probit
transformation should then be applied to the response variable
and then regressed, using least squares regression, on dose or
log-dose. An F Test for linearity should be conducted to
determine whether the chosen regression technique adequately
describes the experimental data.
(ii) Calculate the ratio of the mean shell growth for each
group of test oysters (exposed to each of the test chemical
concentrations) to the mean shell growth of the group of control
oysters. From these data the concentration-response curve is
-------
EG-5
August, -1982
drawn and an EC 50 along with the 95 percent confidence limits on
the value are determined from the curves. The mean measured
concentration of test chemical should be used to calculate the EC
50 and to plot the concentration-response curve.
(6) [Reserved]
(d) Test conditions — (1) Test species — (i) Selection.
(A) The Eastern oyster, Crassostrea virginica, should be used as
the test organism.
(B) oysters used in the same test should be 30 to 50
mill-imeters in valve height and should be as similar in age
and/or size as possible to reduce variability. The standard
deviation of the valve height should be less than 20 percent of
the mean.
(C) Oysters used in the same test should be from the same
source and from the same holding and acclimation tc.nk(s).
(D) Oysters should be in a prespawn condition of gonadal
development prior to and during the test as determined by direct
or histological observatio-n of the gonadal tissue for the
presence of gametes.
(ii) Acguis ition. Oysters may be cultured in the
laboratory, purchased from culture facilities or commercial
harvesters, or collected from a natural population in an
unpolluted area free from epizootic disease.
8
-------
EG-5
August, 1982
(iii) Acclimation. (A) Oysters should be attended to
immediately upon arrival. Oyster shells should be brushed clean
of fouling organisms and. the transfer of the oysters to the
holding water should be gradual to reduce stress caused by
differences in water quality characteristics and temperature.
Oysters should be held for at least 12 to 15 days before
testing. All oysters should be maintained in water of the
quality to be used in the test for at least seven days before
they are used.
(B) During holding, the oysters should not be crowded and
the dissolved oxygen concentration should be above 60 percent
saturation. The temperature of the holding water should be the
same as that used for testing. Holding tanks should be kept
clean and free of debris. Cultured algae may be added to
dilution water sparingly, as necessary to support life and growth
and such that test results are not affected as confirmed by
previous testing.
(C) Oysters should be handled as little as possible. When
handling is necessary, it should be done as gently, carefully,
and quickly as possible.
(D) A batch of oysters is acceptable for testing if the
percentage mortality over the seven day period prior to testing
is less than five percent. If the mortality is between 5 and
-------
EG-5
August, 1982
10%, acclimation should continue for seven additional days. If
the mortality is greater than 10%, the entire batch of oysters
should be rejected. Oysters should not be used which appear
diseased or otherwise stressed. Oysters infested with mudworms
(Polydora sp.), boring sponges (Cliona cellata) or which have
cracked, chipped, bored, or gaping shells should not be used.
(2) Test facilities—(i) Apparatus. (A) In addition to
normal laboratory equipment, an oxygen meter, equipment for
delivering the test chemical, adequate apparatus for temperature
control, and test-tanks made of chemically inert material are
needed.
(B) Constant conditions in the test facilities should be
maintained as much as possible throughout the test. The
preparation and storage of the test material, the holding of the
oysters and all operations and tests should be carried out in an
environment free from harmful concentrations of dust, vapors and
gases and- in such a way as to avoid cross-contamination. Any
disturbances that may change the behavior of the oysters should
be avoided.
(ii) Dilution water. A constant supply of good quality
unfiltered seawater should be available throughout the holding,
acclimation and testing periods. Natural seawater is
recommended, although artificial seawater with food added may be
10
-------
EG-5
August, 1982
used. In either case, to ensure each oyster is provided equal
amounts of food, the water should come from a thoroughly mixed
common source and should be delivered at a flowrate of at least
i-Vae and preferably five liters per hour per oyster. The flowrate
•should be _+_ 10 percent of the nominal flow. A dilution water is
acceptable if oysters will survive and grow normally for 14 days
without exhibiting signs of stress; i.e. excessive mucus
production (stringy material floating suspended from oysters),
lack of feeding, shell gaping, poor shell closing in response to
prodding, or excessive mortality. The dilution water should have
a salinity in excess of 12 parts per thousand, and should be
similar to that in the environment from which the test oysters
originated. A natural seawater should have a weekly range in
salinity of less, than 10 parts per thousand and a monthly range
in pH of less than 0.8 units. Artificial seawater salinity
should not vary more than 2 parts per thousand nor more than 0.5
pH units. Oysters should be tested in dilution water from the
same origin.
(3) Test parameters (i) Carriers . Stock solutions of
substances of low aqueous solubility may be prepared by
ultrasonic dispersion or, if necessary, by use of organic
solvents, emulsifiers or dispersants of low toxicity to
oysters. When such carriers are used the control oysters should
11
-------
EG-5
August, 1982
be exposed to the same concentration of the carrier as that used
in the highest concentration of the test substance. The
concentration of such carriers should not exceed 0.1 ml/1.
(ii) Dissolved oxygen. The dissolved oxygen concentrations
should be at least 60 percent of the saturation value and should
be recorded daily.
(iii) Lo ad i ng . The loading rate should not crowd oysters
and should permit adequate circulation of water while avoiding
physical agitation of oysters by water current.
(iv) Temperature. The test temperature is 20°C 4^ 1°C.
Temporary fluctuations (less than 8 hours) within 15°C to 25°C
are permissible. Temperature should be recorded continuously.
(v) pH. The pH should be recorded twice weekly in each test
chamber.
(e) Reporting . In addition to the reporting requirements
prescribed in Part 792--Good Laboratory Practice Standards, the
report should contain the following:
(1) The source of the dilution water, the mean, s-tandard
deviation and range of the salinity, pH, temperature, and
dissolved oxygen during the test period.
(2) A description of the test procedures used (e.g. the
flow-through system, test chambers, chemical delivery system,
12
-------
EG-5
August, 1982
aeration, etc.).
(3) Detailed information about the oysters used, including
the age and/or size (i.e. height), source, history, method of
confirmation of prespawn condition, acclimation procedures, and
food used.
(4) The number of organisms tested, the loading rate, and
the flowrate.
(5) The methods of preparation of stock and test solutions,
and the test chemical concentrations used.
(6) The number of dead and,live test organisms, the
percentage of organisms that died, and the number that showed any
abnormal effects in the control and in each test chamber at each
observation period.
(7) The 96-hour shell growth measurements of each oyster;
the mean, standard deviation and range of the measured shell
growth at 96 hours of oysters in each concentration of test
substance and control.
(8) The calculated 96 hour EC 50 and its 95 percent
confidence limits and the statistical methods used to calculate
these values.
(9) When observed, the 96 hour observed no-effect
concentration (the highest concentration tested at which there
were no mortalities, abnormal behavioral or physiological effects
13
-------
EG-5
August, 1982
and at which shell growth did not differ from controls).
(10) A graph of the concentration-response curve based on
the 96 hour chemical concentration and shell growth measurements
upon which ihe EC 50 was calculated.
(11) Methods and data records of all chemical analyses of
water quality parameters and test substance concentrations,
including method validations and reagent blanks.
(12) Any incidents in the course of the test which might
have influenced the results.
(13) A statement that the test was carried out in agreement
with the prescriptions of ' the test guideline given above
(otherwise a description of any deviations occuring).
14
-------
EG-6
August, 1982
OYSTER BIOCONCENTRATION TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-6
Guidelines for Testing Chemicals August, 1982
OYSTER BIOCONCENTRATION TEST
.(a) Purpos e. This guideline is to be used for assessing the
propensity of chemical substances to bioconcentrate in tissues of
estuarine and marine molluscs. This guideline describes a
bioconcentration test procedure for the continuous exposure of
Eastern oysters (Crassostrea virginica) to a test substance in a
flow-through system. The United States Environmental Protection
Agency (US EPA) will use data from this test in assessing the
hazard a chemical may present to the environment.
(b) Def initions . The definitions in section 3 of the Toxic
Substances Control Act (TSCA) and in Part 792--Good Laboratory
Practice Standards are applicable to this test guideline. The
following definitions also apply:
(1) "Acclimation" is the physiological compensation by test
organisms to new environmental conditions (e.g., temperature,
s alinity , pH) .
(2) "Bioconcentration" is the net accumulation of a chemical
directly from water into and onto aquatic organisms.
(3) "Bioconcentration factor (BCF)" is the quotient of the
concentration of a test chemical in tissues of aquatic organisms
at or over a discrete time period of exposure divided by the
concentration of test chemical in the test water at or during the
same time period.
-------
August, 1982
(4) "Depuration" is the elimination of a test chemical from
a test organism.
(5) "Depuration phase" is the portion of a bioconcentration
test after the uptake phase during whii.ii the organisms are -in
flowing water to which no test chemical is added.
(6) "EC 50" is that experimentally derived concentration of
a chemical in water that is calculated to induce shell -deposition
50 percent less than that of the controls in a test batch of
organisms during continuous exposure within a particular period
of exposure (which should be stated).
(7) "Loading" is the ratio of the number of oysters to the
volume (liters) of test solution passing through the test chamber
per hour.
(8) "Steady-state" is tt j time period during which the
amounts of test chemical being taken up and depurated by the test
oysters are equal, i.e., equilibrium.
(9) "Steady-state bioconcentration factor" is the mean
concentration of the test chemical in test organisms during
steady-state divided by the mean concentration of the test
chemical in the test solution during the same period.
(10) "Umbo" is the narrow end (apex) of the oyster shell.
(11) "Uptake" is the sorption of a test chemical into and
onto aquatic organisms during exposure.
-------
EG-6
August, 1982
(12) "Uptake phase" is the initial portion of a
bioconcentration test during which the organisms are exposed to
the test solution.
(13) "Valve height" is the greatest linear dimension of the
oyster as measured from the umbo to the ventral edge of the
valves (the farthest distance from the umbo).
(c) Test procedures--(1) Summary of the test. Oysters are
continuously exposed to a minimum of one constant, sublethal
concentration of a test chemical under flow-through conditions
for a maximum of 28 days. During this time, test solution and
oysters are periodically sampled and analyzed using appropriate
methods to quantify the test chemical concentration. If, prior
to day 28, the tissue concentrations of the chemical sampled over
three consecutive sampling periods have been shown to be
statistically similar (i.e., steady-state has been reached), the
uptake phase of the test is terminated, and the remaining oysters
are transferred to untreated flowing water until 95 percent of
the accumulated chemical residues have been eliminated, or for a
maximum depuration period of 14 days. The mean test chemical
concentration in the oysters at steady-state is divided by the
mean test solution concentration at the same time to determine
the bioconcentration factor (BCF). If steady-state is not
reached during 28 days of uptake, the steady-state BCF should be
-------
EG-6
August, 1982
calculated using non-linear parameter estimation methods.
(2) [Reserved]
(3) Range-finding test. The oyster acute toxicity test is
used to determine the concentration levels to be uses--! in the
oyster bioconcentration test.
(4) Definitive test. (i) The following data on the test
chemical should be known prior to testing:
(A) Solubility in water.
(B) Stability in water.
(C) OctanolTwater partition coefficient.
(D) Acute toxicity (e.g. propensity to inhibit shell
deposition) to oysters.
(E) The validity, accuracy and minimum detection limits of
selected analytical methods.
(ii) At least one or more concentrations should be tested to
assess the propensity of the compound to bioconcentrate. The
concentrations selected should not stress or adversely affect the
oysters and should be less than one-tenth the EC 50 determined in
either the range-finding or 96-hour definitive test in the Oyster
Acute Toxicity Test Guideline (USEPA 1981). The test
concentration should be less than the solubility limit of the
test substance in water and should be close to the potential or
expected environmental concentration. The limiting factor of how
-------
EG-6
August, 1982
low one can test is based on the detection limits of the
analytical methods. The concentration of the test material in
the test solution should be at least ten times greater than the
detection limit in water.
(iii) If it is desirable to document that the potential to
bioconcentate is independent of the test chemical concentration,
at least two concentrations should be' tested which are at least a
factor of 10 apart.
;
(iv) 'TO determine the duration of this test, an estimation
of the uptake phase should be made prior to testing based upon
the water solubility or octanol-water partition coefficient of
the test chemical. This estimate should also be used to
designate a sampling schedule.
(v) The following criteria should be met for a valid test:
(A) If it is observed that the stability or homogeneity of
the test chemical cannot be maintained, then care should be taken
in the interpretation of the results and a note made that these
results may not be reproducible.
(B) The mortality in the controls should not exceed 10
percent at the end of the test.
(C) The dissolved oxygen concentration should be > 60
percent of saturation throughout the test.
(D) There should be evidence that the concentration of the
-------
EG-6
August, 1982 '
chemical being tested has been satisfactorily mainained (e.g.
within 80 percent of the nominal concentration) over the test
period.
(E) Results are invalid and the test should be repeated if
the oysters spawn during the test.
(F) Temperature variations from 20°C should be held to a
min imum.
(vi) The following methodology should be followed:
i 11 ! i i
(A) The test should not be'! started until the test chemical
delivery system has been observed to be functioning properly and
the test chemical concentrations have equilibrated (i.e. the
concentration does not vary more than 20 percent),. Analyses of
two sets of test solution samples taken prior to test initiation
should document this equilibrium. At initiation (time 0', test
solution samples should be collected immediately prior to the
addition of oysters to the test chambers.
(B) The appropriate number of oysters should be brushed
clean and should be randomly distributed to each test chamber.
The oysters should be spread out equidistant from one another and
placed with the left (cupped) valve down and the unhinged ends
(opposite from umbo) all oriented in the same direction facing
the incoming flow.
(C) Oysters should be exposed to the test chemical during
-------
EG-6
August, 1982
the uptake phase until steady state has been reached or for a
maximum of 28 days. The uptake phase should continue for at
least 4 days. Then the remaining oysters should be transferred
to untreated flowing water and sampled periodically to determine
if depuration of the test chemical occurs. Every test should
include a control consisting of the same dilution water,
conditions, procedures, and oysters from the same group used in
the test, except that none of the test chemical is added. If a
i : ,
carrier is present in the test chamber, a separate carrier
control is required.
(D) Oysters should be observed (and data recorded) at least
daily for feeding activity (deposition of feces ) or any unusual
conditions such as excessive mucus production (stringy material
floating suspended from oysters), spawning, or appearance of
shell (closure or gaping). If gaping is noted, the oyster(s)
should be prodded. Oysters which fail to make any shell
movements when prodded are to be considered dead, and should be
removed promptly with as little disturbance as possible to the
test chamber(s) and remaining live oysters.
(E) For oysters sampled, careful examination of all the
tissues should be made at the time of shucking for any unusual
conditions, such as a watery appearance or differences in color
from the controls.
-------
EG-6
August, 1982
(F) Observations on compound solubility should also be
recorded. These include the appearance of surface slicks,
precipitates, or material adsorbing to the test chamber.
(vii) Sampling. (A) At each of the designated sampling
times, triplicate water samples and enough oysters should be
collected from the test chamber(s) to allow for tissue analyses
of at least four oysters. The concentration of test chemical
should be determined in a minimum of 4 oysters analyzed
I ' : •' ' I ! I '
individually at each sampling period. If individual analysis is
not possible, due to limitations of the sens.itivity of the
analytical methods, then pairs, triplicates or more oysters may
1 i
be pooled to constitute a sample for measurement. A similar
number of control oysters should also be collected at each sample
point, but only those collected at the first sampling period and
weekly thereafter, should be analyzed. Triplicate control water
samples should be collected at the time of test initiation,and
weekly thereafter. Test solution samples should be removed from
the approximate center of the water column.
(B) At each sampling period the appropriate numbers of
oysters are removed and treated as follows:
(_!_) The valve height of each oyster should be measured.
(_2_) Oysters should be shucked as soon as practical after
removal and should never be refrigerated or frozen in the
-------
EG-6
August, 1982
shell. The shell should be opened at the hinge, the adductor
muscle severed and the top valve removed. The remaining adductor
muscle should be severed where it attaches to the lower valve and
the entire oyster removed.
(_3_) The shucked oysters should then be drained three
minutes, blotted dry, weighed and analyzed immediately for the
test chemical. If analyses are delayed, the shucked oysters
should be wrapped individually in aluminum foil (for organic
I
analysis) or placed in plastic or glass containers (for metal
analysis) and frozen.
(C) If a radiolabelled test compound is used, a sufficient
number of oysters should also be sampled at termination to permit
identification and quantitation of any major (greater than 10
percent of parent) metabolites present. It is crucial to
determine how much of the activity present in the oyster is
directly attributable to the parent compound.
(5) Test results (i) Steady-state has been reached when
the mean concentrations of test chemical in whole oyster tissue
for three consecutive sampling periods are statistically similar
(F test, P=0.05). A BCF is then calculated by dividing the mean
tissue residue concentration during steady-state by the mean test
solution concentration during this same period. A 95 percent
confidence interval should also be derived from the BCF. This
-------
EG-6
August, 1982
should be done by calculating the mean oyster tissue
concentration at steady-state (XQ) and its 97.5 percent
confidence interval Xo ± t (S.E.) where t is the t statistic at
P=0.025 and S.E. is the one standard error of the mean. This
calculation would yield lower and upper confidence limits- (Lo and
UQ). The same procedure- should be used to calculate the mean and
97.5 percent confidence interval for the test solution concen-
trations at steady-state, Xs ± t(S.E.), and the resulting upper
: |;< !
and lower confidence limits (Ls and Us)• The 95 percent
confidence interval of the BCF would then be between LO/US' ajnd
I1
U0/Lg . If steady-state was not reached during the maximum 28 day
uptake period, the maximum BCF should be calculated using the
i
mean tissue concentration from that and the previous sampling
day. An uptake rate constant should then be calculated using
appropriate techniques. This rate constant is used to estimate
the steady-state BCF and the time to steady-state.
(ii) If 95 percent elimination has not been observed after
14 days depuration then a depuration rate constant should also be
calculated. This 'rate constant is used to estimate the time to
95 percent elimination.
(iii) Oysters used in the same test should be 30 to 50
millimeters in valve height and should be as similar in age
and/or size as possible to reduce variability. The standard
10
-------
EG-6
August, 1982
deviation of the height should be less than 20 percent of the
mean (N=30).
(6) Analytical measurements. (i) All samples should be
analyzed using USEPA methods and guidelines whenever feasible.
The specific methodology used should be validated before the test
is initiated. The accuracy of the method should be measured by
the method of known additions. This involves adding a known
amount .of 'the test chemical to three water samples taken from an
i
i ' i
aquarium containing dilution water and a number of oysters equal
to that to be used in the test. The nominal concentration of
these samples should be the same as the concentration to be used
in the test. Samples taken on two separate days should be
analyzed. The accuracy and precision of the analytical method
should be checked using reference or split samples or suitable
corroborative methods of analysis. The accuracy of standard
solutions should be checked against other standard solutions
whenever possible.
(ii) An analytical method should not be used if likely
degradation products of the test chemical, such as hydrolysis and
oxidation products, give positive or negative interferences,
unless it is shown that such degradation products are not present
in the test chambers during the test. Atomic absorption
spectrophotometric methods for metal and gas chromatographic
11
-------
EG-6
August, 1982
methods for organic compounds are preferable to colorimetric
me thods.
(iii) In addition to analyzing samples of test solution at
least one reagent blank should also be analyzed when a reagent is
used in tht analysis.
(iv) When radiolabelled test compounds are used, total
radioactivity should be measured in all samples. At the end of
the uptake phase, water and tissue samples should be analyzed
s i • ; : i : i :!
using appropriate methodology to identify and estimate tne amount
of any major (at least 10 percent of the parent compound)
degradation products or metabolites that may be present.
(d) Test condi tions--(1) Tes t species . (i) The Eastern
oyster, Crassostrea virginica, should be used as the test
organism.
(ii) Oysters used in the same test should be 30 to 50
millimeters in valve height and should be as similar in age
and/or size as possible to reduce variability. The standard
deviation of the valve height should be less than 20 percent of
the mean.
(iii) Oysters used in the same test should be from the same
source and from the same holding and acclimation tank(s).
(iv) Oysters should be in a prespawn condition of gonadal
development prior to and during the test as determined by direct
12
-------
EG-6
August, 1982
or histological observation of the gonadal tissue for the
presence of gametes.
(v) Oysters may be cultured in the laboratory, purchased
from culture facilities or commercial harvesters, or collected
from a natural population in an unpolluted area free from
epizootic disease.
Cvi) The holding, and acclimation of the oysters should be as
follows:
(A) Oys'ters should be attended to immediately upon
arrival. Oyster shells should be brushed clean of fouling
organisms and the transfer of the oysters to the holding water
should, ,be gradual to reduce stress caused by differences in water
quality characteristics and temperature. Oysters should be held
for at least 12 to 15 days before testing. All oysters should be
maintained in water of the quality to be used in the test for at
least seven days before they are used.
(B) During holding, the oysters should not be crowded and
the dissolved oxygen concentration should be above 60 percent
saturation. The temperature of the holding waters should be the
same as that used for testing.. Holding tanks should be kept clean
and free of debris. Cultured algae may be added to dilution
water sparingly, as necessary to support life and growth, such
that test results are not affected, as confirmed by previous
13
-------
EG-6
August, 1982
testing. Oysters should be handled as little as possible. When
handling is necessary, it should be done as gently, carefully,
and quickly as possible.
(C) A batch of oysters is acceptable for testing if the
percentage mortality ovei the seven day period prior to testing
is less than five percent. If the mortality is between 5 and 10
percent, acclimation should continue for seven additional days.
If the mortality is greater than 10 percent, the entire batch of
I _ ! , \\ | I :
oysters should be rejected. Oysters' should not be used which
appear diseased or otherwise stressed. Oysters infested with
mudworms (Polydora sp.), boring sponges (Cliona cellata) or which
have cracked, chipped, bored, or gaping shells should not be
us ed.
(2) Facilities--(i) Apparatus. (A) An oxygen1meter,
equipment for delivering the test chemical, adequate apparatus
for temperature control, test tanks made of chemically inert
material and other normal laboratory equipment are needed.
(3) Constant conditions in the test facilities should be
maintained as much as possible throughout the test. The
preparation and storage of the test ma terial, , the holding of the
oysters and all operations and tests should be carried out in an
environment free from harmful concentrations of dust, vapors and
gases and in such a way as to avoid cross-contamination. Any
14
-------
EG-6
August, 1982
disturbances that may change the behavior of the oysters should
be avoided.
(ii) Dilution water A constant supply of good quality
unfiltered seawater should be available throughout the holding,
acclimation, and testing periods. Natural seawater is
recommended, although artificial seawater with food (algae) added
may be used. In either case, to ensure each oyster is provided
equal amounts of food, the water should come from a thoroughly
mixed common source and should be delivered at a flow rate of at
least one, ,and preferably five liters per hour per oyster. The
I
flowrate should be _+_ 10 percent of the nominal flow. A dilution
water is acceptable If oysters will survive and grow normally
over th^ 'period in which the test is conducted without exhibiting
signs of stress, i.e. excessive mucus production (stringy
material floating suspended from oysters), lack of feeding, shell
gaping, poor shell closing in response to prodding, or excessive
mortality. The dilution water should have a salinity in excess
of 12 parts per thousand, and should be s imiliar to that in the
environment from which the test oysters originated. A natural
seawater should have a weekly range in salinity of less than 10
parts per thousand and a monthly range in pH of less than 0.8
units. Artificial seawater should not vary more than 2 parts per
thousand nor more than 0.5 pH units. Oysters should be tested in
15
-------
EG-6
August, 1982
dilution water from the same origin.
(3) Test parameters — (i) Carriers . Stock solutions of
substances of low aqueous solubility may be prepared by
ultrasonic dispersion or, if necessary, by use of organic
solvents, emulsifiers or dispersants oi low toxicity to
oysters. When such carriers are used, the control oysters should
be exposed to the same concentration of the carrier as that used
in the highest concentration of the ties t substance. The
I i , I i ,
concentration of such carriers should not exceed 0.1 ml/1.
(ii) Dissolved oxygen. The dissolved oxygen 'concentrations
should be at least 60 percent of the air saturation value and
should be recorded daily.
(iii) Loading . The loading rate should not crowd oysters
and should permit adequate circulation of, water while avoiding
physical agitation of oysters by water current.
(iv) Temperature. The test temperature should be 20°C +_
1°C. Temporary excursions (less than eight hours) within 15°C to
25°C are permissible. Temperature should be recorded continu-
ously.
(v) pH. The pH should be recorded twice weekly in each test
chamber.
(e) Reporting. In addition to the reporting requirements
prescribed in Part 792—Good Laboratory Practice Standards, the
16
-------
EG-6
August, 1982
report should contain the following:
(1) The source of the dilution water, the mean, standard
deviation and range of the salinity, pH, temperature and
dissolved oxygen during the test period.
(2) A description of the test procedures used (e.g. the
flow-through system, test chambers, chemical delivery system,
aeration, etc.).
(3) Detailed information about the oysters used, including
i j I • , ! i
age, and/or size (i.e. height), weight (blotted dry), source,
history, method of conf irjna tion of prespawn condition,
acclimation procedures ,and food used.
(4) The number of organisms tested, loading rate and
flowrate. t
(5) The methods of preparation of stock and test solutions
and the test chemical concentrations used.
(6) The number of dead and live organisms, the percentage of
oysters that died and the number that showed any abnormal affects
in the control and in each test chamber at each observation
period.
(7) Methods and data records of all chemical analyses of
water quality parameters and test chemical concentrations,
including method validations and reagent blanks.
17
-------
EG-6
August, 1982
(8) Description of sampling, sample storage (if required)
and analytical methods of water and tissue analyses for the test
chemical.
(9) The mean, standard deviation and range of the concentra-
tion of test chemical in the, test solution and oystet tissue at
each sampling period.
(10) The time to steady-state.
(11) The steady-state or maximum BCF and the 9'5 percent
i • i ' i
confidence Limits.
(12) The time to 95 percent elimination of accumulated
residues of the test chemical from test oysters.
(13) Any incidents in the course of the test which might
have influenced the results.
(14) If the test was not done in accordance with the
prescribed conditions and procedures, all deviations should be
described in full.
(f) References. U.S. Environmental Protection Agency.
1981. Oyster Acute Toxicity Test Guideline, Toxic Substances
Control Act, section 4. Office of Pesticides and Toxic
Substances, Washington, D.C.: U.S. Environmental Protection
Agency.
18
-------
ES-3
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
OYSTER ACUTE TOXICITY TEST AND BIOCONCENTRATION TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
TABLE OF CONTENTS
Subject Page
I. Purpose 1
II. Scientific Aspects 1
General 1
Test Procedures 6
Range Finding 6
Acute Test 6
Bioconcentration Test 7
Definitive Test 8
Acute Test 8
Biocentcation Test 10
Analytical 15
Water Quality 15
Collection of Test Solution Samples 15
Test Substance Measurement 16
Test Data 18
Analysis 18
Acute Toxicity Test 18
Bioconcentration Test 23
Temperature Measurements 28
Test Conditions 28
Test Species 28
Selection 28
Sources 30
Size 31
Condition 31
Maintenance of Test species 33
Feeding 33
Facilities 35
General 35
-------
Subject Pag
Construction Materials 36
Test Substance Delivery System 36
Test Chambers and 'Loading 38
Flow-through System 39
Cleaning 40
Dilution Water 40
Carriers 44
Environmental Conditions 45
Dissolved Oxygen (See Section 2.1.3, 45
Dilution Water)
Temperature 45
Light 48
Salinity (See Seiction 2.1.3, 48
Dilation Water)
Reporting 48
III. Economic Aspects 48
IV. References 51
-------
Office of Toxic Substances ES-3
August, 1982
I. Purpose
The purpose of this document is to provide the
scientific background and rationale used in the development
of Test Guidelines EG-5 and EG-6 which uses the Eastern
oyster ,Crassostrea Virginia, to evaluate toxicity and bio-
concentration of chemical substances. The Document provides
an account of the logic used in the selection of the test
methodology^ procedures' and conditions prescribed in the
Test Guidelines. Technical issues and practical
considerations relevant to the Test Guidelines are
discussed. In addition!,; estimates of the cost of conducting;
the tests are provided.
II. jcientific Aspects
A. General
The test guidelines represent a synthesis of testing
procedures and the current laboratory practices of various
researchers. Increased interest and research in aquatic
toxicology and bioconceatration and their use as monitoring
tools has led to the need for standardized procedures for
testing the sublethal responses of marine and estuarine
bivalve molluscs. Mortality testing is not practical
because bivalves are able to close their valves and seal
themselves off from environmental stress for long periods of
time. The results of such tests would be difficult, if not
impossible, to interpret. A more useful test is the shell
deposition test, which employs concentrations that will
produce an adverse effect, but will not cause the animal to
close up. The shell deposition test is intended to provide
a short-term assessment of the hazard which a test chemical
-------
ES-3
August, 1982
may present to oysters, and to serve as a range-finding test
for the bioconcentration test. Therefore, the test should
be of short duration, and'be similar in conditions to the
bioconcentration test.
Butler et al. (1960) demonstrated that shell growth in
juvenile oysters could be employed as a sensitive method for
the continuous monitoring of physiological stress occur ring
in bivalves exposed to various concentrations o£
pesticides. In Butler's studies, shell growth was used as a
measure of reversible inhibitory effect. The advantage of:
shell1 growth li^s in the ability to us^, the shell as an
ongoing physiological stress monitor without the need for
-periodic sacrificing of organisms. In 'addition, the -test is
rapid, reliable, reproducible and requires no specialized
equipment or personnel training.
The method first developed by Butler which utilized
shell deposition as a bioassay technique has been
successfully employed by numerous researchers (Schuster and
Priagle 1969, Tinsman and Maurer 1974, Frazier 1976, :onger
et al. 1978, Cunningham 1976, Epifanio and Mootz 1976, Lowe
et al . 1972). Epifanio (1979) showed that growth of hard
and soft tissues in oysters was closely coupled, ,as
determined by correlation analysis. This indicates that
shell deposition serves as a useful indicator of oyster
physiological response rather than just as a singular
response to calcium carbonate de-position. Studies conducted
by Conger et al . (1978) indicated a statistically
significant difference (P<_ 0.001) in the level of inhibition
of shell deposition in oysters which were subjected to 0.25
mg/1 cadmium. Schimmel et al. (1976) compared a 50 percent
-------
ES-3
August, 1982
median shell deposition reduction (SC50) in Crassostrea
virginica and the LC50 in five fish species. In this study,
shell deposition was at least as sensitive a test as those
employing fish. In tests conducted by Butler (1965) the
shell deposition test was more sensitive than other acute
toxicity tests employing other estuarine organisms. The
oyster shell deposition methodology is also described in
Standard Methods (APHA 1975), Bioassay Procedures for the
Ocean Disposal Permit Program (US EPA. 1978) and American
Society for Testing and Materials (ASTM 1980).
: Interest.in bioconcentration began!with the discovery
that levels of heavy metals in many animals were much higher
than in the surrounding water. In oysters, the phenomenon
was observed early in this century (Hiltner and Wickman
1919), and elaborated upon by subsequent workers (Galtsoff
1942, Chipman et al . 1958). The increased impacts of
hydrocarbons and various organic compounds, such as
insecticides, in recent years, led to field studies of
bioconcentration by oysters of such chemicals (Stegeman and
Teal 1973, Butler 1967, Hansen et al. 1976, Brodtmann 1970).
Laboratory testing of oyster bioconcentration is a
relatively recent development. Schim>nel et al. (1977),
Bahner et al. (1977), Frazier (1979a,b), Hansen et al.
(1976), Lee et al. (1978), Parrish et al. (1976), and
Stegeman (1974) have all investigated bioconcentration using
oysters in controlled laboratory settings. However, the
experimental methodology of each investigator was often
substantially different.
Other flow-through testing, particularly concerned with
sublethal effects on fish and bivalves in the late 1960's
-------
ES-3
August, 1982
and early 1970's led to a general recognition of the need
for standardization in flow-through testing.. Several
authors proposed methodologies (Sprague '196 9, Cairns and
Dickson 1973, Esvelt and Connors 1971, Lichatowich et al.
1973, LaRoche et al . 1970). These were synthesized into the
bioassay section of Standard Methods (APHA 1975) and
sections of the Ocean Disposal Bioassay Manual (Butler and
Lowe 1978). However, neither publication considered
bioconcentration. Bioconcentration methodologies have
emerged only in the last few years, and have been drafted as
proposed standards by the .American Society .for Testing and
Materials (ASTM 1980).
The tes t. guidelines adapt, to the extent possible, the
procedures of Standard Methods, EPA and ASTM to the specific
requirements of the Eastern oyster, Crassos trea vi rg inia
Gmelin.
Many industrial chemicals have not been previously
tested by standard aquatic bioassay methods 'and, as a
result, cannot be classified as to their toxicological
properties or propensity to bioconcentrate.
The oyster shell deposition test provides information on
the effects of short-term exposure of the test oysters to
the test chemical under controlled conditions (ASTM
1980b). Continuous administration of the test chemical in
this 96-hour flow-through system represents a oractial
simulation of chemical spills of effluent discharges to non-
motile organisms which are incapable of avoiding the
perturbation (APHA 1975). As such, the oyster shell
deposition test is particularly useful for evaluating the
short-term toxicity of specific substances or wastes on
-------
ES-3
August, 1982
marine molluscs. This test is employed primarily as an
appropriate range-f inding test for the more complex
bioconcentration test. As a range-finding test, it provides
information on the upper limit of exposure that is not
anticipated to cause adverse effects during the
bioconcentration test.
Oysters, as filter feeders, can be exposed to relatively
large amounts of a potential toxicant. .This is because the
oyster pumps large volumes of water and removes both living
and non-living particulate matter from that water. A
potential toxicant can be accumulated in the oyster itissues
in concentrations much greater than occur in the ambient
water or particulate matter. This accumulation, known as
bioconcentration, has been demonstrated tor a number of
petrochemicals (Anderson and Anderson 1976, Anderson 1978,
Bahner et al . 1977, Lee et al. 1978, Stegeman 1974),
pesticides (Brodtmann 1970, Butler 1967, Parrish et al.
1976, Schimrnel et al. 1977), and metals (Frazier 1975, 1976,
1979 a,b). The contaminated organism can, in turn, pass its
body burden of toxicant on to the next trophic level in a
concentrated form. Since humans are major consumers of
oysters, the potential for oysters to bioconcentrate a
potentially toxic substance is of additional concern. In
addition, bioconcentration of a substance by oys tars may be
an indication that the substance is biologically active and
could affect other elements of the aquatic system.
The bioconcentration test provides an estimate of that
potential. The results of the test can provide a basis for
decisions concerning what concentrations, if any, of the
test chemical in water may be bioconcentrated to potentially
-------
ES-3
August, 1982
hazardous concentrations in the aquatic biota.
B . Test procedures
1. Range-Finding Test
a. Acute test
For the oyster acute toxicity test, a range-finding test
is recommended to determine the appropriate concentrations
of test chemical to be used for a definitive test when the
acute toxicity of the substance is unknown or cannot be
elucidated -from existing toxicity data. This approach
should minimize the possibility that an inappropriate
concentration series will; be utilized in. the definitive test
and under certain circumstances may even preclude the need
to conduct the definitive test. In" order to minimize the
cost and time required to obtain the requisite data nominal
concentrations are permitted,, test duration may be
shortened, replicates are not required, and other test
procedures and conditions are relaxed.
The range-finding test (or other available information')
needs to be accurate enough to ensure that dose levels in
the definitive test are spaced to result in concentrations
above and below the EC50 values for shell deposition, IF
the substance has no measurable effect at the saturation
concentration (at least 1000 mg/1), it is considered
relatively non-toxic to oysters and definitive testing is
deemed unnecessary. In all cases, the range-finding test is
conducted to reduce the expense involved in having to repeat
a definitive test because of inappropriate test chemical
concentrations.
In the range-f inding test, groups of five or more test
oysters are exposed to a broad range of concentrations of
-------
ES-3
August, 1982
the substance. Sufficient concentrations should be tested
such that the concentration which inhibits shell deposition
by 50 percent of the control organisms can be approximated.
The number of concentrations will normally range from 3-6
depending upon the shape of the toxicity curve for that
substance and prior knowledge of its approximate toxicity.
Only concentrations less than the solubility limit in water
are tes ted .
b. Bioconcentrati.on Test
The oyster acute toxicity test is used as the range-
finding! test for the oyster bioponcentr^ition test. The
concentration of test chemical in the test solution should
not stress, irritate, or otherwise adversely effect the
organisms during the bioconcentration test. To meet this
criteria,, the ASTM (1980) recommends that the highest
concentration be no more than one-tenth the 96 hour ECcn
based on reduced shell deposition.
If stress, irritation, or other adverse effects are
observed, the bioconcentration test should be repeated at a
lower concentration.
In the bioconcentration test, it would be most useful
for the hazard and risk assessment processes to use an
exposure concentration that approximates the expected or
estimated environmental concentration. One should take
care, however, that the selected concentration is at least
three times above its detection limit and will allow
quantification of the residues in tissue. Test
concentrations of 1-10 ug/1 would be appropxiate for many
compounds.
-------
ES-3
August, 1982
2. Definitive test
a. Acute Test
The specific requirements of the definitive oyster acute
toxicity test (USEPA 1980) are the analytical rleterminations
of chemical concentrations, the unbiased selection of
oysters for each treatment, the use of controls, the
assessment of test validity, and the recording, analysis,
and presenta-tion of data. These requirements assure that
the chemical concentration - oyster response relationship is
accurately known, that chemical effects are not coa founded
{ay'differential oyster sensitivity, and' that the
relationships are clearly presented. Reporting the
occurrence of such effects as abnormal shell movement and
feeding behavior provide qualitative data that further
assist the assessment of toxicity.
The results of a definitive test are used to calculate
the 96 hour EC50 and the concentration-response relationship
of the test chemical and. the test oysters. If the con-
centrations of test chemical which produce no effect, a
partial inhibition of shell deposition, and 100 percent
inhibition have been determined during the range-finding
test, then five or six test chemical concentrations should
be sufficient to estimate the appropriate EC50 value in a
definitive test. In some cases however, to obtain two
partial inhibitions bracketing the 50 percent level, it may
be necessary to test, 8-10 concentrations.
The slope of the concentration-response curve provides
an indication of the range of sensitivity of the test
oysters to the test chemical and may allow estimations of
lower concentrations that will affect the test organism.
-------
ES-3
August, 1982
For example, if the slope of the concentration-response
curve is very steep, then a slight increase in concentration.
of the test chemical will affect a much greater portion of
the test oysters than would a similar increase if the slope
of the curve was very shallow. The slope of the
concentration-response curve reveals the extent of
sensitivity of the test oysters over a range o£
concentrations .
The exposure of two or more replicate groups having a
minimum of 20 oysters each, to each test chemical
coqcentration is required in the guideline. : That minimum is
based on an optimum number of test oysters ne.eded for
statistical confidence, equipment requirements, and
practical considerations of handling the test organisms.
At least two replicates should be included in order to
demonstrate the level of precision in the data and indicate
the significance of variations. Test chambers holding
replicate groups should have no water connections between
them. The distribution of test oysters to the test chambers
should be randomized to prevent bias from being introduced
into the test results.
The exposure' time of 96 hours in the oyster acute
toxicity test guideline is specified in order to permit a
comparison of data developed through the use of this test
guideline with the acute toxicity data in the published
literature (see Section 1.5) The use of the 96-hour exposure
period was proposed initially in 1951 by an aquatic bioassay
committee (Doudoroff et al. 1951) and was selected, in large
part, as a matter of convenience since it is easily
scheduled within the five-day work week. The 96-hour
-------
ES-3
August, 1982
exposure period is also required in the flowing seawater
toxicity test using oysters as a bioassay procedure for the
US SPA Ocean Disposal Permit Program (US SPA 1978).
b. Bioconcentration Test
The bioconcentration test guideline recommends that the
uptake phase last no more than 28 days and the depuration
phase last no more than 14 days for a total maximum test
duration of: 42 days. This is based on the experience of
researchers who have found that, generally speaking,
substances are either rapidly taken up or very slowly taken
up. Kirieger eif^al. (19719) demonstrated attainment of
steady-state for antipyrine uptake in less than 90 minutes
using mussels. Schimmel et al . (1978) showed that oysters
reached uptake equilibrium with respect to sodium
pentachlorophenate in 4 days. On the other hand, Stegeman
(1974) postulated that while low molecular weight
hydrocarbons are rapidly taken up and released, high
molecular weight compounds are taken up and released much
more slowly and, in fact, may never be completely
eliminated. Hydrocarbons apparently reach equilibrium with
the lipid fraction of the animal, so that the physiological
state of the organism has a great influence on
bioconcentration and depuration.
Because of the role of the lipid fraction in modifying
bioconcentration, it is possible to generate estimates of
the bioconcentration factor of organic chemicals from a
knowledge of their lipophilic nature. Veith et al. (1979)
analyzed the correlation between the n-octanol/water
partition coefficient (P), a commonly used measure of a
substance's lipophilic nature, and the experimentally
10
-------
ES-3
August, 1982
derived bioco'tcentration factor (BCF). They show that the
log BCF and the log P are linearly related by the equation:
log BCF = 0.85 log P - 0.70
They suggest that the high correlation of the equation
(r2 = 0.897) means that the log BCF can be estimated to
within an order of magnitude for substances having a broad
range of partition coefficients. Approximately 5 percent of
the substances tested had low log BCFs despite high log P,
thus fiajlling outs jlde of the general Delation. However, as
Veith et al . (1979) point out, none of the substances with
high BCF values had. low log P values. This means that use
of • the relationship should not lead to an underestimation of
the . 'Die-concentration factor. (
Chiou et al . (1977) present support for estimating the
partition coefficient from the aqueous solubility. Their
relationship states that:
log P = 5.00 - 0.670 log S
where S is the aqueous solubility in micromol/liter. They
found that the log P values for 34 organic substances ranged
from 1.26 (for phenoxyacetic acid) to 6.72 (for 2, 3, 4, 2',
4', 5',-PCB).
On the basis of these known relationships between
solubility, lipophilic nature and bioconcentration, Neeley
(1978) developed the equations for estimating the times to
steady-state. The estimates are based on fish, but are
aoolicable to molluscs. In the bioconcentration test, the
11
-------
ES-3
August, 198,2
exposure period should be long enough to demonstrate that
steady-state has been reached.
Before starting a bioconcentration test, an estimation
of the BCF or the time to steady-state should be made in
order to avoid running the test for the maximum time
period. A summary by Kenaga and Goring (1980) presents data
and methods to estimate the BCF. The two naos t commonly used
factors for predicting bioconcentration potential are water
solubility and octano'l-water partitioning. Water solubility
can be determined empirically in the laboratory, oc in some
cades, taken f-rom the1 literature (Chioa et al. 1977; Kenaga
and Goring 1980). Octanol-water partition coefficients can
be determined empirically, estimated by reverse-phase high
pressure liquid chroma tog rap'ny according to Veith et al.
(1979), calculated according to Leo et al. (1971) or taken
from the literature (Chiou et al . 1977, Hanch et al . 1972,
Kenaga and Goring, 1980). However, some of the reported
data are highly variable and may not be appropriate for use-
An estimate of, the time to steady-state (S- in hours) can
be estimated from the water solubility or octanol-water
partition coefficient using the equations developed by ASTM
(1980): S=3.0/antilog (0.431 log.W-2.11) or S=3.0/antilog (-
0.414 log P + 0.122) where W = water solubility (mg/1) and
P=octanol-water partition coefficient.
Presented below is a summary of data correlating various
exposure times to the corresponding estimates of the
partition coefficient and BCF.
12
-------
ES-3
August, 1982
Log P
BCF
1,585
8,710
33,113
120,226
316,228
524,807
933,254
3.2
3.94
4.52
5.08
5.5
5.72
5.97
105
446
1387
4150
10,000
14,521
23,686
2
4
7
12
18
22
28
Log BCF
.02
2.65
3.1'4
4.62
4.0
4.16
4.37
Log BCF was estimated using the equation of Veith et al.
(1970) 'where log BCF=0.85 log.p-0.70.
Based on the estimate of the time to steady state, one
of the following sampling scheme's raay be used to generate
appropriate data.
Sampling Days
Test
Period/ Sa<4
34-14
.S> 15-21
S>21
Exposure
1 ^
/:b
1
2
3
4
Depuration
lb
gb
12b
1
4b
1
3
7
10
12
14
1
2
6
1
3
7
10
14
18
22
1
3
7
10
1
3
7
10
14
21
28
1
3
7
10
14
a. length estimated time to steady state in days,
b. hours.
13
-------
ES-3
August, 1982
There are two methodologies in use today to estimate
bioconcentration potential; the kinetic approach and the
steady-state approach and both are based on research
conducted with fish. Bishop and Maki (1980) and Hamelink
(1977) give a review of both, Using the kinetic approach,
Bishop and Maki (1930), Branson et al. (1975), Cember et al .
(1978) and Krzeminsky (1977) proposed the use of first-order
kinetic expressions from relatively short (_<_ 5 days) fish
exposures, and a subsequent depuration period, to calculate
uptake and depuration rate constants. These rate constants
are then us^u to (estimate the BCF ait th.e .taJnjs jOf apparent
steady-state, and the time to 50 percent elimination. The
steady-state method, in more widespread use, exposes fish
for a longer period of tine until steady-state in the tissue
is experimentally observed (Barrows et al. 1980, Bishop and
Maki 1980, Veith et al. 1979) and continues with a
depuration phase until SOpercent or 95 percent elimination
has been observed. The estimation of bioconcentration using
the kinetic approach cannot account and adjust for changes
in the rates of uptake and depuration such as those observed
by Barrows et al. (1980) and Melancon and Lech (1979). The
use of the kinetic approach also requires access to a
sophisticated computer system, apparatus not readily
available to many laboratories.
Although Bishop and Maki (1930) nnd Branson et al .
(1975) have shown excellent agreement between estimates of
bio-concentration factors for some compounds using both
approaches, the agency has recommended a modified steady-
state method for determination of bioconcentration. The
empirical nature of the data, the relative ease with which
14
-------
ES-3
Januay, 1982
the test can be performed and the number of researchers and
laboraLo ries that have performed such tests make this test
more appropriate at this time. As the data base for
comparisons of BCFs between the two methods grows, the
kinetic approach may become more useful and valuable. Under
TSCA the Agency is required to review all tosh guidelines
annually, and in the future the Agency will consider
adopting the kinetic approach.
3. Analytical
a. Water Quality Analysis
Measurement of certain water quality parameters of the
dilution water such as dissolved oxygen, temperature,
salinity and pH is important. Quantification of these
parameters at the beginning, during, an-] citr. the end of the
exposure period for flow-through tests is necessary in order
to determine if the water quality varied during the test.
If significant variation occurs, the resulting data should
be interpreted in light of the estimated toxicity value's. A
decrease in dissolved oxygen indicates that the flow rate
should be increased.
b. Collection oE Test Solution Samples
The objective of the recommended sampling procedure is
to obtain a representative sample of the test solution for
use in measuring the concentration of the test chemical.
Although there is mixing in the test chamber, material can
concentrate near the sides and bottom of the chamber due to
physical or chemical properties of the substance, or to
interactions with organic materials associated with the test
animals. For this reason,, water samples should be taken
near the center of the test chamber. The handling and
15
-------
ES-2
August, 1982
storage of the samples requires care to present the loss of
the test chemical from the sample before analysis.
Standardized methods should be used in collecting the
samples and performing the analyses to develop chemical and
physical data. Appropriate sources foe such methodology
include, but are not limited to Hedgpeth 1966, Strickland
and Parsons 1972, AOAC 1975, APHA 1975, USEP^ 1974, and ASTM
1979.
c . Test Chemical Measurement
The actual substance concentration used in the
definitive test should be determined y/i th the ,best available
analytical precision. Analysis of stock solutions and test
solutions just prior to use will minimize problems with
storage (e.g., formation of degradation products,
adsorption, transformation, etc.). Nominal concentrations
are not adequate for the purposes of the definitive tests.
If definitive testing is not required- because the substance
elicits an insufficient response at the 1000 mg/1 level in
the range-E lading test, the concentration c substance in
the test solution should be determined to confirm the actual
exposure level. The pH of the test solution should be
measured prior to testing to determine if it lies outside of
the species' optimal range. This test guideline does not
include pH adjustment cor the following reasons: the use of
acid or base may chemically alter the test chemical making
it more or less toxic, the amount of acid or base needed to
adjust the pH may vary lirom one test solution concentration
to the next, and the effect the test chemical has on o.H may
indirectly affect the physiology of the test oysters.
16
-------
ES-3
August,- 1982
To assess and quantify any possible changes in test
chemical concentration, whenever a malfunction of the
toxicant delivery system is detected, all potentially
affected test chambers should be sampled at that time.
If the measured concentrations of dissolved test
chemical are 50 percent more oc less than the nominal
concentration, steps should be taken to dete cni .n^ the cause
for this deviation. A sample of the stock solution as well
as influent samples to various test chambers should be
analyzed to determine if the reduction in test chemical
occurs prior to delivery of hthe test solution to the
aquaria. If results of these analyses indicate that the
proper amounts of test sbstance ate entering the test
chambers, then the total test chemical concentration should
be measured in at least the chambers containing the highest
test chemical concentration. These data will give
indications if the difference between nominal and measured
test concentrations is due to volatilization or degradation
of the test chemical, or to insolubility of the test
chemical in the dilution water.
If the toxicant delivery system has been properly
calibrated and the oysters randomly introduced into each
test chamber, the measured differences between replicates at
each concentration should be less than 20 percent. If the
differences exceed this, the test should be repeated.
The concentrations of test chemical measured after
initiation should be within 30 percent of the concentrations
measured prior to introduction of the oysters. If the
difference exceeds this, the test should be repeated using a
higher flow rate.
17
-------
ES-3
August, 1982
Use of reliable and validated analytical techniques and
methods is essential to the usefulness of the test data in
assessing the environmental hazard of the substance.
Significant variation in the measured concentrations lessens
the value of the toxicity data generated.
4. , Test Data
a. Analys is
A coherent theory of the dose-response relationship, on
which acute toxicity tests are based, was introduced by
Bliss (1935), and is widely accepted today. This theory is
based on four assumptions: , . i ;
(1) Response is a positive function of dosage, i.e. it
-is expected that increasing treatment rates should
produce increasing responses.
(2) Randomly selected animals are normally distributed
with respect to their response to a toxicant.
(3) Due to homeos tas is , response magnitudes are pro-
portional to the logarithm of the dosage, i.e. it takes
geometrically increasing dosages (stresses) to produc
arithmetically increasing responses (strains) in test
animal populations.
(4) In the case of a direct dosage of animals, their
resistance to effects is proportional to body mass.
Stated another way, the treatment needed to produce a
given response is proportional to the s ize of the
animals treated.
b. Acute Toxicity Test
Oyster shell deposition data have been analyzed by
Cunningham (1976) and Schimmel et al. (1976, 1978).
Cunningham (1976) evaluated all shell growth data to the
18
-------
SS-3
August, 1982
P _<_0.05 level of confidence using an analysis of variance
coupled with a Duncan Multiple-Hinge test. A detailed
summary of the procedure for ranking the means and computing
the Duncan Multiple Range (DMR) values is given in Steel and
Torrie (1960).
Schimmel et al . (1976) developed an EC50, which is that
concentration of toxicant which produces a 50 percent median
shell deposition reduction in test oysters as compared to
control oysters. During other studies, Schimmel et al.
(1978), analyzed oyster shell deposition data by linear
regression with probit transformation to determine the EG50
and 95 percent confidence intervals.
Two types of .statistical techniques should be employed
for analyzing oyster shell deposition data: 1) analysis of
variance and 2) linear or non-linear regression. The test
design that is assumed is control, carrier control (if
solvent carrier is utilized), and five test concentrations
giving a total of seven treatments. For each treatment, 20
similar-sized oysters are tested for 96 hours. At the end
of the test, each oyster's shell deposition is measured and
recorded, giving 20 separate growth rsponses for each
treatment.
At this point, it is appropriate to conduct a one-way
ANOVA on these data to determine if there is a significant
effect on shell deposition due to the treatments (test
concentrations). A significant F value (P less than, or
equal to 0.05) would indicate such an effect and should be
following with an appropriate post hoc test (e.g., the
studentized Neuman-Keuls ' or Duncan's multiple range tests,
or Dunnetts ' or Williams' pairwise comparison tests). These
tests are designed to indicate
19
-------
ES-3
August, 1982
which test concentrations caused significant effects. For
some situations this information may be all that is
necessary; i.e., proof that a statistically 'significant
effect has occured due to the test substance. On the other
hand, if the AN OVA shows no significant effect due to
treatments, then the criteria requiring effects on both
sides of 50 percent will not have been met. If the control
and control with carrier are different, then there are
severe test problems that should be rectified. Either the
solvent is toxic at the concentrations tested, or there is
large variability among ;oysters, probably undia,gnosed
disease or improper test apparatus.
The second method of -analysis to be utilized is
regression—eithor linear with possible data transformations
or non-linear least squares. Growth data collected from an
oyster shell deposition test is dose-response in which the
responses are graded (or continuous) as opposed to quantal
(discrete or binomial). Due to this fact, the distribution
of measurements at each concentration level is generally
assumed to be normally distributed and the response curve is
sigiaoid (slant S shaped); in the center of the response
curve, the curve is typically relatively straight, while at
each end of the curve, the curve becomes asymptotic to the
100 percent (control growth or no-effect) and 0 percent (no
growth or full-effect) levels. Use of the guideline's
proposed test design causes the following to occur: two
controls and one no-effect concentration groups 60 oyster
growth values at the no-effect end of the rsponse curve.
The highest or full-effect concentration groups 20 oyster
growth values at the other: end of the response curve. That
20
-------
ES-3
August, 19:32
leaves 20 growth values in each of three treatments to
describe the linear central portion of the curve. Linear
regression, if used, should depend only on data from the
three central treatments, since 0 and 100 percent responses
may be far out on the curved ends of the response curve.
Improper u-se of linear regression on data from all
treatments (will likely overestimate the EC5Q and widen the
associated confidence interval, especially if the highest
test concentration -was chosen to be very high as compared
with the other test concentrations. On the other: hand, if
all 5 concentrations' (provide partial response, then simple
linear regression on growth data is an appropriate model if
the fit is reasonable.
The alternate approaches to straight linear regression
are: 1) regular probit analysis regression (using maximum
likelihood or minimum ), 2) various transformations prior
to least squares linear regression, and 3) non-linear
regress ion.
Probit analysis assigns relatively small weights to
response values near 0 and 100 percent. This is one of the
primary reasons why this analysis is acceptable for use on
dose-response data that contain no-effect and full-effect
concentrations. Although probits do not exist for 0 a;vl 100
percent effects, they replaced with close estimates and used
in the regression calculations. Another reason i!o r using
the probit transformation is that it linearizes the
integrated normal signmoid curve. If all five test
concentrations provide partial responses, then one can
likely expect probit analysis to give reliable results when
estimating the EC50 and it3 confidence interval provided the
21
-------
ES-3
August, 1982
fit is reasonable. Generally, oro'oits are regressed on log-
dose as opposed to dose.
The major drawback to using probit analysis in this
situation is that the method does not make full use of the
continuous aspect of the response variable. Probit analysis
only requires that the response variable,be quantal.
Consequently, the approach literally wastes information by
not using it.
The probit -trans format ion can be applied to the response
variable and then regressed, using least squares regression,
on dose or log-dose. This approach has the same i ad vantage
as probit analysis in that it tends to linearize sigmoid-
type curves. Therefore, it is .appropriate to utilize data
with response rates at or near 0 and 100 percent. In
addition, this approach makes use of the fact that th-.e
response variable is a continuous measure. This approach,
when the fit is reasonable, should give the most reliable
EC5Q estimate .and possibly a narrower confidence interval
than the other approaches.
Several other transformational approaches that might be
tried (when the probit transformation regressed on dose or
log-dose doesn't fit the data) are regress response on log-
dose, response on the square root of dose or response on the
inverse of dose.
If none of the above-mentioned linear regression
transformations produce an acceptable linear function, then
a non-linear sigmoid function such as: GROWTH = a/(1 +
•OCDOSE) or GROWTH = I/(a 4- b-c*DOSE) could be fit to the
data. The problem with using nonlinear functions is that in
these cases three parameters, a, b, and c (rather than two
22
-------
ES-3
August, 1982
a-3 in simple linear models), should be estimated. This
generally requires more data for comparable fits. However,
these functions will fit the data.
For each of the above me thods--s imple linear, probi. t
analysis, linear regression via the various transformations,
and non-linear regress ion--general ized lack-of-fit tests can
be conducted to determinei whether th^ chosen regression
technique adequately describes the experimental data. Since
there are twenty shell growth values foe each treatment
(test concentration), the appropriate statistical procedure
(except in , t;he case of probit analysis) is tto: conduct an "F
Test for Linearity." The comparable test for probit
analysis is the Chi-square goodness-of-f it test. If the
computed F value for linearity is large, then the linear
regression does not adequately describe the <^ta and the
EC^Q value and confidence interval, estimates are suspect.
If the computed F value is acceptable (i.e., P less than or
equal to .05 or P less than or equal to .10), them there is
no reason to doubt that the data have been sufficiently
described with the regression function and it would be
appropriate to compute the EC50 and confidence intervals
required by the test guideline,
c. Bioconcentration Test
The bioconcentration data (tissue test substance
concentration) should be determined and recorded separately
for each oyster, if possible, and of course also identifying
the test chamber from which each oyster was taken.
Certainly under the conditions necessitated by some chemical
analyses where large amounts of tissue are needed for an
analyses, this may not be possible. However, for a
23
-------
ES-3
August, 1932
bioconcentration test, duplicate samples are a necessity for
establishing whether steady-state has been reached. It is
not uncommon for bioconcentration data to vary half an order
of magnitude from sample to sample. Therefore, there should
be duplicate sample measurements for each sample period.
Duplicate sample values are required for computing whether
steady-state has bean reached and for accurate computation
of uptake and depuration rates, regardless of the
statistical methods used. A minimum of four or more sample
values for each sample period is recommended. The oysters
sampled at each period Erom each tes,t chamber should ]be
individually analyzed. The control oysters en be pooled
before analysis unless the chemical of interest or its"
metabolites are Pound or are expected in the control oyster
samples, since the controls serve only to identify
accidental and unknown contamination of test oysters fro.fi
uncontrolled sources.
The variance of each sample period is likely to increase
as the tissue concentrations increase, thus for statistical
purposes multiple oyster samples at each sample period is
necessary for determining when stead-state is reached for
calculating a suitable 95 percent confidence interval.
An estimate of the time to steady state, the steady-
state BC?1, and the time to 95 percent elimination should be
made for each compound tested. If steady-state has not been
observe3 during the maximum 28 day exposure period or if 95
percent elimination has not been achieved during 14 days
depuration, data generated during these tests should be used
to estimate these values. The BIOFAC program developed by
Blau and Agin (1978) uses nonlinear regression techniques to
24
-------
ES-3
August, L982
estimate the uptake and depuration rate constant, the
steady-state BCF, the time to reach 90 percent of steady-
state, the time to reach 50 percent elimination and the
variability associated with each estimate.
To date, there is no one specific method recommended for
identifying the time to 9 5 percent elimination of
accumulated residues. However, it is still of value to have
it reported. The problems associated with calculating this
95 percent point are:
. 1.) identifying the shape of the depuration ;qurve as to
whether it is linear or curvilinear;
2) if it is curvilinear, what curve best fits the
data; and
3) are the data sufficiently good to allow
extrapolation to estimate the 95 percent point?
Bioconcentration data is best displayed as log or
natural log (In) of the measured residue concentration on
the vertical axis and time (linear) on the horizontal
axis. The uptake curve will be exponential and increasing
until leveling of f; at steady-state. This uptake curve is
well' represented by the standard kinetic uptake function
Residue = Cone. * '"'i/<<2 * (l-eK2 t) •
This function has been shown to accurately represent
most uptake data and has been used to determine uptake rates
for oysters. However, there is no general function that
cons istently and adequately represents the depuration curve;
25
-------
ES-3
August, 1982
an appropriate choice should be made based on each data
set. The common description of the observed problem is that
chemicals partition within the oyster into different tissues
(compartments) that depurate the chemical differentially,
thus causing the depuration curve to be more complex and to
vary for different data. It is important that the curve
fits the data reasonably well since extrapolation is usually
required to obtain the 95 percent depuration point.
Generally, a non-linear parameter estimation- statistical
model can be used to describe the depuration data.
Since the same curve does not typically fit da.ta of thjLs
'type, a goodness-of-f it test should be conducted. If such a
test were completed successfully, then extrapolation using .
the equation is more reasonable -
The final Incision to be made is what 95 percent
depuration level is to be reported? The reference value is
the steady-state bioconcentration value; it is chosen as 100
percent uptake (0 percent depuration) and is normally
reported in ug or mg per g or kg of tissue. At this point
tfio options are available, 1) calculate 95 percent of the
steady-state value in concentration units, or 2) calculate
95 percent of the steady-state value in log units. The more
acceptable method is the latter. The following example will
illustrate the <:li \L terence in the methods.
Assume steady-state oyster concentration equals 500 ug
chemical pe r g oyster (500 pom). Using the linear
method for computing the 95 percent depuration
endpoinc: 95% X 500 ppm = 475 ppm; and 500 ppm - 475
ppm = 25 ppm is the endpoint. The time required for
depuration to 25 ppm would be reported. Using the log
26
-------
ES-3
August, 1932
method for computing the 95 percent depuration
endpoint: 95% X In (500 ppm) = .95 X 6.21 = 5.90; and
6.21 - 5.90 = .31. Then the antilog of .31 = e-31 =
1.36 ppm is the endpoint. The time required for
depuration to 1.36 ppm would be reported. The last
value, 1.36 ppm, is the actual 95 percent reduction
(depuration) endpoint. This is because we are dealing
with a In- dose vs. time relationship 'and all
computations and comparisons should be made on the In
transformed data with final back transformation to
normal! [units for Reporting. If 25 ppm were reported,
the 25 pom endpoint would represent only 48 percent
depuration (.48 X 6,21 = 2.98; 6.21 - 2.98 = 3.23; and
e3,23 _ 25
In view of this example it is clear that the exact
method of computing the 95 percent endpoint should be
reported along withg the time required for depuration to
this point.
In summary, the following procedures should be utilized
for analyzing oyster bioconcentration data:
1) Accurately tabulate and quality assure the residue
data, exposure concentrations, and sampling
procedures and periods .
2) Compute the desired multi-compartment kinetic
and/or non-linear parameter statistical equation
using log r as Hue and linear time data.
3) Plot the resulting curve (s) and data points.
27
-------
ES-3
August,- 1982
4) Conduct a lack-of-fit test to determine whether the
resulting equation(s) satisfactorily describe the
data.
5) If satisfied that extrapolation is reasonable,
compute the steady-state concentration, BCF, and
95 percent depuration endpoint using log
transformed data. Back transform the endpoints to
original concentration units for reporting.
The resulting constants (K]_, K2/ etc.) are required for
constructing the computed curve and for estimating the time
to 95 dercenfc depuration.i:
d . Temperature Measurements
In order to substantiate that temperature was maintained
within specified limits, it will be necessary to measure and
record temperature throughout the test. Requisite
instrumentation is readily available, easy to maintain, and
should not increase complexity or costs of the test.
Temperatures should be recorded hourly to prevent any severe
fluctuations in temperature that might affect growth
processes and/or chemical uptake.
B . Test Conditions
1. Test Species
a. Selection
The Eastern oyster, Crassostrea Virginia (Gmelin),
serves as a valuable indicator of biologically damaging
pollutants in estuaries due to. a number of important
characteristics. First, the oyster is a long-lived
sedentary filter feeder that is unable to move away from
exposure to environmental contaminants nor close its shell
for excessively long periods of time to avoid exposure. It
28
-------
ES-3
1982
accumulates organic (both biological and chemical) and
inorganic substances from aquatic ecosystems in a manner
which accurately reflects environmental changes and quality
(GaltsofE 1964).
Second, the oyster is economically important as a
commercial and recreational fishery resource and as a human
food source. For example, 1977 commercial landings were
valued at $52 million (Council on Environmental Quality
1979).
Third, the oyster occurs naturally over a wide
geographic; range and is locally abundant from Maine to the
Gulf of Mexico. This wide geographic range allows
comparative studies of control and exposure organisms under
differing environmental conditions.
Fourth, the oyster is readily cultured throughout its
life cycle under controlled conditions (Maurer .-ind Price
1967, Epifanio and Mootz 1976, Loosanoff and Davis, 1963).
In a properly equipped and maintained facility, the oyster
is a hardy species that can be maintained for long periods
of time with minimal effort.
Fifth, numerous morphological, physiological and
pathological studies have been completed on the oyster
(Wilbur and Yonge 1964, Galtsoff'1964 and Sindermann
1970). It has been claimed that the oyster is the best
known, most studied marine organism (Galtsoff 1964).
Sixth, Crassostrea virginica has been used extensively
as a bioassay organism because of its known sensitivity to a
wide variety of toxicants (LaRoche et al . 1973). It has
been shown to be an affective bioconcentrator of aromatic
hydrocarbons (Lee et al. 1978, Anderson 1978), the i:ise<:--
29
-------
ES-3
August, 1982
ticicles kepone (Bahner et al. 1977, Hansen et al. , 1976),
DDT (Lowe et al . 1971, Butler 1967) and chlordane (Parrish
et al. 1976), and various heavy metals (Frazier 1975, 1976,
1979 a,b) . In addition, most specific responses of oysters
to their environment have been studied and juantified
including shell deposition rate, breeding temperatures,
glycoge:'i content, salinity requirements, numbers of
reproductive cells, diseases and predators, and soft/hard
tissue ratios.
Although no forma.1 comparison of bioconcentration
factors among bivalves! has ibeen toublish^d, Butler! (1967) j
presented data which show that, in general, oysters
bioconcentrate insecticides to a greater d«-uree than most
other common bivalves. Average five-day bioconcentration
factors Cor seven pesticides ranged from 500 to 700 for the
hard clam, marsh clam, and asiatic clam to 1200 for the
oyster and 3000 for the soft shell clam. Sutler; concluded
that, on the basis of its greater bioconcentration factors,
the large body of knowledge concerning its biology, its
sessile nature and its extensive range, the oyster makes an
excellent biological monitor.
The voluminous literature on oyster biology is scattered
through numerous scientific publications. However, the
following references serve as suitable entries into the
field: Galtsoff (1964), LaRoche et al. (1973), Sparks
(1972), and Loosanoff and Davis (1963).
b. Sources
Oysters may be cultured in the laboratory, purchased
from culture facilities, or collected from a natural
population in an unpolluted area, free from epizootic
30
-------
33-3
August, 1982
disease. Procedures for collecting, transporting and
holding oysters are described in APHA, 1975. All oysters
• used for a particular test should be from the same source.
Test oysters should not have been used in a previous test,
either in a treatment or in a control.
c. Size
The test guidelines recommend using oysters between 30
and 50 mm in height. This range represents a synthesis of
the wide range of sizes reported in the 11 nerature • Various
workers have used oysters as small as 29 mm (Parrish et al.
1976) and as large as 120 mm (Scott and pliddaugh 19,78).
Typically, however, experimental oysters have ranged from 40
to 60 mm.
Butler and Lowe (1978) and APHA.(1975) recommend using
small (25 to 50 mra) oysters because they are active over a
wider range of temperatures and because they need less
space. The ASTM (1980) recommends 40 to 60 mm. Therefore,
in light of past experience and current recommendations, a
size range of 30 to 5.0 mm is justified.
d . Co ad i t io n
Oysters should be in a prespawn condition of gonadal
development prior to and during the test. A test is
unacceptable if oysters spawn during the test. ' For this
reason, and in cons iderntion of test temperature, it. is
recommended that oysters from natural areas be collected and
tested in the spring of the year. Gonadal condition of
oysters should be more certain in stocks obtained from
culture facilities. Prespawn condition should be confirmed
by measuring the condition and gonadal index of a randomly
selected representative sample of oysters to be used for
31
-------
ES-3
August, 1982
testing by the method of Scott and Middaugh (1978) and by
prep.-iri.ag and examining histological sections of tissues
from the same oysters by the method of Tripp (1974) -to
determine gonadal condition and to additionally ensure the
population is not diseased.
Although several authors utilized oysters from natural
populations for bioassay tests (Schimmel et al. 1978, Rawls
1977), other investigators utilized laboratory-reared
oysters (Conger et al. 1978, Cunningham and Tripp 1973) . As
demonstrated by Scott and Middaugh (1978), determining the
physiological condition of oysters is very important in
minimizing test variabilty- The depletion of gametes and
glycogen that occurs during spawning would certainly make
bioconcentration data impossible to analyze, and thus .should
be avoided by ensuring test oysters remain in a prespawn
cond it Ion.
Oysters collected from a natural population should be
collected at those times known to be free Eco-n inCluences of
recent spawning, such as the s-pring of the year. Gamete
production can be monitored by gros^ observation of
individual oysters and semiquanti t:a tive measurements of
-jonad development can be made by the method of Tripp
(1974). Oysters which are laboratory-reared should be
examined prior to use to determine if their growth and shell
thickness is consistent with known population averages
(Pruder and Bolton 1978).
Oysters are susceptible to a number of pathogens that
aay result in epizootics and mass mortalities in both
natural and cultural populations (Sindermann 1970). It is
necessary to determine that purchased oysters do not
32
-------
ES-3
August, 1982
originate from epizootic disease areas.
In addition to taking the physiological condition of the
oysters into account, examinations for parasitism and
disease should be conducted to ensure to the investigators
that the response of the oysters to the toxicant is not
influenced by such factors. Common oyster diseases and
procedures for their assessment are outlined in Cheng
(1970), Couch et al. (1974), Galtsoff (1964), Sindermann
(1970) and. Sparks (1972).
Oysters with shells heavily infested with mud worms
(Polydora webster.i) should not be used., the mud worm forms
black areas on the inner faces of oyster shells and make the
shells brittle (MacKenzie and Shearer 1961). Heavily
Infested oysters may become weakened and eventually die
(Roughley 1922, 1925). Oysters can be protected from
mudworms to some extent if they are reared off the bottom
(Loosanoff and Engle 1943).
2. Maintenance of Test Species
a. Feeding
The test guidelines permit supplemental deeding if
natural plankton concentrations are too low to support
oyster growth, or if artificial seawater systems are used.
This statement, of course, leaves open the question of what
plankton concentrations are adequate. This question cannot,
at present, be answered with any degree of accuracy.
Spifanio, et al . (1975) discussed the relationship between
filtration rate and algal densities and concluded that the
variability caused by temperature, animal size, particle
composition, and density prevents an accurate specification
of oyster nutrition. Their discussion presents information
33
-------
ES-3
August, 1982
suggesting that at 20 °C oysters feed most efficiently at
algal densities around 2 x 10^ cells per ml. However, there
are apparently no studies of the minimum necessary algal
densities at various temperatures for various sizes of
an imal.
In the absence of adequate information on oyster
nutrition in the wild, the best policy would appear to be
test any source water suspected to be inadequate to supply
growth. Juvenile oysters should be held in the testing
system to determine rates of growth over an' extended period
of time. This will give an estimate of the system's ability
to meet the lemands of the flow-through bioassay.
It should be pointed out that most estuarine and
nearshore waters will contain adequate quantities of
phytoplan^ton during the period when water temperatures are
suitable for testing. It is particularly true if the
testing facility is located near an area which supports
natural oyster populations. Therefore, it is unlikely that
food availability will be a major factor when ambient water
of suitable quality is used.
If supplemental feeding is necessary, the methods and
materials employed by experimental aquacultute facilities
should be utilized. Basically, -these consist of culturing
two or three algal species - Isochrysis galbana, Monochrysis
lu theri and Thai las ios lira pseudonna have been used
successfully by the University of Delaware (Epifanio et al.
1975 Epifanio and Mootz 1976) - to be fed to the oysters
either as the sole ration, or as a supplement to the natural
algal flora. Although the actual algal culture presents no
particular difficulties, the additional manpower ^nd capital
34
-------
ES-3
August, 1982
costs of supplemental feeding make it an undesirable
strategy. Ukeles (1971) has reviewed the nutritional
requirements of shellfish.
3. Facilities
a. General
The requirements for facilities as set forth in the test
guidelines are intended to ensure that the conditions in the
test chambers arrs as uniform as possible and that the actual
concentrations of test chemical in the test chambers are
similar to the intended concentrations.
; ,The test, gu ide lines require that flowing seawater be
i i ' i i
utilized. Static test design cannot be utilized due to
problems in maintaining the oysters in a state of good
health. The flow-through system more closely simulates the
natural exposure process, eliminating problems associated
with accumulation of organic material (and associated
bacteria which could lower dissolved oxygen) and toxic
metabolic products. Test chemicals are more thoroughly
mixed in a flow-through system and problems of so rut ion are
reduced .
Galtsoff (1964) found that oysters held in flowing
seawater at Woods Hole, Massachusetts, deposited a median of
1.4 milligrams of shell material per centimeter squared of
shell surface per day during the growing season. With
sufficient, suitable phytoplankton food in the dilution
v/ater, Epifanio et al. (1975) found that a small oyster
between 30 and 50 millimeters in height may deposit as much
as 1.0 millimeter oT peripheral new shell per day. Most
laboratory systems which have been designed to hold and
study the toxic response of oysters have employed a minimum
35
-------
ES-3
August, L982
flow-through volume of £ Lve liters of water per oyster per
hour (Butler et al. 1960).
b. Construction Materials
Due to hh^ toxicity of many heavy metals at low
concentrations (US SPA 1976) and the ability of metal pipe,
galvanized sheeting, laboratory t equipment, etc,- to leach
metals into water, no metal other than stainless steel
(preferably S316) should be used. In the same manner, un-
aged plastic!zed plastic (PVC) should not be used due to the
high toxicity of a main component, di-2-ethyl-hexyl
phthalate (Mayer and^Sanders 1973) and t-'ne ability of DEHP
to leach into aquaria systems from these materials
(Carmignani and Bennett 1976). To avoid any possible stress
due to exposure to low levels of metals, phthalatos, and
other potential contaminants, #316 stainless steel, glass
and pe rn.uo cocarbon plastics should be us id whenever
possible and economically feasible. If other materials
should be used, conditioning to a continuous flow of heated
dilution water should be performed for a minimum of 40
hours .
c. Test Substance Delivery System
To maximize the accuracy and precision of test results
developed through the use of this test guideline, the
quantity of test chemical introduced by the test chemical
delivery system should be as constant as possible from one
addition of test chemical to the n.^xt, fluctuations in the
quantity of test chemical introduced into the test chamber
may result in abnormally high or low response value (e.g.
EC50's) of the test organisms and in a ^ider spread of
response values in replicate tests. The greater the
36
-------
ES-3
1982
variation in the quantity of test chemical introduced, the
greater the potential for abnormalities and spread oE the
response values.
Variations in the quantity of dilution water entering
the test chambers during a gi/'en time interval may also
r'i'l
create undesirable difference's in test conditions between
test chambers. The concentrations of dissolved oxygen and
test chemical in a test chamber, for example, Qay decrease
more rapidly in chambers having lower flow rates.
Differences between test chambers' in the concentration of
dissolved oxygen, test phenical, .metabolic oroducts ,and
'II ' ' . I ^ I
degradation products, individually or in c n\'i>:<.nation, may
result in response values for the test organisms which are
inaccurate.
•The following criteria presented by Hods on (1979) should
be considered when selecting or designing a toxicant
delivery system: 1) if the delivery of dilution water stops,
so should delivery o the toxicant 2) consistency in
delivery amounts throughout the test periop 3) independence
from electrical failure 4) independence from temperature and
humidity fluctuations 5) capacity to deliver small
quantities 6) ease of construction, with few moving parts
and 7) ease of operation.
Any one of several toxicant devices can be used as long
as it has been shown to be accurate and reliable throughout
the testing period. The greater the variation in the
quantity of test chemical introduced, the greater the spread
of response values measured during testing. Syringe
injector systems (Barrows et al. 1980, Spehar et al. 1979),
metering pump systems (Veith et al . 1979), and modified
37
-------
ES-3
August, 1982
proportional dilators (Macek et al. 1975, Neeley et al.
1974) have been reported to be successfully used.
The solubility of the test compound should also be taken
into account in selecting an appropriate delivery system.
If the compound can be solubilized in water, a device
capable of delivering amounts of test solution greater than
1 milliliter (nl) will probably be needed. If a carrier ij
required, a system capable of accurately delivering small
amounts, less than 100 micro-liters (ul), will probably be
required to minimize the carrier concen-tration in the test
jsolution.
Each system should be calibrated prior to starting the
test to verify that the correct proportion of test chemical
to dilution water is delivered to the appropriate test
chambers ,
d . Test Chambers and Loading
Flexibility is allowed in the design of test chambers as
long as adequate space is provided Oor test oysters to meet
loading requirements. As a guideline ase the US EPA Bioassay
Procedures for the Ocean Disposal Permit Program Manual
(USEPA 1978), which recommends glass or fiberglassed wood
containers measuring 64 x 33 x 10 cm deep (25 x 14x4 inches)
to provide adequate space for 20 oysters. Such containers
permit adequate circulation of the water, while avoiding
physical agitation of- the oysters by the water current.
These containers hold about 18 L at 75% capacity and at a
flow rate of 100 L hour"1, will provide 5 L of water hour -1
oyster ~^-. Small oysters were reported to (ieed and grow
readily under these conditions.
Silicone adhes ive is the preferred bonding agent for
38
-------
fiS-3
August, 1982
constructing test chambers. It is inert, and the solvent it
generally contains (acetic acid) is easily washed away or
volatilized from the system. A minimum amount of the
adhesive should contact the test solution because it may
absorb test materials. If large amounts of, \:he adhesive are
needed for strength, it should be applied tb>' the outs ides of
chambers and apparatus to minimize contact.
e . Flow-through System
The test guidelines require that flowing seawater be
• '. i
utilized. Static test design cannot be utilized due to
problems, in maint.fining the oys.ters in a state pf. good
I 'if ' '
.health. The flow-through system more closely simulates the
natural iexposure process, eliminating problems associated
with accumulation of organic material (and associated
bacteria, w'n ich could lower dissolved oxygen) and toxic
metabolic products. Test chemicals are more thoroughly
i
mixed in a flow-through system and problems of so rot ion are
reduced.
Galtsoff (1964) found that oysters held in flowing
seawater at Woods Hole, Massachusetts, deposited a median of
1.4 milligrams of shell material per centimeter squared of
shell surface per day during the growing season. With
sufficient, suitable phytoplankton food in the dilution
water, Epifanio et al. (1975) found that a small oyster
between 30 and 50 millimeters in height may deposit as much
as 1.0 millimeter of peripheral new shell per day- Most
laboratory systems which have been designed to hold and
study the toxic response of oysters have employed a minimum
flow-through volume of five liters of water per oyster pec
hour (Butler et al. 1960).
39
-------
ES-3
August, L982
f. Cleaning
Before use, test systems should be cleaned to remove
dust, dirt, and other debris and any residues that may
remain from previous use of the system. Any of these
substances may affect the resales of a test by sorption of
test materials or by exerting an adverse oflfiyct on test
organisms. New chambers should be cleaned to remove any
diet or: chemical residues remaining from manufacture or
accumulated during storage. Detergent is used to remove
hydrophobia or lipid-like substances. Acetone is used for
the same purpose and to remove any dotergent residues. It
i l i i ! I ' '
is important to use pesticide-free aob>rie to prevent the
contamination of the chambers with pesticides which
influence the outcome of the test. Nitric acid is used to
clean metal residues from the system. A final thorough
rinse with water washes away the nitric acid residues. At
the end of a test, test systems should be washed in
preparation for the next test. It is easier to clean the
equipment before chemical residues and organic matter become
embedded or absorbed into the equipment.
g. Dilution Water
The test guidelines require the availability of an
adequate and dependable supply of clean, unfiltered
estuarine or ocean water. This water should not deviate
substantially from the desired temperature and salinity
ranges (APHA 1975, Pruder and Bolton 1978). General
requirements for a water supply and water system are
described in Standard Methods (APHA 1975) and in Epifanio
and Mootz (1976). If necessary, artificial seawater may be
used for limited studies (Conger et al. 1978). However, the
40
-------
,ES-3
August,- l'HV2
large volumes required to maintain the flow and loading plus
the no?t*3 to add food to maintain growth makes use of
artificial seawater problematic. Refer to Spotte (1979) for
methods- to prepare and mix large volumes of artificial
seawater. p
' 'I
The flow-through system should supply at least one litler
of water per oyster pec hour. The ASTM Proposed Standard
Practice (ASTM 1980b) recommends one liter per hour,
although this preliminary recommendation has been questioned
in the review process, and is likely to be changed.
Behind the AST'l cec, omio.endat|ipn is a factor which, has
only recently received the attention it merits, namely the
problem of waste disposal. Dilution water containing a
potentially hazardous t[es t chemical cannot be discharged
directly into natural waters. Some form of preliminary
treatment is necessary. As the amount of dilution water
increases, treatment facilities and costs go up. These
considerations dictate the use of the lowest flo rate ../a Ic'a
will support oyster growth.
The recommendation that the flow rate be at least one
liter per oyster per hour is clearly at the lower end of the
range of reported values. Parrish et al . (1976) exposed
oysters to chlordane in 7.5 liters of water per oyster per
hour. Scott et al . (1979) used five liters per houc.
Schimmel et al. (1973) supplied approximately six liters per
oyster per hour to test sodium pentachloro-ohenate, but 11
liters to test lindane and 3HC (Schimmel et al. 1977).
Scott and Middaugh (1978) used at least five liters per
oyster per hour in their acclimation tanks. Standard
Methods ( APHA 1975) advises holding spawning stock in at
41
-------
ES-3
August, 1982
least seven liters per animal per hour, but those are larger
oysters (7.5 to 15 cm) than are used in the test
guidelines. For test oysters, "a minimum of five liters per
hour of seawater per oyster provides adequate growing
conditions" (APHA 1975). However, recent experience at the
US EPA Gulf Breeze Laboratory (Schimmel, personal
communication) and at private facilities (Parrish, personal
communication) has demonstrated the effectiveness of using
one liter. Since that flow rate is economical and has been
proven effective, its use in .the test guidelines is
justified. The one liter figure is a minimum; if a facility
; I i I I f '
can support a higher flow rate, it should by all means use
the higher flow. It is important that the Plow of water be
I
constant. If the flow is interrapr.-d, the oysters will
quickly deplete the food supply in the stagnated water and
will increase the levels of metabolic .vastes. These factors
-will Couse the oysters to close their shells, which will
invalidate the tests.
Many previous tests utilizing the- oyster as a test
organism were conducted at salinity regimes native to the
testing facility. Thus, tests conducted by Schirnniel et al .
(1978) were conducted at a salinity range between 18 and 23
parts per thousand, whereas those completed by Cunningham
(1976) were run at salinities close to full-strength
seawater (34 parts per thousand).
Salinity is difficult to control in a natural flow-
through system, and controlling it introduces an element of
artificiality to the test. Therefore, the dilution water
should be drawn from an area where the expected range of
salinity is within the oyster's optimum.
42
-------
ES-3
August, 1982
Regardless of the system used, salinity should be
monitored daily during the test. If significant changes
occur, due to severe weather changes or system malfuncion,
for example, the validity of the test is challenged and the
test should be repeated.
To avo 3 possible inconsistencies and inaccuracies in
test resul 3, healthy oysters are needed for use in toxicity
tests. There is also a need to determine that the dilution
water, whatever its source, is 'able to maintain the oysters
to be used in a healthy condition for the duration of the
holding and testing periods. . .
J I i = '•• , i ! I ' '
An appropriate way to make that determination is to
place oysters in the dilution water for an extended period
of time and observe their behavior, growth and
development. Ideally, those observations should be made by
an experienced oyster biologist familiar with certain stress
reactions which are difficult for an untrained observer to
identify.
Part icula'te matter and gas bubbles, if present in the
dilution water, may clog the toxicant delivery system used
in flow-through tests. Gas bubbles also may cause excess ive
loss of volatile test chemicals. Either circumstance may
alter the concentration of test chemical to which the test
oysters are exposed. To avoid this problem an apparatus
capable of removing particulate 'matter or gas bubbles from
the dilution water may be required, I?, the dilution water
is heated prior to use, it may also be necessary to de-
saturate the water from >100% of oxyjen saturation. Penrose
and Squires (1976) describe a suitable apparatus for this.
43
-------
ES-3
August, 1982
An adequate supply of dissolved oxygen should be
available to the oysters. To facilitate this, the dilution
water or holding water should be at 90-100% of oxygen
saturation prior to delivery to the holding tanks or test
sys tern.
h. Carriers ;
Carriers may be used to aid in the dissolution of test
compounds into dilution water only after significant efforts
to dissolve it in dilution water or dilution water stocks
have failed. Schoor (1975) believes that the use of a
carrier may interfere with the uptake of the>test compqund
by the test organism; if the carrier molecules affect the
adsorption of the test compound at the gill surface, a
I
change in the rate of transport into the test organism may
result. The author,also states thai: \.:'\-i use of a carrier
may increase the concentration OL compound in the, test
solution above solubility by creating a stable wat(er
, i
emu Is ion.
Since there is little information available on the
effects of carriers on oysters, follow precautionary usage
procedures that have been established with fish. When a
carrier is used, triethylene glycol (TEG), dimethyl
formamide (DMF), or acetone may be used. The solvents
should be tried in the order stated due to their relative
toxicity to fathead minnows as reported by Cardwell et al .
(manuscript 1930). The minimum amount should be used and
the concentration of TEG should not exceed 80 mg/1, the MATC
(maximum acceptable toxicant concentration) value.
Concentrations of DMF and acetone should not exceed 5.0
mg/1, the MATC for DMF. Although there is no MATC value foe
44
-------
ES-3
igur, t, 1982
acetone, its acute toxicity is similar to that of DMF.
Ethanol should not be used due to its tendency to
stimulate the excessive growth of bacteria in the test
chambers .
4. Environmental Conditions
a. Dissol\ 'd Oxygen (See Section on
Dilution Water)
b. Temperature
It is desirable to standardize the range of test
temperatures to the extent possible so as to avoid
variability between different laboratories in widespread
I ^ ' i i t i >
geographic areas. Also, tests on some specific substances
will ,vary significantly at temperatures as -auch as 10°C
apart, a range commonly experienced between, for instance,
New York and South Carolina. Waldichuk (1974) presented
data showing such a phenomenon In the case of cadmium.
Gun tec (1957) set forth the relationship between oyster
growth and temperature, as determined by regional
location. Based on Gunter's data and work conducted by
Butler (1953), it is clear that regional temperature
differences should be taken into account in es tablishing
experimental bioassay conditions. Most bioassay tests
utilizing the oyster were conducted at ambient temperatures
of the dilution water at the testing facility. Tests
conducted by Schimmel et al. (1978) were conducted at
ambient temperatures between 7 and 9°C; those completed by
Cunningham (1976) were run at times at temperatures as high
as 25°C.
Within its natural range, Crassostrea virginica is found
at great temperature extremes, ranging from 1°C to 46°C
45
-------
ES-3
August, 1982
(Epifanio et al. 1976). However, the optimal temperature of
oyster growth has been reported as between 15°C and 25°C in
the Gulf Coast (Collier 1954) and between 13 °C and 22 °C in
Long Island Sound (LoosanoEE and Nomeiko 1949). Although
oysters in the southern range, such as in the Gulf of
Mexico, have a greater growth per year, he maximum growth
per day occurs during the summer in oysters located in the
mid-Atlantic region. Loos an of f; (1958) found that oysters
from Long Island have maximum ciliary (feeding) activity at
approximately 25°C. Cunningham (1976) and others observed
declining shell growth|during cold ambient water temperature
periods. Laboratory studies conducted by Eoifanio and Mootz
(1976) utilized a controlled range between 16° and 26°C
throughout the year. The ASTM (198Ob) proposed standards
for tests with oysters call for a test temperature between
3°C and 28°C. Butler and Lowe (1978) recommend' that the
source water be between 15°C and 30°C. The test guidelines
specify 20°C (with a permissable short term deviation within
15°-25°C) because it maximizes filtration, growth, and is
sufficiently low that oysters can be maintained in a
prespawn condition. In addition, it will tend to diminish
tha influence of disease on test results because the fungus,
Dermocystidium marina, causes high levels of mortality above
25°C (Hewatt and Andrews 1957).
According to Stickney (1979), C. viginica requires a
temperature range of 21 to 27°C. for spawning and can be
conditioned to spawn within about 6 weeks in the winter if
exposed to a temperature range of 23 to 24°C.. even though
spawning naturally occurs in the spring.
The 20°C test temperature was selected as a compromise,
45
-------
ES-3
August, 1932
striking a balance between the temperature that would
maximize the physiological activities of the oyster (i.e.
ciliary movement, water transport, metabolism, etc.) which
would thereby enhance the uptake and exposure to the test
solution on the one hand, and on the other hand assuring
that the oysters would not spawn on the other hand. The
ideal physiological-maximum temperature Cor the oyster is
approximately 25°C, but oysters spawn aoove 20°C. Since the
oyster approaches • naximum -phys iological activities at 20°C,
and since this temperature does not induce oysters to spawn,
it was selected. It is realized that temperature variation
T ! ' : ! : 1
may occur in controlling the volumes of water r«j:n. red in
flow-through systems. Since prolonged exposure of .oysters
to temperatures above 20°C may induce spawning, it is
preferred .that variations in test temperature be held to a
minimum or held to temperatures below 20°C. The effect of
temperature on the oyster is discussed in Galtsoff (1964).
A standardized temperature is ordinarily desirable in
bioassay testing. This is because the toxicity of many
substances varies with temperature (Tucker and Leitzke 1979,
Frazier 1979a, Waldichuk 1974).
The effects of sudden temperature changes on organisms
may range from death to temporary impairment of
physiological functions, depending on the acclimation
temperature, the magnitude of the temperature change, the
temperature tolerance of the species, and the circumstances
and duration of the exposure. To avoid any undue stress,
accurate temperature control devices should be used to both
maintain constant temperatures, and to gradually increase or
decrease the temperature during acclimation procedures.
47
-------
ES-3
August, 1982
Such mechanisms have been described by Defoe (1977) and
Lemke and Dawson (1979).
c. Light
The duration and intensity of light are not important
environmental variables in oyster tests. Oysters t-^/e no
visual organ ancl ace not sensitive to light (Galtsoff,
1964). Therefore, no lighting regime is required.
d. Salinity^ (See Section on Dilution Water)
C. Reporting
The sponsor should submit to the Agency all data
developed during the test that are suggestive or predictive
of oyster toxicity and bioconcen'cration. In addition,
information on water quality, experimental design,
equipment, and oyster condition are required beause these
data have a bearing on the validity of the test. If testing
specifications are followed, the sponsor should report that
specified procedures were followed arul present the
results. If alternative procedures were used instead of
those recommended in the test guideline, then the protocol
used should be fully described and justified.
Test temperature, chemical concentrations, test data,
concentration response curves, and statistical analyses
should all be reported. The justification for this body of
information is contained in this support document. If
species other than the recommended were used, the rationale
for the selection of the other species should be provided.
III. Economic Aspects
The US EPA awarded a contract to Enviro Control, Inc.
(1930) to provide the Agency with an estimate of the cost
for performing the oyster acute toxicity and
48
-------
ES-3
August, 1982
bioconcentration tests according to the test guidelines.
•Two estimates were provided; a protocol estimate and a
laboratory price survey estimate,
The protocol best estimate for the oyster acute toxicity
test was $1480. This estimate was prepared by separating
the guideline into individual tas^s and estimating th' hours
to accomplish each task. Hourly rates were then appl ed to
yield .a total direct labor charge. An overhead rate of 115
percent, other direct costs (for laboratory supplies and
reagents) of $75.00, a general and administrative rate of 10
percent,- and a; fee ofi 20 percent */ere then added tol the
direct labor -charge to yield the final estimate.
Enviro Control estimated that differences in salaries,
equipment, overhead costs and other factors between
laboratories would result in as much as ^ 0 percent variation
from this estimate. Consequently they estimated that test
costs could range from $740 to $2221.
The laboratory price survey best estimate was $900.00
for the oys'er acute toxicity test. Two laboratories
supplied estimates of their costs to perform the tests
according to this guideline. These costs ranged from
$700.00 to $1100.00. The reported estimate is the mean
value calculated from the individual costs.
The protocol best estimate for the oyster
bioconcentration test was $7680. This estimate was prepared
by the same method of the oyster acute toxicity test, with
the exception that the other direct costs totaled $250. The
test cost was estimated to range from $3840 to $11,520.
The laboratory price survey best estimate was $8092 for
the oyster bioconcentration test. Four laboratories
49
-------
ES-3
August, 1982
supplied estimates ranging from $4,000 to $LO.-000. The
reported estimate is the meaa value calculated from the
individual costs.
50
-------
ES-3
August, 1982
IV. REFERENCES
AOAC. 1975. Association of Official Analytical
Chemists. Official methods for analysis of the
Association of Official Analytical Chemists. 12th
Edition. Washington, DC: Association of Official
Analytical Chemists.
APHA. 1975. American Public Health Association, Water
Pollution Control Federation. Standard methods for the
examination of water and wastewater. 14th Edition. New
York: American Public Health Association.
Anderson RD and Anderson JW. 1976. Oil bioassays with
the American oyster, Crassostrea virginica (Gmelin)
Proc. Nat. Shelfish Assoc. 65:38-42.
1 '
Anderson RS. 1978. Benzo(a) pyrene'metabolism in the
American oyster Crassostrea virginica, Ecol. Res. Series
U.S. EnvironmentliT Protection Agency, Gulf Breeze, FL:
25 pp. #EPA-600/3-78-009.
ASTM. 1979. American Society for Testing and
Material. Annual book of ASTM standards. Part 3.,
Water. Philadelphia, PA: American Society for Testing
and Material .
ASTM. 1980a. American Society for Testing and
Materials. Standard practice for conducting basic acute
toxicity test with fishes, macroinvertebrates, and
amphibians. Philadelphia, PA: American Society for
Testing and Materials. E729-80.
ASTM. 1980b. American Society for Testing and
Materials. Proposed Standard practice for conducting
bio-concentration tests with fishes and saltwater
bivalve molluscs. Draft No. 10 August 22, 1980.
Bahner LH, Wilson AJ, Sheppard JM, Patrick JM, Goodman
LR, Walsh GE. 1977. Kepone accumulation, transfer and
loss through estuarine food chains. Ches. Sci.
18(3):299-308.
51
-------
ES-3
August, 1982'
Barrows ME, Petrocelli SR, Macek KJ. 1980.
Sioconcentration and elimination of selected water
pollutants by bluegill sunfish ( Lepomis macrochirus ) .
In: Ann Arbor Science Pub., Inc. Dynamics, exposure and
hazard assessment of toxic chemicals Hague R. Ann
Arbor, MI.
Bishop WE and Maki AW. 1980. A critical comparison of
two bioconcentration test methods. In: Aquatic
toxicology. Eaton JG, Parish PR, and Hendricks AC, eds .
Philadelphia, PA: American Society for Testing and
Materials.
Blau GE and Agin GL. 1978. A users manual for BIOFAC: A
computer program for characterizing the ratio of uptake
and clearance of chemical in aquatic organisms.
Midland, fMI: Dow Chemical Co.
I ! 'i
Bliss CI. 1935. The calculation of the dosage-
mortality curve. Ann. Appl. Biol. 22:134-307-
Branson DR, Blau GE, Alexander HC, ,Neely WB. 1975.
Bio-concentration of 2,2, 4,4 - tetrachlorobiphenyl in
rainbow trout as measured by an accelerated test. Tran.
Am. Fish. Soc. 4:785-792,
Brodtmann NV. 1970. Assimilation of 1,1,1 trich oro-
2,2-bis (p-chlorophenyl) ethane (DDT) by Crassos t ;ea
virginica. Bull. Environ. Contam. Toxicol. 5:455-462.
Butler PA. 1953. Oyster growth as affected by
latitudi-nal temperature gradients. Com ml. Fish. Rev.
352-7-12.
. 1965. Reaction of estuarine molluscs to
some environmental factors. Publ. Health. Serv. Publ.
999-WP-25.
. 1967. Pesticide residues in estuarine
molluscs. In: Nat Symp. on Estuarine Pollution,
Stanford Univ., 1967. Gulf Breeze, FL: U.S. Bureau
Commercial Fisheries Biol. Lab, 107-121.
Butler PA and Lowe JI. 1978. Flowing seawater toxicity
test using oysters (Crassostrea virginica); In: Bioassay
procedures for the Ocean Disposal Permit Program Gulf
Breeze, FL: US. Environmental Protection Agency. EPA-
600/9-78-010.
52
-------
ES-3
August, 1982
Butler PA, Wilson AJ, Rick AJ. 1960. Effects of pesti-
cides on oysters. Proc. Nat. Shellfish. Assoc. 51:23-
32.
Cairns J and Dickson JKL, eds . 1973. Biological methods
for the assessment of water quality. Philadelphia, PA:
American Society for Testing and Materials. ASTM-STP
528.
Cardwell RP, Foreman DG, Payne TR, Wilbur OJ. 1980.
Acute and chronic toxicity of four organic chemicals to
fish. Manuscript.
Carmignani GM and Bennett JP. 1976. Leaching of
plastics used in closed aquaculture systems.
Aquaculture 7:89-91.
Cember H, Curtis, EH, Blaylock BG. 1978. Mercury -biocon-
centra'tion in fish: temperature and concentration
effects. Environ. Pollution 17:311-319.
Cheng TC. 1970. Marine molluscs as hosts for
symbioses: with a review of known parasites of
commercially important species. Vol. 5. In: Russell
F3, ed. Advances of marine biology. New York: Academic
Press .
Chiou CT, Freed UH, Schmedding DW, Kohnert RL. 1977.
Partition coefficient and bioaccumulation of selected
organic chemicals. Env. Sci. Tech. 1L(5): 475-478.
Chipman WA, Rice TR, Price TJ. 1958. Uptake and
accumu-lation of radioactive zinc by marine plankton,
fish and shellfish. U.S. Fish Wildl. Serv. Fish. Bull.
58:279-292.
Collier A. 1954. A study of the response of oysters to
temperature, and some long range ecological
interpretaions . Nat. Shellfish. Assoc. 1953:13-38.
Conger KA, Swift ML, Reeves JB III, Lakshmanan S.
1978. Shell growth of unfed oysters in the
laboratory: a sublethal bioassay system for
pollutants. Life Sciences. Vol. 22, Pergamon Press,
N.Y.: pp. 245-254.
53
-------
ES-3
August, 1982
Couch J, Gardner G, Harshberger JC, Tripp MR, Yevich
PP. 1974. Histological and physiological evaluations
in some marine feuna. In: Marine Bioassays, Workshop
Proceedings 1974. Washington, D.C.: American Petroleum
Institute, Environmental Protection Agency, Marine
Technology Society, pp. 156-173.
Council on Environmental Quality. 1979. Environmental
Quality - 1979. The tenth annual report of the Council
on Environmental Quality, Dec. 1979., Washington, DC:
U.S. Government Printing Office.
Cunningham PA and Tripp MR. 1973. Accumulation and
depuration of mercury in the American oyster,
Crassostrea virginica. Mar. Biol.20(l):14-19.
Cunningham PA. 1976. Inhibition of shell growth in the
presence of mercury ; a1 ndj s ubs equent recovery pf 'puvenile
oysters. Proc. Nat. 'Shellfish. Assoc:J 66:1-5. T
Defoe DL. 1977. Temperature safety device for aquatic
laboratory systems. Prog. Fish Cult. 39:131.
Doudoroff P, Anderson BG, Burdick GE, ;Galtsoff ,PS, Hart
WB, Patrick R, Strong ER, Suber EW, Van Horn WM.
1951. Bioassay Methods for the evaluation of acute
toxicity of industrial wastes to fish. Sewage Ind .
Wastes 23:1380-97.
Enviro Control Inc. 1980. Cost analysis methodology
and protocol estimates: ecotoxicity standards.
Submitted to Office of Regulatory Analysis. Washington,
DC: U.S. Environmental Protection Agency.
Epifanio CE, Sirna R, Pruder G. 1975. Mariculture of
shellfish in controlled environments: a prognosis.
Aqua-culture 5:227-241.
Epifanio CE. 1979. Growth in bivalve molluscs:
nutritional effects of two or more species of algae in
diets fed to the American oysters, Crassostrea virginica
(Gmelin) and the hard clam, Mercenaria mercenarTa
(L.). Aquaculture 18:1-12.
Epifanio CE and Mootz CA. 1976. Growth of oysters in a
re-circulating maricultural system. Proc. Nat.
Shellfish Assoc. 65:32-37.
54
-------
ES-3
August, 1982
Esvelt LA and Connors JD. 1971. Continuous-flow fish
bioassay apparatus for municipal and industrial
effluents. In: Esvelt LA, Kaufman WJ, Selleck RE eds.
Toxicity removal from municipal wastewaters. Sanitary
Research Engineering Laboratory. Berkeley, CA: Univ.
Calif. Berkeley.
Fingerman M. 1959. Effects of temperature and salinity
on ciliary activity in the oyster. Technical Note No.
50, Comml. Fish. Rev. 21(2):10-11.
Frazier JM. 1975. The dynamics of metals in the
American oyster, Crassostrea virginica. I. Seasonal
effects. Ches. Sci. 16(3):162-171.
. 1976. The dynamics of metals in the
American oyster, Crassostrea virginica. II.
Environmental effects. Ches. Sci. 17(3):188-197.
. 1979a. Bioaccumulation of' cadmium in
marine organisms. Env. Health Perspectives 28:75-79.
. 1979b. Factors influencing the
bioaccumula-1ion of metals in the American oyster,
Crassostrea virginica. In: Intern. Conf. of Management
and Control "of Heavy Metals in the Env. Sept. 18-21,
1979. London, England: Imperial College.
Galtsoff PS. 1942, Accumulation of manganese and
sexual cycle in Ostrea virginica. Physiol. Zool.
15:210-215.
1964. The American Oyster, Crassos trea
virginica Gmelin. U.S. Fish. Wildl. Serv- Fish. Bull.
64:1-480.
Gunter G. 1957. Temperature. In: Hedgpeth JW ed .
Treatise on marine ecology and paleoecology, Geological.
Society of America, Memoir 67, Vol. 1. Baltimore, MD:
Waverly Press, pp. 159-184.
Hamelink JL. 1977. Current bioconcentration test
methods and theory. In: Mayer FL', and Hamelink JL, eds.
Aquatic toxicology and hazard evaluation, ASTM STP 634,
Philadelphia, PA: American Society for Testing and
Materials, pp. 149-161.
55
-------
Augus t,
ES-3
1982
Hansen DJ, Wilson AJ, Niramo DWR, Schimmel SC, Bahner LH,
Huggett R. 1976. Kepone: hazard to aquatic
organisms. Science 193(4253): 528.
Hansch C, Leo A, Nikaitani D J. 1972. On the additive-
constitutive character of partition coefficients. Org.
Chem. , 37: 3090-3092.
Hedgpeth JW. 1966. A treatise on marinev .ecology and
oaleoecology. Vol. I. Ecology. National; j Research
Council. Washington, DC: National Academyji of Sciences,
1296 pp. ''
Hewatt WG and Andrews JD. 1957- Temperature control
experiments on the fungus disease Derrnocystidium marinum
of oysters. Proc. Nat. Shellfish. Assoc. 46:129-133.
Hiltner RJ and Wickman HJ.
Biol.,,Chem. 38s205-221.
1919. Zinc in oysters. J.
Hodson PV. 1979. Metering device
bioassays with aquatic organisms.
41:1294l31.
for toxicants used
Prog. Fish Cult.
in
Kenaga EE and Goring CAI. 1980. Relationship between
water solubility, soil-sorption, octanol -water
partitioning, and concentration of chemicals in biota.
In: Eaton J, Parrish PR, Hendricks AC eds. Aquatic
toxicology, ASTM STP 707. Ame lean Society for testing
and materials. In press.
Krieger RL, Gee S J, Lira LO, Ross JH, Wilson A, Alpers C,
Wellings SR. 1979. Disposition of toxic substances in
mussels (Mytilus californianus) ; Preliminary metabolic
and histological studies. In: Khan M A Q ed. Pesticide
and xenobiotic metabolism in aquatic organisms.
Washington, DC: American Chemical Society, pp. 259-277.
Krzeminski SF, Gilbert JT, Ritts JA. 1977- A pharmaco-
kinetic model for predicting pesticide residues in
fish. Arch. Environ. Contam. Toxicol. 5:157-166.
LaRoche G, Eisler R, Tarzwell CM. 1970. Bioassay pro-
cedures for oil and oil dispersant toxicity evaluation.
J. Water Pollut. Cont. Fed. 42:1982.
56
-------
ES-3
August, 1982
LaRoche G, Gardner GP, Eisler R, Jackin E, Yevich PP,
Zaroogian G. 19.73. Analysis of toxic responses in
marine poikilotherms. In: Glass GE, ed. Bioassay
techniques and environmental chemistry. Ann Arbor, MI:
Ann Arbor Science Publishers, pp. 199-216.
Lee RF, Gardner WS, Anderson JW, Blaylock JW, Barwell-
Clarke J. 1978. Fate of polycyclic aromatic hydrocar-
bons in controlled ecosystem enclosures. Env. Sci.
Tech. 12(7):832-843.
Lemke AE and Dawson WF. 1979. Temperature monitoring
and safety control device. Prog. Fish Cult. 41:165-166.
Leo A, Hansch C, Elkins D. 1971. Partition
coefficients and their uses. Chem. Reviews 7:525-616.
Lichatowich JW, O'Keefe PW, Strand JA, Temple ton WL.
1973. Development of methodology|and apparatus for the
bioassay of oil. In: Proc. Joint1 Conf. Prevention and
Control of Oil Spills. Washington, DC: American
Petroleum Institute, U.S. Environmental Protection
Agency, U.S. Coast Guard.
Loosanoff VL and Davis HC. 1963. Rearing of bivalve
molluscs. Adv. Mar. Biol. 1:1-136.
Loosanoff VL and Nomejko CA. 1949. Growth of oysters,
0. virginica, during different months. Biol. Bull.
~9~7:82-94.
Loosanoff V. 1958. Some aspects of behavior of oysters
at different temperates. Biol. Bull. 114(1) :57-70 .
Lowe JI, Wilson PD, Rick AJ, Wilson AJ. 1971. Chronic
exposure of oysters to DDT, toxaphene and parathion.
Proc. Nat. Shellfish. Assoc. 61:71-79.
Lowe JI, Parrish PR, Patrick JM, Forester J. 1972.
Effects of polychlorinated biphenyl Aroclor 1254 on the
American oyster, Crassostrea virginica. Mar. Biol.
17(3): 209-214.
57
-------
ES-3
August, 1982
Macek KJ, Barrows ME, Krasny RF, Sleight BH III.
1975. Bioconcentration of 14 C-pesticides by bluegill
sanfish during continuous aqueous exposure. In: Veith
GD, Konasewich DE, ed . Symposium on structure-activity
correlation in studies of toxicity and bioconcentration
with aquatic organisms. Windsor, Ont: International
Joint Commission, Great Lakes Research Advisory Brd.,
op. 119-141.
Mayer FL and Sanders HO. 1973. Toxicity of phthalic
acid esters in aquatic organisms. Environ. Health
Perspective 3:153-157.
Maurer D and Price KS. 1967. Holding and spawning
Delaware Bay oysters (Grasses trea virginica) out of
season. I. Laboratory facilities for retarding
spawning. Proc. Nat. Shellfish.. Assoc. 158:71-77.
Mela.n^on MJ Jr. andj,Lech JJ.l' 1979. Uptake,
biotrans formation, dispos itio.ny and elimination of 2-
methylnaphthalene and naphthalene in several fish
species. In: eds. . Aquatic toxicology. ASTM STP 667,
Marking LL and Kimerl'e RA, eds. Philadelphia, PA:
American Society of Testing and. Materials, pp. 5-22.
I
Neely WB. 1978. Estimating rate constants for uptake
and clearanc of chemicals by fish. Presented at Gordon
Conf. : Reactions and Fates of Organics in Natural
Waters.
Neely WB, Branson DR, Blau GF. 1974. Partition
coefficient to measure bioconcentration potential of
organic chemicals in fish. Environ. Sci. Technol.
8:1113-1115.
Parrish PR, Schimmel SC, Hansen DJ, Patrick JM Jr and
Forester J. 1976. Chlordane: Effects on several
estuarine organisms. J. Toxicol. En'v . .Health. 1:485-
494.
Penrose WR and Squiras JM. 1976. Two devices for
removing supersaturating gases in aquarium systems.
Trans. Amer. Fish. Soc. 105( 1): 116-118 .
58
-------
ES-3
August, 1982
Pruder GD and Bolton ET. 1978. System configuration
and performance: bivalve molluscan mariculture. In:
Avault JW, ed. Proc. Ninth Ann. Mtg. World Mariculture
Soc., pp. 747-759.
Rawls CK. 1977. Field studies of shell regrowth as a
bioindicator of eastern oyster (Crassostrea virginica
G.) response to 2, 4-D BEE in Maryland tidewaters. Ches.
Sci. 18(3) -.266-211.
Rubinstein NI, Wilkes FG , D'Asaro CN, Sommers C.
1980. The effects of contaminated sediments on repre-
sentative estuarine species and developing benthic
communities. In: Baker RA, ed. Contaminants and
Sediments. Vol. 1. Ann Arbor, MI: Ann Arbor Science
Publ., pp. 445-461.
Schimmel SC, Patrick Jr. JM, Faas LF. 1978. Effects of
sodium pentaqhlorophenate on several estuarine animals;
toxicity, uptake and depuration. In: Ra KR ed. Penta-
chlorophenol. New York: Plenum Publ. Co., pp. 147-155.
Schimmel SC, Patrick Jr. JM, Forester J. 1976.
Heptachlor: toxicity to and uptake by several estuarine
organisms. J. Toxicol. Environ. Health l(b):955-65.
. 1977. Toxicity and bioconcentration of BHC
and lindane in selected estuarine animals. Arch.
Environ. Contam. Toxicol. 6: 355-363.
Schoor WP- 1975. Problems associated with low-
solubility compounds in aquatic toxicity tests:
theoretical model and solubility characteristics of
Aroclor® 1254 in water. Water Research 9:937-944.
Scott GI, Kling'ensmith S, Middaugh DP. 1979.
Unpublished research on chlorination studies using
Crassostrea virginica. In: Research Review, 1979. Gulf
Breeze,FL: EPA Environmental Research Lab, pp. 31.
Scott GI and Middaugh DP. 1978. Seasonal chronic
toxicity of chlorination to the American oyster,
Crassostrea virginica (G .) In: Water chlorination:
environmental impact and health effects. Vol. 2. Ann
Arbor, MI: Ann Arbor Science, pp. 311-328.
59
-------
ES-3
August, 1982
Shuster C and Pringle B. 1969. Trace metal
accumulation by the American oyster, Crassos trea
virginica. Proc. Nat. Shellfish. Assoc. 61:71-79.
Sindermann C. 1970. Principal diseases of marine fish
and shellfish. New York: Academic Press.
Sparks AK. 1972. Invertebrate pathology., non-communi-
cable diseases. New York: Academic Press.
Spehar RC, Holcombe GW, Carlson RW, Drummond RA, Yount
JD, Pickering QH. 1979. Effects of pollution on
freshwater fish. J. Water Pollut. Control. Fed.
51(6) :1616-1694.
Spotte S. 1979. Fish and invertebrate culture, water
management in closed systems. 2nd ed. New York: John
Wiley and Sons . I :
Spraque JB. 1969. Measurement of pollutant toxicity to
fish. In: Bioassay methods for aauipe toxicity. Water
Res. 3:793-321.
Steel R and Torrie J. 1960. Principles and procedures
of statistics. New York: McGraw-Hill Book Co.
Stegeman J. 1974. Hydrocarbons in shellfish
chronically exposed to low levels of fuel oil. In:
Vernberg FJ and Vernberg WB, eds . Pollution and
physiology of marine organisms. New York: Academic
Press. pp. 329-348.
Stegeman JJ and Teal JM. 1973. Accumulation, release
and retention of petroleum hydrocarbons, by the oyster
Crassostrea virginica. Mar. Biol. 22:37-44.
Strickland JDH and Parsons TR. 1972. A practical' hand-
book of seawater analysis, Bulletin 167. Ottawa,
Canada: Fisheries Research Canada.
Tinsman J and Maurer D. 1974. The effects of thermal
effluent on the American oyster Crassostrea virginica
(Gmelin) in Indian River Bay, Delaware. College of
Marine Studies. Newark, DL: Univ. of Delaware, NTIS
#PB246-410.
60
-------
ES-3
August, 1982
Tripp MR. 1974. Effects of organophosphate pesticides
on adult oysters (Crassostrea virginica). In: Vernberg
F'J and Vernberg WB, eds. Pollution and physiology of
marine organisms. New York: Academic Press, pp. 225-
236.
Tucker RK and Leitzke JS. 1979. Comparative toxicology
of insecticides for vertebrate wildlife and fish.
Pharmac. Ther. 6:167-220.
Ukeles R. 1971. Nutritional requirements in shellfish
culture. In: Price KS and Maurer DL (eds). Proceedings
of the Conference on Artificial Propagation of
Commercially Valuable Shellfish - Oysters. Univ. of
Delaware, Newark, DL. pp. 43-64.
USEPA. 1974. U.S. Environmental Protection Agency.
Methods for chemical analysis of water and wastes.
Environmental Monitoring and Support Laboratory, Office
of Research and Development. Cincinnati, OH: U.S.
Environmental Protection Agency. EPA-625/6-74-00.
USEPA. 1975. U.S. Envi ronmental Protection Agency.
Methods for acute toxicity tests with fish, macro-
invertebrates, and amphibians. Corvallis, OR: U.S.
Environmental Protection Agency. EPA-660/3-75-009.
USEPA. 1976. U.S. Environmental Protection Agency.
Quality criteria for water. Washington, D.C.: U.S.
Environmental Protection Agency.
USEPA. 1978. U.S. Environmental Protection Agency.
Bioassay procedures for the ocean disposal permit
program. Gulf Breeze, FL: U.S. Environmental Protection
Agency. EPA-600/9-78-010.
USEPA. 1980. U.S. Environmental Protection Agency.
Oyster acute toxicity test standard. Toxic Substances
Control Act., Section 4. Office of Pesticides and Toxic
Substances, Washington, D.C.: U.S. Environmental
Protection Agency.
Veith GD, DeFoe DL, Bergstedt BU. 1979. Measuring and
estimating the bioconcentration factor of chemicals in
fish. J. Fish. Res. Board Can. 36:1040-1048.
61
-------
ES-3
August, 1982
Waldichuk M. 1974. Some biological concerns in heavy
metal pollution. In: Vernberg FJ and Vernberg WB eds .
Pollution and physiology of marine organisms. New York:
Academic Press.
Wilbur KM and Yonge CM. 1964. Physiology of
Mullusca. Academic Press, New York.
62
-------
EG-7
August, 1982
PENAEID SHRIMP ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-7
Guidelines for Testing Chemicals August, 1982
PENAEID SHRIMP ACUTE TOXICITY TEST
(a) Purpose . Th'is guideline is intended for use in
developing data on the acute toxicity of chemical substances and
mixtures ("chemicals") subject to environmental effects test
regulations under the Toxic Substances Control Act (TSCA) (Pub.L.
94-469, 90 Stat. 2003, 15 U.S.C. 2601 et seq. ) . This guideline
prescribes tests using penaeid shrimp as test organisms to
develop dajta on this acute toxicity of chemicals. The United
States Environmental Protection Agency (EPA) will use data from
these tests in assessing the hazard of a chemical to the aquatic
environment.
(b) Def initions. The definitions in section 3 of the Toxic
Substances Control Act (TSCA), and in Part 792—Good Laboratory
Practice Standards apply to this test guideline. The following
definitions also apply to this guideline:
(1) "Death" means the lack of reaction of a test organism to
gentle prodding.
(2) "Flow-through" means a continuous passage of test
solution or dilution water through a test chamber, holding or
acclimation tank with no recycling.
(3) "LC50" means that experimentally derived concentration
-------
EG-7
August, 1982
of test substance that is calculated to have-killed 50 percent of
a test population during continuous exposure over a specified
period of time.
(4) "Loading" means the ratio of test organism biomass
(grams, wet weight) to the volume (liters) of test solution in a
test chamber.
(c) Test procedures — (1) Summary of the test. Prior to
testing, the bottoms of the test chambers are covered with 2-3 cm
of sand and 'then filled with appropriate volumes of dilution
water. The flow is adjusted to the rate desired to achieve
loading requirements. Penaeid shrimp are introduced into the
test chambers according to the experimental design. The shrimp
are maintained in the test chambers for a period of 3-7 days
prior to the beginning of the test. The test begins when the
test substance is introduced into the test chambers. The rate of
flow is adjusted to maintain the desired test substance
concentration in each chamber. The shrimp are,observed during
the test; dead shrimp are counted, removed, and the findings
recorded. Dissolved oxygen concentration, pH, temperature,
salinity, test substance concentration and other water quality
characteristics are measured at specified intervals in selected
test chambers. Data collected during the test are used to
develop concentration-response curves and LC50 values for the
-------
EG-7
August, 1982
test substance.
(2) [Reserved]
(3) Range-finding test. (i) A range-finding test should be
conducted to determine the test substance concentrations to be
used for the definitive test.
(ii) The shrimp should be exposed to a series of widely
spaced concentrations of test substance (e.g. 1, 10,, 100 mg/1,
etc.) .
(iii) A minimum of five penaeids should be exposed to
each concentration of test substance for up to 96 hours. No
replicates- are required and nominal concentrations of the
chemical are acceptable.
(4) Definitive test. (i) The purpose of the definitive
test is to determine the concentration-response curves and the
48- and 96- hour LC50 values with the minimum amount of testing
beyond the range-finding test.
(ii) A minimum of 20 shrimp per concentration should be
exposed to five or more concentrations of the chemical chosen in
a geometric series in which the ratio is between 1.5 and 2.0
(e.g., 2, 4, 8, 16, 32 and 64 mg/1). An equal number of shrimp
should be placed in two or more replicates. If solvents,
solubilizing agents or emulsifiers have to be used, they should
be commonly used carriers and should not possess a synergistic or
antagonistic effect on the toxicity of the test substance. The
-------
EG-7
August, 1982
concentration of solvent should not exceed 0.1 ml/1. The
concentration ranges should be selected to determine the
requested concentration-response curves and LC50 values. The
concentration of test substance in test solutions should be
analyzed for chemical concentration prior to use and at
designated times.
(iii) Every test should include controls consisting ofi the
• l i ' \
same dilution water, conditions, procedures and shrimp from the
same population or culture container, except that none of the
chemical is added.
(iv) The dissolved oxygen concentration, temperature,
salinity and pH should be measured at the beginning of the test
and at 24, 48, 72 and 96 heirs in each test chamber.
(v) The test duration is 96 hours. The test is unacceptable
if more than 10 percent of the control organisms die or appear to
be stressed or diseased during the 96 hour test period. Each
test chamber should be checked for dead shrimp at 3, 6, 12, 24,
48, 72 and 96 hours after the beginning of the test.
Concentration-response curves and 48- and 96- hour LC50 values
should be determined along with their 95 percent confidence
limits.
-------
EG-7
Augus't, 1982
(vi) In addition to death, any abnormal behavior or
appearance should also be reported.
(vii) • Distribution of shrimp among test chambers should be
randomized. In addition, test chambers within the testing area
should be positioned in a random manner or in a way in which
appropriate statistical analyses can be used to determine the
variation due to placement.
(viii) The concentration of dissolved test substance (that
which passes 'through a 0.45 micron filter) in the test chambers
should be measured as often as is feasible during the test. The
concentration of test substance should be measured:
(A) in each chamber at the beginning of the test and at 48
and 96 hours after the start of the. test;
(B) in at least one chamber containing the next to the lowest
test substance concentration at least once every 24 hours during
the tes t;
(C) in at least one appropriate chamber whenever a
malfunction is detected in any part of the test substance
delivery system. Among replicate test chambers of a treatment
concentration, the measured concentration of the test substance
should not vary more than 20 percent.
(5) [Reserved]
-------
EG-7
August, 1982
(6) Analytical measurements (i) Test chemical. Deionized
water should be used in making stock solutions of the test
substance. Standard analytical methods should be used whenever
available in performing the analyses. The analytical method used
to measure the amount of test substance in a sample should be
validated before beginning the test by appropriate laboratory
practices. An analytical method is not acceptable if likely
degradation products of the test substance, such as hydrolysis
1 ' i
and oxidation products, give positive or negative interferences
which cannot be systematically identified and corrected
mathematically.
(ii) Numerical The number of dead shrimp should be counted
during each definitive test. Appropriate statistical analyses
should provide a goodness-of-f it determination for the
concentration-response curves. A 48- and 96- hour LC50 and
corresponding 95 percent intervals should be calculated.
(d) Test conditions — (1)" Test species — (i) Selection.
This test should be conducted using one of three species of
penaeid shrimp: Penaeus aztecus (brown shrimp), Penaeus duorarum
(pink shrimp), or Penaeus setiferus_ (white shrimp). Post-larval
juvenile shrimp should be utilized. Shrimp may be reared from
eggs in the laboratory or obtained directly as juveniles or
adults. Shrimp used in a particular test should be
-------
EG-7
August, 1982
of similar age and be of normal size and appearance. Shrimp
should not be used for a test if they exhibit abnormal behavior
or if they have been used in a previous test, either in a
treatment or in a control group.
(ii) Acclimation. During acclimation, shrimp should be
maintained in facilities with background colors and light
intensities similar to those of the testing areas. In addition,
any change in the temperature and chemistry of the dilution water
used for holding and acclimating the test organisms to1 those of
the test should be gradual. Within a 24 hour period, changes in
water temperature should not exceed 1°C, while salinity changes
should not exceed 2 °/oo.
(iii) Care and handling. Upon arrival at the test facility,
the shrimp should be transferred to water closely matching the
temperature and salinity of the transporting medium. Shrimp
should be held in glass tanks of 30 liter capacity or larger. No
more than 22 to 24 shrimp should be placed in a 30 liter tank
unless the flow-through apparatus can maintain dissolved oxygen
levels above 60 percent of saturation. With species of the genus
Penaeus, a minimum flow rate of 7.5 1/g body weight day should be
provided. Larger flows, up to 22 1/g body weight day, may be
desirable to insure dissolved oxygen concentrations above 60
percent of saturation and the removal of metabolic products. The
7
-------
EG-7
August, 1982
period of acclimation to ambient laboratory conditions should be
at least 4-7 days .
(iv) Feeding . Penaeid shrimp should not be fed during
testing. Every two or three days during the. acclimation period,
shrimp should be fed fish pieces approximately 1 cm^. Uneaten
food should be removed daily.
(2) Facilities--!i) Apparatus. (A) Facilities which may
be needed to perform this test include: flow-through tanks for
I 'l ! ' i
holding and acclimating penaeid shrimp;1 a mechanism for
controlling and maintaining the water temperature and salinity
during the holding period; apparatus for straining particulate
matter, removing air bubbles, or aerating water when necessitated
by water quality requirements; and an apparatus providing a 12-
hour light and 12-hour dark photoperiod with a 15-to-30 minute
transition period. Facilities should be well ventilated, free of
fumes and free of all other disturbances that may affect test
'organisms.
(B) two to three centimeters of acid-washed sand, free of
excess organic matter, should be placed in the bottom of test
chambers .
(C) Test chambers should be loosely covered to reduce the
loss of test solution or dilution water due to evaporation,
mimimize entry of dust and other particles and prevent escape of
the shrimp.
-------
EG-7
August, 1982
(ii) Cleaning. Test substance delivery systems and test
chambers should be cleaned before each test following standard
laboratory practices.
(iii) Construction materials. Materials and equipment that
'! contact test solutions should be chosen to minimize sorption of
test chemicals from dilution water and should not contain
substances that- can be leached into aqueous solution in
i '
quantities that can affect test results.
! ; L i • i • ' i
( iv) Dilution water. (A) Natural or artificial seawa'ter is
acceptable as dilution water if shrimp will survive in it without
signs of stress, such as unusual behavior or discoloration.
Shrimp should'be acclimated and tested in dilution water from the
same origin.
(B) Natural seawater should be filtered through a five
micrometer filter with a pore size < 20 microns prior to use in a
test.
(C) Artificial seawater can be prepared by adding
commercially available formulations or by adding specific amounts
of reagent-grade chemicals to deionized water. Deionized water
with a conductivity less than 1 u ohm/cm at 12°C is acceptable
for making artificial seawater. When deionized water is prepared
from a ground or surface water source, conductivity and total
organic carbon (or chemical oxygen demand) should be measured on
each batch.
-------
EG-7
August, 1982
(v) Test substance delivery system. Proportional diluters,
metering pumps or other suitable systems should be used to
deliver test substance to the test chambers. The system used
should be calibrated before each test. Calibration includes
determining the flow rate through each chamber and the
concentration of the test substance in each chamber. The general
operation of the test substance delivery system should be checked
twice daily during a test. The 24-hour flow through a test
• I 1 i •
chamber should be equal to at least five times! the volume of the
test chamber. During a test, the flow rates should not vary more
than 10 percent among test chambers or across time.
(3) Tes t parameters . Environmental parameters of the water
contained in test chambers should be as specified below:
(i) Temperature of 23 ± l°c.
(ii) Dissolved oxygen concentration between 60 and 105
percent saturation. Aeration, if needed to achieve this level,
should be done before the addition of the test substance. All
treatment and control chambers should be given the same aeration
treatment.
(iii) The number of shrimp placed in a test solution should
not be so great as to affect results of the test. Loading
requirements will vary depending on the flow rate of dilution
water. The loading should not cause the dissolved oxygen
concentration to fall below the recommended levels.
10
-------
EG-7
August, 1982
( iv) Photoperiod of 12 hours light and 12 hours darkness,
with a 15-30 minute transition period.
(v) Salinity of 20 ± 3 Percent.
(e) Reporting. The sponsor .should submit to the EPA all
VI
data developed by the test that are suggestive or predictive of
acute toxicity and all other toxicological manifestations. In
addition to the general reporting requirements prescribed in Part
i '
792--Good Laboratory Practice Standards, the reporting of test
i i i , i
data should include the following
(1) The nature of the test, laboratory, name of the
investigator, test substande and dates of test should be
supplied.
(2) A detailed description of the test substances should be
prr/ided. This information should include the source, lot
number, composition, physical and chemical properties and any
carrier or additives used.
(3) Detailed information about the shrimp should be
provided: common and scientific names, source of supply, age,
history, weight, acclimation procedure and feeding history should
be reported.
(4) A description of the experimental design including the
number of test solution concentrations, number of replicates and
11
-------
EG-7
August, 1982
number of shrimp per replicate should be provided.
(5) The source of the dilution water, its chemical
characteristics (e.g., salinity) and a description of any
pretreatment.
(6) A description of the test chambers, the depth and volume
of solution in the chamber, the number of organisms per
treatment, the number of replicates, the loading, the lighting,
the test substance delivery system and flow rate expressed as
volume additions per 24 hours.
(7) The concentration of the test substance in each test
chamber before the start of the test and at the end.
(8) The number of dead shrimp and measurements of water
temperature, salinity, and dissolved oxygen concentration in each
test chamber should be recorded at the designated times.
(9) Methods and data records of all chemical analyses of
water quality and test substance concentrations, including method
validations and reagent blanks.
(10) Recorded data for the holding and acclimation periods
(temperature, salinity, etc.).
(11) Concentration-response curves should be fitted to
mortality data collected at 24, 48, 72 and 96 hours. A
statistical test of goodness-of-f it should be performed.
(12) For each set of mortality data, the 48- and 96- hour
12
-------
EG-7
August, 1982
LC50 and 95 percent confidence limits should be calculated on the
basis of the average measured concentration of the test
substance. When data permits, LC50 values with 95 percent
confidence limits should t'd computed for 24 and 72 hour
observations .
(13) The methods used in calculating the concentration-
response curves and the LC50 values should'be fully described.
13
-------
ES-4
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
PENAEID SHRIMP ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
TABLE OF CONTENTS
Subject Page
I. Purpose 1
II. Scientific Aspects 1
Test Procedures 1
General 1
Range-Finding Test 2
Definitive Test 2
Test Conditions 3
Test Species 3
Selection 3
Sources • 6
Maintenance of Test.Species 7
!i :
Handling and Acclimation 7
Feeding 9
Facilities 12
General 12
Construction Materials 13
Test Substance Delivery System 14
Test Chambers 16
Cleaning of Test System 17
Dilution Water 17
Controls 19
Carriers 20
Randomization 20
Environmental Conditions 21
Dissolved Oxygen 21
Light 21
Temperature and Salinity 22
Reporting 23
III. Economic Aspects 26
IV. References 28
-------
Office of Toxic Subs tan-ces S3-4
August, 1982
Technical Support Document for Penaeid Acute Toxicity Test
I. Purpose
The purpose of this document is to provide the
scientific background and rationale used in the development
ol: Test Guideline EG-7 which uses Penaeid shrimp to evaluate
the toxicity of chemical substances. The Document provides
an account of the scientific evidence and an explanation of
the logic used in the selection of the test methodology,
procedures and conditions prescribed in the Test
Guideline, Technical issues and practical 'considerations
relevant to thd:;T]est Guideline ^re discussed. In addition,
estimates of the cost of conducting the test are provided.
11. Scientific Aspects
A. Test Procedures
1. General
A flow-through bioassay technique was chosen because of
several distinct advantages over' static exposure methods.
Continuously flowing seawater not only simulates the natu~ al
exposure process but, when used as a laboratory tool,
eliminates problems associated with the accumulation of
organic matter and toxic metabolic products. Flow-through
techniques should be used with materials which have a high
oxygen demand, are highly volatile, are unstable in aqueous
solution, are readily biodegradable, or are removed from
test solutions in significant amounts by the test
organisms. Toxicants flowing through this system are more
thoroughly mixed and loss due to sorption to sediments and
feces is minimized, Flow-through techniques for. holding and
acclimation of shrimp provides a smooth transition to actual
tes ting .
-------
ES-4
August, 1982
For acute tests, 96 hours is a convenient interval of
time for starting and completing a test in a normal five-day
work week, and is better than shorter periods for estimating
cumulative and other chronic effects. Because set-up is the
most expensive portion of a test, a 96-hour test is only
slightly more expensive than 24 or 48 hour tests. Yet
additional data on the LCSO's over time and the observation
of other abnormal effects that do not appear in shorter
tests are gained for this slight increase in cost. Although
the 48 hour test can reduce costs, the 96-hour toxicity test
was selected for the penaeid tes t, gu idelines because of
: i ' ! I ]
greater probability for determining the incipient LC50
(threshold limit for acute toxicity) through extension of
the toxicity curve.
2. Range-Finding Tes t
The concentration range for the definitive tests is
normally , chosen based on the results of a range-finding
test. Range-finding tests with penae id shrimp are usually
short-term (24-96 hour) flow-through bioassays which utilize
fewer organisms per test substance concentration than
required for the definitive test. In all cases, the range-
finding test is conducted to reduce the expense involved
without having to repeat a definitive test due to
inappropriate test substance concentrations.
3. Definitive Test
The concentration range for the definitive test is
chosen based on results on the range-finding test. By using
a minimum of five test substance concentrations, partial
kills both above and below the median 50 percent mortality
level are probable and will help define the concentration-
-------
ES-4
August, 1982
response curve. The more partial kills, the better the
definition of the concentration-response curve. The slope
and shape of the concentration-response curve can allow
insight into the mode of action of a chemical and will allow
estimation of the effects of lower concentrations upon the
test organisms. In addition, by having partial kill data, a
greater array of statistical methods can be used to
determine an LC50 value.
• A sample size of 20 shrimp permits several combinations
of replicates and sample sizes to be used. The use of
replicate samples allows an analysis of variance to be
performed on the results.
Measurements of test substance concentrations at
designated periods during the flow-through tesf: allows
documentation of real test concentrations at appropriate
periods under acute conditions.
Chemical and physical parameters (temperature, pH,
dissolved oxygen, and salinity) are recorded at specified
times to permit evaluation of the biological conditions
present for shrimp survival in test water.
Specified observations on mortality characteristics are
designed to allow an adequate evaluation of dose-response
effects in acute penaeid tests. In addition, these defined
observation times allow greater comparability of dose-
response data between different chemicals and laboratories.
B. Test Conditions
1. Test Species
a. Selection
The prime considerations in the selections of test
organisms for toxicity tests are: (a) their sensitivity to a
-------
ES-4
August, 1982
wide spectrum of test substances; (b) their geographical
distribution and abundance; (c) their recreational, economic
and ecological importance; (d) their availability as test
organisms, and existence of established culture. Penaeid
shrimp have become the most valuable marine species
harvested from U.S. Coastal waters by commercial fisherman
(Temple, 1973). In 1972, an estimated 190.6 million pounds
of shrimp were harvested Erom coastal waters; 87 percent of
this harvest was from the Gulf. During 1974, the Louisiana
brown shrimp catch alone was 27.4 million pounds, and was
va,lued at 13 million dollars, (Temple, 1973,; Knudsen et al.,
1 i ' ' ' I I ! I i I
1976).
Perhaps the most important quality of .penaeid shrimp for
toxicity testing is their consistently high sens.itivity to
test substances. In virtually all comparative toxicological
studies in the laboratory, penae id. shrimp proved the most
sensitive marine organism to a variety of toxins. Pink
shrimp have been used,repeatedly in the last 10 years (Lowe
1971, Tagatz 1975; Schimmel 1979; Parrish 1976; Nimmo and
Banner 1976); white shrimp and brown shrimp have also been
used successfully in toxicological tests (Nimmo and Bahner
1974; Curtis 1979).
Penaeid shrimp are also sensitive at sublethal doses of
toxins; this allows the maximum amount of information to be
gleened from each test. For example, pink shrimp were one
of a selected group of estuarine animals used to assess the
effects of mirex leaching into the environment; toxicity was
latent and became more apparent with increasing length of
exposure. Under stress they showed darker coloration, loss
of equilibrium and a cessation oi: burrowing behavior,
-------
E3-4
August, 1982
measures which provided valuable sub-lethal effects data
(Tagatz et al. 1975) .
In juvenile pink shrimp exposed to Mirex, Lowe (1971)
observed the first case of delayed toxicity. Mortality at
the end of seven days was 25 percent, but increased to 100
percent by day 11 even after the shrimp were removed to
mirex-free water. In a flow-through acute toxicity test,
Schimmel, et al. (1979) found pink shrimp especially
sensitive to the .insecticides EPN and leptophos; in this
case there was 20 percent mortality at non-detectable
(nominal,) concentrations of these insecticides in test
water.
An additional point to consider is the suitability of
the species for 'cultivation, since cultured shrimp are
preferable for use in toxicity testing. Previous history of
an organism is a major variable affecting the potential
response to a test substance. While the tropical species P.
monodon and P. orientalis have been shown to grow most
rapidly under high-density cultivation, these species are
not representative of those organisms residing in U.S.
coastal waters. Penaeus aztecus (brown shrimp) and P.
setiferus (white shrimp) grew significantly longer in low
densities (25m^) than high densities (166m--). However, of
the 9 shrimp species studied, brown shrimp had the second
highest survival rate at high densities (Forster and Beard
1974). In the same study, penaeid shrimp were shown to be
less variable in individual growth rates than Machro'oracium
spp., the freshwater prawn.
Some limitations in the use of penaeid shrimp have been
reported. High mortalities due to cannibalism, cramped test
-------
ES-4
August, 1982
conditions, or control mortality can be expected in culture
(Curtis et al. 1979; Tagatz et al. 1976).
In view of the continued successful use of penaeid
shrimp for toxicity testing in many laboratories, and their
sensitive and varied response to sublethal toxic concen-
trations, they are the species of choice for this flow-
through bioassay- In fact, because of their suitability as
a test organism and their value as an economic resource, a
wealth of literature is available for reference in
developing culture and testing techniques, as well as a
comparative toxicology data base.
b. Sources
Whether collecting organisms for testing or for
culturing purposes, a great deal of care should be taken to
avoid stress and insure survival in transport to the
laboratory. Shrimp should be collected from unpolluted
sources and measurements of water temperature, salinity and
pH should be taken at capture time. '/This allows for
successful acclimation to laboratory conditions. Taking
organisms from areas of known high levels of parasitism,
disease, pollution, or where deformed individuals are found
should be avoided to insure valid results.
The following salient points are emphasized by APHA
(1975) when collecting organisms for bioassay purposes; in
general, great care should be taken to insure the shrimp are
not damaged in the collection, transfer and transporting
process:
(1) When seining or using trawls, make short hauls;
keep gravel, sand and other debris out of net..
-------
ES-4
August, 1982
(2) Do not expose delicate, easily danaged stages to
the air; juvenile shrimp can be transferred with dip
nets .
(3) Do not collect too many animals at one time.
(4) Do not crowd the organisms during transportation
and wa^ :h for signs of stress; observe animals in the
labora* :>ry for additional signs of stress.
Juvenile and adult shrimp are most" easily collected by
hand-held seines or boat trawls.. For test organisms, one
should select shrimp of uniform size and in the post-larval
stages. Dp not mix stages within the test. For culture,
i • i i • ! ' , ' I
purposes, the preferred method is to collect gravid females
and allow them to spawn in the laboratory. It may not yet be
feasible to breed penaeid shrimp prawns in captivity (Walker
1975). As a guide for distinguishing life stages in
penaeids, Rose et al. (1975) has suggested these total
length criteria: juvenile (25 mm); subadult (90 mm); and
adult (140 mm). Shrimp have been success-Cully transported
by motor vehicle in plastic bags or buckets filled with
oxygenated seawater (Mock 1974). It is recommended that
test organisms come from a controlled environment such as a
laboratory maintenance system. This insures that shrimp are
uniform in age, size and exper i-mental history.
2. Maintenance of Test Species
a. Handling and Acclimation
Tanks for holding and acclimation should be identical to
those used for testing, which eliminates further stressing
of shrimp by an additional transfer. Water should be of the
same temperature and salinity as water from the collection
site; a gradual change in water quality parameters should be
-------
ES-4
August, 1982
made to acclimate the shrimp to the test conditions. This
gradual period of acclimation should be at least 7 days; and
up to 2 weeks has been suggested (APHA 1975). It has been
shown that activity responses to tidal rhythms do not Hade
until 7 days of captivity have passed. It is important that
shrimp be similar in their activity cycles before testing
begins (Subrahmanyam 1976).
Following the initial holding period, shrimp should be
randomly assigned to their respective test chambers and held'
there until testing begins, again eliminating further
handling. However, as will often be the case, extra shrimp
will need to be maintained in separate tanks. No more than
22 to 24 shrimp should be kept in a 30 liter tank with a
flow-through mechanism to allow maintenance of dissolved
oxygen (DO) levels above 60 percent oE saturation. Flow
rates should be great enough to remove .aetabolic products
and food bu.ild-up which have been demonstrated to cause high
mortality (Mock 1974). A minimum flow of 7 1/g'day shrimp
should be maintained, while flows up to 22 1/g day may be
needed. Holding tanks should attain preliminary test
condition within 7 days; gradually acclimate shrimp to
salinity and temperature conditions required by the test so
as to minimize stress.
Inspection for parasitism and disease should be made
during the acclimation period; diseased shrimp should never
be used in tests. Methods for detecting and. treating the
following prawn diseases are given in Delves-Broughton
(1976) and should be referreed to when needed; shell
disease, black module, vibriosis, hasplosporidian infection,
filamentous gill growth, filamentous bacteria on eggs, and
-------
ES-4
August, 1982
systematic fungus disease. Spontaneous muscle necrosis is
the result of abrupt changes in temperature and salinity;
treatment is discussed by Lakshmi (1978). When antibiotic
use is necessary, oxytetracycline and oleandonycin are
suggested. They are 99 percent bactericidal at high closes
and do not significantly lepress respiration in shrimp (Chan
•and Lawrence, 1974). If antibiotics are used in the water
of test chambers (during acclimation), they should be
removed before testing begins. This is possible when the
chelator EDTA is substituted at a concentration of 10 mg/1
of seawater (APHA 1975).
* I i
' b. Feeding
During holding and acclimation period juvenile or adult
penaeids may be fed cut-up fish. Fillet from mullet,
grouper, or other abundant species should be cut into pieces
about 1 cm^' and fed, one per shrimp, every 2 or 4 days.
Uneaten food should 'be removed every 24 hours to reduce
fouling. Protozoal stages of shrimp are generally £ed
algae, chiefly the diatoms Thalass ios ira and Skeletohema
(Cook & Murphy, 1969, Cook 1967, Mock 1974). Techniques for
culturing the diatom, and mechanisms for maintaining them in
shrimp rearing tanks are discussed in detail by Mock
(1974). Equipment and procedures for the continuous mass
culture of algae as a food source are also found in APHA
(1975). Add algae as a concentrate either fresh
(centrifuges) or frozen. Do not add algal culture medium to
acclimation or test water since it is toxic to shrimp (Mock
1974). The number of algal cells necessary to rear a
population of larval shrimp during the protozoal stages are
as follows (APHA 1975):
-------
ES-4
August,- 1982
Protozoel I Skeletonema 50,000 cells/ml
Protozoel II Skeletonema 150,000 cells/ml
Protozoel III Tetraselmis 20,000 cells/ml
It was found that the addition of several algal foods
insure higher rates than additions of a single algal species
at comparable concentrations (Mock 1974). When only one
species is used for larval shrimp, it should be Skeletonema
cos tatum.
Brine shrimp (Artemia sp.) nauplii have been used
extensively as food for, the mys i stage through the fourth
1 I
post-larval stage. In tests of food preferenc in brown and
white shrimp, both species demonstrated a preference for
nauplii of brine shrimp (Artemia) when given a choice of
diets. Brown shrimp were more flexible, but still preferred
Artemia. Karim and Aldrich (1976) tested various commercial
foods and brown shrimp preferred Vio Bio Fish Flour and
white shrimp preferred Silvray Fish Feed. It was stressed
that these prepared foods not be recommended for general use
until their effect on survival and growth are demonstrated
to be favorable.
Therefore, Artemia should be used for post-larval
stages. The quantities required are:
Mys is Artemia nauplii 3/ml
Mys is Artemia nauplii 3/ml
Mys is Artemia nauplii 3/ml
Post-larval I-IV Artemia nauplii 3/ml
10
-------
ES-4
August, 1982
A recent study by Johns and Walton (1979) reported that
adult Mysidopsis bahia fed Artemia spp. from San Pablo Bay,
California exhibited increased mortality, did not reproduce
and showed reduced growth rates. In contrast, both juvenile
and adult mys id shrimp fed Artemia spp. strains collected
from Brazil, Australia, Italy and Utah laintained high
survival and cjrowth rates. These resu ts imply that
nutritional quality of Artemia, possibly associated with
pesticide or heavy metal .contamination, can s ignif icnatly
influence test results and, therefore, should be considered.
There are several basic methods oE crustacean
uquaculture: extensive culture (using large outdoor
enclosure); intensive culture (small outdoor tanks,); and
indoor intensive (high-density flow through tanks in the
laboratory). The indoor intensive system is most practical
for use with bioassay techniques because of the relative
ease oC controlling the aquaculture environment. Tanks are
stocked at high densities (for this guideline, not more than
22-24 adults per 30 liter tank). In order to prevent
fouling of the system, it is necessary to circulate water
through the tank to maintain high DO levels. The overflow
test water may overflow to a drain or be recycled through a
biological filter (Walker 1975).
A temperature range of 28°-30°C and a salinity range of
27-35°/oo are most satisfactory for shrimp larval culture.
Since the shrimp will be later used in acute toxicity tests,
they should be acclimated to the presceibed test conditions
by post-larval stages. The specific requirements of each
soecies should be considered.
11
-------
ES-4
August, 1982
Penaeid shrimp can be reared regularly from the egg to
post-larval stage in the laboratory. Cook and Murphy (1969)
have described, in detail, equipment and techniques for
conditioning, spawning and rearing large numbers of shrimp
larvae from eggs. Methods of rearing shrimp larvae for
experimental studies have been described by Cook (1967).
These references should be followed closely.
3. Facilities
a. General
The delivery of constant concentrations of test
substances 'is required to reduce variability in test
I \\
results. Large fluctuations in test substance concentration
will give abnormally high or low responses, depending upon
the mechanism of toxic action. Proportional diluters and
metering pumps (Mount and Brungs 1967) have been found to
provide constant concentrations and are widely used.
Proportional diluters operate on a sequential filling
and emptying of water chambers. The water chambers are
cali-'orated to contain a measured amount of water. Separate
water chambers can be provided Cor toxicant and diluent
waters. Diluent and toxicant waters are mixed in siphon
tubes and delivered to the replicate test chambers. The
cyclic action of the diluent is regulated by a solenoid
valve connected to the inflow dilution. The system is
subject to electrical power failure, so an alternate
emergency power source is recommended.
The proportional diluter is probably the best for
routine use. It is accurate over extended periods of time,
nearly trouble free, and has fail-safe provisions (Lemke et
al. 1978). A small chamber to promote mixing of toxicant-
12
-------
ES-4
August, 1982
bearing water and dilution water should be used between the
diluter and the test chamber for each concentration. Since
replicate chambers are used in this test, separate delivery
tubes can be run from the mixing chamber to each replicate
test chamber.
Calibration of the toxicant delivery,'sys tern should be
checked carefully before and after each test. This should
include determining the flow rate and toxicant concentra-
tion through each test chamber. The general operation of
tne system should be checked daily.
Alterations in the design of the proportional diluter,
sucn as the use of six or more concentrations have been
usef,u^ in some situations (Benoit and Puglisi 1973).
b. Construction Materials
'In an excellent review of potential sourc.v; toe chemical
contamination in the culture system and laboratory, Bernhard
i
and Z^attera (1970), stress the importance of avoiding
chemical contamination in cult ring marine organisms.
Therefore, choice of laboratory equipment on toxicant
testing is critical.
Several materials such as rubber and oolyvinyl chlorides
have been found highly toxic; and should never be used in
culture or testing of marine organisms. Teflon (algoflon),
Perspex, Polyethylene, Tygon, Polypropylene, Polycarbonates
(Makrolor) and Polyester (Gabraster) have been shown to be
non-toxic and suitable for experiments with marine
organisms .
All pipes, tanks, holding chambers, mixing chambers,
metering devices, and test chambers should be made of
materials that minimize the release of chemical coni-.aninants
13
-------
ES-4
August, 1982
into the dilution water or the adsorption of the test
substances. Chemicals that leach from cons truc-tion
materials can. stress test organisms, or possibly act
synergis tical ly or antagonistically with test substances to
give inaccurate results. Generally, undesirable substances
are not leached from oerf luorocarbon plastic, titanium, and
borosilicate glass; in addition, the tendency of these
materials to adsorb substances, is minimal. 'Rubber, copper,
brass, galvanized metal, lead and epoxy resins should not
come in contact with dilution water, s tock "solution, or test
solutions i because of the toxic substances they contain
(US EPA 1975). All containers and pipe need to be
conditioned before use in order to leach and wash away any
undesirable residues that may be present.
c. Test Substance Delivery System
Flow-through systems should have the capability to vary
and maintain water temperature, dissolved oxygen, and
salinity at desired levels during holding, acclimation and
testing. Penaeid shrimp are extremely sensitive to
fluctuations in these parameters, which affect test
validity. Tagatz et al. (1975) reported that a slight (3-
4°C) change in water temperature resulted in significant
increases in the mortality rates of juvenile Peneaus
duorarum exposed to mirex. These mortality increases were
greater than those due to longer (3x) exposure times (Lowe
et al. 1971). Similarly, salinity decreases have been shown
to cause significant increase in mortalities of P. aztecus
exposed to Aroclor (Nimmo and Banner 1975). Combined, or
synergistic effects of dissolved oxygen (DO), temperature
and salinity on the toxicity of toxaphene on pink shrimp
14
-------
ES-4
August, 1982
were identified and are discussed in detail by Courtenay and
Roberts (1973).
Shrimp health and survival ace directly affected by
water quality and handling. Physically stressed organisms
are not valid test subjects. Attention to husbandry f\nd
'•71
routine water quality monitoring are of paramount importance
in prevention of disease (Delves-Broughton and Poupard
1976.). Spontaneous muscle necrosis (exhibited as white foci
on the 4th, 5th and 6th abdominal segments) in brown shrimp
(P. aztecus) was induced in healthy shrimp by over-crowding,
lowering dissolved oxygen (DO) levels, or changing ohys ico-
i : i i I i ' ' '
chemical conditions ( Laks hrnli . et al. 1978). High mortality
in adult bcown shrimp from gas bubble,disease was caused by
supersaturation of .dissolved oxygen (DO) in water (Supplee
and Lightner 1976). This occurred with dissolved oxygen
(DO) levels exceeding 250 percent saturation. Morbidity and
mortality not only hinder the progress of testing, but alter
those toxic effects of concern, thus invalid ting tests.
Salinity and temperature also affect burrowing behavior,
metabolic rate, and cause increased aggression; such
aberrations cause distorted test results. For example,
hyperactivity, was shown in brown shrimp within 30 minutes
of a change from optimum conditions (Lakshmi et al. 1978).
The dilution water should be filtered through a twenty
micrometer filter (or smaller) to sufficiently reduce the
amount of suspended sediments, organic material and
biological organisms (phytoplankton, zooplankton, fungi,
bacteria, etc). This will minimize the confounding of
results associated with the differential sorption of the
test substance on cell walls, clay particles, etc. which in
15
-------
ES-4
August, 1982
turn may enhance or reduce the availability of the test
substance to the shrimp.
Accumulation of gases can cause adverse effects;
therefore, a device for removing air bubbles may be
necessary (Penrose and Squires 1976) . When the dissolved
oxygen (DO) in the dilution water is less than 60 percent,
aeration is suggested. Culturing techniques recommend 70-
100 percent saturation for penae id shrimp (Forster and Beard
1974;' Supplee and Lightner 1976). A device for simulating
natural photoperiod with transitions from light to dark is
suggested so, that conditions can be optimized for shrimp
i i ' ! I I ' ':
(Drummond and Dawson 1970).
d. Test Chambers
Choice of test chamber size should consider both the
needs of the test organism and the requirements of the
test. Chamber size should reflect the appropriate loading
require-ment using the number of organisios specified in the
experimental design. P. aztecus and P. setiferus showed a
significant difference in length attained when grown in low
(25,/m2) and high 166/m2) densities (Forster and Beard
1974). Stress caused by crowding has been shown to induce
latent viral infections in healthy pink shrimp (Couch 1974).
Penae id shrimp are large organisms. Juvenile
individuals range from 0.4 mm up to approximately 25 mm in
length (Rose 1975). In the 60x30x30 cm high container
recommended, 20-22 juvenile shrimp may be housed, as long as
the total live weight is no more than 50 g per chamber. A
substrate of two to three centimeters of organic free sand,
permits the shrimp to burrow. Screens for chamber tops are
recommended (APHA, 1975) to prevent the escape of shrimp
from the test chambers.
16
-------
33-4
August, 1982
e. Cleaning of Test System
Before use, test systems are cleaned to remove dust,
dirt, other debris, and residues that may remain from the
previous use of the system. New chambers should be cleaned
to remove any chemical or dirt residues1 remaining from
manufacture or accumulated during storage and
construction. Detergent is used to remove hydrophobia or
lip id-like substances. Acetone is used for the same purpose
and to remove any detergent residues. It is important to
use pesticide-free acetone to prevent the cpntamination of
the chambers with pesticides. .Nitric acid can be used to
ii',, i i ; i
clean. ;,va tal residues from the system.
At the end of a test, test systems should be washed in
preparation for the next test or: Storage. This will prevent
chemical residues and organic ma,tter from becoming embedded
l i
or absorbed into the equipment.1'
Priming the system with dilut'ion water before use allows
equilibrium to be reached between the chemicals in the wat >r
and the materials of the testing system. The testing system
may sorb or react with substances in the dilution water.
Allowing this equilibrium to be established before exposure
of the test shrimp to the test substance lessens the chances
of water chemistry changes during a test.
f. Dilution Water
A constant supply of dilution water is required to
maintain consistent experimental conditions. An
interruption in flow or changes in water quality parameters
can change the chemistry of the test system and possibly
affect the response of the test population. Therefore, the
results oC a test with variable dilution water quality are
not comparable to tests run under constant conditions and
17
-------
ES-4
August, 1982
the results are more difficult to interpret.
For acute toxicity tests, a minimum criterion for an
acceptable dilution water is that healthy test organisms
will survive for the duration of the acclimation period
without showing signs of stress. Signs of stress in penaeid
shrimp include darkened coloration, cessation of burrowing
behavior, loss of equilibrium, and antennae-chewing (Tagatz
1976; Forster and Beard 1974). Inves ti-ga tors should be
familiar with normal shrimp behavior patterns, as well as
gross physical changes which may occur during testing.
Since shrimp have both estuarine iand marine phases
during their life cycle, the salinity 'of dilution water is
of prime importance. Determination of the desired salinity
was made by, considering the natural habitat characteristics,
laboratory results, and individual species preferences. The
most important test requirement will be to maintain a
constant salinity level for the entire holding and testing
period1. ' It is important also to monitor dissolved oxygen
(DO) levels; they should be kept above 60 percnt of
saturation. The pH of the test solution appears to be less
important to the health of shrimp. Some studies have
suggested a pH of 8.3 to 8.7 for white shrimp (Curtis et al.
1979) and 8.0 for euryhaline species in general (Kester et
al. 1967; Zaroogian et al. 1969),
Natural seawater, obtained from a point source with
similar characteristics to those designated for the test
species, or water from an area where the test organisms were
obtained, is preferrable to artificial sea water. Dilution
water should be of cons.tant quality and should be
uncontaminated. Contaminants may affect the results
directly and indirectly. For example, low levels of
18
-------
ES-4
August, 1982
organochlorine chemicals have been shown to increase the
prevalence of latent viral infections in pink shrimp (Couch,
1975). This is as important during holding as during
tes ting .
If alternatives to reconstituted seawater are used, they
should meet the following specifications for contaminant
levels (US EPA 1975) .
Suspended Solids < 20 mg/1
TOC < 10 mg/1
Un-ionized ammonia < 20 ug/J.
Residual Chlorine < 3 ug/1
Total organophosphorus pe's tic ides < 40 ng/1
Total organochlorine
pesticides plus PCB's < 50 ng/1
Maintaining the desired salinity Un/el in natural waters
often poses a problem. When poss ible,, obtain water from an
area of high salinity and obtain low salinities by adding
either deio'nized or glass distilled water of a satisfactory
quality. To increase salinity, use a strong, natural brine,
which can be obtained by freezing and then partially thawing
seawater. This procedure can be used if limited amounts of
seawater are needed. However, it is recommended that
artificial seawater be used when large quantities of
dilution water are needed (APHA, 1975).
g. Controls
Controls are required for eve cy test to assure that any
effects which are observed are due to the test substance and
not to other factors. These may include effects from
construction materials, environmental factors, vapors,
stressed test organisms, etc.
19
-------
ES-4
August, 1982
Ten percent mortality may be anticipated due to inherent
biological factors. In a test chamber of 20 organisms, this
amounts to two deaths. Any increase above this may be
attributed to conditions of the test. The ten percent
mortality figure is representative of a wide variety of
organisms including both fish and invertebrates captured
from the wild. Capture tends to stress organisms so there
is more likelihood of stress related death. In addition,
invertebrates are generally more -vulnerable to handling
injury. If penaeid shrimp are raised under controlled
conditions, they are generally more healthy than are
I ! I'll
captured organisms, therefore, fewer should die'during a
test because of inherent biological factors.
h. Carriers
Carriers can effect test organisms and can possibly
alter the form of the test substance in water. Therefore,
it is preferable to avoid the use of carriers in toxic i ty
tests unless require'd 'to dissolve the test substance. Since
carriers can stress or adversely effect test organisms, the
amount of carrier should be kept to a minimum. A
recommended maximum is 0.1 ml/L (APHA 1975).
Triethylene glycol and dimethylformamide have been shown
to exert the least influence on test organisms and test
substances of several carriers that have been used in
testing marine organisms. Acetone and ethanol have a
stronger tendency to reduce the surface tension of the water
and therefore decrease oxygen saturation (Veith and Corns tock
1975; Krugel et al., 1978; APHA 1975).
i . Randomiza tion
The positions of test chambers are randomized to prevent
conscious or unconscious biases from being introduced.
20
-------
ES-4
August, 1932
These biases can be in environmental conditions such as
temperature and lighting, shrimp selection and distribution,
diluter system function, etc.
4 . Environmental Conditions
a. Dissolved Oxygen
Large variations in flow rates to chambers will result
in environmental differences between chambers. Par :uae ters
such as dissolved oxygen (DO) and test substance concen-
tration can decrease more rapidly in chambers with low
salinities. High salinities can be decreased by adding
either deionized or glass distilled water of a satisfactory
• ! I I ' i ; '!!
quality.1 To increase salinity, use a strong, natural brine,
which can be obtained by freezing and then partially thawing
seawater.. This procedure can be used if limited amounts of
seawater are needed. However, it is recommended that
artificial seawater be used when large quantities of
dilution water are needed (APHA, 1975).
b. Light
The three species of penaeid shrimp have been shown to
have a nocturnal peak in activity when held in captive
laboratory conditions. Burrowing frequencies and durations
were highest during bright light hours, and shrimp were more
active above the substrate during dark hours. Of the three
species tested, P. setiferus (white shrimp) was least
influenced by the light schedules and was more active than
either pink or brown shrimp, being exposed on the substrate
day and night (Wickham and Minkler, 1975). This is
supported by catch data for shriip which show higher catch
levels in daylight for white shrimp than the other species.
21
-------
ES-4
August, 1982
Cool white fluorescent lighting should be used. White
light has been determined necessary for maintaining a
circadian burrowing pattern. A 15-30 minute transition
period between light and dark cycles is suggested. It
appears that shrimp initiate activity rhythm changes during
this transition period (Bishop and Herrnkind 1976). Thus,
the 12:12 light-dark schedule using white light not only
mimics environmental conditions, but also allows for
equalizing the time organisms spent exposed to test
substances in the water and substrate. Furthermore this
standardization facilitates comparison between tests using
i ; ; j
different species.
c. Temperature and Salinity
Penaeid shrimp occur naturally in,estuarine waters where
temperature and salinity vary over a wider range than in
oceanic waters. A review of the literature of toxicity
testing demonstrates the broad range of conditions over
which penaeid shrimp have been maintained.
Brown shrimp (Penaeus aztecus) occur in waters ranging
from 15°-35°C in temperature and 9-40 °/oo in salinity
(Copeland and Bechtel, 1974). In culture, growth was found
to be optimum between 15-20°C (Temple, 1973).
Pink shrimp Peneaus duorarum) occur naturally in waters
where temperature and salinity vary from 5-38°C and 20-
35°/oo respectively (Copeland and Bechtel, 1974). Optimal
temperatures for growth are in excess of 20°C (Copeland and
Bechtel, 1974).
White shrimp (Penaeus setiferus) are found in water
ranging from 10-40°C and 0-33 °/oo in salinity (Copeland and
Bechtel, 1974). Optimum temperatures for growth have been
22
-------
ES-4
August, 1982
shown to be between 15 and 20°C (Temple, 1973).
Thus there is considerable overlap in temperature and
salinity requirements of the three penaeid species.
Therefore, it is reasonable 'to select a single temperature
and salinity level for testing purposes; tests should be
conducted a a temperature of 2 3 _+ 1°C and a salinity level
of 20 _+_ 2 ° oo to minimize the difficulty in obtain-ing a
suitable source of dilution, water.
Furthermore, minimizing variability in testing
conditions by specifying temperature and salinity conditions
allows greater comparability of inter-laboratory test
•results and' for the development' of a comparative toxicolo'gy
data base. An acceptable method For maintaining desired
temperature and salinity ranges in flow-through bioassays
with marine organisms is described in Banner and Nimmo
(1975).
C . . Reporting
A coherent theory of the dose-response relationship was
introduced by,Bliss (1935), and is widely accepted today.
This theory is based on four assumptions:
(1) Response is a positive function of dosage, i.e., it
is expected that increasing exposure should produce
increasing responses.
(2) Randomly selected animals are normally distributed
with respect to their response to a toxicant.
(3) Due to homeos tas is, response magnitudes are
proportional to the logarithm of the dosage, i.e., it
takes geometrically increasing dosages (stresses) to
produce arithmetically increasing responses in test
animal populations .
23
-------
ES-4
August, 1982
(4) In the case of direct dosage of animals, their
resistance to effects is proportional to body mass.
Stated another way, the treatment needed to produce a
given response is proportional to the size of the
animals treated.
The concentration-response curve, where percent
mortality is plotted as a function of the logarithm of test
solution concentration, can be interpreted as a cumulative
distribution of tolerance within the test population
(Hewlett and Plackett 1979). Experiments designed to
measure tolerance directly (Bliss 1944) have shown that in
i I
most cases tolerance is lognormally distributed within an
experimental population in most cases. Departures from the
lognormal pattern of distribution are generally associated
with mixtures of very susceptible and very resistant
individuals within a population (Hewlett and Plackett
1979). In addition, mixtures of toxicants can produce
i
tolerance curves which deviate significantly from 'the
lognormal pattern (Finney 1971).
If tolerances are lognormally distributed within the
experimental population, the resulting concentration-
response curve will be sigmoidal in shape, resembling *a
logistic population curve (Hewlett and Plackett 1979).
While estimates for the mean lethal dose can be made
directly from the dose response curves, a linear trans-
formation often is possible, using probit (Bliss 1934;
Finney 1971) or logit (Hewlett and Plackett 1979) trans-
formations .
Once the mortality data have been transformed, a
straight line can be fitted to the data points. This line
is more often fitted by eye (APHA 1975), but a least square
24
-------
ES-4
August, 1982
ii-near regression procedure is strongly recommended (Steel
and Torrie 1960). From the regression equation, confidence
limits can be determined for predicted mortality values. An
additional advantage is that the significance of the slope
of the' regress ion line can be determined (Draper and Smith
1966). By using replicate experimental chambers, an
analysis of variance can also be performed to determine
whether deviations of data points from the regression line
are random fluctuations and indicate whether a linear model
is an appropriate representation of the data points (Draper
and Smith 1976).
'While values for the mean lethal dose, LC50,1 can be
estimated graphically from the linearized concentration-
response curve (APHA 1975), other techniques are preferable
since the graphical method does not permit the calculation
of confidence limits.
The probit method (Finney 1971) uses- the probit trans-
formation and the maximum likelihood curve fitting
technique. The Litchfield and Wilcoxon method (1949) is a
modified probit method which does not require partial kills,
as does the unmodified probit method. The log it method
(Ashton 1972) utilizes either the maximum likeli-hood or the
minimum chi-square method (Berkson 1949) to estimate LC50.
The moving average (Thompson 1947) is simple to apply but
depends on the symmetry of the tolerance distributions to
provide accurate estimates.
The moving average method can only be utilized to
calculate the LC50. An additional disadvantage of this
method is that confidence limits for LC50 cannot be
calculated if partial kills are not available.
25
-------
ES-4
August, 1932
The lack of partial kills seriously impairs the utility
of the probit, logit, and moving average methods. In
situations where there are no partial kills, the binomial
test (Siegal 1956) can be used to estimate the confidence
limits around the LC50 -value (Stephan, 1977). The LC50
value can be calculated from the relation:
LC50 = [(A) x (B)]/2
where
A = concentration at which no organisms die
B = concentration where all organisms die
A and' B are the confidence limits of the estimate and
are sign IT leant above the 95 percent level since more than
six test organisms are exposed at each concentration level
(Stephan 1977).
If dose-response data is plotted for each 24 hour
interval throughout the test, the.LCSO determined from each
curve can be plotted as a function of time, yielding an
acute toxicity curve (APHA 1975). This curve approaches the
time axis asymptotically, indicating the final or threshold
value for LC50. The absence of a threshold LC50 may
indicate the need for an acute test of longer duration.
III. Economic Aspects
The agency awarded a contract to Enviro Control, Inc. to
provide us with an estimate of the cost for performing a
flow-through acute toxicity test. Enviro Control supplied
us with two estimates; a protocol estimate and a laboratory
survey estimate.
26
-------
ES-4
August, 1982
Protocol Estimate
range mean
Acute $510-$1529 $1019
This estimate was prepared by separating the guidelines
into individual tasks and estimating the hoars used to
accomplish each task. Hourly rates were then applied to
yield a total direct labor charge. An overhead rate of 115
percent, other direct costs of $105, a general and
administrative rate of 10 percent and a fee of 20 percent
we'rre then added to the direct labor charge to yield the
final estimate.
Laborato cy Survey Es tima te
range mean
Acute $1000-$1450 $1234
The laboratory survey estimates were based on two
laboratory estimates.
27
-------
ES-4
August, 1982
IV. REFERENCES
Ashton WD. 1972. The log it transformation. New York:
Hafner Publishing Co.
Benoit DA and Puglisi FA. 1973. A simplified flow-splitting
chamber and siphon for proportional diluters. Water Res.
7:1915-1916.
Berks on J. 1949. The minimum Chi-square and maximum
likelihood solution in terms of d. linear transform, with
particular reference to bioassay. J. Amer. Stat.'Assoc.
44:273-278.
Bernhard M and'Zattera A. 1970. The importance of avoiding
chemical contamination for a successful cultivation of marine
organisms. Helgo,. Weiss. Meres. 20:655-675. ,: ,
Biship JM and Herrnkind WF. 1976. Burying and molting of
pink shrimp, Penaeus duorarum (Crustacea: Penaeidae) under
selected ohotoperio'ds of white light and ultraviolet light.
Biol. Bull. 150(2):163-182,
Bliss CI. 1935. The calculation of the dosage-mortality
curve. Ann. Appl. Biol. 22:134-307.
Chan E and Laurence A. 1974. Effect of antibiotics on the
respiration of the post-larval brown shrimp, Penaeus aztecus
Texas. Sci. 25:134.
Clark SH and Caillouet CW. 1975. Di'el fluctuations in
catches of juvenile brown and white shrimp in a Texas
esturine canal. Contri. Marine Sci. 19:119-124.
Cook HL and Murphy MA. 1969. , The culture of larval Penaeid
shrimp. Trans. Amer. Fish. Soc. 98:751,
Copeland BJ and Bechtel TJ. 1974. Some environmental limits
of six Gulf Coast estuarine organisms. Contri. Mar. Sci.
18:169-204.
Couch JA. 1974. An enzootic nuclear polyhedrosis virus of
pink shrimp: ultrastructure, prevalence, and enhancement. J.
Invert. Path. 24:311-331.
28
-------
ES-4
August, 1982
Courtenay 'WR and Roberts- MH. 1973. Environmental effects on
toxaphene toxicity to selected fishes and crustaceans.
Ecological Research Series No. EPA-R3-7 3-03 5. April 1973,
73pp.
Curtis MW, Copeland TL, Ward CH. 1979. Acute toxicity of 1 2
industrial chemicals to freshwater and saltwater organisms.
Water Res. 13:137-141.
DeFoe DL. 1975. Multichannel toxicant injection system for
flow-through bioassays. J. Fish Res. Bd . Canada 32:544-546.
Del ves-Broughton J and Poupard CW. 1976. Disease problems
of prawns in recirculation systems in the U.K. Aquaculture
7:201-217.
Draper NR and. Smith H, 1966., Applied regression analysis.
Ne York: Joh'n Wiley and Sons. ' !'
Drummond RA and Dawson-WF. 1970. An inexpensive method for
simulating die'l pattern of lighting in the laboratory. Trans.
Amer. Fish Soc. 99:434-435.
Forster: JRM and Beard TW . 1974. Experiments to assess the
suitability of nine species of prawns for intensive
cultivation. Aquaculture. 3:355-368.
Hansen DJ, Schimmel SE, Matthews E. 1974. zwoirUnce of
Aroclor 1254 by shrimp and fishes. Bull. Environ. Contam. and
Tox. 12(2) :253-256.
Hewlett PS and Plackett RL. 1979. The interpretation of
quantal responses in biology. Baltimore, MD: University Park
Press .
Karim M and Aldrich DW . 1976. Laboratory study of the food
preferenc of post-larval brown shrimp, Penaeus aztecus ( Ives )
and White shrimp _P_. setiferus (Linneaus). Balgladesh J. Zool.
4(1) :1-11.
Kester DR, Dredall IW, Connors DN, Pytokowicz RM . 1967.
Preparation of artificial seawater. Limnol. Oceanog . 12:176-
179".
29
-------
ES-4
August, 1982
Knudsen EE, Harke WH, Mackler JM. 1976. The growth rate of
marked juvenile brown shrimp, Penaeus aztecus, in a semi-
impounded Louisiana coastal marsh. Proc Gulf. Caribbean Fish
Inst. 29:144-159.
Krugel J, Jenkins D, Klein SA. 1978. Apparatus for the
continuous dissolution of poorly water-soluble compounds for
bioassays. Water Res . 12:269-272.
Lakshmi GH, Venkataramiah A, Howse HD. 1978. Effects of
salinity and temperature changes on spontaneous muscle
necrosis in Penaeus aztecus Ives. Aquaculture. 13:35-43.
Lasker R and Theilacker GH. 1965. Maintenance of euphausid
shrimp in the laboratory. Limnol. Oceanogr. 10:287-288.
Lemke AE, Brungs WA, Halligan BJ. 1978. Manual for
construction and operation of tpxicity-testing proportional
diluters. EPA Report1 No. 600/3-78-072.
Lowe JI, Parrish RR, Wilson AJ, Wilson PD, Duke AT. 1971.
Effects of Mirex on selected estuarine organisms. Trans. 36th
N, Aner. Wildlife and Nat. Res. Conf., Gulf Breeze. Contrb.
No. 124.
Martosubroto P. 1974. Fecundity of pink shrimp, Penaeus
duorarum Surloenroad. Bull. Mar. Sci. 24(3):606-627.
Mock CR. 1974. Larval culture of penaeid shrimp at the
Galveston Biological Laboratory. NOAA Tech. Rep. NMFS
Circular. 388:33-40.
Mount DI and Brungs WA. 1967. A simplified dosing apparatus
for fish toxicology studies. Water Res. 1:21-29.
Nimmo DR and Bahner LH. 1974. Some physiological conse-
quences with polychlorinated biphenols and salinity stress in
Penaeid shrimp. In: Pollution and Physiology of Marine
Organisms. New York: Academic Press.
Nimmo DWR and Bahner LH. 1976. Metals, pesticides and
PCB's; toxicity to shrimp singly and in combination. In:
Estuarine Processes. Vol. I. New York: Academic Press, op.
523-532.
30
-------
ES-4
August, 1982
Parrish PR, Couch JA, Forester J, Patrick JM, Cook GH.
1973.. Dieldrin: effects on several estuarine organisms,
Proc. 27th Ann. Conf. of S.E. Assoc. Game Fish Commiss ioa^
pp. 427-434.
Parrish PR, Schimmel SC, Hans en D J, Patrick JM, Forester J.
1976. Chlordane : effects on several estuarine organisms. J,
Tox. Environ. Health. 1:435-494.
Penrose WR and Squires WR. 1976. Two devices for removing
supersaturated gases in aquarium systems. Trans Aii. Fish
Soc. 105(1):116-118.
Rose CD. 1975. Extensive culture of Penaeid shrimp in
Louisiana salt-marsh impoundments. Trans Am. Fish Soc. 2:296-
307. .
Schimmel SC, Patrick JM, Eaas LF. 1978. Effects of sodium
pe.itachlorophenate on several estuarine animals': Toxicity,
uptake and depuration. In: Rao RK, ed . Pentachlorphenol.
New York: Plenum Publishing Co.
Schimmel SC, Hamaker TL, Forester J. 1979. Toxicity aad
bioconcentration of EPN and leptophosin estuarine animals
USEPA/ERL Gulf Breeze. Contribution No. 354.
Siegel. 1956. Nonparametric statistics for the behavioral
sciences. New York: McGraw-Hill, Publ. Co.
Subrahmanyam CB. 1976. Tidal and diurnal rhythms of
locomotory activity and oxygen consumption in the pink
shrimp, Penaeus duoranum. Contrib. Marine Sci. 20:123-132.
Supplee VC and Lightner, DV. 1976. Gas-bubble disease due
to oxygen supersaduration in raceway - reared California
brown shrimp. Prog. Fish Cult. 38:158-159.
Tagatz ME, Borthwick PW, Forester J. 1975. Seasonal effects
of leached mirex on selected estuarine animals. Arch. Environ
Contam. Tox. 3:371-332,
Tagatz ME, Borthwick PW, Invey JM, Knight J. 1976, Effects
of leached mirex on experimental communities of estuarine
animals. Arch. Environ Contam. Tox. 14:435-442.
31
-------
ES-4
August, 1982
Temple- RF. 1973. Shrimp research at the Galves ton-
Laboratory of the Gulf Coast Fisheries, Center. U.S. Marine
Fish Ser. Mar. Fish. Rev, 35(3-4), 16-20.
Thompson WR. 1947. Use of moving averages and interpolation
to estimate median effective dose. I. Fundamental formulae,
estimation and error, and relation to other methods.
Bacteriol. Rev. 11:115-145.
US EPA. U.S. Environmental Protection Agency. 1975. Methods
for acute toxicity tests with fish, Macroinver-tebrates, and
Amphibians. Duluth, MM: EPA-660/3-75-0 09. U.S.
Environmental Protection Agncy.
Veith GD and Corns tock VM. 1975. Apparatus for continuously
saturi.ng water with hydropho'oic organic chemicals. J. Fish
Res, Bd. Canada. Vol. 32:1849-1851.
[ • I.
Walker A. 1975. Crustacean aquaculture. Proc. Nutr'. Soc.
34:6.5-73.
Wickham DA and Minkler FC. 1975. Laboratory observations on
daily patterns of burrowing and locomotor activity of pink
shrino, Penaeus duorarum, brown shrimp. Penaeus aztecus , and
white shrimp, Penaeus setiferus. Contrib. Marine Sci.19:21-
43.
Zaroogian GE, Pesch G, Morrison G. 1976, Formulation of an
artificial seawater media suitable for oyster larvae
development. Am. Zool. 9:11-41.
32
-------
EG-8
August, 1982
ALGAL ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances EG-8
Guideline for Testing Chemicals August, 1982
ALGAL ACUTE TOXICITY TEST
(a) Purpose. The guideline in this section is intended for
use in developing data on the acute toxicity of chemical
substances and mixtures ("chemicals") subject to environmental
effects test regulations under the Toxic Subs.tances Control Act
(TSCA) (P.L. 94-469, 90 Stat. 2003, 15 U.S.C. 2601 et seq.).
This guideline prescribes test procedures and conditions using
freshwater and marine algae to develop data on the phytotoxicity
: J
of' chemicals. The United States Environmental Protection Agency
(US EPA) will use data from these tests in assessing the hazard of
a chemical to the environment.
(b) Def initions . The definitions in Section 3 of the Toxic
Substances Control Act (TSCA) and the definitions in Part 792--
Good Laboratory Practice Standards apply to this test
guideline. The following definitions also apply to this
guideline:
(1) "Algicidal" means having the property of killing algae.
(2), "Algistatic" means having the property of inhibiting
algal growth.
(3) "ECx" means the experimentally derived chemical
concentration that is calculated to effect X percent of the test
-------
EG-8
August, 1982
criterion.
(4) "Growth" means a relative measure of the viability of an
algal population based on the number and/or weight of algal cells
per volume of nutrient medium or test solution in a specified
period of time.
(5) "Static system" means a tesrt container in which the test
solution is not renewed during the period of the test.
(c) Test procedures — (1) Summary of the test. (A) In
preparation for the test, fill test containers with appropriate
Mi ' '[
volumes of nutrient medium and/or'test solution. Start the test
by introducing algae into the test.and control containers in the
growth chambers. Environmental conditions within the growth
chambers are established at predetermined limits.
(3) At the end of 96 hours enumerate the algal cells in all
containers to determine inhibition or s t iinulation of growth in
test containers compared to controls. Use data to define the
concentration-response curve, and calculate the EC-10, EC-50, and
EC-90 values.
(2) [Reserved]
(3) Range-finding test. (i) A range-finding test should be
conducted to determine if:
(A) definitive testing is necessary
(B) test chemical concentrations for the definitive test.
-------
EG-8
August, 1982
(ii) Algae are exposed to a widely spaced (e.g., log
interval) chemical concentration series. The lowest value in the
series, exclusive of controls, should be at the chemical's
detection limit. The upper value, for water soluble compounds,
should be the saturation concentration. No replicates are
required; and nominal concentrations of the chemical are
acceptable unless definitive testing is>not required.
(iii) The test is performed once for each of the recommended
algal species or selected alternates. Test chambers should
contain! equal volumes of test solution and approximately l! x 104
Selenastrum cells/ml or 7.7 x 104 Skeletonema cells/ml of test
solution. The algae should be exposed to each concentration of
test chemical for up to 96 hours. The exposure period may be
shortened if data suitable for the purposes of the range-finding
test can be obtained in less time.
(iv) Definitive testing is not necessary if the highest
chemical concentration tested (water saturation concentration or
1000 mg/1) results in less than a 50 percent reduction in growth
or if the lowest concentration tested (analytical detection
limit) results in greater than a 50 percent reduction in growth.
(4) Definitive test. (i) The purpose of the definitive
test is to determine the concentration response curves, the EC-
10's, EC-50's, and Ec-90's for algal growth for each species
-------
EG-8
August, 1982
tested, with a minimum amount of testing beyond the range-finding
tes t.
(ii) Algae should be exposed to five or more concentrations
of the test chemical in a geometric series in which the ratio is
between 1.5 and 2.0 (e.g., 2, 4, 6, 8, 16, 32 and 64 mg/1).
Algae should be placed in a minimum of three replicate test
containers for each concentration of test chemical and control.
More than three replicates may be required to provide sufficient
quantities of test solution for determination of test substance
concentration at the end of the test. ' Each test chamber should
contain equal volumes of test solution and aporo-ximately 1 x 10^
Selenastrum cells ml"1 or 7.7 x 104 Skeletonema cells/ml of test
solution. The chemical concentrations should result in greater
than 90 percent of algal growth being inhibited or stimulated at
the lowest concentrations of test substance compared to controls.
(iii) Every test should include a control consisting of the
same nutrient medium, conditions, procedures, and algae from the
same culture, except that none of the test substance is added.
If a carrier is present in any of the test chambers, a separate
carrier control is required.
(iv) The test begins when algae from seven to ten-day-old
stock cultures are placed in the test chambers containing test
solutions having the appropriate concentrations of the test
-------
EG-8
August, 1982
substance. Algal growth in controls' should reach the logarithmic
growth phase by 96 hours (at which time the number of algal cells
should be approximately 1.5 x 106/ml for Skeletonema or 3.5 x
10°/ml for Selenastrum). If growth in controls does not reach
this logarithmic phase within this 96-hour period, the test is
invalidated and should be repeated. At the end of 96 hours the
algal growth response (number or weight of algal cells/ml) in all
test containers and controls should be determined by an indirect
(spectrophotometry, electronic cell counters, dry weight, etc.)
or a direct (actual microscopic cell count) method. Indirect"
methods should be calibrated by a direct microscopic count. The
percentage inhibition or stimulation of growth for each
concentration, EC-10, EC-50, EC-90 and the concentration-response
curves are determined from these counts.
(v) At the end of the definitive test, the following
additional analyses of algal growth response should be performed:
(1)' Determine whether the altered growth response between
controls and test algae was due, to' a change in relative cell
numbers, cell sizes or both. Also note any unusual cell shapes,
color differences, flocculations, adherence of algae to test
containers, or aggregation of algal cells.
(2) In test concentrations where growth is maximally
inhibited, algistatic effects may be differentiated from
-------
EG-8
August, 1982
algicidal effects by the following two methods:
(A) Add 0.5 ml of a 0.1 percent solution (we ight/ volume) of
Evans blue stain to a one milliliter aliquot of algae from a
control container and to a one milliliter aliquot of algae from
the test container having the lowest concentration of test
chemical which completely inhibited 'algal growth (if algal growth
was not completely inhibited, select an aliquot of algae for
staining from the test container having the highest concentration
of test chemical which inhibited algal growth). Wait ten to
1 ' i' i
thirty minutes,' examine microscopically, and determine the
percent of the cells which stain blue (indicating cell
mortality). A staining control should be performed concurrently
using heat-killed or formaldehyde-preserved algal cells; 100
percent of these cells should stain blue.
(B) Remove 0.5 ml aliquots of test solution containing
growth-inhibi ted algae from each replicate test container having
the concentration of test substance evaluated in (2) (I) above.
Combine these aliquots into a new test container and add a
sufficient volume of fresh nutrient medium to dilute the test
chemical to a concentration which does not affect growth.
Incubate this subculture under the environmental conditions used
in the definitive test for a period of up to nine days, and
observe for algal growth to determine if the algistatic effect
-------
EG-8
August, 1982
noted after the 96-hour test is reversible. This's ubculture test
may be discontinued as soon as growth occurs.
(5) [Reserved]
(6)- Analytical measurements — (i) Chemical . (A) Glass
distilled or deionized water should be used in the preparation of
the 'nutrient medium. The pH of the test solution should be
measured in the control and test containers at 'the beginning and
at the end of the definitive test. The concentration of test
chemical in the test containers should be determined at the
beginning and end of th4 definitivb test by standard analytical
methods which have been validated prior to the test. An
analytical method is unacceptable if likely degradation products
of the chemical, such as hydrolysis and oxidation products, give
positive or negative interference.
(B) At the end of the test and after aliquots have been
removed for algal growth-response determinations, microscopic
examination, mortal staining, or subculturing, the replicate test
containers for each chemical concentration may be pooled into one
sample. An aliquot of the pooled sample may then be taken and
the concentration of test chemical determined. In addition, the
concentration of test chemical associated with the algae alone
should be determined. Separate and concentrate the algal cells
from the test solution by centrifuging or filtering the remaining
-------
EG-8
August, 1982
pooled sample and measure the test substance concentration in the
algal-cell concentrate.
(ii) Numerical . Algal growth response (as percent of
inhibition or stimulation in the test solutions compared to the
controls) is calculated at the end of the test. Mean and
standard deviation should be calculated and plotted for each
treatment and control. Appropriate statistical analyses should
provide a goodness--of-f it determination for the concentration
response curves. The concentration response curves are plotted
1 I ! • 1 • ' :' I
using the mean measured tes't solution concentrations obtained at'
the end of the test.
(d) Test condi tions--( 1) Tes t species . Species of algae
recommended as test organisms for this test are the freshwater
green alga, Selenastrum capr icornu turn, and the marine diatom,
Skeletonema costatum. Algae to be used in acute toxicity tests
may be initially obtained from commercial sources and
subsequently cultured using sterile technique. Toxicity testing
should not be performed until algal cultures are shown to be
actively growing (i.e. capable of logarithmic growth within the
test period) in at least two subcultures lasting seven days each
prior to the start of the definitive test. Al.l algae used for a
particular test should be from the same source and the same stock
culture. Test algae should not have been used in a orevious
-------
EG-8
August, 1982
test, either in a treatment or a control.
(2) Facilities — (i) General. (A) Facilities needed to
perform this test includes: a growth chamber or a controlled
environment room that can hold the test containers and will
maintain the air temperature, lighting intensity and photoperiod
specified in this test guideline; apparatus for culturing and
enumerating algae; a source of distilled and/or deionized wa'ter;
and apparatus for carrying out analyses of the test chemical.
(B) Disposal facilities should be adequate to accommodate
spent glassware, algae and test solutions at the end of the test
and any bench covering, lab clothing, or other contaminated
ma terials.
(ii) Test containers. Erlenmeyer flasks should be used for
test containers. The flasks may be of any volume between 125 and
500 ml as long as the same size is used throughout a test and the
test solution volume does not exceed 50 percent of the flask
volume .
(iii) Cleaning and sterilization. New test containers may
contain substances which inhibit growth of algae. They should
therefore be cleaned thoroughly and used several times to culture
algae before being used in toxicity testing. All glassware used
in algal culturing or testing should be cleaned and sterilized
prior to use according to standard good laboratory practices.
-------
fiG-8
August, 1982
(iv) Conditioning. Test containers should be conditioned by
a rinse with the appropriate test solutions prior to the start of
the test. Decant and add fresh test solutions after an appro-
priate conditioning period for the test chemical.
(v) Nutrient medium. (A) Formulation and sterilization of
nutrient medium used for algal culture and preparation of test
solutions should conform to those currently recommended by the
U.S. EPA for freshwater and marine algal bioassays. No chelating
agents should be included in the nutrient medium used for test
solution preparation. Nutrient medium should be freshly prepared
for algal testing, and may be dispensed in appropriate volumes in
test containers and sterilized by autoclaving or filtration. The
pH of the nutrient medium should be 7.5 for Selenas trum and 8.1
for Skeletonema at the start of the test and may be adjusted
prior to test chemical addition with 0. IN NaOH or HC1.
(B) Dilution water used for preparation of nutrient medium
and test solutions should be filtered, deionized or glass
distilled. Saltwater for marine algal nutrient medium and test
solutions should be prepared by adding a commercial, synthetic,
sea salt formulation or a modified synthetic seawater formulation
to distilled/deionized water to a concentration of 30 parts per
t nou s a nd .
(vi) Carriers . Nutrient medium should be used in making
10
-------
EG-8
August, 1982
stock solutions of the test chemical. If a carrier other than,
nutrient medium is absolutely necessary to dissolve the chemical,
the volume used should not exceed the minimum volume necessary to
dissolve or suspend the chemical in the test solution.
('3) Test parameters. (A) The test temperature should be
maintained at 24°±i°c for Selenastrum and 20°±1°C for
Skeletonema. Temperature should be recorded hourly during the
test.
(B) Test chambers containing Selenas trum should be
illuminated Continuously and those containing Skeletonema should
be provided a 14-hour light and 10-hour dark photoperiod with a
30 minute transition period under fluorescent -lamps providing 300
± 25 uEin/m^ sec (approximately 400 ft-c) measured adjacent to
the test chambers at the level of test solution.
(C) Stock algal cultures should be shaken twice daily by
hand. Test containers should be placed on a rotary shaking
apparatus and oscillated at approximately 100 cycles/rain for
Selenastrum and at approximately 60 cycles/min for Skeletonema
during the test. The rate of oscillation should be determined at
least once daily during testing.
(D) The pH of nutrient medium in which algae are subcultured
should be 7.5 for Selenastrum and 8.1 for Skeletonema, and is not
adjusted after the addition of the algae. The pH of all test
11
-------
EG-8
August, 1982
solutions and controls should be measured at the beginning and
end of the test.
(E) Light intensity should be monitored at least daily
during the test at the level of the test solution.
(e) Reporting . The sponsor should submit to the EPA all
data developed by the test that are suggestive or predictive of
acute phytotoxici ty. In addition to the general reporting
requirements prescribed in Part 792—Good Laboratory Practice
Standards , the following s.hould be reported:
(i') Detailed information about thle test organisms,' including
the scientific name, method of verification, and source;
(ii) A description of the test chambers and containers , the
volumes of solution in the containers, the way the test was begun
(e.g. conditioning, test substance additions, etc.), the number
of replicates, the temperature, the lighting, and method of
incubation, oscillation rates, and type of apparatus;
(iii) The concentration of the test chemical in the control
and in each treatment at the end of the test and the pH of the
solutions ;
(iv) The number of algal cells in each, treatment and control
and the method used to derive these values at the -beginning and
end of the test; the percentage of inhibition or stimulation of
12
-------
EG-8
August, 1982
growth relative to controls; and other adverse effect in the
control and in each treatment;
(v) The 96-hour EC-10, EC-50 and EC-90 values and their 95-
percent confidence limits, the methods used to derive these
values, the data used to define the shape of the concentration-
response curve and the goodness-of-fit determination;
(vi) Methods and data records of all chemical analyses of
water quality and test substance concentrations, including method
valid a tig 03 and reagent blanks;
III! i
(vii) The results of any optional analyses such as:
microscopic appearance of algae, size or color changes, percent
mortality, of cells and the fate of subcultured cells, the
concentration of test substance associated with algae and test
solution supernate or filtrate;
(viii) If the range-finding test showed that the highest
concentration of the chemical tested (not less than 1000 mg/1 or
saturation concentration) had no effect on the algae, report the
results and concentration and a statement that the chemical is of
minimum phytotoxic concern;
(ix) If the range-finding test showed greater than a 50
percent inhibition of algal growth at a test concentration below
the analytical detection limit, report the results,
concentration, and a statement that the chemical is phytotoxic
below the analytical detection limit.
13
-------
ES-5
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
ALGAL ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
TABLE OF CONTENTS
Contents Page
I. Purpose 1
II. Scientific Aspects 2
Test Procedures 3
General 3
Range-finding Test 6
Definitive Test 7
Analytical Measurements 9
Test Conditions 10
Test Species 10
Facilitites 14
Tes 11 Containersj 15
Cleaning and Sterilization 15
Conditioning 16
Nutrient Medium 16
Environmental Conditions 17
Reporting 20
III. Economic Aspects 20
IV. References 22
-------
Office of Toxic Substances ES-5
August, 1982
TECHNICAL SUPPORT DOCUMENT FOR ALGAL ACUTE TOXICITY TEST
Purpos e
The purpose of this document is to provide the
scientific background and rationale used in the development
of Test Guideline EG-8 which uses freshwater and marine
algae' to evaluate the acute toxicity of chemical
substances. The Document provides an account of. the
scientific evidence and an explanation of the logic used in
the selection of the test methodology, procedures and
conditions prescribed in the Test Guideline. Technical
issues and practical corts iderations | relevant to the Test
Guideline are discussed. In addition, estimates of the cost
of conducting the tests are provided.
11. Scientific Aspects
A. Test Procedures
1. General . A balanced growth of algae in the
aquatic environment is essential, but extremes in
productivity may be detrimental to other organisms. Some
algae .are able to inhibit or stimulate the growth of other
algae, for example Selenastrum can inhibit Microcys tis
growth in eutrophic water (Toerien et al. 1974). Inhibition
of algal growth would alter the food web and r.educe the
productivity of ecosystems. The' toxic effect of ' a chemical
or other inhibitor may increase the susceptibility of algae
to other environmental stresses (Fisher and Wurster 1973).
Stimulation of algal growth may cause an algal bloom which
may have negative aesthetic effects; may adversely affect
commercial sport fisheries (Lightner 1978, Lovell 1979) and
recreation; may impart unpleasant taste to drinking water;
may release substances deleterious to aquatic animals,
-------
ES-5
August, 1982
and/or may indirectly kill aquatic organ is ins by creating
anoxic conditions (Shilo 1964, Schwimmer and Schwimmer
1967). Stimulation of algal growth, while primarily a
problem in eutrophic freshwaters, has created serious
ecological problems in the open ocean as well. In the
spring of 1976 and extending into the fall, there was an
extensive algal bloom, dominated by Ceratium tripos, located
off the New Jersey coast. The bloom, together with a dearth
of storm activity, anomalous surface wind conditions, and
unusually warm sea surface temperatures resulted in a huge
atoxic, areat. 100 miles long ar^d 40 mijles wide which,had a,
severe impact on the finfish and shellfish populations in
the area. The immediate effects on commercial and sport
fishes, lobsters, and shellfish were not entirely known.
However, an estimated 59,000 metric tons of surf clams were
killed (representing twice the annual U.S. harvest), and .up
to 50% of other shellfish populations sampled were killed.
One commercial trawler reported up to 75% of fish collected
were dead. It was predicted that these mortalities would
affect recruitment, population size and harvests for years
to come (Sharp 1976).
Another more commonly known phenomenon is the adverse
effect caused by stimulated growth of tox.igenic marine
algae. Frequently explosive mass development of these
organisms in the form of blooms and tides occur, resulting
in fish kills, contaminated shellfish, and outbreaks of
paralytic shellfish poisonings in humans. (Shilo 1964,
Taylor and Seliger 1979).
Even when toxigenic organisms are not present in
sufficient concentrations to affect human health, red tides
-------
E3-5
August, 1982
may reduce the market for shellfish because of adverse
publicity (Council on Environmental Quality, 1979).
Furthermore, the high concentrations of phytoplankton that
occur during blooms can be harmful to shellfish because the
rate of water transport by molluscs is reduced and feeding
ceases (Galtsoff 1964).
Algal growth was selected to measure phytotoxici ty for
the following reasons:
o The selection of phytoplanktonic algae for toxicity
testing is based upon their importance in aquatic
ecosystems. Algae were one pfj the first) cellular
life forms, dating as far back as 3.1 billion years
in the fossil record (Bold and Wynne 1978). and are
numerous today. Because phytoplankton are
ubiquitous, it is usually the case that most marine
and freshwater ecosystems are based upon the
primary production of phytoplankton (Stern and
Stickle 1978). Primary production is of prime
significance to estuarine energetics since the
primary producers are at the base of the food
web. In estuaries phytoplankton are the main
primary producers in the water (Vernberg 1977).
o Algae convert inorganic carbon to organic carbon
and liberate oxygen during photosynthesis. Thus,
they are primary producers of food and energy for
the lower trophic-level herbivores which in turn
provide food for the upper trophic-level
carnivores, generally fishes (Vance and Maki
1976). Some species fix nitrogen, required for the
growth of vascular plants. Therefore, much of the
-------
ES-5
August, 1982
food people eat and the oxygen they breathe are the
result of algal productivity.
Inferences may be drawn from laboratory tests for
inhibition or stimulation of algal growth as to the
extent to which a chemical substance can interfere
with primary productivity and nutrient cycling in'
lakes, streams, estuaries, and oceans. Further
inferences may be, drawn from algal bioconcentration
data as to the potential of a chemical substance to
b ioaccumulate in food chains. However, in the
natural environment there are top many factors
! I ' . ! : I !
acting to regulate algal populations which cannot
be simulated in a simple laboratory test. The real
value of the test guideline is to determine
thresnold toxicity values and to evaluate the
relative toxicity of test substances to one another
under rigidly controlled conditions.
Algal testing has been" well established in the
literature. In 1967, the EPA began developing
algal assays for evaluating the ecological effects
of pollution to the environment. Initially
designed for considering problems associated with
eutrophication (Maloney and Miller 1975), algal
assays have also been used to define the toxic
effects of heavy metals (Davies 1978), pesticides
(Schauberger and Wildman 1977, Walsh and Alexander
1980), oil spills (Corner 1978, Fisher and Wurster
1973, O'Brien and Dixon 1976, Vandermeulen and
Ahern 1976), chemical substances (US EPA 1978 a,b,c,
Harding and Phillips 1978), dyes (Little and
-------
ES-5
August, 1982
Chillingworth 1976), complex industrial wastes
(USEPA 1978d, Walsh and Alexander 1980, Walsh et
al. 1980) and natural organic components of fresh
and marine water (Prakash and Rashid 1968). Over
the years, extensive use of this test has
sufficiently refined it- to qualify as a standard
method to measure water quality. Algal assays are
recommended for use by the APHA (1975) USEPA (1977,
1978 a,b,c,d) and are currently under review by the
American Society for Testing and Materials.
Further discussion on the validity of applying
algal assays in water quality assessment is found
in Fitzgerald (1975); Joint Industry/Government
Task Force on Eutrophication (1969); Leischman et
al (1979); USEPA (1978b) Miller et al. (1978);
Murray et al . (1971); Reynolds et al. (1974); and
USEPA (1971, 1975a).
o The algal growth method is 1) relatively rapid, 2)
inexpensive, 3) capable of being performed by
persons with minimal technical training and 4)
reproducible, using large numbers of organisms with
sufficient replication and precision.
The test procedure involves assessment of algal growth •
in test chambers relative to controls by requiring a
quantitative determination of algal cell numbers, and by
recommending a) a qualitative appraisal of algal numbers and
size by means of microscopic observation, and b) a
determination of viability of growth-inhibited algae by
means of mortal staining coupled with microscopic
observation and/or subcultur ing . The test procedure is
-------
ES-5
August, 1982
simple because it requires, only the combination of set
amounts of test substance, nutrient medium and algae, and
then monitoring the growth response 96 hours later. At the
end of 96 hours a further assessment of growth and viability
is recommended.
In the test the following procedures are required:
o Algal growth should be logarithmic at the beginning
of the test and algal number should be determined.
o The number of algae should be determined at the end
of the test.
o The concentration of chemical in the ' test solution
I : ! • I ' . i '111
should be determined at the. beginning and end of
the test and the concentration of chemical
associated with the algal cells should also be
determined .
o growth and bioconcentration data should be
subjected to statistical analyses.
These requirements will ensure consistency and will
minimize variability of the test results. The test also
recommends testing of algicidal and/or algistatic chemical
effects.
2. Range-Finding Test
It is recommended that a range-finding test be conducted
prior to the definitive test in those instances where no
information is available or can be elucidated on the
phototoxicity of the test chemical. This approach should
minimize the possibility that an inappropriate concentration
series will be utilized in the definitive test and under
certain circumstances may even preclude the need to conduct
the definitive test. In order to minimize the cost and time
-------
ES-5
August, 1982
required to obtain the requisite data nominal concentrations
are permitted, test duration may be shortened, replicates
are not required and other test procedures and conditions
are relaxed.
If test results indicate that the chemical is non-toxic
or very toxic to algae and if definitive testing is not
conducted, it is necessary to ascertain that the control
algae have attained a logarithmic growth rate by 96 hours
and that the test was conducted at the specified incubation
temperature. These verifications establish that the algae
tested were viable and that the test was properly conducted.
In some situations there may be enough information
available on toxicity to select the appropriate concen-
tration without a range-finding test. The range-finding
test (or other available information) needs to be accurate
enough to ensure that dose levels in the definitive test, are
spaced to result in concentrations above and below the EC-10
and EC-50 values for algal growth and mortality. If the
chemical has no measurable effect at the saturation
concentration (at least 1000 mg/1), it is considered
relatively nontoxic to algal growth and definitive testing
for effects on these processes is deemed unnecessary. In
all cases, the range-finding test is conducted to reduce the
expense involved with having to repeat a definitive test
because of inappropriate test chemical concentrations.
3. Definitive Test
The specific requirements of the definitive test are the
analytical determinations of chemical concentrations, the
unbiased selection of algae for each treatment, the use of
controls, the assessment of test validity, and the
-------
ES-5
August, 1982
recording, analysis, and presentation o£ data. These
requirements assure that the chemical concentration - algae
response relationship is accurately known, that chemical
effects are not confounded by differential algal growth and
that the relationships are clearly present. Reporting the
occurrence of such abnormal effects as irregular cell size
or shape, clumping, loss of chlorophyll, cell mortality, or
other unusual effects provides qualitative data that further
assist the assessment of phytotoxicity.
The purpose of the definitive test is to determine the
EC-10, EC-50 and concentration-response jcurves for algal
1 ! ' ' ' I !
growth for each specie's tested with a minimum of testing
beyond the range-finding test. The concentration range for
the definitive test is based upon the results of the range-
finding for that species. It is probable that each of the
species tested may have a different estimated EC-50 based on
the range-finding test and that more than five
concentrations of a test substance in a geometric series may
be needed to properly describe the dose-response
relationship for either species being tested. By testing a
minimum of five concentrations in a series per species
the dose-response relationship will be better defined. The
slope and shape of the dose-response curve can give an
indication of the mode of action of the chemical and will
allow estimatiion of the effects of lower concentrations on
the algae.
The primary observations - number of algae per chemical
and determination of the actual chemical concentrations
employed in the definitive test, are needed to accurately
describe the dose-response curve from which the EC-10 and
EC-50 are calculated.
-------
ES-5
August, 1982
The recommended experimental design is the randomized
complete block. As discussed by Hammer and Urquhart (1979),
it is essential that the investigator randomly assign test
containers 'to treatments to assure that each aliquot of
algae has the same chance of receiving any of the treatments
(exposure level of test chemical). To account for variation
within the growth chamber and to increase the sensitivity
for detecting treatment differences, small square blocks
should be delineated in the growth chamber with
randomization of treatment within blocks. Replication
should occur over growth chambers, (of the same type) as, in
many cases, a wi thin-growth chamber estimate of residual
variance badly underestimates the between chamber estimate
(Hammer and Urquhart 1979). This means that differences
between growth chambers are often greater than differences
between growth and environmental conditions within chambers.
4. Analytical Measurements
The actual chemical concentration used in the definitive
test should be determined with the best available analytical
precision. Analysis of stock solutions and test solutions
just prior to use will minimize problems with storage (e.g.,
formation of degradation products, adsorption,
transformation, etc.). Nominal concentrations are adequate
for the purposes of the range-finding test. If definitive
testing is not required because the chemical elicits an
insufficient response at the 1000 mg/1 level in the range-
finding test, the concentration of chemical in the test
solution should be determined to confirm the actual exposure
level.
-------
ES-5
August, 1982
The pH of the test solution should be measured prior to
testing to determine if it lies outside of the species
optimal range. While it is recognized that algae may grow
over a broad range of hydrogen-ion concentrations and
typically exhibit a pH optima for logarithmic growth, this
test guideline does not include pH adjustment for the
following reasons: the use of acid or base may chemically
alter the test substance making it more or less toxic, the
amount of acid or base needed to adjust the pH may vary from
one test solution concentration to the next, and the effect
the test chemical has on pH may indirectly affect growth and
ii ! '
development of the algae. Therefore1, the pH of each test
solution should be determined and compared to the acceptable
range for growth and development of the test algae.
The data obtained in bioassays are usually expressed as
standard response curves in which growth response of the
test species is plotted against the concentration of the
test chemical. The manner of expressing algal growth
response varies considerably. For this guideline algal
growth responses are expressed as direct measurements of
number of algae per ml of solution. The statistical
analysis ( goodness-of-f it determination) facilitates
accurate calculations of EC-10 and EC-50 as well as
providing confidence limits for the concentration (dose)-
response curve.
B . Test Conditions
1. Test Species
Both Salenastrum capr icornu turn and Skeletonema cos tatum
have a number of useful characteristics as listed below,
10
-------
ES-5
August, 1982
which are necessary for an algal species to be used in
bio as says (Toerien et al. 1971):
(a) broad nutrient- response (grows both in
oligotrophi'c and euthropic waters).
(b) distinct shape
(c) uniform size
(d) divide distinctly
(e) do not attach to glass or surface
(f) stay in suspension with slight agitation
(g) cells do not clump (aggregate)
(h) grow at a maximum rate ,in a short time in a
1 ' I ' ! !
medium simple to constitute
(i) do not excrete autotoxins
(j) cells are easy to count by both direct or
indirect methods.
Selenastrum capr icornu turn is an excellent laooratory
freshwater organism, easy to culture and count, and is both
sensitive and consistent in its response to a wide range of
nutrient levels (Payne and Hall 1979).
When included in multisoecies toxicity screening tests,
Selenas trum has been found to be a comparably sensitive
species. Maki and Macek (1978) found this to be true in an
environmental safety assessment for a nonphosphate detergent
builder. Selenastrum was as sensitive to trinitrotoluene as
the copepod, Trigriopus californicus, and was twice as
sensitive as oyster larvae (Smock et al. 1976). Selenastrum
was as sensitive as Daphnia and the fathead minnow to eight
preparations of synfuels (Greene, personal communication).
In a study of the toxicity of 56 dyes to Selenastrum and
fish (fathead minnows), basic dyes do not markedly inhibit
11
-------
ES-5
August, 1982
algal growth, and "of special significance, however, is the
rather startling correlation between results of algal assays
and the results of fish bioassays" (Little and Ch illingworth
1974). Greene (personal commun iation) analyzed the results
of this study and found the algae appear more sensitive than
fish to 35 of the dyes tested while the fish were only more
sensitive to seven of the dyes tested. In a recent test
conducted on 35 chemicals on the EPA priority pollutant list
by EG & G Bionomics (Parrish, personal communication), there
were no significant differences in the EC-O's between
Selenas trum a.nd Skeletqnema, Daphnia and bluegill fish,
I i ' .I
Leporgis macrochirus. Selenas trum was significantly more
sensitive than sheepshead minnow. In another 2 tests EG & G
performed for Monsanto Industrial Chemical Co. (1979a,b)
evaluating two phthalate esters (Santicizer 60 and 711),
Selenas trum was as sensitive as Microcys t is aerugenos a,
Navicula pelliculosa, Skeletonema costaturn and Dunaliella
tert iolecta. Palmer (1969) has extensively reviewed the
algal literature and has ranked the 60, most pollution
tolerant genera as reported by 165 authors. In comparing
two green algae often used in algal toxicity testing,
Chiorella and Scenedesmus to Selenas trum, great variation is
found. Of the 60 genera, Scenedesmus was the fourth most
tolerant, Chlorella was the fifth most tolerant, but
Selenastrum was the fifty-seventh most tolerant. This
analysis is borne out by recent results obtained by Green
(personal communication) in testing effluent toxicity to
algae. He found that Chlorella and Scenedesmus are
generally more resistant to industrial effluents and both
were naturally present in 100% effluents (eight submitted by
12
-------
E3-b
August, 1982
the US EPA Industrial Environmental Laboratory, Research
Triangle Park, Raleigh, North Carolina). Selenastrum only
grew when the effluents were diluted to 1—10% of the
original concentration (which supported" Chlorella and
g'.:anedesmus growth). This was also the case in another
'•• r
Affluent which contained 1.7 mg/1 cyanide. Both Chlorella
and Scenedesmas grew in it, but Selenas trum grew only when
the effluent was diluted to 1% or less. Chlorella has also
recently been shown to be much less sensitive to toxics than
i
Daphnia or fish (Xenaga and Molenaar, 1979).
While it is recognized that numerous marine algae are
• ' . . I ' 'J ' . : ' !
sensitive to toxicants (North et al. 1972); heavy metals
(Davies 1978), simple organics (benzene, cresol, hexane,
phenol and toluene), various inorganics (Cl, CN, Hg) and
complex wastes (industrial sewage, salfite waste liquor,
detergent), and petroleum compounds (Corner 1978),
Skeletonema costatum was selected for use in the toxicity
test guideline. This species has been frequently reported
on. in the bioassay literature (US Army 1978), and is a
recommended bioassay organism (APH 1975, US EPA 1977a, b,
1978, Gentile and Johnson 1974).
The testing procedure for Skeletonema has recently
proven useful for the evaluation of the relative potential
hazards of a compound or a complex waste by providing data
for the calculation of the EC-50 or SC-20 (Walsh and
Alexander 1980, Walsh et al. 1980). Skeletonema was as
sensitive to the 35 priority pollutants and two phthalate
esters as Selenastrum in multi-species toxicity screening
tests, as in the previously described studies.
13
-------
ES-5
August, 1982
Skeletonema was found to be more sensitive (at lOppb) to
growth inhibition effects induced by PCB's than two
freshwater algae (Euglena gracilis and Chlamydomonas
reinhardt ii ) and two other marine algae ( Thalass ios ira
pseudonana, and Dunaliella tertiolecta) (Mosser et al.
1972) .
Skeletonema cos ta turn was also more sensitive (growth
inhibited) at lower concentrations of wastewater
chlorination products ( 3-chlorobenzoic acid, 5-chlorouracil,
4-chlororesorcinol, 3-chloroohenol and Captan) than
Dunaliella tertiolecta and -Poirphyr idium, sp. (Sikkja and
Butler 1977) .
Skeletonema and Selenastrum are specified for testing
toxicity of pesticides (Subpart J, Pesticide Registration
Guidelines). Additional justification for selection of
these test species is provided in these guidelines (see FR
45 (214): 72948-72978).
Other species may be substituted for either of these two
species when appropriate. Some freshwater or marine species
which are of concern or have a significant ecological role
may constitute a more crucial risk population. If so, those
species of particular ecological or economic value should be
selected. The rationale for selection of alternative
species should be discussed with the Agency and/or supported
in the report of findings.
2. Facilities
a. General
The test requires a growth chamber or temperature
controlled enclosure capable of maintaining a uniform
temperature of 24° + 1°C if Selenastrum is tested or
14
-------
E3-5
August, 1982
20° +_ 1°C if Skeletonema is tested. Other facilities
typically needed include standard laboratory glassware,
culture flasks, work areas to clean and prepare equipment
and to measure chemical concentrations and algal growth and
proper dispos a],'; facilities . Without these^facilities, the
testing cannot'jibe adequately conducted.
•b. Test Containers
Sterile Erlenmeyer flasks are recommended as test and
culture containers. Any flask volume may be used between
.
125-500 ml.' However, it is imperative that flasks of the
same volu;me be used throughout the test. Hannon and
[ ' | ! |l " !
Patouillet (1979) found a marked difference (2.6x)in mercury
toxicity for marine algae, Phaeodactylum tricornutum,
depending1 on the surface : volume ratio of the culture
vessel. Flasks should be stoppered with sterile plugs (such
as foam rubber or cotton stoppers) which will prevent
possible bacterial' contamination yet allow air flow-
c. Cleaning and Sterilization
Standard good laboratory practices are recommended to
remove dust, dirt, other debris, and organic and inorganic
residues from the test containers and other glassware and
supplies should be washed and sterilized to prevent
contamination.
Algal cells are discarded at the end of a test. Algae
are capable of considerable adaptation to the toxic effects
of antime taboli tes and antibiotics, such as streptomycin,
penicillin, chloramphenicol, sulfanilimide and sodium
selenate (Kumar 1964).
It is important to avoid contamination of algal cultures
by bacteria. Bacteria may metabolize high molecular weight
15
-------
ES-5
August, 1982
organic compounds to produce carbon dioxide and/or cofactors
that stimulate growth of Selenastrum (Tison and Lingg 1977,
Sachdev and Clesceri 1978). Consequently axenic cultures of
algae should be maintained by proper sterile culture
techniques as well as growing and testing algae in sterile
containers and nutrient medium.
d. Conditioning
Test containers are to be rinsed with appropriate test
solutions prior to the beginning of•the toxicity tests.
This method should allow for sorption of the test substance
to the t^st container, thereby saturating- the container
1 1 • i
surface so that no further interactions of 'test substance
will take place when new test solution is added and the test
begins. Hannan and Patouillet (1979) found that up to 50%
of mercury could be lost to adsorption to vessel walls in a
two-day toxicity test. Therefore, with proper conditioning
all the test substance in the test solution should be
available to test algae and any results will reflect an
accurate concentration response.
e. Nutrient Medium
The nutrient medium recommended in the test guideline,
are those currently recommended by the US EPA for use in
bioassays (US EPA 1977, 1978a,b,c, Walsh and Alexander 1980,
Walsh et al . 1980) .
Use of the nutrient media under the test conditions will
ensure maximum growth rates (i.e., logarithmic) in test
algae and controls. Selenastrum and Skeletonema will divide
2-3 times per day (Nielsen 1978, Lewin and Guillard 1963,
USEPA 1971b). This should enhance exposure of test algae to
the test substance because algal cells in this growth phase
16
-------
ES-5
August, 1982
absorb and metabolize substances at a rapid rate (Fogg
1965). Shiroyama et al . (1973) found maximum phosphorus and
nitrogen uptake occurred in the first five days of growth.
Many media used for culturing algae contain a chelating
agent, usually EDTA, to keep,f':.ucronutrients in so.lution.
However, a medium containing'n'a chelating- agent is less than
ideal for testing toxicants because chelators can increase
or decrease toxicity and can add uncertainty to the test
results (Payne 1975, Fogg 1965, Prakash and Rashid 1968,
I
Bender 1970, Giesy 1974, Lin and Schelske 1979, Barber and
iRyfc'her 1969, Johnston 1964, Droop 1960,,1962; Eyster 1968,
Erickson et al. 1970).
3. Environmental Conditions
Selenas trum and Skele tonema will grow over a wide
• i ,
temperature range, from less than 5°C to 35°C (Claesson and
1 i
Forsberg 1978), and between 13°C and 30°C (Fogg 1965),
respectively. The temperature selected for toxicity testing
using Selenas trum w .s 24°C because luxury uptake of ammonia
nitrogen, maximum specific growth rate, and sensitivity to
phenol occur at that temperature (Reynolds et al. 1974,
1975a, 1975b 1976). The test temperature 20°C selected for
Skele tonema is recommended in other toxicity testing manuals
(USEPA 1978a,c) and in recent publications (Walsh and
Alexander 1980, (Walsh and Alexander 1980, Walsh et al .
1980) .
Algae require light for photosynthesis and growth.
Fitzgerald (1975) and Miller et al. (1978) have shown that
light intensity will affect the rate of growth of
Selenas trum. As practically all the provisional algal assay
procedure (Joint Industry/Government Task Force 1969)
17
-------
ES-5
August, 1982
development work was done on Selenastrum at 400 ft-c, it was
not seen as necessary to make a change (USEPA 1978b).
Continuous lighting of algal cultures is required for
Selenastrum in the test guideline. While this does not
reflect environmental conditions, it does maximize testing
for toxicity. Practically all toxicity tests using
Skeletonema have recommended split day/night lighting (USEPA
1978a, 1978c, Walsh and Alexender 1980, Walsh et al .
1980). For the sake of consistency, it was not seen as
necessary to make a change in the procedure.
The test guideline; requires a test solution pH of 7.5
for Selenas trum because it maximizes growth. Selenas trum
grows between pH 4 and 10 (Brezonik et al . 1975) and
maximally between pH 7 and 9.6 (Claesson and Forsberg
1978). Maximum adenosine triphosphate (ATP) (i.e., energy
production) occurs in Selenastrum cultured between pH 7.5
and 8 (Brezonik et al. 1975). The pH selected for testing
with Skele tonema, 8.1» was selected because it is
recommended by other toxicity testing manuals (USEPA 1978a)
and in recent publications (Walsh and Alexander 1980, Walsh
et al. 1980) and approximates the natural oceanic pH. The
pH should be adjusted as. exactly as possible to the test pH
because fluctuations in pH affects toxicity.
The purposes of oscillating the cultures are to enhance
e-xposure of algal cells to test substances and to enhance
dissolution and solubilization of test substances in the
test solution. Turbulence created by shaking algal cultures
is important to enhance the transfer of dissolved substances
between the media and the cells. Munk and Riley (1952)
showed that this transfer is faster if nutrients are
18
-------
E3-5
August, 1982
continually renewed adjacent to the cell by movement of the
medium.
Oscillating test containers is also analogous to wind
and wave induced mixing of natural waters. This agitation
and mixing serves to maximize algal exposure to the test
substance.
Temperature, light intensity, pH and oscillation rate
are all recorded as specified in the test guideline to
ensure that the environmental conditions of the test are
> 1 I
me t.
Temperature should be recorded at least hourly to ensure
1 'II i ' , I !
that it does not exceed the specified limits. Inexpensive
growth chambers are available which are equipped with
adequate recording instruments or chambers may be equipped
with ones at minimal cost. Severe fluctuations in
temperature may affect algal growth and/or subsequent
i
chemical uptake or metabolism.
Light intensity readings at t a surface of the solutions
may be made manually and ensure that all containers are
receiving equal light. Light variations will affect algal
growth so'daily recordings are necessary to maintain uniform
and constant, r.adiation. The pH is measured at the beginning
and end of the test as an indication of effects of test
chemical additions and subsequent algal metabolism on the
hydrogen-ion concentration. This will indicate if the test
solution is outside of the algal pH optima for growth as
well as show what pH variations may exist between chemical
concentrations .
19
-------
ES-5
August, 1982
C. Reporting
The sponser should submit to the Agency all data
developed during the test that are suggestive or predictive
of phytotoxicity. If testing specifications are followed,
the sponsor should report that specified procedures were
followed and present the results. If alternative procedures
were used instead of those recommended in the test
guideline, then the protocol used should be fully described
and justified.
Test temperature, chemical concentrations, test data,
concentration-response curves, and statistical analyses
should all be reported. The justification for this body of
information is contained in this support document. If algal
species other than the two recommended were used, the
rationale for the selection of the other species should be
provided.
III. Economic Aspects
The Agency awarded a contract to Enviro Control, Inc. to
provide an estimate of the cost for performing an acute
toxicity test using freshwater algae according to the
Guideline. Enviro Control supplied two estimates; a
protocol estimate and a laboratory survey estimate.
The protocol estimate was $1760. This estimate was
prepared by identifying the major tasks needed to do a test
and estimating -the hours to accomplish each task.
Appropriate hourly rates were then applied to yield a total
direct labor charge. An estimated average overhead rate of
115%, other direct costs of $400, a general and
administrative rate of 10%, and a fee of 20% were then added
to the direct labor charge to yield the final estimate.
20
-------
ES-5
Augus t, '1982
Environ Control estimated that differences in salaries,
equipment, overhead costs and other factors between
laboratories could result in as much as 50% variation from
this estimate. Consequently, they estimated that test costs
could range from $873 to $2636. f-
The laboratory survey estimate was $1465, the mean ofi.
the estimates received from eight laboratories. The
estimates ranged from $430 to #3600 and were based on the
costs to perform the test according to the Guideline.
Although a cost analysis was 'not performed for a test
us ing , mar ine algae, the procedures used are similar, to the
freshwater algal test!and the costs should ba similar.
21
-------
Augus t,
ES-5
1982
IV. References
APHA. 1975. American Public Health Association, American
Waters Works Association, and Water Pollution Control
Federation. Standard methods for the examination of water
and wastewater, 14th ed. Washington, D C: American Public
Health Association.
Barber RT, Ryther JH. 1969. Organic chelators factors
affecting primary productivity in the Cromwell Current
upwelling. J. Exp. Mar. Biol. 3:191-99.
Bender ME. 1970. On the significance of metal complexing
agents in secondary sewage effluents. Environ. Sci.
Technol'. ' 4:520.
Bold HC and Wynne MJ
Englewood Cliffs, NJ;
1978. Introduction
Prentic-Hall, Inc.
to the algae.
Brezonik PL, Browne FX, Fox JL. 1975. Application of ATP
to plankton biomass and bioassay studies. Water Res. 9:155-
162.
Claesson A and Forsberg A. 1978.
with one or five species minitest.
Limnol. 21:21-30.-
Algal assay procedures
Mitt. Internat. Verein.
Corner EDS,
plankton.
compounds .
1978. Pollution studies with
Part 1. Petroleum hydrocarbons
Adv. Mar. Biol. 15:289-380.
marine
and related
Council on Enviromental Quality- 1979. Ecology and living
resources: coastal ecology and shellfish. Environmental
Quality 1970, 10th Annual Report of the Council on
Environmental Quality, December 1979. Washington, D,C: US
Government Printing Office.
Davies AG. 1974. The growth kinetics of Isochrys is galbana
in cultures containing sublethal concentrations of mercuric
chloride. J. Mar. Biol. Assn. U.K. 54:157-169.
Davis AG. 1978. Pollution studies with marine plankton.
Part II. Heavy metals. Adv. Mar. Biol. 15:381-508.
22
-------
ES-b
August, 1982
Droop Mr. 1960. Some chemical considerations in the design
of synthetic culture media for marine algae. Botanical
Marina 2:231-246.
Erickson SJ, Lackie N, Maloney TE. 1970. A screening
technique for estimating copper toxicity to estuarine
phytoplankton. J. Water Poll". Contr. Fed. 42:270-278.
Eyster C. 1968. Microorganic and raicroinorganic
requirements for algae. In: Jackson DF, ed . Algae, man,
and, the environment. New York: Syracuse University Press,
pp. 27-3.6.
Fisher NS and Wurster CR. 1973. : Individual and combined
effects of temperature and polychlorinated bioh'enyls on the
growth of three species of phytoplankton. Environ. Poll.
5:205-212. : » | i i
. 1975. Factors affecting the algal assay
procedure. Washington, DC: US EPA.
Fogg GE. 1965. Algal cultures and , phytoplankton ecology.
Madison, WI: University of Wisconsin Press. '
Galtsoff PS. 1964. The American Oyster, Grasses trea
virginica Gmelin. U.S. Fish. Wildl. Serv^ Bull. 64:1-480
Gentile JH, Johnson MW. 1974. Marine phytoplankton In:
Marine bioassays, workshop proceedings concened Dy G. Cox.
Washington, DC: Marine Technology Society. pp. 128-143.
Giesy JP. 1974. The effects of humic acids on the growth
and the uptake of iron and phosphorus by the green algae
Scenedesmus obliguus Kuetz. Ph.D. thesis. East Lansing,
MI: Michigan State Un ivers ity .
Hammer PA and Urquhart NS. 1979. Precision and
replication: Critique II. In: Tibbits TW and Kozlowski TT,
eds. Controlled environment guidelines for plant
research. New York: Academic Press, pp. 368.
Hannan PJ and Patouillet C. 1979. An algal toxicity test
and evaluation of adsorption effect. J. Water Poll. Contr.
Fed. 51:834-840.
23
-------
ES-5
August, 1982
Harding LW and Phillips JH. 1978. Polychlorinated biphenyl
(PCB) effects on marine phytoplankton photosynthesis and
cell division. Mar. Biol. 49:93-101.
Johnston R. 1964. Seawa'ter, the natural medium of
phytoplankton. 2. Trace metals and chelation and general
discussion. J. Mar. Biol. Ass. U.K. 44:87-109.
Joint Industry/Government Task Force on Eutrophication.
1969. Provisional algal assay procedure. Joint
Indus t./Gov. Task Force Eutroph. , New York: 6 2p.
Kenega EE and Molenaar R J. 1979. Fish and Daphnia toxicity
as surrogates for aquatic vascular plant and algae Environ.
Sci. Technol. 13:1479-1488.
Leischraan AA/j Green JC|, Miller WE. 1979,. Bibliography . of
literature pertaining to the genus Selen'as tram. Corvallis,
OR: U.S. Environmental Protection Agency. US EPA-6 00/9-7 9-
021.
Lewin JC and Guillard R. 1963. Diatoms. Ann Rev.
Microbiol. 7:373-414.
Lightner DV, 1978. Possible toxic effects of the marine
blue-green alga Spirulina subsalsa, on the blue shrimp,
Penaeus_ s tyliros tr is . J. Invert. Path. 32:139-150.
Lin KG and Schelske CL. 1979. Effects of nutrient
enrichment, light intensity and temperature on growth of
phytoplankton from Lake Huron. Duluth, MN: U.S.
Environmental Protection Agency. EPA-600/3-76-0 75.
Little LW and Ch illingworth MA. 1974. Effect of 56
selected dyes on growth of the green alga Selenastrum
capricornutum. New York: American Dye Manufacturers
Institute, Inc., pp.2-14.
Lovell RT. 1979. Fish culture in the U.S. Science
206:1368-72.
Maki AW and Macek KJ. 1978. Aquatic environmental safety
assessment for a nonphosphate detergent builder. Environ.
Sci. Technol. 12:573-580.
24
-------
ES-5
August, 1982
Maloney TE and Miller WE. 1975. Algal assays: development
and application. STP 573. Philadelphia, PA: American
Society Eor Testing and Materials, pp. 344-354.
Miller WE, Greene JC, Merwin EA, Shiroyama T. 1978. Algal
bioassay techniques for pollution evaluation. In: Toxic
materials in the aquatic environment. Seminar at Water >
Resources Res. Inst., Spring 1978, corvallis, OR: Oregon
State University SEMIN WR 024-78, pp. 9-16.
Monsanto Industrial Chemicals Company. 1979a. TSCA sec.
8(d) submission 3DHQ-1078-0 234. EG & G Bionomics data on
Santicizer 711, 1978. Washington, D.C.: Office of Toxic
Substances, U.S. Environmental Protection Agency.
. 1979b. TSCA sec. 8(d) submission 8DHQ-1078-
,0285. 103 & G Bionomics data on Sajn.t(icizer 160, 1978.
Washington, D.C.: Office of Toxic Substances, U.S.
Environmental Protection Agency.
•Mosser'JL., Fisher NS, Teng TC, Wurster CF. 1972.
Polychlorinated biphenyls: toxicity to certain
phytoplankters. Science 175:191-192.
Munk WH and Riley GA. 1952. Absorption of nutrients by
aquatic plants. J. Mar. Res. 11:215-240.
Murray L, Scherifig J, Dixon PS. 1971. Evaluation of algal
assay procedures - PAAP Batch Test. J. Water Poll. Contr.
Fed. 43:1991-2003.
Nielsen ES. 1978. Growth of plankton algae as a function
of N-concentration, measured by means of a batch technique.
Mar. Biol. 46:185-189.
North WJ, Stephans GC, North BB. 1972. Marine algae and
their relation to pollution problems, In: Ruivo M, ed.
Marine pollution and sea life. London: Fishing News (Books)
Ltd., pp. 330-340.
O'Brien Py and Dixon PS. 1976. The effects of oils and oil
components on algae: a review. Br. Phycol. 11:115-142.
Payne AG. 1975. Application of the algal assay procedure
in biostimulation and toxicity testing. In Middlebrooks
E J, Falkenborg DH, Maloney TE, eds . Biostimulation and
nutrient assessments. Ann Arbor, MI: Ann Arbor Science
Publishers, pp. 3-28.
25
-------
ES-5
August, 1982
Payne AG and Hall RH 1979. A method for measuring algal
toxicity and its application to the safety assessment of new
chemicals. In: Marking LL, Kimerle RA, eds. Aquatic
toxicology. ASTM Special Technical Publication 667,
Philadelphia, PA: American Society for Testing and
Materials, pps. 171-180.
Prakash A and Rashid MA. 1968. Influence of humic
substances on the growth of marine phytoplankton:
d inof lagellates . Limnol. Oceanog. 13:598-606.
Reynolds JH, Middlebrooks E J, Porcella DB, Greeney WJ.
1974. A continuous flow kinetic model to predict the effect
of temperature on the toxicity of -waste to algae.
Springfield, VA: U.S. Dept. Commerce, NTIS PB-23194.
1975a. Effjeicts
-------
ES-5
August, 1982
Shiroyama T, Miller WF, Greene JC. 1973. Effect of
nitrogen and phosphorus on the growth of Seleaastrum
capricornutum. Washington, D.C.: U.S. Environmental
Protection Agency. U.S. EPA-660/3-75-034.
Sika HC and Butler GL. 1977. Effects of selected
wastewater chlorination products and captan on marine
algae. Corvallis, OR: U.S. Environmental Protection Agency
E- \-600/3-77-029.
S ock LA, Stoneburner DL, Clark JR. 1976. The toxic
effects of trinitrotoluene (TNT) and its primary degradation
products on two species of algae and the fathead minnow.
Water Res. 10:537-543.
'Stern WM and Stick WB. 1978. Effects of turbidity and
suspended material injaquatic environments. Washington, DC:
US Department of Commerce. NTIS, #AD-A056-035.
Taylor DL Seilger HH, ed. 1979. Toxic dinoflagellate
blooms. New York: Els evier/North-Holland.
Tis,on DL and Lingg AJ. 1977.' Algal bacterial mutualism in
a simulated aquatic environment. Ann. Mtg . Am. Soc.
Microbiol. 77:237.
Toerien DF, Huang CH, Radimsky J, Pearson EA, Scherf ig J.
1971. Final Report, Provisional Algal Assay Procedures.
EPA/ Water Pollution Control Research series, SERL Report.
Springfield, VA: National Technical Information Service.
US Army, Office of Chief of Engineers. 1978.
Considerations in Conducting Bioassays. Technical Report D-
78-23 of the Dredged Material Research Program. Vicksburg,
MS: U.S. Army. DAWC 39-73C-0134.
US EPA. 1971a. U.S. Environmental Protection Agency.
Algal assay procedure: bottle test. Corvallis, OR: U.S.
Environmental Protection Agency.
. 1971b. Environmental Protection Agency.
The interlaboratory precision test: an eight laboratory
evaluation of the provisional algal assay procedure bottle
test. Corvallis, OR: U.S. Environmental Protection Agency.
27
-------
ES-5
August, 1982
1973. U.S. Environmental Protection
Agency. Biological field and laboratory methods for
measuring the quality of surface waters and effluents.
Cincinnati, OH: U.S. Environmental Protection Agency. EPA
670/4-73-001.
._ . 1974. U.S. Environmental Protection
Agency. Marine algay assay procedure: bottle test.
Corvallis, OR: U.S. Environmental Protection Agency-
. 1975. U.S. Environmental Protection
Agency. DDT: a review of scientific and economic aspects of
the decision to ban its use as a pesticide. Washington,
D.C. U.S. Environmental Protection Agency. EPA-540/1-75-
022.
. 1977., U.S. Environmental Protection
Agency/Corps of Engineers Technical Committee on criteria
for dredged and fill material. Ecological' evaluation of
proposed discharge of dredged material into ocean waters.
Vicksburg, MS: U.S. Environmental Protection Agency.
._ 197Sa'. U.S. Environmental Protection
Agency Ocean Disposal Bioassay Working Group. Bioassay
procedures for the ocean disposal pe'rmit program. Gulf
Breeze, FL: U.S. Environmental Protection Agency.
. .1978-b. U.S. Environmental Protection
Agency. The Selenastrum capr icornu turn Printz algal assay
bottle test: Experimentalal design, application, and data
interpretation protocol. Corvallis, OR: U.S. Environmental
Protection Agency, EPA-600/9-78-018.
_. 1978c. U..S. Environmental Protection
Agency. IERL-RTP Procedures Manual: Level 1 Environmental
assessment, 2nd ed. Research Triangle Park, NC: U.S.
Environmental Protection Agency. EPA-600/7-78-201.
. 1978d. U.S. Environmental Protection
Agency. Environmental Assessment: source test and
evaluation report--Chapman low-Btu gasification. EPA-600/8-
78-202, Research Triangle Park, NC: U.S. Environmental
Protection Agency.
Vance BD and Maki AW. 1976. Bioconcentration of Dibrom by
Stigeoclonium pachydermmn. Bull. Environ. Contain.
Toxicol. 15:601-607.
28
-------
ES-5
August, 1982
Vandermeulen JH and Ahern TP- 1976. Effect of petroleum
hydrocarbons on algal- physiology: review and progress
report. INL Lockwood APM, ed. Effects of pollution on
aquatic organisms. Cambridge, Eng: Cambridge University
Press, PP. 107-125.
Vernberg F J. 1977. Characterization of the natural estuary
in terms of energy flow and pollution impact. In: USEPA.
Proc. Conf. on estuarine pollution control and assessment.
Vol. I & II. Springfield, VA: NTIS, pps. °9-39. EPA 44/1-
77-007.
Walsh GE and Alexander SV. 1980. A ma'rins- algal bioassay
method results with pesticides and industrial wastes. Water
Air Soil Pollut. 13:5-55.
Walsh GE, Bahner LH, Horning WB. 1980. ,Toxicity of textile
mill effluents to freshwater and estuarine algae,
crustacean!, and1 fishes. Environ.: Poll. (Sdr A) 21:16:9-179.
29
-------
EG-9
August, 1982
FISH ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances 53-9
Guideline for Testing Chemicals August, 1982
FISH 'ACUTE TOXICITY TEST
(a) Purpose. This guideline may be' used to develop
data on the acute toxicity of- chemical substances and
mixtures ("chemicals") subject to environmental effects test
regulations under the Toxic Substances Control Act (TSCA)
(P.L. 94-469, 90 Stat. 2003, 15 U.S.C. 2601 _e_t. seq.-) . This
guideline prescribes tests to be used to develop data on the
acute toxiqi,ty of chemicals to, fish. The United Stages
Environmental Protection Agency (EPA) will use data from
these tests in assessing the hazard of a chemical to the
environment.
(b) Def initions. The definitions in Section 3 of the
To,xic Substances Control Act (TSCA), and the definitions in
"Good Laboratory Practice Standards for Physical, Chemical,
Persistence, and Ecological Effects Testing" (Proposed Part
772, Subpart B, Section 772.110-2) apply to this test
guideline. The following definitions also apply to this
guideline:
(1) "Acclimation" means the physiological compensation
by test organisms to new environmental conditions (e.g.,
temperature, hardness, pH) .
-------
EG-9
August, 1982
(2) "Acute toxicity test" means a method used to
determine the concentration of a substance that produces a
toxic effect on a specified percentage of test organisms in
a short period of time (e.g., 96 hours). In this guideline,
death is used as the measure of toxicity.
(3) "Carrier" means a solvent used to'dissolve a test
substance prior to delivery to the test chamber.
(4) "Conditioning" means the exposure of construction
materials, test chambers, and testing apparatus to dilution
water or to test solutions prior to the start of a test in
order to minimize the sorption of the test substance onto
the tes t- facilities or the leaching of substances from the
test facilities into the dilution water or test solution.
(5) "Death" means the lack of opercular movement by a
test fish.
(6) "Flow-through" means a continuous or an
intermittent passage • of test solution or dilution water
through a test chamber, or a holding or acclimation tank
with no recycling.
(7) "Incipient LC50" means that test substance
concentration, calculated from experimentally-derived
mortality data, that is lethal to 50 percent of a test
population when exposure to the test substance is continued
-------
'EG-9
August, 1982
until the mean increase in mortality does not exceed 10% in
any concentration over a 24 hour period.
(8) "LC50" means that test substance concentration,
calculated from experimentally-derived mortality data, that
is lethal to 50 percent of a test population during
continuous exposure over a specified period of time.
(9) "Loading" means the ratio of fish biomass (grams,
wet weight) to the volume (liters) of test solution in a
11. i i
test chamber or passing through it in a 24 hour period.
(10) "Static" means the test solution is not renewed
during the period of the test.
(11) "Test solution" means the test substance and the
dilution water in which the test substance is dissolved or
suspended .
(c) Test procedures--( 1) Summary of the test. (i)
Test chambers are filled with appropriate volumes of
dilution water. If a flow-through test is performed, the
flow of dilution water through each chamber is adjusted to
the rate desired.
(ii) The test substance is introduced into each test
chamber. In a flow-through test, the amount of test
substance which is added to the dilution water is adjusted
to establish and maintain the desired concentration of test
substance in each test chamber.
-------
EG-9
August, 1982
(iii) Test fish which have been acclimated in
accordance with .the test design are introduced into the test
and control chambers by stratified random assignment.
(iv) Fish in the test and control chambers are observed
periodically during the test; dead fish are removed at least
twice each day and the findings are recorded.
(v) The dissolved oxygen concentration, pH, temperature
and the concentration of test substance are measured at
i
intervals in 'selected test chambers.
(vi) Concentration-response curves and LC50 values for
the tes t-s ubs tance are developed from the mortality data
collected during the test.
(2) [Reserved]
(3) Range finding test. If the toxicity of the test
substance is not already known, a range finding test should
be perfonfied to determine the range of concentrations to be
used in the definitive test. The highest concentration of
test substance for use in the range finding test should not
exceed its solubility in water or the permissible amount of
the carrier used.
(4) Definitive tes t. (i) A minimum of 20 fish should
be exposed to each of five or more test substance
concentrations. The range of concentrations to which the
-------
EG-9
August, 1982
fish are exposed should be such that in 96 hours there are
at least two partial mortality exposures bracketing 50%
survival.
(ii) For exposure to each concentration of a test
substance, an equal number of test fish should be placed in
two or more replicate test chambers. The distribution of
individual fish among the test chambers should be
randomized .
(iii) Every test should include a control consisting of
the same dilution water, conditions, procedures,. and fish
from the same group used in the test, except that none of
the test substance is added.
(iv) Mortality data collected during the test are used
to calculate a 96-hour LC50. The 24-, 48-, and 72-hour
values should be calculated whenever there is sufficient
mortality data to determine such values. If the 96-hour
LC50 is less than 50% of the estimated 48-hour LC50 in a
flow-through test, the test should be continued until the
mean increase in mortality at any test concentration does
not exceed 10% over a 24-hour period or until 14 days.
(v) Test fish should not be fed while they are being
exposed to the test substance under static conditions or
during the first 96 hours of flow-through testing. If the
-------
EG-9
August, 1982
test continues past 96 hours, the fish should be fed a
suitable food at a maintainance level every other day
beginning on test day 5. Any excess food and the fecal
material should be removed when observed.
(5) Test results. (i) Death is the primary criterion
used in this test guideline to evaluate the toxicity of the
test substance.
(ii) In addition to death, any abnormal behavior such
as, but not limited to, erratic swimming, loss of reflex1,
increased excitability, lethargy, or any changes in
appearance or physiology such as discoloration, excessive
mucous production, hyperventilation, opaque eyes, curved >
spine, or hemorrhaging should be recorded.
(iii) Observations on compound solubility should be
recorded. The investigator should report the appearance of
surface slicks, precipitates, or material adhering to the
sides of the test chamber.
(iv) Each test and control chamber should be checked
for dead fish and observations recorded at 24, 48, 72, and
96 hours after the begining of the test or within one hour
of the designated times. If the test is continued past 96
hours, additional observations should be made every 24 hours
until termination.
-------
EG-9
August, 1982
(v) The mortality data is used to calculate LCSO's and
their 95% confidence limits, and to plot concentration-
response curves for each time interval whenever sufficient
data exists. The methods recommended for use in calculating
LCSO's include probit, logit, binomial, and moving average
ang le .
(•vi) A test is unacceptable if more than 10 percent of
the control fish die or exhibit abnormal behavior during a
]>.[:', ] , , I
96-hour testJ 'If a flow-through test is continued past 96
hours, the maximum allowable additional mortality is' 10
percent.
(6) Analytical measurements — (i) Water quality
analys is. (A) The hardness, acidity, alkalinity, pH,
conductivity, TOC or COD, and particulate matter of the
dilution water should be measured at the beginning of each
static test and at the beginning and end of each flow-
through test. The month to month variation of the above
values should be less than 10% and the pH should vary less
than 0.4 units.
(B) During static tests, the dissolved oxygen
concentration, temperature and pH should be measured in each
test chamber at the beginning of the test and as often as
needed thereafter to document changes from the initial
-------
EG-9
August, 1982
levels. The test solution volume should not be reduced by
more than 10% as a result of these measurements.
(C) During flow-through tests, dissolved oxygen,
temperature and pH measurements should be made in each
chamber at the beginning of the test and every 48 hours
thereafter until the end of the test.
(i i) Collection of samples for measurement of test
substance. Test solution samples to be analyzed for the
test substance should be taken midway between the top,
bottom, and sides of the test chamber. These samples should
not include any surface scum or material dislodged from the
bottom or sides. Samples should be analyzed immediately or
handled and stored in a manner which minimizes loss' of test
substance through microbial degradation, photodegradation,
chemical reaction, volatilization, or sorption.
(iii) Measurement of test substance. (A) For static
tests, the concentration of dissolved test substance (that
which passes through a 0.45 micron filter) should be
measured at a minimum in each test chamber at the beginning
(0-hour, before fish are added) and at the end of the
test. During flow-through tests, the concentration of
dissolved test substance should be measured as follows:
-------
EG-9
August, 1982
(JJ in each chamber at 0-hour;
(2) in each chamber at 96-hours and every 4 days
thereafter as long as the test is continued and;
(_3_) in at least one appropriate chamber whenever a
malfunction is detected in any part of the test substance
delivery system.
(B) Filters and their holders used for determining the
• dissolved test substance concentrations should be prewashed
With several volumes of distillled water and undergo a final
rinse with test solution. Glass or stainless steel filter
holders are best for organic test substances, while plastic
holders are best for metals. The sample should be filtered
within 30 minutes after it is taken from the test chamber.
(C) The analytical methods used to measure the amount
of test substance in a sample should be validated before
beginning the test. The accuracy of a method should be
verified by a method such as using known additions. This
involves adding a known amount of the test substance to
three water samples taken from a chamber containing dilution
water and the same number and species of fish as are used in
the test. The nominal concentration of the test substance
in those samples should span the concentration range to be
used in the test. Validation of the analytical method
-------
EG-9
August, 1982
should be performed on at least two separate days prior to
starting the test.
(D) An analytical method is not acceptable if likely
degradation products of the test substance give positive or
negative interferences, unless it is shown that such
degradation products are not present in the test chambers
during the test.
(E) In addition to analyzing samples of test solution,
at least one reagent blank, containing all reagents used,
should also be analyzed.
(F) If the measured concentrations of dissolved test
substance are considerably lower (e.g. <50 percent) than the
nominal concentrations, the total test substance
concentration should be measured in the highest test
concentration.
(G) Among replicate test chambers, the measured
concentrations should not vary more than 20%. The measured
concentration of the test substance in any chamber during
the test should not vary more than 30 percent from the
measured concentration prior to initiation of the test.
(H) The mean measured concentration of dissolved test
substance should be used to calculate all LCSO's and to plot
all concentration-response curves.
10
-------
EG-9
August, 1982
(d) Test conditions--!1) Test species. (i)
Selection. The test species for this test are the rainbow
trout (Salmo gairdneri), bluegill (Lepomis macrochirus) and
fathead minnow (Pimephales promelas). The particular
species of fish to be used will be prescribed in the test
rule .
(ii) Age and Condition of Fish. (A) Juvenile fish
should be used. Fish used in a particular test should be
the same age and be of normal size and appearance for their
age. The longest fish should not be more than twice the
length of the shortest.
(B) All newly acquired fish should be quarantined and
observed for at least 14 days prior to use in a test.
(C) Fish should not be used for a test if they appear
stressed or if more than five percent die during the 48
hours immediately prior to the test.
(iii) Acclimation of test fish. (A) If the holding
water is not from the same source as the test dilution
water, acclimation to the dilution water should be done
gradually over a 48-hour period. The fish should then be
held an additional 14 days in the dilution water prior to
testing. Any changes in water temperature should not exceed
3°C per day- Fish should be held for a mininum of 7 days at
the test temperature prior to testing.
11
-------
EG-9
August, 1982
(B) During the final 48-hours of acclimation, fish
should be maintained in facilities with background colors
.and light intensities similar to those of the testing area
and should not be fed. '
(2) Facilities — (i) General. Facilities needed to
perform this test include:
(A) flow-through tanks for holding and acclimating
fish,
'(B) A'mechanism for controlling and maintaining the
wate-r temperature during the holding, acclimation and test
periods,
(C) Apparatus for straining particulate mater, removing
gas bubbles, or insufficient dissolved oxygen, respectively,
(D) Apparatus for providing a 16 hour light and 8 hour
dark photoperiod with a 15- to 30-minute transition period,
(E) Chambers for exposing test fish to the test
substance;
(F) A test substance delivery system for flow-through
tests .
(ii) Construction materials. Construction materials
and commercially purchased equipment that may contact the
stock solution, test solution, or dilution water should not
contain substances that can be leached or dissolved into
12
-------
EG-9
August, 1982
aqueous solutions in quantities that can alter the test
results. Materials and equipment that contact stock or test
solutions should be chosen to minimize sorption of test
chemicals. Glass, #316 stainless steel, and perfluorocarbon
plastic should be used whenever possible. Concrete,
fiberglass, or plastic (eg.PVC) may be used for holding
tanks, acclimation tanks, and water supply systems, but they
should be used to remove rust particles. Rubber, copper,
brass, galvanized metal, epoxy glues, and lead should not
come in contact with the dilution water, stock solution, or
test solution.
(iii) Test substance delivery system. In flow-through
tests, diluters, metering pump systems or other suitable
devices should be used to deliver the test substance to the
test chambers. The sysstem used should be calibrated before
each test. Calbration includes determining the flow rate
through each chamber and the concentration of the test
substance delivered to each chamber. The general operation
of the test substance delivery system should be checked
twice daily during a test. The 24-hour flow rate through a
test chamber. During a test, the flow rates should not vary
more than 10 percent from one test chamber to another or
from one time to any other.
13
-------
EG-9
August, 1982
( iv) Test chambers. Test chambers made of stainless
steel should be weled, not soldered. Test chambers made of
glass should be fused or bonded using clear silicone
adhesive. As little adhesive as possible should be left
exposed in the interior of the chamber.
(v) Cleaning of test system. Test substance delivery
systems and test chambers should be cleaned before each
test. They should be washed with detergent and then rinsed
in sequence with clean water, pesticide-free acetone, clean
water, and five percent nitric 'acid, followed by two or more
changes of dilution water.
(vi) Dilution water. (A) Clean surface or ground
water reconstituted water, or dechlorinated tap water is
acceptable as dilution water if the test fish will survive
in it for the duration of the holding, acclimating, and
testing periods without showing signs of stress, such as
discoloration, hemorrhaging, disorientation or other unusual
behavior. The quality of the dilution water should be
constant and should meet the following specifications
measured at least twice a year:
14
-------
EG-9
August, 1982
Subs tance Maximum
Concentration
Particulate matter 20 rag/liter
Total or-ganic carbon or 2 mg/liter
chemical oxygen demand 5 mg/liter
Un-ionized ammonia lug/liter
Residual chlorine 1 ug/liter
Total organochloring pesticides 50 ng/liter
Total organocholorine pesticides
plus polychlorinated bephenyls • 50 ng/liter
(PCBs) or organic1 chlorine ' 25 hg/li|:er
(B) The concentration of dissolved oxygen in the
dilution water should be between 90 and 100 percent
saturation; 9.8-10.9 mg/1 for tests with trout, and 8.0-8.9
rag/1 for tests with bluegill or fathead minnow at sea
level. If necessary, the dilution wa±er can be aerated
before the addition of the test substance. All
reconstituted water should be aerated before use. Buffered
soft water should be aerated before but not after the
addition of buffers.
(C) If disease organisms are present in the dilution
water sufficient numbers to cause infection, they should be
killed or removed by suitable equipment.
15
-------
EG-9
August, 1982
(D) Glass distilled or carbon filtered deionized water
with a conductivity less than 1 micromho/cm is acceptable
for use in making reconstituted water. If the reconstituted
water is prepared from a ground or surface water source,
conductivity, and total organic carbon (TOG) or chemical
oxygen demand (COD) should be measured on each batch.
(vii) Carriers. (A) Distilled water should be used in
making stock solutions of the test substance. If the stock
vollufne howeve'r is more than 10% :Of the iest solutidn volumeJ
dilution water should be used. If a carrier is absolutely
necessary to dissolve the test substance, the volume used
should not exceed the minimun volume necessary to dissolve
or suspend the test substance in the test solution. If the
test substance is a mixture, formulation, or commercial
product, none of the ingredients is cons idered a carrier
unless an extra amount is used to prepare the stock
solution.
(3) Triethylene glycol and dimethyl formamide are the
perferred cariers, but acetone can also be used. The
concentration of triethylene glycol in the test solution
should not exceed 80 mg/1. The concentration of dimethyl
formamide or acetone in the test solution should not exceed
5/0 mg/1.
16
-------
EG-9
August, 1982
(3) Test parameters--( i) Loading. The number of fish
placed in a test chamber should not be so great as to affect
the results of the test. The loading should not be so great
that the test substance concen- trations are decreased by
more than 20 percent due to uptake by- the fish. In static
tests, loading should not exceed 0.5 grams of fish per liter
of solution in the test chamber at any one time. In flow-
through tests loading should not exceed 0.5 grams of fish
per1 liter of test solution passing thrbugh the! chamber in 24
hours. These loading rates should be sufficient to maintain
the dissolved oxygen concentration above the recommended
levels and the ammonia concentration below 20 ug/1.
(ii) Dissolved oxygen concentration. (A) During
static tests with rainbow trout the dissolved oxygen in each
test chamber should be greater than 5.5 mg/1. In tests with
bluegill and fathead minnows, the DO should be maintained
above 4.5 mg/1.
(B) During flow-through tests the dissolved oxygen
concen- tration should be maintained above 8.2 mg/1 in tests
with trout and above 6.6 mg/1 in tests with bluegills or
fathead minnows.
(iii) Temperature. The test temperature should be 22
_fl°C for bluegill and fathead minnows, and 12± 1°C for
17
-------
EG-9
August, 1982
rainbow trout. The temperature should be measured at least
hourly in one test chamber.
(iv) Light. A 16-hour light and 8-hour dark
photooeriod with a 15- to 30-minute transition period should
'::!•
''i!i
be maintained. |j
(e) Reporting . The sponsor should submit to the EPA
all data developed by the test that are suggestive or
i
predictive of totficity. In addition to the reporting
requirements1 prescribed in the Good Laboratory Practice
Standards for Physical, Chemical, Persistence, and'
' I
Ecologjical Effects Testing, the rep'orted test data should
include the following:
(A) The source of the dilution water, a, description of
any pretreatment, and the measu ed hardness, acidity,
alkalinity, pH, conductivity, TOC or COD and particulate
matter.
(B) A description of the test chambers, the depth and
volume of solution in the chamber, the specific way the test
was begun (e.g., conditioning, test substance additions),
and for flow-through tests, a description of the test
substance delivery system.
(C) Detailed information about the test fish, including
the scientific name and method of verification, average
18
-------
EG-9
August, 1982
weight (grams, wet weight), standard length, age, source,
history, observed diseases, treatments and mortalities,
acclimation procedures, and food used.
(D) The number of replicates used, the number of
organisms per replicate, the loading rate, and the flow rate
for flow-through tests.
(E) The measured DO, pH and temperature and the
lighting regime.
(F) The solvent used,the test substance concentration1
in the stock solution, the highest solvent concentration in
the test solution and a decription of the solubility
determinations 'in water and solvents if used.
(G) The concentration of the test substance in each
test chamber just before the start of the test and at all
subsequent sampling periods.
(H) The number of dead and live tests organisms, the
percentage of organisms that died, and the number that
showed any abnormal effects in the control and in each test
chamber at each observation period.
(I) The 96-hour LC50, and when sufficient data have
been generated, the 24-, 48-, 72-, and incipient LC50
values, their 95 percent confidence limits, and the methods
used to calculate the LC50 values and their confidence
1imi ts.
19
-------
EG-9
August, 1982
(J) When observed, the observed no effect concentration
(the highest concentration tested at which there were no
mortalities or abnormal behavioral or physiological
effects).
(K) The concentration-response curve at each
observation period for which a LC50 was calculated.
(L) Methods and data records of all chemical analyses
of water quality parameters and test substance
concentrations, including mqthod validations andireagerit
blanks.
20
-------
ES-6
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
FISH ACUTE TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Table of Contents
Subject Page
I. Purpose 1
II. Scientific Aspects 1
Test Procedures 1
Range Finding Test 1
Definitive Test 3
Time-dependent vs. Time-independent Test 4
Static vs. Flow-through Test 6
Length of Exposure 8
Tes t Res u Its 9
Analytical Measurements 10
Water Quality Analysis , 10
Collection of Test Solution Samples 11
Test Substance Measurement 12
Test Conditions 14
Test Species 14
Selection 14
Sources 16
Maintenance of Test Species 18
Age and Condition 18
Care and Handling 20
Acclimation 21
Facilities 22
General 22
Construction Materials 25
Test Substance Delivery System 26
Test Chambers 28
Cleaning of Test System 29
Dilution Water 30
Carriers 37
Environmental Conditions 38
-------
Subject Page
Loading 38
Temperature 41
Light 42
Reporting 42
III. Economic Aspects 45
IV. References 55
11
-------
Office of Toxic Substances SS-6
August, 1982
Technical Support Document for Fish Acute Toxicity Test
I. Purpose
The purpose of this document is to provide the
scientific background and rationale used in the development
of Test Guideline EG-9 which uses rainbow trout, bluegill or
fathead minnow to evaluate the acute toxicity of chemical
substances to fish. The Document provides an account of the
scientific evidence and an explanation of the logic used in
the selecion of the test methodology, procedures and
conditions prescribed in the Test Guideline. Technical
issues and practical considerations are discussed. In
addition', estimates of the cost of conductiing the test are
provided .
11. Scientific Aspects
A. Test Procedures
1. Range Finding Test
A range finding test is recommended for determining the
appropriate concentrations of test substance to use in
performing a definitive acute toxicity test. In the range
finding test, groups of five or more test fish are exposed
to a broad range of concentrations of the substance. Enough
concentrations should be tested such that concentration
lethal to approximately 50% of the organisms can be
ascertained. The number of concentrations will normally
range from 3-6 depending upon the shape of the toxicity
curve for that chemical and prior knowledge of its
approximate toxicity. Only concentrations less than the
solubility limit in water are tested. The exposure period
used in the range finding test can be as short as 24 hours
or as long as 96 hours. If an exposure period less than 96
hours is used, the test substance concentration range
-------
ES-6
August, 1982
selected for the definitive test may have to be adjusted to
make allowances for a greater potential toxicity after 24-
hours of exposure. If no mortalities at a test
concentration equal to the solubility limit are produced, no
additional higher concentrations need to be tested.
2 . Definitive Test
a. General
The results of a definitive test are used to calculate
the 96-hour LC50 and the incipient LC50 when appropriate,
and the concentration-response relationship of the test
substance and the test fish. If the concentrations of test
' ' ' . ' ! I '
substance which produce no effect, a partial kill, and 100
percent mortality have been determined during the range
finding te's t, then five or six test substance concentrations
should be sufficient to estimate the appropriate LC50 values
in a definitive test. In some cases however, to obtain two
partial kills bracketing the 50% mortality level, it may be
necessary to test 8-10 concentrations.
The slope of the concentration-response curve provides
an indication of the range of sensitivity of the test fish
to the test substance and may allow estimations of lower
concentrations that will affect the test organisms. For
example if the slope of the concentration-response curve is
very steep, than a slight increase in concentration of the
test substance will affect a much greater portion of the
test fish than would a similar increase if the slope of the
curve was very shallow. The slope of the concentration-
response curve reveals the extent of sensitivity of the test
fish over a range of concentrations.
The exposure of a minimum of 20 fish, divided into two
-------
ES-6
August,.1982
or more replicate groups , to each test substance
concentration is required in the guideline. That minimum is
based on an optimum number of test fish needed for
statistical confidence, equipment requirements, and
practical considerations of handling the test organisms.
At least two replicates should be included in order to
demonstrate the level of pecision in the data and indicate
the significance of variations. Test chambers holding
replicate groups should have no water connections between
them. The distribution of test fish to the test chambers
should.be randomized to prevent bias from being introduced
' ' I I i'l
into the test results.
Fish should not be fed during the test for two reasons.
i
Firsit, fecal matter which may accumulate can result in a
decrease in the dissolved oxygen concentration in the test
cha'mber. Second, some test substances can physically bind
to1 the uneaten food or fecal matter, thus making a portion
oC the test substance unavailable for uptake by the fish.
An occurrence of either of these conditions could produce
unreliable test data.
b. Time-dependent vs. Time-independent
In time-dependent tests, fish are continuously exposed
to a series of concentrations for a specified period of
time, usually 96-hours, at which time the LC50 is
calculated. In time-independent tests (TI's), fish are
continously exposed to a series of test concentrations until
such a time when no additional mortalities are expected to
result from continued exposure. The LC50 calculated at this
time is termed the incipient LC50. This same value has also
been termed the ultimate median tolerance limit, the lethal
-------
ES-6
August, 1982
threshold concentration or the asymptotic LC50 by many
researchers. Sprague (1969) has stated that the incipient
LC50 is the "most useful single criterion of toxicity." In
a review of 375 toxicity tests by the author, a lethal
threshold clearly had not been reached in 42 tests while in
122 tests th threshold was reached in four days or
longer. Eat n (1970) proposed the use of TI's as the
resulting 50% effect concentration is dependent upon the
response of the test fish and not based on an arbitrary time
period.
Use of TI's greatly increases one's ability to evaluate
I . • , ! ' I ' : ' 1 "
the chronicity of compounds as discussed by Tucker and
Leitzke (1979). The chronicity of a compound, or the degree
to which a compound effects additional mortalities over a
prolonged period of time, is assessed by comparing LCSO's
over time, e.g. 96/48 hour LC50 or 10 day incipient LC50/96
hour LC50. Those compounds with relatively high ratios
(i.e. >0.5) would not be expected to cause chronic
effects. If only 96-hour toxicity tests were performed, the
toxicity of those compounds whose mode of action requires at
least 3-4 days to begin to express toxicity would be grossly
underes t imated.
It should be remembered that for compounds that do not
express chronicity, i.e. there is little additional
mortality during the last 48 hours of the test, the test
will be a 96-hour toxicity test. Consequently, the more
costly testing for the estimation1 of an incipient LC50 will
be performed only for compounds for which additional testing
will probably be performed during the hazard or risk
assessment process.
-------
ES-6
August, 1982
c. Static vs. Flow-through
In the static system, the test substance and dilution
water are not changed during the 96-hour exposure period.
In promulgating test rules, the Agency will prescribe the
use of a static or flow-through system, or both, for fish
acute toxicity tests for developing, cost effectively, those
data essential to assessing the risk to the environment of a
given chemical.
Each of these methods of exposing fish to a test
substance offers certain advantages and presents certain
disadvantages not shared by the other. Theistatic exoosure
.•I ' ' !|! ! i I !
system requires less equipment and set up time, and
therefore is a less expensive test.. On the other hand, in a
I
flow-through system, loss of the test substance due to
uptake by the fish, degradation, or to volatilization is
minimized, and metabolic products toxic to the test
organisms (e.g. ,ammonia) do not build up. The concentration
i i
of dissolved oxygen in the test chamber can also be
maintained above the level that might stress the fish-.
Because of these features, the Agency will specify the
use of the flow-through method in testing the toxicity of
chemicals which volatilize or degrade rapidly, which reduce
the dissolved oxygen concentration within the test chamber,
or which are taken up by the test organisms at a rate that
significantly lowers the concentration of the test substance
within the test chamber. In additon to developing data
needed to determine the 96-hour LC50 and the concentration-
response curve for such test substances, a flow-through
exposure may be continued to get information on the
potential chronicity of the compound. By design, static
-------
E3-6
August, 1982
tests can not be continued past 96 hours and still yield
reliable results.
In a static system, since the test substance is not
renewed, the concentration of the chemical to which test
fish are exposed does not remain constant if the chemical
adheres to the test chambei walls, vaporizes readily,
degrades rapidly, or is re dily taken up by the test fish.
The static method of exposure however, can be used' to
develop toxicity data for those substances which are not
subject to a significant reduction in concentration during
the exposure period. Static toxicity ; data in combination
I ^ ! i i | | i ,
with data developed through the use of a flow-through test
can also be used to detect and evaluate the toxicity of
metabolities and degradation products. If for instance, the
96-hour LC50 from a static test is less than that from a 96
hour flow-through test, it can be ass'.umed that more toxic
metabolites or degradation products were formed during the
static test,
The Agency forsees a need for both flow-through and
static test methods and each method will be considered in
the developement of a test rule. The chemical nature of the
test substance, its use and the nature of its release into
the environment ( e .g . ,continuous or intermittent) will be
considered. The Agency does not, however, assume that data
developed through the use of one of these test methods can
substitute for data developed through the use of the other,
since evidence exists in the literature to show1 that the
toxicity of some test substances for test organisms may be
10 times greater in flow-through than in static exposures
(Mauck et al. 1976).
-------
ES-6
August, 1982
d. Length of Exposure
The rainiraium exposure time of 96 hours in the fish acute
toxicity test standard is specified in order to permit a
comparison of data developed through the use of this test
guideline with the large base of acute toxicity data in the
published literature. The 96-hour exposure period is
advocated by various groups concerned with establishing
uniformity in testing methodology (APHA 1975, ASTM 1980,
Committee on Methods... 1975). Most of the recently
published data on the acute toxicity of chemicals to
freshwater fish were developed using. a96-hour exposure ,
period (Brugns et al . 1977, Me Kim et al. 1976, Soehar et -al .
1979, 1980). The use of the 96-hour exposure period was
proposed initially in 1951 by am aquatic bioassay committee
, (Doudoroff et al. 1951) and was selected, in large part, as
a -natter of convenience since i't is easily scheduled within
the five-day work week. Only when there are indications of
i i
chronicity during a 96-hour test will tne test period be
extended. The previously cited studies indicate that this
is not a frequent occurrence.
3. Test Results
While death is the primary endpoint in these tests, any
behavioral or physiological changes in the fish such as
erratic swimming, lost of reflexes, increased excitability,
lethargy, discoloration, excessive mucous production,
hyperventilation, opaque eyes, curved spine, hemorrhaging or
any other observed effects should be recorded.
Quantification of such observations at test substance
concentrations not causing lethality are useful in
identifying and assessing potential chronic lethal effects
-------
ES-6
August, 1982
at these lower test concentrations.
Mortality data at 24, 48, 72 and 96 hours and at
termination if the test exceeds 96 hours, should be
presented for each time interval. Whenever sufficient data
exists at these.time intervals,an LC50 and 95% confidence
interval should be calculated and a graph of percentage
mortality - concentration prepared. When more .than one
LCSO's at various time intervals should be prepared.
The recommended methods of LC50 calculation include the
probit, logit, binominal and moving average angle methods.
The data obtained Erom each test will determine which method
is the most appropriate for that data':$et.
4. Analytical Measurements
a . Water Quality Analysis
Measurement of certain water quality parameters of the
dilution water such as hardness, parti'culate matter,
alkalinity, acidity, conductivity, TOG, and pH is important.
Quantification of these parameters at the beginning and end
of the exposure period of flow-through tests is necessary in
order to determine if the water quality varied during the
test. If significant variation occurs, the resulting data
should be interpreted in light of the estimated toxicity
values.
In Static Systems the dissolved oxygen concentration and
pH should be measured in each test chamber at the beginning
of the test just prior to addition of the fish and then as
often as needed to document any subsequent changes from the
initial levels .
In Flow-through the dissolved oxygen concetration (DO)
and pH of the test solution in each chamber should be
-------
ES-6
August, 19'82
determined at the beginning of the test and every 48 hours
thereafter until the end of the test. A decrease in DO
indicates that the flow rate should be increased.
b. Collection of Test Solution Samples
The objective of the recommended sampling procedure is
to obtain a representative sample of the test solution for
use in measuring the concentration of the test substance.
Although there is mixing in the test chamber, especially in
flow-through tests, material can concentrate near the sides
and bottom of the chamber due to physical or chemical
properties | ;of the substance, or to interactions with1 organic
material associated with the test animals. For this reason,
water samples should be taken near the center of the test
chamber. The handling and storage of the samples requires
care to prevent the loss of the test substance from the
sample before analysis.
c . Test Substance measurement
In Static Systems the concentration of dissolved test
substance should be measured in each test chamber at least
at the beginning and end of each test. If the reduction in
test s ubs tance .concentration exceeds 50%, the test should be
repeated at a lower loading rate, or a flow-through test
should be performed.
In Flow-through Systems the test substance concentration
should be determined in each test chamber at 0 and 96 hours
and if the test continues past 96 hours, at least every four
days up to and including the day of termination. To further
assess and quantify any possible changes in test substance
concentration, whenever a malfunction of the toxicant
delivery is detected, all potentially affected test chambers
-------
ES-6
August, 1982
should be sampled at that time.
If the measured concentrations of dissolved test
substance are 50 percent more or less than the nominal
concentrations, steps should be taken to determine the cause
for this deviation. A sample of the stock solution as well
as samples of the influent to various test chambers should
be analyzed to determine if the reduction in test substance
occurs prior to delivery of the test solution to the
aquaria. If results of these analyses indicate that the
proper amounts of test substance are entering the test
chambers, then the total test substance concentration should
be measured in at least the chambers containing the highest
test substance concentration. These data will give
indications if the difference between nominal and measured
test concentrations is due to volatilization or degradation
of the test substance, or to insolubility of the test
substance 'in the dilution water.
If the toxicant delivery system has been properly
calibrated and the fish randomly introduced into each test
chamber, the measured differences between replicates at each
concentration should be less than 20%. If the differences
exceed this, the test should be repeated.
The concentrations of test substance measured after
initiation should be within 30% of the concentrations
measured prior to introduction of the fish. If the
difference exceeds this, the test should be repeated using a
higher flow rate.
Use of reliable and validated analytical techniques and
methods is essential to the usefulness of the test data in
assessing the environmental hazard of the chemical.
10
-------
ES-S
August, 1982
Significant variation in the measured concentrations lessens
the value of the toxicity data generated.
B, Test Conditions
1. Test Species
a. Selection
The species of fish selected for use in this test
guideline are the rainbow trout, bluegill and fathead
minnow. Adelman and Smith (1976) listed the following as
the important criteria to use in the selection of a
"standard fish" for bioassays : 1) relatively constant
response to a broad range of toxicants when tested under
similar conditions 2) available in large quantities with
close quality control 3) eas.ily handled for bioassay
purposes 4) easily transported 5) continuous availability of
the desired size and 6) capable of successful completion of
a life cycle in 1 year or less. These species meet all
criteria but the last; only the fathead minnow can complete
a full life cycle in less than one year.
The main reason why these three species were selected is
because there is a very large toxicity data base with each,
and all three are readily available and require little
expertise in maintaining healthy populations. All three are
widely distributed in the United States, and are either
ecologically or economically important (Scott and Grossman
1973, Kitchell et al. 1979).
Studies on relative sensitivity of the three species
have been performed and indicate that rainbow trout are
generally the most sensitive and fathead minnow the least
sensitive to a variety of test substances. Kenage (1979)
compiled LC50 data on 20 pesticides with all three
11
-------
ES-6
August, 1982
species. With 12 of these compounds, the trout was the most
sensitive and the fathead minnow the least. In five cases
the bluegill was the most sensitive and in one case the
fathead minnow was the most sensitive. The LCSO's were
similar between the three species for the remaining two
compounds. The LCSO's for the fathead minnows were
generally 6x those for the trout and 2x those for
bluegill. As it was not stated if the tests were performed
under comparable conditions these values only approximate
relative sensitives of the species.
Nevins and Johnson (1978) tested three phosphate ester
mixtures with all three Species of fish undler identical
static conditions and two compounds under flow-through
conditions. In all five cases, rainbow trout were the most
sensitive-, but only in 2 cases were the fathead minnows the
least sensitive. ,
Folmar et al. (1979) performed static acute toxicity
tests with technical grade glyphosate, the formulated
pesticide Roundup® surfactant with the above 3 fish
species. In these tests however, the fathead minnow was the
mos t sens itive .
From these data it is clear that it can not be assumed
that rainbow trout will be the most sensitive species.
Although fathead minnows are generally the least sensitive
of the three species, their small size, ease of culture, and
short life cycle make them the easiest to work with. Their
extensive use in early life stage testing and full-chronic
testing (Macek and Sleight 1977, McKim 1977) adds to the
importance of their role is aquatic toxicology programs.
Although rainbow trout can also be used in early life stage
12
-------
ES-6
August, 1982
and full or potential chronic toxicity testing, the- expense
is only occasionally warranted.
b. Sources
All three species are readily available at the
appropriate test sizes from commercial fish suppliers.
There are some suppliers that now specialize in culturing
fish under controlled conditions just for toxicity1
testing. Rainbow trout can be purchased and readily shipped
to researchers as either eyed eggs or as finger lings already
the appropriate size for'testing. Trout should oe purchased
only from suppliers that have been state-certified to raise
disease-free' f ish. ; '• • [
There are many suppliers of bluegill throughout the
country that will readily air-freight fish. As t'hese fish
are not amenable to artificial fertilization or spawning in
the laboratory, figerlings are normally shipped.
Researchers need to carefully select their suppliers, as
many are known to have little concern for providing disease
and parasite-free fish.
Fathead minnows can be purchased as eggs or juveniles,
or cultured in the laboratory in a brood unit (U.S. EPA
1971). In light of past problems with the health of fish
received from some suppliers, it is recommended that
researchers rear their own fatheads for toxicity testing.
Whenever fish are to be used for a test or a set of
tests, all fish used for that test should be from the same
source and held under similar conditions prior to testing to
minimize variability. Alexander and Clarke (1978) performed
toxicity tests with 2 strains of rainbow trout exposed to 13
ing/1 dodecylsodium sulfate. The median survival time of the
13
-------
ES-6
August, 1982
Idaho strain exposed to phenol was significantly less than
the median survival time of the Nisqually strain. Response
times for the other compounds were similar between
strains. In a second set of tests with jus.t phenol, no
difference in LC50 values was found between Idaho and
Nisqually strains, but a significant difference was found
between Idaho and Manx strains.
2. Maintenance of Test Species
a. Age and Condition
The age, and consequently the size of rainbow trout,
bluegill, and fathead minnows was selected based on the ease
I : '' I ' !
of handling and testing fish of this size. All'fish used in
the same test s'hould be as similar in size as possible to
limit the effects due to size differences.
The health and condition of fish used in acute toxicity
tests is an important consideration. Diseased or stressed
fish may increase the sensitivity of the fish to the
toxicant. Iwama and Greer (1980) performed 96-hour acute
toxicity tests with Coho Salmon (Oncorhynchus kisutch) that
had been exposed to, and contracted, a mild state of
bacterial kidney disease (-Conynebacterium salmoninus). When
diseased fish were exposed to pentachlorophenate, the
estimated LC50 was 39 ug/1, significantly lower than.the
LC50 of 65 ug/1 for healthy fish.
Prior exposure to contaminants may also effect the
response of test fish to a toxicant. Bills et at. (1977)
performed acute toxicity tests with several compounds using
rainbow trout that had previously been exposed to PCB's
(Aroclor 1254). Previously exposed trout that had body
burdens of 3.4 ug/g of Aroclor 1254 had significantly
lower LCSO's when exposed to two of the test compounds,
14
-------
' ES-6
August, 1982
cyanide and chromium. Trout with body burdens of only 0-46
ug/1 of PCB's were .more sensitive to cyanide with a LC50 of
66 ug/1 versus 90 ug/1 for "clean fish."
Alexander and Clarke (1978) performed static acute
toxicity tests with rainbow trout that had been exposed to
40 ug/1 of total residual chlorine for 24 hours immediately
prior to testing with phenol. Trout previously exposed to
chlorine had a 48 hour LC50 of -7.7 mg/1 for phenol,
significantly lower than the LC50 of 10.1 mg/1 for non
chlorine- exposed trout.
Based on these data it is recommended that all fish used
in aquatic toxicity testing contain no more than 0.5 ug/1
i I : ' ' '
PC3, and not be1 exposed to any contaminants during holding.
b. Care and Handling
Upon arrival at the laboratory, fish or eggs should
immediately be cared for to prevent additional stress from
crowding during transport. The test organisms should be
gradually transferred to the holding water at the testing
facility as soon as possible.
Alexander and Clarke (1978) performed acute toxicity
tests with rainbow trout and five different potential
reference toxicants to determine what effects starvation,
changes in temperature, and crowding would have on the
median survival time (MST) for each toxicant. Trout that
were starved for 15-21 days before separate exposures to
phenol and sodium pentachlorophenate had significantly lower
MST's than fed fish. There were no differences in MST's
among fish exposed to sodium azide, copper sulfate, or
dodecylsodium sulfate. When trout that had been held at
10°C were subjected to temperature decreases of 1-5°C over
15
-------
ES-6
August, 1982
24 hours and then immediately transferred to test chambers
at 15°C, only fish exposed to phenol had a significantly
lowe MST than fish gradually acclimated to 15°C.
Temperature stress did not alter the MST for the remaining
four compounds.
To determine the potential effects of crowding, one
group of trout was held at a high density of 3.3 g/1 day,
compared to the normal holding density of 0.6 g/1 day- Only
fish exposed to sodium azide and sodium pentachlorophenate
had significantly lower MST's due to crowding.
In subsequent toxicity tests with just phenol,'the
authors determined that trout with holding mortalities as
high 'as 7-9% and 18% had LC'50's similar to those: generated
with trout with only a 1-2% holding mortality.
Although the above research does not present conclusive
evidence on the role of feeding, crowding, and temperature
changes during holding, it does demonstrate that at least
for some test substances these variables should be
controlled and optimized to prevent possible differences in
test fish response to toxicant exposure.
c. Acelimation
Brauhn and Schoettger (1975) found .that fish that had
become accustomed to unrestricted swimming in rearing ponds
underwent intense competition for food and swimming space
when placed in confined holding tanks. Rainbow trout
appeared to be less affected by restricted space than
bluegill and fathead minnows. The authors also recommended
that fish be maintained in holding tanks with color
backgrounds and light intensitites similar to those in the
testing area to prevent additional stress when transferred
16
-------
ES-6
August, 1982
to testing chambers.
Although the importance of temperature acclimation to
the test temperature is still unclear, a maximum gradual
change of 3°C/day is recommended at this time. Peterson and
Anderson (1969) concluded that complete acclimation to
temperature, based on changes in locomotor activity and
oxygen consumption, requires approximately two. weeks before
metabolism is back to normal. They also determined that the
rate of change was more important than the amount of change.
Changes in the hardness of water to which the fish are
exposed should also be controlled. Lloyd (1965) determined
that trout transferred from hard to soft water needed at
; i , i . | I i i ,
leas't 5 ' days of acclimation to the soft water before their
response to a toxic metal was the same as the response of
fish continually held in soft water.
3. Facilities
a. General
Facilities needed to perform this test include: (1)
flow-through tanks-for holding and acclimating fish, (2) a
mechanism for controlling and maintaining the water
temperature during the holding, acclimation, and test
periods, (3) apparatus for straining particulate matter,
removing gas bubbles, or aerating the water when water
supplies contain particulate matter, gas bubbles, or
insufficient dissolved oxygen, respectively, (4) an
apparatus for providing a 16-hour light and 8-hour dark
photoperiod with a 15- to 30-minute transition period, (5)
chambers for exposing test fish to the test substance, and
(6) a test substance delivery system.
17
-------
ES-6
August, 1982
Flow-through tanks, into which a continuous or
intermi'ttant flow of water occurs, should be used for
holding and .acclimating test fish. The renewal of the- water
in flow-through tanks minimizes the accumulation of
metabolic products such as ammonia. The build up of organic
matter within the tanks might provide a nutrient source for
bacteria present in the water. Bacteria using oxygen to
metabolize and decompose the organic matter in the tank
could then reduce the dissolved oxygen concentration of
water. Decreased dissolved oxygen concentrations as well as
the accumulation of ammonia could increase'the likelihood of
disease in the test fish (Brauhn and Schoettger 1975). The
'use of diseased fish in acute toxicity tests could result in
the development of inaccurate and unreliable data.
The effects of sudden temperature changes on fish may
range from death to temporary impairment of physiological
functions, depending on the acclimation temperature, the
magnitude of the temperature change, the temperature
tolerance of the species, and the circumstances and duration
of the exposure. To avoid any undue stress, accurate
temperature control devices should be "used to both maintain
constant temperatures, and to gradually increase or decrease
the temperature during acclimation procedures. Such
mechanisms have been described by DeFoe (1977) and Lemke and
Dawson (1970).
Particulate matter and gas bubbles, if present in the
dilution water, may clog the toxicant delivery system used
in flow-through tests. Gas bubbles also may cause excessive
loss of volatile test substances. Either circumstance may
alter the concentration of test substance to which the test
18
-------
E3-6
August, 1982
fish are exposed. To avoid this problem an apparatus
capable of removing particulate matter or gas bubbles from
the dilution water may be required. If the dilution water
is heated prior to use, it may also be necessary to de-
saturate the water from >100% of oxygen saturation. Penrose
and Squires (1976) describe a suitable apparatus for tnis.
An adequate supply of dissolved oxygen should be
available to the fish. To facilitate this, the dilution
water or holding water should be at 90- 100% of oxygen
saturation prior to delivery to the holding tanks or test
system.
The duration, and intensity of light ar.e environmental
1 , i i :
variables which could possibly influence the results of
acute toxicity tests. Any possible variations in test data
due to differences in light conditions can be minimized by
using uniform light conditions during testing. A device
capable of regulating photoperiods and the transitions from
light to darkness and darkness to light has been described
by Drummond and Dawson (1979).
b . Construction Materials
Due to the toxicity of many heavy metals at low
concentrations (U.S. EPA 1976) and the ability of metal
pipe, galvanized sheeting, laboratory equipment, etc. to
leach metals into water, no metal other than stainless steel
(preferably #316) should be used. In the same manner, un-
aged plasticized plastic (Tygon® tubing) should not be used
due to the high toxicity of a main component, di-2-ethyl
hexyl phthalate (Mayer and Sanders 1973) and the abilitiy of
DEHP to leach into aquaria systems from plastics (Carmignani
and Bennett 1976). To avoid any possible stress due to
19
-------
ES-6
August, 1982
exposure to low levels of metals, phthalates, and other
potential contaminants, #316 stainless steel, glass and
perfluorocarbon plastics (e.g. Teflon©) should be used
whenever possible and economically feasible. If other-
materials should be used, conditioning to a continuous flow
of heated dilution water should be performed for a minimum
of 48 hours.
c. Test Substance Delivery System
To maximize the accuracy and precision of test results
developed through the use of this test guideline, the
quantity of test substance introduced by the test substance
delivery system should be as constant as possible from one
addition of test substance to the'next. Fluctuations in the
quantity of test substance introduced into the test chamber
may result in abnormally high or low response value (e.g.
LCSO's) of the test organisms and in a wider spread of
response values in replicate tests. The greater the
variation in the quantity of test substance introduced, the
greater the potential for abnormalities and spread of the
response values.
Variations in the quantity of dilution water entering
the test chambers during a given time interval may also
create unders irable differences in test conditions between
test chambers. The concentrations of dissolved oxyen and
test substance in a test chamber, for example, may decrease
more rapidly in chambers having lower flow rates.
Differences between test chambers in the concentration of
dissolved oxygen, test substance, metabolic products and
degradation products, individually or in combination may
result in response values for the test organisms which are
i naccurate.
20
-------
ES-6
August, 1982
Any one of several toxicant delivery devices can be used
as long as it has been shown to be accurate and reliable.
Various modifications of proportional diluters have been
used by Auwarter (1977), DeFoe (1975), Mount and Brungs
(1976), and Ozburn and Smith (1977). A manual for their
construction and operation has been prepared by Lemke et al.
(1978). A metering pump system has also been used .by
Chandler and Partridge (1975), as have saturator systems
(Krugel et al. 1978, Veith and Corns tock 1975).
The following criteria presented "by Hodson (1979) should
be considered when selecting or designing a toxicant
delivery system; 1) the delivery of the toxicant should s too
j ! . j I • i • j , ,
if delivery of the dilution water stops, 2) it should be '
consistent in delivery amounts 'throughout the test period,
3) independent of electrical failure, 4) independent of
temperature and humidity fluctuations, 5) capable of
delivering small quantities, 6) easy to construct with few
moving parts and 7) easy to operate.
The solubility of the test compound should also be taken
into account in selecting an appropriate delivery system.
If the compound can be solubilized in water, a device
capable of delivering amounts of test solution greater than
1 millilter (ml) will probably be needed. If a carrier
should be used, a system capable of accurately delivering
small amounts, less than 100 microliters (ul), will probably
be required to minimize the carrier concentration in the
test solution.
Each system should be calibrated prior to starting the
test to verify that the correct proportion of test substance
to dilution water is delivered to the appropriate test
chambers .
21
-------
ES-6
August, 1982
d. Test Chambers
Test chambers should be constructed of stainless steel,
perfluorocarbon plastic, or glass. Stainless steel should
be welded, and the glass bonded with silicone adhesive. If
adhesive is used, the amount exposed to the test solution
should be minimized to limit the sorption of test
materials. The size, shape and depth of the test chambers
are not imporant, as long as the volume accomodates the
loading requirements. The chamber however, should be
sufficiently large and contain enough water such that the
fish are not stressed due to crowding.
' e> Cleaning of' the Test System
Before use, test systems should be cleaned to remove
dust, dirt, and any other debris or residues that may remain
from previous use of the system. Any ^ of these substances
may affect the results of a test by sorption of test
materials or by exerting an adverse effect on test
organisms. New chambers should be cleaned to remove any
dirt or chemical residues remaining from manufacture or
accumulated during storage. Detergent is used to remove
hydrophobia or lipid-like substances. Acetone is used for
the same purpose and to remove any detergent residues. It
is important to use pesticide-free acetone to prevent the
contamination of the chambers with pesticides which
influence the outcome of the test. Nitric acid is used to
clean metal residues from the system. A final thorough
rinse with water washes away the nitric acid residues. At
the end of a test, test systems should be washed in
preparation for the next test. It is easier to clean the
equipment before chemical residues and organic matter become
22
-------
ES-6
August, 1982
embedded or absorbed into the equipment.
Conditioning the flow-through system with dilution water
before it is used in a' test allows an equilibrium to be
established between any substance in the water and the
materials of the test system. Since the test system may
sorb or react with substances in the dilution water,
allowing this equilibrium to become established before the
test begins lessens the chances of changes in water
chemistry occurring during at test. Even after extensive
washing, new facilities still may contain toxic residues.
The best way to determine if toxic residues remain is to
test for their presence by maintaining or rearing the test
fish species in the facility for a period of time equal to
or exceeding the time required to complete a test.
f. Dilution Water
A constant supply of good quality dilution water is
needed to maintain consistent experimental conditions during
testing. A change in water quality during a test may alter
the response of the test fish to the test solution. Most
research on the effects of water quality have centered
around the effects of changes in pH and total hardness on
the acute and subacute toxicity of compounds. Mauck et al .
(1977) performed static, acute tests with bluegill and
Mexacarbate at various pH's. They observed that Mexacarbate
was 38 times more toxic at a pH of 9.5 and 5 times more
toxic at a pH of 8.5 than at a pH 7.5. They ascertained
however that these large increases in toxicity were mostly
caused by the rapid hydrolization of the parent compound to
more toxic breakdown products, and not to an increase in
sensitivity of the fish at the higher pH levels.
23
-------
ES-6
August, 1982
In another study of the effects of pH, Mauck et al.
(1976) found that the acute toxicity of five pyrethroids did
not change as the pH of the dilution water was increased
from 7.5 to 9.5. The toxicity of pyrethrum extract however
decreased as the LC50 increased.from 41 rag/1 at a pH of 6.5
to (37 mg/1 at a pH of 9.5.
In a study with the formulated herbicide, Roundup®,
Folmar et al. (1979) determined that the herbicide was five
times more toxic to rainbow trout as the pH increased ' from
6.5 to 7.5. Additional increases to pH's of 8.5 and 9.5 did
not further increase the toxicity. When bluegill were
similarly tested, there was only a 2 fold increase in
I • : • I i
toxicity between a pH of 6.5 and 7.5.
When the toxicity of nitrite was tested at different pH
levels, an inverse trend relationship was observed; toxicity
decreased with increasing pH (Wedemeyer aad Yasutake
1978). In static tests with steelhead trout (Salmo
gairdneri) the toxicity decreased 8-fqld for 5 g fish and 3-
fold for 10g fish when the pH was increased from 6.0 to 8.0.
To estimate the potential chronic effects of reduced pH
on freshwater fishes, chronic toxicity tests were performed
with the fathead minnow by Mount (1973) and with the brook
trout (Salvelinus fontinalis) by Menendez (1976). The
results of both studies were similar; hatchability of eggs
was reduced at pH levels <6.5. Both authors recommended
that the pH of water should be above 6.5 to fully support
the growth and reproduction of these fishes.
Much work has been performed studying the ameliorating
effects of increased hardness on the toxicity of heavy
metals to freshwater fish (Carrol et al. 1979, Hoi combe and
24
-------
SS-6
August, 1982
Andrew 1978, Howarth and Sorag.ue et al. 1976), but much less
has been done with organic compounds. In the study by Mauck
et al. (1977) the toxicity of Mexacarbate to brown trout
(Sal mo trutta) sac fry and Coho salmon (Oncorhynchus
kisutch) finger lings did not change when the hardness of the
dilution water was increased 'from 40-48 mg/1 to 160-180
mg/1.
Mauck et al . (1976) found only slight variations in LC50
values when they performed static, acute, toxicity tests
with bluegill and five pyrethroids at hardnesses of 10-13,
40-4.8, 160-180, and 280-320 mg/1. The LC50 of pyrethrum was
however significantly reduced from 62 to 46.5 mg/1 as the
hardness was1 increased from 10-13 to 281J-320 mg/1.
Wedemeyer and Yasutake (1978) found that the toxicity of
nitrite to 5g steelhead trout decreased 24 times as the
hardness was increased from 25 to 300 mg/1.
Although the reported data demonstrate that relatively
large differences in the pH and hardness of the dilution
water (> 2x) can effect the toxicity of a compound, it is
not known what role, if any, small changes or even large
gradual changes (> 2x) will have on the acute toxicity of
compounds.
Brungs et al . (1976) performed a fathead minnow chronic
toxicity test with copper using water collected downstream
from a sewage treatment plant as the dilution water.
Throughout testing, the hardness varied from 88-352 mg/1,
akalinity from 50 to 248 mg/1, pH from 7.5 to 8.5, DO from
5.0 to 13.0 and temperature from 0 to 30°C. When results
from this test were compared to the results of a similar
chronic test performed with constant quality dilution water
25
-------
ES-6
August, 1982
(Mount 1978), it was shown that the variations in water
quality had little or no effect on the toxicity of copper.
McLeay et al. (1979) performed static acute toxicity
tests with rainbow trout and a pulp and paper-mill effluent
using 1Q>-different dilution waters. Dilution water hardness
( |
ranged from 5 to 400 mg/1 , pH from 6.4 to 8.4, conductivity
from 15 to 778 umhos/cm, and alkalinity from 11 to 392
mg/1. The 24 hour LC50 values using the 10 waters ranged
from 4.4 to 15.6% effluent and was pH related. After
adjusting the pH of each dilution water to 6.5, the
variation in the LCSO's was reduced to a range of 4.4 to
6.'9% effluent, indicating little effect due to th'e other
measured and non-measured dilution water characteristics.
Mattson et al . (1976) performed static, acute toxicity
tests with five organic compounds and fathead minnows in
Lake Superior water and in reconstituted soft water and
found no differences. Although no data on measured water
quality paramaters of each dilution water were presented, the
similarity of data from tests done in two obviously
different dilution waters is noteworthy.
Although there is little data demonstrating that changes
in the quality of the dilution water during testing will
affect the test results, the dilution water should be kept
as constant as possible during testing to minimize such a
risk.
A dependable source of clean surface or well water
usually will provide water having greater consistency in its
chemical makeup than water from a municipal water supply.
Municipal water may originate from several sources which
differ in chemical makeup. In addition, municipal water
26
-------
ES-6
August, 1982
frequently is treated chemically as part of a purification
process. Since the proportions in which water from
different sources are mixed, and since the chemical
treatment given water during the purification process may be
different from time to time, the chemical makeup of
municipal water may vary considerably. Reconstituted water,
while theoretically more consistent from batch to batch than
either surface or ground water or municipal water, may in
some instanc.es lack trace minerals required by some species
oi: fish. Cairns (1969) performed many acute toxicity tests
on some compounds with both reconstituted water and natural
water and found that the data generated: from the tes ts: in
natural water were not consistent or reproducible whereas
the results from the tests with reconstituted water were
consistent. Of more concern, however, is the prohibitive
expense of continuously preparing reconstituted water for
use in fish-holding and flow-through toxicity tests.
Fish culturists do not know all of the conditions
required to maintain healthy fish, nor do they know all of
the components and combination of components in water that
adversely affect the health of fish (Brauhn and Schoettger
1975). Nevertheless, to avoid possible inconsistencies and
inaccuracies in test results, healthy fish are needed for
use in toxicity tests. There is, therefore, a need to
determine that the dilution water, whatever its source, is
able to support the fish species to be used in a healthy
condition for the duration of the holding and testing
periods.
An appropriate way to make that determination is to
place young fish of a sensitive species, preferably the one
27
-------
ES-6
August, 1982
to be used in subsequent tests, in the dilution water for an
extended period of time and observe their behavior, growth
and development. Ideally, those observations should be made
by an experienced fish culturist familiar with certain
stress reactions which are difficult for an untrained
observer to identify.
Surface and ground water may vary cons iderably in their
chemistry depending upon the season of the year and
precipitation patterns. Variations in the chemistry of
surface water may involve the quantity of "particulate
matter, dissolved organic and inorganic chemicals, un-
ionized ammonia, residual chlorine' and vaclou's other
contaminants. As an indication of uniformity of the
dilution water used in the toxicity tests, it Is recommended
in the guideline that certain water chemistry parameters be
measured at least twice a year, or more frequently if it is
suspected that one or more of those parameters has changed
significantly The water chemistry parameters singled out
and the maximum acceptable concentrations listed for these
parameters are among those generally accepted as substances
and concentrations which do not adversely affect freshwater
fish (APHA 1975, ASTM 1980). Recognizing that some
variation in water chemistry is normal in natural surface
waters/ a 10 percent fluctuation from month to month in
water hardness, akalinity, and conductivity, and a variance
of 0.4 pH units is accepted as suitable.
g. Carriers
A carrier may be used to aid in the dissolution of a
test compound into dilution water only after significant
efforts to dissolve it in dilution water of dilution water
28
-------
ES-6
August, 1982
stocks have failed. Schoor (1975) believes that the use of
a carrier may interfere with the uptake of the test compound
by the test organisms'; if the carrier molecules affect the
adsorption of the test compound at the gill surface, there
will be a resultant change in the rate of transport into the
test organism. The author also states that the use of a
carrier may increase the concentration of compound in the
test solution above solubility by creating a stable water
emu Is ion.
When a carrier is required, triethylene glycol (TEG),
dimethyl formamide (DMF) or acetone may be used. The
solvents should be tried in the order stated due to their
relative toxicity to fathead minnows as reported by Cardwell
et al . (manusc rioi-. 1980). The minimum amount should be used
and the concentration of TEG should not exceed 80 mg/1, the
MATC (maximum acceptable toxicant concentration) value.
Concentrations of DMF and acetone should not exceed 5.0
mg/1, the MATC for DMF. Although there ,is no MATC value for
acetone, its acute toxicity is similar to that of DMF.
Ethanol should not be used due to its tendancy to
stimulate the excessive growth of bacteria in the test
chambers .
4. Environmental Conditions
a. Loading
In the static tests, the loading should not be so high
to deplete the dissolved oxygen or result in significant
depletion of the toxicant due to uptake of the chemical by
the fish. A maximum loading of 0.5 g/1 will generally be
sufficient for compounds that do not have a high
bioconcentration potential, or are not likely to reduce the
29
-------
ES-6
August, 1932
dissolved oxygen concentration due to degradation. For
other chemicals, flow-through tests should be performed.
A maximum loading rate of 0.5 g/1 day flow-through tests
should be sufficient to maintain proper dissolved oxygen and
ammonia concentrations, and the prop^-f concentration of the
i ,
test chemical in the test solution. ',Ji
In a flow-through study by Blanchard et al. (1977) a
loading of 1.9 g/1 day was not sufficient to prevent loss of
14 C-sec-butyl-4-chlorodiphenyloxide from the test water.
The concentration of test substance decreased more than 50%
during the first 12 hours of exposure and,did not return to
I i I ! i i ' ' | ' .)
the expected, concentration until after 72 hours. '
In flow-through .studies with 2 strains of rainbow trout,
Alexander and Clarke (1978) tested phenol at three different
loading rates of 0.7, 1.4, and 2.6 g/1 day and found no
significant differences in MST's between the three rates Eor
each strain. These data indicate that at least for 'phenol
loading up to 2.6 g/1 day i: not an important factor.
b. Dissolved Oxygen
The level of dissolved oxygen maintained in a test
chamber can influence the sensitivity of test organisms to a
test substance. Increased acute toxicity of hydrogen,
cyanide was observed in various fish species with the
dissolved oxygen concentration was below 5 mg/1 or
approximatly 60% saturation at 25°C (Smith et al. 1978).
Fathead minnow growth was inhibited at a dissolved oxygen
concentration between 5.0 mg/1 and 7.3 mg/1 at a temperature
range of 15-2 5°C, equivalent to approximately 65% saturation
(Brungs 1971).
30
-------
ES-6
August, 1982
In order to provide a minimum margin of safety in acute
toxicity tests, a dissolved oxygen concentration greater
than 50% of saturation is recommended as a minimum.
c. Temperature
Since fish are poikilothermic, nearly all their
biochemical processes are affected by the water temperature
to which they are exposed. Prosser (1973) states that for
every 10°C rise in temperature, the metabolism of fish
normally increases by a factor of two. Therefore, it is
likely that the toxic effects of chemicals can be
temperature dependent.
During 96-hour tests with mercuric chloride and rainbow
trout, MacLeod and Pessah (1973) noted that increased
toxicity was directly related to an increase in
temperature. Similar results were seen for the herbicide
Roundup® (Folmar et al. 1979) and quinaldine sulfate
(Marking and Olson 1975). It should be recognized,
however,that _not all chemicals exhibit a temperature related
variance to acute toxicity (Smith and Heath 1979).
The optimal temperature at which acute toxicity tests
were conducted has yet to be identified, but there are some
temperatures which have undergone wide spread use and
acceptance. In accordance with these practices the Agency
recommends that acute toxicity tests with the fathead minnow
and bluegill be performed at 22± 1°C and tests with rainbow
trout be performed at 12° ± 1°C.
d. Light
Although light is recognized as a potentially important
environmental variable, very few studies have been performed
evaluating its potential effects. McLeay and Gordon (1978)
31
-------
ES-6
August, 1982
found no difference in the toxicity of a pulpmill affluent
between tests performed with 8/16 and 16/8 hour light/dark
photoperiods. To avoid any possible effects from extraneous
light sources in the laboratory, the recommended photoperiod
is 16-hours light, 8-hours dark, with a 15-30 minut;^?-
/ j
transition period. ',i:j
!|
C. Reporting
A coherent theory of the concentration-response
relationship was introduced by Bliss (1935), and is widely
accepted today. This theory is based on four assumptions :
(a) resoonse is a positive function of dosage, i.e., it is
; t I , i
expected that increasing treatment rates should produce
increasing responses, (b) randomly selected animals are
normally distributed with respect to their .sensitivity to a
toxicant, (c) due to homeos tas is/response magnitudes are
proportional to the logarithm of the dosage (stresses) to
produce arithemtically increasing responses (strains) in
test animals populations, (d) in the case if direct dosage
of animals, their resistance to effects ii. proportional to
body mass. Stated another way, the treatment needed to
produce a given response is proportional to the size of the
animals treated .
The concentration-response curve, in which percent
mortality is plotted as a function of the logarithm of test
substance concentration, can be interpreted as a cumulative
distribution of tolerance within an experimental population
(Hewlett and Plackett 1979). Experiments designed to
measure tolerance directly (Bliss 1944) have shown that
tolerance, in most cases, is lognormally distributed within
a population. Departures from the lognormal pattern of
32
-------
ES-6
August, 1982
distribution are generally associated with mixtures of very
susceptible and very resistant individuals (Hewlett and
Plackett 1979). In addition, mixtures of toxicants can
produce tolerance curves which deviate significantly from
the lognormal pattern (Finney 1971).
If tolerance is lognormal ly distributed within the test
population, the resulting concentration-response curve will
be signmoidal in shape, resembling a logistic population
curve (Hewlett and Plackett 1979). While estimates for the
median lethal dose can be made directly from the
concentration-response curves,a linear transformation often
is possible, using probit (Bliss 1934, Finney 1971) or logit
(Hewlett and Plackett 1979) transformations.
Once the mortality data have been transformed, a
straight line can be fitted to the data points. Although
this line is most often fitted by eye (APHA 1975), a least
squares linear regression procedure is strongly recommended
for this purpose (Steel and _Torrie 1960). From the
regression equation, confidence limits can be determined for
predicted mortality values. An additional advantage is that
the significance of the slope of the regression line can be
determined (Draper and Smith 1966). By using replicate
tests, and analysis of variance can be performed to
determine whether deviations of data points from the
regression line are random fluctuations or indications that
a linear model is an inappropriate representation of the
data points (Draper and Smith 1966).
While the values for the median lethal dose, LC50, can
be estimated graphically from linearlized concentration-
response curve, other techniques are preferable since the
33
-------
ES—b
August, 1982
graphical method does not permit the calculation of a
confidence limit (APHA 1975),. The probit method uses probit
transformation and the minimum likelihood curve fitting
technique (Finney 1971). The Litchfield and Wilcoxon method
is a modified probit methods which does not require partial
kills, as does the unmodified probit method (Litchfield and
Wilcoxon 1949). The logit method utilizes either the
maximum likelihood or the minimum Chi square method to
estimate the LC50 (Ashton 1972, Berkson 1949). ' The moving
average method is simple to apply but depends on the
symmetry of the tolerance distribution to provide accurate
i i 11 I i • '
estimates (Thompson 1947). It1 'cannot' be1 util ized to
calculate any concentration level other than the,LC50. An
additional disadvantage is that confidence limits for the
LC50 cannot be calculated if nq partial kills are available.
The lack of partial kills seriously impairs the utility
of the probit, logit, and moving average methods. Tn
situations where there are no partial kills the binomial
test (Siegel 1956) can be used to estimate the conf idenc .-
limits around the LC50 value (Stephan 1977). The LC50 value
can be calculated from the relation
LC50 = (A B) 1/2
Wh e r e
A = concentration at which no organisms die
B = concentration where all organisms die
A and B are the confidence limits of the estimate and
34
-------
ES-6
August, 1982
ara significant above the 95 percent level if six or more
tests organisms are exposed at each concentration level.
If dose-response data are plotted for each 24 hour
interval throughout the test, the LC50 determined from each
curve can be plotted as a function of time, yielding an
acute toxicity curve (APHA 1975). This curve may approach
the time axis asymptotically, indicating the final or
threshold value of the LC50. The absence of a threshold
LC50 may indicate the need for a test of longer duration.
The LC50 value has limited utility, since a number of
substances with entirely different toxicity characteristics
i I i
'can produce identical LC50 numbers'. The difference will
therefore be in the slope of the concentration-response
curve (Casarett and Doull 1975).
The majority of response data will produced a near-
linear regression line. Yet very valuable information is
gained when the regression line is found to deviate
significantly from a straight line. For example, in fish
bioassays, the concentration-response line can appear
straight from the one percent to the 40 percent effect level
and then bend abruptly to the horizontal. Above a certain
level of test substance concentration no further mortality
of fish occurs. Further increments of test s'u'ostance simply
precipitate from solution and become unavailable to fish. A
low slope or broken regression line can occur when the
experimenter has inadvertently mixed two populations of
experimental animals (markedly different in their
susceptibility) together at each treatment level.
35
-------
ES-6
August, 1982
III. Economic Aspects
The Agency awarded a contract to Enviro Control, Inc. to
provide us with an estimate of the cost for performing
static and flow-through acute toxicity tests according to
this Guideline. Enviro Control supplied us with two
estimates; a protocol estimate and a laboratory survey
es t imate.
The protocol estimate was $621 for a static tes t, and
$743 for a flow-through test. These estimates were prepared
by separating the Guideline into individual tasks and
estimating the hours to accomplish each task. Hourly rates
were then applied to yield a total direct labor charge. 'An
overhead rate of 115%, other i direct costs of i$40 for, sitjatic
and $50 for flow-through tests, a general and' adminis trative
rate of 10%, and a fee of 20% were then added, to the direct
labor charge to yield the final estimate.
I
Enviro Control estimated that differences| in salaries,
equipment, overhead costs and other factors between
laboratories could result in as much as 50% variation from
this estimate. Consequently they estimated that test costs
could range from $310 to $931 for static tests and $372 to
$1115 for flow-through tests
The laboratory survey estimate was $471 for static tests
and $795 for flow-through tests. Five laboratories supplied
estimates of their costs to perform the tests according to
this Guideline. These costs ranged from $300 to $625 for
static tests and $550 to $1250 for flow-through tests. The
reported estimate is the mean value calculated from the
individual costs.
36
-------
ES-6
August, 1982
IV. REFERENCES
Adelman IR, Smith LL Jr. 1976. Fathead minnows
(Pimephales promelas and goldfish (Carassius auratus)
as standard Irish in bioassays and their reaction to
potential reference toxicants. J. Fish. Res. Board Can.
-33: 209-214.
Alexander DG, Clarke R Me V. 1978. The selection and
limitations of phenol as a reference toxicant to detect
differences in sensitivity among groups of rainbow trout
(Salmo gairdneri). Water Reseach 12: 1085-1090.
APHA. 1975. American Public Health Association,
American Water Works Association, and Water Pollution
Control Federation. Standard Methods for Examination of
Water, ;and Was,tewater, 14th ed. New York. American
Publi'c Health Association.
Aston -WD. 1972. 'The Loyit Transformation. New York:
Hafner Publishing Co.
ASTM. 1980. American Society for Testing and
Materials. New Standard Practice for Conducting Basic
Acute Toxicity Test with Fishes, Macroinvertebrates, and
Amphibians. E 729-80.
Auwarter AG . 1977- A f low-through system for study
interactions of two toxicants on aquatic organisms.
In: Mayer FL, Hamelink JL, eds . Aquatic Toxicology and
Hazard Evaulation. ASTM STP 634. American Society for
Testing and Materials: pp. 90-98.
Berkson J. 1949. The minimum Chi-square and maximum
likelihood solution in terms of a linear transform, with
particular reference to bioassay. J. Am., Stat. Assoc.
44: 273-278.
Bills TD, Marking LL, Olson LS 1977- Effects of the
residues of the oolychlorinated biphenyl: Aroclor 1254
on the sensitivity of rainbow trout to selected
environmental contaminants. Prog. Fish-Cult. 39: 150.
37
-------
ES-6
August, 1982
Blanchard FA, Takahashi IT, Alexander EC, Bartlett EA.
1977. Uptake, clearance and bioconcentration of 14C-
Sec-butyl-4-chlorodiphenyl oxide in rainbow trout.
In: Mayer FL, Hamelink JL, eds . Aquatic Toxicology and
Hazard Evaluation. ASTM STP 634. . American Society for
Testing and Materials: pp. 162-177.
Bliss CI. 1934. The method of probits. Science 79:
38-39.'
Blish CI. 1935. The calculation of the dosage-
mortality curve. Ann. Biol. 22: 134-307.
Bliss CI. 1944. The U.S. P. colla'ooratove rat assay for
digitalis. J. Amer. Pharm. Ass. 33: 25-245.
Braughn JL, Schoettger RA. 1975. Aquisition and
culture or research fish: rainbow trout, fathead
minnow, channel catfish, and bluegill. Corvallis,
Oregon: U.S. Environmental Protection Agency. EPA-
660/3-75-011.
Brungs WA. 1971. Chronic eEjects of low dissolved
oxygen concentrations on the fathead minnow (Pimephales
pcomelas). J. Fish. Res. Board Can. 28(8): 1119-1123.
Brugns WA, Geckler JR, Cast M. 1976. Acute and chronic
toxicity of copper to the fathead minnow in a surface
water of variable quality. Water Research 10: 37-43.
Brungs WA, McCormick JH, Neiheisel TW, Spehar RL,
Stephan CE, Stokes GN. 1977. Effects of pollution on
freshwater fish. J. Water Pollut. Control. Fed.'
51(6): 1425-1493.
Cairns J Jr. 1969. Fish bioassays - reproducibil i ty
and rating. Revista de Biologia 7:7-13.
Carmignani GM, Bennett JP. 1976. Leaching of plastics
used in closed aquaculture systems. Aquaculture 7:
89-91.
Cardwell RP, Foremn DG, Panye TR, Wilbur OJ. 1930.
Acute and chronic toxicity of four organic chemicals to
dish. Manuscript.
-------
Augus t,
ES-6
1982
Carrol JJ, Ells SJ, Oliver WS. 1979. Influences of
hardness constituents on the acute toxicity of
to brook trout (Saluvelinus fontinalis). Bull
SnvironiTi. Contain. Toxicol. 22 : 575-5 81 .
cadmium
Casarett LJ, Doull J. 1975. Toxicology, the Basic
Science of Poisons. New York: MacMillan Publishing Co.
Chandler JH Jr, Partridge SK. 1975. A'solenoid acuated
chemical-metering apparatus foe use in flow-through
toxicity tests. Prog. Fish-Cult. 37(2): 93-95.
Committee on Methods for Toxicity Tests with Aquatic
Organisms. 1975. Methods for acute toxicity tests with
fish, macroinvertebrates, and amphibians. Corvallis,
Oregon: U.S. Environmental Protection Agency. EPA-
660/3-75-009.
eFoe DL. 1975. Multichannel toxicant injection system
Eor flow-through bioassays. J. Fish. Res. Board Cananda
32: 544-546.
DeFoe FL. 1977. Temperature safety device for aquatic
laboratory systems. Prog. Fish-Cult. 39: 131.
Doudoroff P, Anderson BG, Burdick GE, Galtsof PS, Hart
WB, Patrick R, Strong ER, Suber EW, Van Horn WM.
1951. Bioassay Methods for the evaluation of acute
toxicity of industrial wastes to fish. Sewage Ind.
Wates 23: 1330-97.
Draper NR,
Analys is.
Smith H. 1966. Applied Regression
New York: John Wiley and Sons.
Drummond RA, Daws on
for simulating diel
laboratory. Trans.
WF. 1970.
patterns of
Amer. Fish.
An inexpensive method
lighting in the
Soc. 99: 434-435.
Eaton JG. 1970. Chronic inalathion toxicity to the
bluegill (Lepomis macrochirus Rafiesque). Trans. Amer
Fish. Soc. 103: 729-735.
Finney AJ. 1971.
University Press.
Probit Analysis. London: Cambridge
39
-------
Augus t,
ES-6
1982
Folmar LC, Sanders HO, Julin AM. 1979. Toxicity of the
herbicide glyphosphate and several of its formulations
to fish and aquatic invertebrates. Arch. Environm.
Con tarn. Toxicol. 8: 269-278.
Hewlett PS, Plackett RL. 1979. The interpretation of
quantal responses in biology. Baltimore: University
Park Press.
Hods on PV. 1979. 'Metering device
bioassays with aquatic organisms.
41: 129-131.
for toxicants used
Prog. Fish-Cult.
in
Hoi comb GW, Andrew RW. 1978. The acute toxic ity of
zinc to rainbow and brook trout. Duluth, Minnesota:
U.S. Environmental Protection Agency. EPA-600/3-78-094
Howarth RS, Sprague.JB. 1978. Copper lethality to
rainbow 'trout in waters of various hardness and pH.
Water Research 12: 455-462.
Iwama GK,
infection
to juvenile coho
109: 290-292.
Greer GL. 1980
on the toxic ity
salmon.
Effect of bacterial
of sodium pentachlorophenate
Trans. Amer. Fish. Soc.
Kenaga EE. 1979. Acute and chronic toxicity of 75
pesticides to various animal species. Down to Earth
35: 25-31.
Kitchell JF, O'Neill RV, Webb D, Gallepp GW, Ba'rtell SM,
Koonce JF, Ausmus BS. 1979. Consummer Regulation of
Nutrient cycling. Bioscience 29: 28-34.
Krugel S, Jenkins D, Klein SA. 1978. Apparatus for the
continuous dissolution of poorly water-soluble
components for bioassays. Water Research 12: 269-272.
Lemke AS, Dawson WF. 1979. Temperature monitoring and
safety control device. Prog. Fish-Cult. 41: 165-166.
Lemke AE, Brungs WA, Halligan BJ. 1978. Manual for
construction and operation of toxicity testing
promotional diluters. Duluth, Minnesota: U.S.
Environmenal Protection Agency- EPA-600/3-73-072.
40
-------
Augus t,
ES-6
1982
Litchfield JT Jr, Wilcoxon F. 1949. A simplified
method of evaluating dose-effect experiments. J. Pharm.
Exp. Ther. 96: 99-1113.
Macek KJ, Sleight BH III. 1977- Utility of toxicity
tests with embryos and fry of fish' in evaluating hazards
associated with the chronic toxicity of chemicals to
fishes. In: Mayer FL, Hamelink JL, eds. Aquatic
Toxicology and Hazard Evaluation. ASTM STP 634.
American Society for Testing and Materials: pp. 23-34.
MacLeod JC, Pessah E.
mercury accumulation,
rainbow trout, (Salmo
Can. 30: 485-491H
1973. Temperature effects on
toxicity and metabolic rate in
ga irderi). J. Fish. Res. Board
Marking LL, Olson LE. 1975. Toxicity of the lampricide
3-trifluoromethyl-4-nitrophenol (TFM) to nontaryet fish
in static tests; Investigations in fish control No.
60. Washington, D.C.: U.S. Fish and Wildlife Service,
Department of the Interior.
Mattson VR, Arthur JW, Walbridge CT. 1976. Acute
toxicity of selected organic compounds to fathead
minnows. Duluth, Minnesota: U.S. Environmental
Protection Agency. EPA-600/3-76-0 97.
Mauck WL, Olson LE, Hogan JW. 1977. Effects of water
quality on deactivation and toxicity of Mexacarbate
(Zectran®) to Fish. Arch. Environm. Contam. Toxicol.
6: 385-393.
Mauck WL, Olson LE, Marking LL. 1976. Toxicity of
natural pyrethrins and five pyrethroids to fish. Arch.
Environm. Contain. Toxicol. 4: 18-29.
Mayer FL, Sanders HO. 1973.
esters in aquatic organisms.
Prespective 3: 153-157.
Toxicity of phthalic acid
Environmental Health
McCarthy LS, Henry JAC, Houston AH. 1978. Toxicity of
cadmium to goldfish (Carass ius auratus ) in hard and soft
water. J. Fish Res. Board Can. 35. 35: 42.
41
-------
ES-6
August, 1982
MeKim JM. 1977. Evaluation of tests with early-life
stages of fish for predicting long-term- toxicity. J.
Fish. Res. Board Can. 34: 1148-1154.
McKim JM, Anderson RL, Benoit DA, Spehar RL, Stokes
3NI. 1976. Effects of pollution of freshwater fish. J.
Water Pollut. Control Fed. 43: 1544-1620.
McLeay D J, Gordon MR. 1978. Effect of seasonal
photoperiod on acute toxic responses of juvenile rainbow
trout (Salmo garidneri) to pulpmill effluent. J. Fish.
Res. Board CaTT! 3T: 1388-1392.
McLeay DJ, Walden CC, Munro JR. 1979. Influence of
dilution water on the toxicity of kraft pulp and paper
mill effluent including mechanisms of effect. Water
Research 13: 151-158.
Menendez R. 1976. Chronic effects of reduced pH on
brook trout (Salvelinus fontinalis ) . J. Fish. Res.
Board Can. 33: 118-123.
Mount DI. 1966. The effect of total hardness and pH on
acute toxicity of zinc to fish. Air and Water Pollution
Int. J. 10: 49-56.
Mount DI. 1968. Chronic toxicity of copper to fathead
minnows (Pimephales oromelas Rafinesque). Water
Research 2: 215-223.
Mount DI. 1973. Chronic effect of low pH on fathead
minnow survival, growth and reproduction. Water
Research 7: 937-993.
Mount DI, Brungs WA. 1967. A simplified dosing
apparatus for fish toxicology studies. Water Research
1: 21-29.
Nevins MJ, Johnson WW. 1978. Acute toxicity of
phosphate ester mixtures to invertebrates and fish.
Bull. Environm. Contain. Toxicol. 19: 250-255.
Ozburn GW, Smith AD. 1977- A mechanical toxicant
injector for flow-through toxicity tests. In: Mayer
FL, Hamelink JL, eds. Aquatic Toxicology and Hazard
Evaluation. ASTM STP 634. American Society for Testing
and Materials. pp. 85-89.
42
-------
' ES-6
August, 1932
Penrose WR, Squires WR. 1976. Two devices for removing
supersaturating gases in aquarium systems. Trans. A/ner.
Fish Soc. 105(1): 116-118.
Peterson RH, Anderson JM . 1969. Influence of
temperature change on spontaneous locomotor activity and
oxygen consumption of a tl an tic salmon ( Sal mo solar)
acclimated to two temoeratures . J. Fish Res. Board Can.
26: 93;';109.
'I'1'
Prosser'1 CL . 1973. Comparative Animal Physiology.
Philadelphia: W.B. Saunders Co.
Sauter S, Buxton K, Macek KJ, Petrocelli SR. 1976.
Effects of exposure to heavy metals on selected
freshwater 'fish. Duluth, Minnesota: U.S. Environmental
Protection Agency. EPA-60Q/3-76-10 5 .
i , i ' • ,] : ;
Schoor WP. 1975. ' Problems associated with low-
solubility compounds in aquatic toxicity tests:
theoretical model and solubility characteristics of
Arolor® 1254 in water. Water Research 9: 937-944.
Scott WB, Grossman EJ . 1973. Freshwater Irishes of
Canada. Bulletin 184, Fisheries Research Board of
Canada.
Siegel S. 1956. Nonparame tr ic Statistics for the
Behavioral Sciences. New York: McGraw-Hill.
Smith HT, Schreck CB, Maugham OE. 1978. Effect of
population density and feeding rate on the fathead
minnow (Pimephales promelas ) . J. Fish. Biol. 12: 449-
455.
Smith MJ, Heath AG . 1979. Acute toxicity of copper,
chromate, 2inc, cyanide to freshwater fish: effects of
different temperatures. Bull. Environm. Contam.
Toxicol. 22: 113-119.
Spehar RC, Holcom'oe GW, Carlson RW , Drummond RA, Yount
JD, Pickering QH. 1979. Effects of pollution on
freshwater fish. J. Water Pollut. Control. Fed.
51(6): 1616-1694.
43
-------
ES-6
August, 1982
Spehar RC, Carlson RW, Lemke AE, Mount DI, Pickering QH,
Snarski VM. 1930. Effects of pollution on freshwater
fish. J. Water Pollut. Control. Fed. 52: 1703-1768.
Sprague J3. 1969. Measurement of pollutant toxicity to
fish. I. Bioassay methods for acute toxicity. Water
Research 3: 794-821.
Steele RGD, Torrie JH. 1960. Principles and Procedures
of Statistics. New York: McGraw Hill.
Stephan CE. 1977. Methods for calculating an LC50.
In: Mayer FL, Hamelink JL, eds. Aquatic Toxicology and
Hazard Evaluation. ASTM STP 634. American Society for
Testing and Materials. pp. 65-84.
Thompson WR. 1947. Use of moving averages and
interpolation to estimation and error, and relation to
'ot'her methods. Bacterial Rev;' 12: 115-145.
Tucker RK, Leitzke JS. 1979. Comparative toxicology of
insecticides for vertebrate wildlife and fish. Pharmac.
Their. 6: 167-220 =
US EPA. 1971. U.S. Environmental Protection Agency.
Tentative plans for the design and operation of a
fathead minnow stock culture unit. Natl . Water Quality
Lab., Duluth, Minn.
USEPA. 1976. U.S. Environmental Protection Agency.
Quality Criteria for water.
USEPA. 1979. U.S. Environmental Protection Agency.
Toxic substances control. Discussion of oremanuf acture
testing policy and technical issues; request for
comment. Fed. Regist. March 16, 1979. 44: 16240-
16292.
USEPA. 1930. U.S. Environmental Protection Agency.
Office of Pesticide Programs. Guidelines for
registering pesticides in the United States. Subpart
E. Hazard evaluation: wildlife and aquatic
organisms. Draft, November 1930.
44
-------
ES-6
August, 1982
Wedemeyer GA, Yasutake WT. 1978. Prevention and
treatment of nirite toxicity 'in juvenile steelhead trout
(Salmo gairdneri). J. Fish. Res. Board Can. 35: 822-
82T.
Veith GD, Comstock VM. 1975. Apparatus for
continuously saturating water y/ith hyroohobic organic
chemicals. J. Fish. Res. Board Can. 32: 1334-1851.
45
-------
EG-10
August, 1982
FISH BIOCONCENTRATION TOXICITY TEST
OFFICE OF TOXIC SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
Office of Toxic Substances E3-10
Guideline for Testing Chemicals August, 1982
FISH BIOCQNCENTRATION TEST
(a) Purpose. This guideline is intended to be used for
assessing the propensity of, chemical substances to bioconcentrate
! I
in freshwater fish. This guideline describes a bioconcehtration
test procedure for the continuous exposure of fathead minnows
(Pimephales promelas) to a test substance in a flow-through
system. The United States Environmental Protection Agency (EPA)
williiulse data from this test i|n assessing the haza;rd a chemical
may present to the environment.
I
(b) Def initions . The definitions in section '3 of the Toxic
Substances Control Act (TSCA) >and the definitions in Part 792--
Good Laboratory Practice Standards) are applicable to this test
guideline. The following definitions also app /:
(1) "Acclimation" is the physiological compensation by test
organisms to new environmental conditions (e.g. temperature,
hardness, pH) .
(2) "Bioconcentration" is the net accumulation of a
substance directly from water into and onto aquatic organisms.
(3) "Bioconcentration factor (3CF)" is the quotient of the
concentration of a test substance in aquatic organisms at or over
a discrete time period of exposure divided by the concentration
-------
EG-10
August, 1982
in the test water at or during the same time period.
(4) "Carrier" is a solvent used to dissolve a test substance
prior to delivery of the test substance to the test chamber.
, \
(5) "Depuration" is the elimination of a test substa-nce from
''l'
a test organism.
(6) "Depuration phase" is the portion of a bioconcentration
test after the uptake phase during which( the organisms are in
flowing water to which no test substance is added.
i: : i ! ' I ' 'I
(7) "Dilution water" is the water to which the test
substance is added and in which the organisms undergo exposure.
1 (8) "Loading" is the ratio of fish biomass (grams, wet
weight) to the volume (liters) of test solution passing through
I i
the test chamber during a 24-hr, period.
(9) "Organic chlorine" is the chlorine associated with all
chlorine-containing compounds that elute just before lindane to
just after mirex during gas chremote-graphic analysis using a
halogen detector.
(10) "Organochlorine pesticides" are those pesticides which
contain carbon and chlorine such as aldrin, DDD, DDE, DDT,
dieldrin, endrin, and heptachlor.
(11) "Steady-state" is the time period during which the
amounts of test substance being taken up and depurated by ' the
test organisms are equal, i.e., equilibrium.
-------
EG-10
August, 1982
(12) "Steady-state bioconcentration factor" is the mean
concentration of the test substance in test organisms during
steady-state divided by the mean concentration in the test
solution during the same period.
(13) "Stock solution" is the concentrated solution of the
test substance which is dissolved and introduced into the
dilution water.
(14) "Test chamber" is the container in which the test
'' I
organisms are maintained during the pest period.
(15) "Test solution" is dilution water containing the
dissolved test substance to which test organisms are exposed.
(16) "Uptake" is the sorption of a test substance into a 3
onto aquatic organisms during exposure.
(17) "Uptake phase" is the initial portion of a
bioconcentration test during which the organisms are exposed to
the test solution.
(c) Test procedures — (1) Summary of the test. (i) Fathead
minnows are continuously exposed to at least one constant
sublethal concentration of a test substance under flow-through
conditions for a maximum of 28 days. During this time, test
solution and fish are periodically sampled and analyzed using
-------
EG-10
' August, 1982
appropriate methods to quantify the test substance
concentration. If prior to day 28, the tissue concentrations of
the substance sampled over three consecutive sampling periods
have b^en shown to be statistically similar (i.e., ^teady-s tate
.'II1
has been reached), the uptake phase of the test maybe terminated
and the remaining fish transferred to untreated flowing water
until 95 percent of the accumulated residues have been
eliminated, or for a maximum depuration period of 14 days.
i 11 ' I '' ' ' ( \
(ii) The mean test substance concentration in the fish at
steady-state is,'divided by the mean test solution concentration
at the same time 'to estimate the bioconcentration factor (BCF).
I
(iii) If steady-state is not reached during 28 days of
uptake, the steady-state BCF is calculated using non-linear
parameter estimation methods.
(2) [Reserved]
(3) [Reserved]
(4) Definitive test—(i) Background information. The
following data on the test substance should be known prior to
tes ting :
(A) Its solubility in water.
(B) Its stability in water.
(C) Its octanol-water partition coefficient.
(D) Its acute toxicity to fathead minnows.
-------
EG-10
August, 1982
(E) The validity accuracy and minimum detection limits of
the proposed analytical methods.
(ii) Selection of test concentration. (A) At least one
concentration should be tested to assess the propensity of the
compound to bioconcentrate. The concentration selected should
not stress or adversely affect the fish and should be less than
one-tenth the 96-hr or incipient LC50 determined from a flow-
through test with fathead minnows. The test concentration should
be less than the solubility limit of the compound in water and
close to the potential or expected environmental concentration.
The limiting factor of how low one can test is based on the
detection limit of the analytical methods. The concentration of
the test material in the test solution should be at least 3 times
greater than the detection limit in water.
(B) If it is desired to document that the potential to
bioconcentrate is independent of the test concentration, at least
two concentrations should be tested that are at least a factor of
10 apart.
(iii) Estimation of test duration. (A) An estimate of the
length of the uptake and depuration phases should be made prior
to testing. This will allow the most effective sampling schedule
to be determined. The uptake phase should continue until steady-
state has been reached, but need not be longer than 28 days. The
test should continue for at least 4 days.
-------
EG-10
August, 1982
(B) The time to steady state (S in hours) can be estimated
from the water solubility or the octanol-water partition
coefficient using the following equations:
S=3.0/antilog (0.431 log W-2.11) or,
S=3.0/antilog (-0.414 log P + 0.122)
wh e r e
W = water solubility (mg/1 and
P = octanol-water partition coefficient
Based upon the estimate of,i the time to steady state, one of
I1
the following sampling schemes may be used, to generate the
appropriate data.
I
Time to Steady-State in Days
Test period S<4 S>4<14 S>15<21 S>21
Sampling Days
Exoosure la
6a
1
2
3
4
Depuration la
6a
12a
1
4a
1
3
7
10
12
14
1
2
4
6
1
3
7
10
14
18
22
1
3
7
10
1
3
7
10
14
21
28
1
3
7
10
14
a=nours
-------
EG-10
August, 1982
(C) The depuration phase should.continue until at least 95
percent of the accumulated test substance and metabolites have
been eliminated, but no longer than 14 days.
(iv) Test initiation. (A) The test should not be started
un* 11 the test substance delivery system has been observed to be
fu.ctioning properly for at least 48 hours. This time should be
sufficient to allow the test substance concentration to become
equilibrated with the test exposure system. Analyses of two sets
of ,test solution samples ,taken prior to test initiation should
document this equilibrium (i.e., the concentrations do not vary
more than 20% from each other). At initiation (time 0), test
solution samples should be collected immediately prior to the
addition of f'ish to the test chambers.
(B) The appropriate number of fathead minnows should be
impartially distributed to each test chamber up to five at a time
until the appropriate numbers have been distributed. The exact
number of test organisms depends upon the expected length of
testing, sample size, and the number of additional specialized
analyses to be performed at termination.
(v) F e ed i ng . (A) Fish should be fed once a day throughout
the uptake and depuration phases. Feeding should always be done
just after sampling to minimize the effects of the test substance
-------
EG-10
August, 1982
oresent in the gut when sampling. Fish should be fed the same
food at a similar quantity as they received during holding and
acclimation.
(B) Uneaten food and fecal material should be removed from
the test aquaria within 30 minutes after fee.ding to minimize
uptake of test substance by the food or feces.
(vi) Observations. (A) Observations on fish appearance and
behavior should be made and recorded daily. Any abnormal
behavior such as erratic swimming, lethargy, increased ',
excitability, or any changes in appearances or physiology such as1
discoloration, hypervent ilation or opaque eyes should bei
recorded .
(B) Observations on compound solubility should also be
recorded. These include the appearance of surface slicks,
precipitates, or material adsorbing to the test chamber.
(vii) Water quality measurements. The water temperature and
dissolved oxygen concentration should be recorded at least daily
and the pH twice weekly in each test chamber during uptake and'
depuration.
(viii) Sampling procedures . (A) At each of the designated
sampling times, triplicate water samples and enough fish should
be collected from the exposure chamber(s) to allow for at least
four fish tissue analyses. A similar number of control fish
-------
EG-10
August, 1982
should also be collected at each sample point, but only fish
collected at the first sampling period and weekly thereafter
should be analyzed. Triplicate control water samples will be
collected at the time of test initiation and weekly thereafter.
Test solution sam les should be removed from the approximate
center of the water column.
'(3) At each sampling period, the appropriate number of fish
is netted and removed from each test chamber. Care should be
' ! I ! !
taken hot to sample the weakest and consequently usually the
smallest fish, especially during the first few sampling periods,
to prevent biasing the test results. Each fish is pithed,
blotted dry and then frozen at <-10°C if not analyzed within four
hours .
(C) At termination, an extra set of fish should be sampled
and eviscerated for quantifying the residues in the viscera and
carcass. If a radio-labelled test compound is used, a sufficient
number of fish should be sampled at termination to permit
identification and quantitation of any major (>10% of parent)
metabolites present. It is crucial to determine how much of the
activity present in the fish is directly attributable to the
parent compound.
(5) Test results — (i) Biological. (A) The maximum
allowable mortality of fish is 10 percent per week. If more than
-------
EG-10
August, 1982
10 percent "of the fish in the control or test chamber(s) die
during any week of testing, the test should be repeated.
(B) Steady-state has been reached when the mean
concentrations of test substance in whole fish tissue taken on
three consecutive sampling periods are statistically similar (F
test, P=0.05). A BCF is then calculated by dividing the mean
tissue residue concentration during steady-state by the mean test
solution concentration during this same period. A 95 percent
! ' ] i ll: ||
confidence interval should also be derived for the BCF. This can
be done by calculating the mean fish tissue concentration ati
steady state (Xf ) and its 97.5 percent confidence interval, +- t
(3.E.), where t is the t statistic = 0.025 and S.E. is one
i
standard error of the mean. This calculation would yield lower
and upper confidence limits (Lf and Uf) . The same procedure can
be used to calculate the mean and 97.5 percent confidence
interval from the test solution concentrations at steady-state,
Xs _+. fc (S.E.), and the resulting upper and lower confidence
limits (Lg and Us) . The 95 percent confidence interval of the
BCF would then be between Lf/Us and Uf/Ls•
(C) If steady-state was not reached during the 28 day uptake
period, the maximum BCF should be calculated using the mean
tissue concentration from that day and the mean water
concentration from that and the previous sampling day. An uptake
10
-------
EG-10
August, 1982
rate constant should then be calculated using appropriate
techniques, such as the BIOFAC program developed by Blau and Agin
(1978). This rate constant will allow the estimation of a steady
state BCF and the estimated time to steady-state.
(D). If 95 percent eliminati n has not been observed after 14
days depuration, then a depuration rate constant should be
calculated. This rate constant will allow estimation of the time
to 95% elimination. -'
I. • : I ! • ii : i •
(11) Analytical . (A) All samples should be analyzed using
EPA methods and guidelines whenever feasible. The specific
methodology used should be validated before the test is
initiated. The accuracy of the method should be measured by the
method of known additions. This involves adding a known amount
of the test substance to three water samples taken from an
aquarium containing dilution water and a number of fish equal to
that to be used in the test. The nominal concentration of these
samples should be the same as the concentration to be used in the
test. Samples taken on two separate days should be analyzed.
The accuracy and precision of the analytical method should be
checked using reference or split samples or suitable
corroborative methods of analysis. The accuracy of standard
solutions should be checked against other standard solutions
whenever possible.
11
-------
EG-10
August, 1982
(B) An analytical met'hod is not acceptable if likely
degradation products of the test substance, such as hydrolysis
and oxidation products, give positive or negative interferences,
unless it is shown that such degradation products are not present
in the test c ambers during the test. Atomic absorption
spectrop ho tome trie methods for metals and gas chroma tog rap hie
methods for organic compounds are preferable to colorimetric
me thods .
(C) In addition to analyzing samples of test solution,
at
least one reagent blank should also be analyzed when a reagent is
used in the analysis.
(D) When radiola'oel led test compounds are used, total
radioactivity should be measured in all samples. At the end of
the uptake phase, water and tissue samples should be analyzed
using appropriate methodology to identify and estimate the amount
of any major (_>_ 10 percent of the parent compound) degradation
products or metabolites that may be present.
(6) [Reserved]
(d) Test conditions — (1) Test species (i) Selection.
(A) The fathead minnow (Pimephales promelas) should be used as
the test organism.
(B) Immature fish should be used. They should be young
enough so as not to mature during the test. Fish used in the
same test should be as similar in size
12 ,
-------
August, 1982
as possible to reduce variability. The standard deviation o'f the
weight should be less than 20 percent of the mean (N= 30).
(C) Fish used in the same test should be from the same
supplier or culture unit and from the same holding and
acclimation tank(s).
(D) Fathead minnows should not be used if they appear
diseased or otherwise stressed or'if more than 5 percent die
during the 48 hours prior to testing. Diseased fish should be
I . i :
discarded or treated and held for a 'minimum of 14 days before
tes ting .
(ii) Care and handling. (A) Fish purchased from a
commercial source should be attended to immediately upon
arrival. Transfer of the fish from the shipping to the holding
water should be gradual to reduce stress caused by differences in
watei quality characteristics and temperature. Fish should be
quarantined 'and observed for at least 14 days prior to testing.
(3) During holding, the fish should not be crowded and the
dissolved oxygen concentration should be above 60 percent
saturation. Holding tanks should be kept clean and free of
debris. Fish should be fed at least once a day with a food which
will support their survival and growth.
(C) Fish should be handled as little as possible. When
handling is necessary, it should be done as gently, carefully,
13
-------
EG-10
August, 1982
and quickly as possible using dip nets made of small mesh nylon,
silk, bolting cloth, plankton netting, or other similar knotless
materials. Handling equipment should be sterilized between uses
by autoclaving, treating with an iodophor or with 200 mg
hypochlorite/liter.
(iii) Acclimation. If the holding water is not from the
same source as the test dilution water, acclimation to the
dilution water should be done gradually over a 43-hour peroid.
I I ! ' i
The fish should then be held' an additional 14 days in the
dilution water prior to testing. Any changes in water
temperature should not exceed 3°C per day. Fish should be held
for a minimum of 7 days at the test temperature prior to testing.
( iv) Loading. The number of fish placed in each test
chamber and the flow rate through the test chamber should be such
that the uptake of the test substance by fish upon introduction
into the test solution does not reduce the measured concentration
of the test solution by more than 20 percent of the concentration
measured before the fish were introduced. The loading should not
exceed 0. Ig fish per liter of test solution delivered over any 24
hour period, and the minimum turnover rate should be 6 aquaria
volumes per 24 hours. For some compounds, loading rates less
than O.lg/1 may be needed to prevent a substantial loss of test
substance as a result of fish uptake.
14
-------
EG-10
August, 1982
(2)' Facilities — (i) Dilution water. (A) A constant supply
of good quality water should be available throughout the holding,
acclimation and testing periods. Although unadulterated well
water is recommended, de-chlorinated tap water or reconstituted
soft water may be used. A dilution water is acceptable if
fathead minnows will survive and grow normally for 60 days
without exhibiting signs of stress, i.e., discoloration, lack of.
feeding, poor response to external stimuli, or lethargy.
I
I. ] '
(B) The total hardness, alkalinity, pH, specific
conductance, temperature and dissolved oxygen concentration of
the dilution water should be determined weekly. The pH should
not vary more 0.4 units and the other parameters more than 10
percent on a monthly basis.
(C) Reconstituted soft water, if used, should be prepared by
adding 4.8 g NaHC03, 3.0 g CaS04 2H20, 3.0 g MgS04/ and 200 mg
KC1 to each 100 1 of deionized or glass distilled water, or to
dechlorinated tap water with a total residual chlorine
concentration less than 1 ug/1. In all cases the specific
conductance at 25°C of the water source should be less than 1
micromho/cm.
(D) All water should be extensively aerated prior to use if
the dissolved oxygen concentration is less than 90 percent of
15
-------
EG-10
August, 1982
saturation. If the concentration of dissolved gases exceeds 110
percent of saturation, the excess gases should be removed using
appropriate apparatus.
(E) The quality of the dilution water should be constant and
should meet the following specifications me.sured at least twice
a year.
Substance Maximum Concentration
Particulate matter , 20 mg/liter
Total organic carbon 2 mg/liter
or
Chemical oxygen demand 5 mg/liter
Un-ionized ammonia 1 ug/liter
^esidual chlorine 1 ug/liter
Total organophosphorus pesticides 50 ng/liter
Total organochlorine pesticides plus
polychlorinated biphenyls (PCBs) 50 ng/liter
or
Organic chlorine 25 ng/liter
Copper, cadmium or zinc 10 ug/liter
16
-------
EG-10
August, 1982
(ii) Construction materials. Materials and equipment that
contact dilution water, stock solutions or test solutions should
not leach or absorb substances. Glass, #316 stainless steel and
perfluorocarbon plastics (e.g. Teflon ®) should be used whenever
possible. Concrete, unplas tic ized plastics and fiberglass may be
used for holding and acclimation tanks and in the water supply
system, but they should be thoroughly conditioned before use by
1 > !
rising with a continuous flow of water > 25°C for 48 hours. The
use of flexible tubing should be avoided as phthalate esters
leach from these materials. Cast iron pipe may be used but
filters will be needed to remove rust particles. Rubber, copper,
brass, galvanized metal, and epoxy glue should not come in
contact with dilution water, stock solutions or test solutions.
(iii) Fish holding and acclimation. (A) Tanks are needed
for holding and acclimating fathead minnows prior to testing.
The number and size of tanks needed depends upon the amount of
testing to be performed and the availability of fish of the right
age. A constant supply of good quality dilution water should be
supplied to all tanks. The volume required depends upon the
holding temperature and the number of fish being held, but the
flow should be great enough to maintain a dissolved oxygen
concentration > 60 percent of saturation.
17
-------
EG-10
August,. 1982
(3) Temperature control apparatus are needed to maintain the
desired holding and acclimation temperatures. Apparatus controls
should be able to maintain temperatures within 1°C of the
appropriate temperature. If the water is heated, care should be
taken to avoid supersaturation of gases in the v, .ter.
(iv) Testing apparatus. (A) Test chambers can be made from
welded #316 stainless steel or from double strength glass joined
with clear silicone adhesive. The size, shape' and depth of the
I ' | • i ;
test chambers are not important as long as they accommodate the
loading requirements.
(B) The test substance delivery system used should
accommodate the physical and chemical properties of the test
substance and the selected exposure concentration. The apparatus
used should accurately and precisely deliver the appropriate
amoun^ of stock solution and dilution water to the test
chambers. The introduction of the test substance should be done
in such a way as to maximize the homogeneous distribution of the
test substance throughout the test chamber.
(C) The dilution water should be delivered to an elevated
headbox from which it can flow by gravity to the test substance
delivery system. Use of a headbox facilitates a constant
delivery rate and heating or cooling of the water to the
approximate test temperature prior to delivery. Water in the
13
-------
EG-10
August, 1982
headbox may also be eas ily 'aerated or degassed as the situation
dictates .
(v) Cleaning o£ test apparatus. Delivery systems and test
chambers should be cleaned before and after each use. If there
is obvious absorption of a test substance by the silicone,
adhesive, those applicable parts of the delivery system should be
discarded.
(3) Test parame ters--( i ) Dissolved oxygen. The dissolved
" - - - - . ! ^
oxygen concentration in each chamber should be greater than 5.3
mg/1 (60 percent of sea-level saturation at 22°C) throughout
testing.
(ii) Temperature. The test temperature should be 22 ±i°c.
Temporary excursions (< 8 hours) to 20 or 24°C are permissible.
(iii) Lighting . A photoperiod of 12 hours light and 12
hours dark with a 15-30 minute transition period is recommended.
(iv) Test substance. The name and purity of the test
substance to be tested will be specified in the test rule.
Radio-labelled 'compounds should not be used unless there are no
suitable, validated, analytical techniques to measure unlabelled
test substance in fish, or the costs of these analytical
techniques are very high.
(v) Carrier use. Whenever possible, the test substance
should be added directly to the dilution water or from a water
19
-------
EG-10
August, 1982
stock solution. With compounds having a low water solubility, it
may be necessary to prepare test solutions using a carrier.
The carriers to be used, in order of preference are: triethylene
glycol (TEE), dimethyl formamide (DMF) and acetone. The amount
used should be kept to a minimum and s.hould not exceed 80 mg/1 UT
the test solution for TB3 and 5.0 mg/1 for DMF and acetone.
(e) Reporting. In addition to the information required in
Part 792--Good Laboratory Pracice Standards, the report should
! i ' i
contain the following:-
(1) The source 'Of the dilution water, 'its mean monthly
chemical characteristics (total hardness, alkalinity, pH,
specific conductance, temperature and D.O.) and a description of
any pretreatment.
(2) Detailed information about the fathead minnows used,
including age, mean and standard deviation wet weight (blotted
dry) and standard length, source, history of disease, parasites
and treatment, acclimation procedures, and food used.
(3) The number of organisms tested, loading rate and volume
additions per 24 hours.
(4) The percentage mortality of control fish and fish in
each exposure chamber and any observed abnormal behavioral or
physiological effects.
(5) The method of stock solution preparation including
20
-------
EG-10
August, 1982
nominal and measured concentrations and solvent used.
(6) The mean, standard deviation and range of the
temperature, dissolved oxygen concentration and pH during the
test period.
(7) 'Photoperiod length and light intensity.
(8) Description of sampling and analytical methods for water
and tissue analyses.
(9) The mean, standard deviation and range of the
concentration of test compound in the test solution and.fish
tissue at each sampling period.
(10) The time to steady-state.
(11) The steady-state or maximum BCF and the 95% confidence
limits.
(12) The time to 95 percent elimination of accumulated
res idues .
(f) References . Blau GE , Agin CL. A users manual for
BIOFAC: A computer program for characterizing the ratio of
uptake and clearance of chemicals in aquatic organisms. Dow
Chemical Co. March 15, 1978.
21
-------
ES-7
August, 1982
TECHNICAL SUPPORT DOCUMENT
FOR
FISH BIOCONCENTRATION TEST
OFFICE OF TOXIC .SUBSTANCES
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
TABLE OF CONTENTS
Page
I. Purpose 1
II. Scientific Aspects 1
General 1
Test Procedures 1
Test Substance Concentrations 4
Duration of Test 5
Test Conditions 8
Test Species 8
Selection 8
Maintenance of Test Species 10
Acclimation 10
Facilities 10
Dilution Water 10
Construction Materials 13
Testing Apparatus , 13
Cleaning 14
Carriers 15
Environmental Conditions 16
Loading 16
Dissolved Oxygen 17
Temperature 17
Light 19
Reporting 19
III. Economic Aspects 19
IV. References 21
-------
Office of Toxic Substances E3-7
Augus t, 1982
Technical Support Document for Fish Bioconcentratio'n. Tes t
I. Purpose
The purpose of this support document is to provide the
scientific background and rationale used in the developmen-t
of Test Guideline EG-10 to evaluate the bioconcentration
potential of chemical substances in fish. The Document
provides an account of the scientific evidence and an
explanation of the logic used in the selection of the test
methodology, procedures and conditions prescribed in the
Test Guideline. Technical issues and practical
cphs1 iderati'ons relevant to the Test Guideline are'
discussed. In addition, estimates of the cost of conducting
the test are provided.
11. Scientific Aspects
A. Test Procedures
1. General
Fish are nearly unbiquitous inhabitants of freshwater
and marine environments. In addition to their economic and
recreational value, fish occupy on essential position in
aquatic food chains, feeding on various forms of plant and
animal life in the aquatic environment and, in turn, being
eaten by some other aquatic or terrestrial consumer.
Through these trophic or feeding interactions, nutrient and
energy exchanges occur which are needed to maintain the
ecological stability of the aquatic environment and sustain
the food chains of commercially and recreationally valuable
fish. Because they are critical links in these food chains,
certain species of fish which have no direct commercial or
recreational value in themselves are essential to the well-
being of economically important fisheries. A chemical which
-------
ES-7
August, 1982
i-3 highly cumulative could destroy "an economically or
-------
ES-7
August, 1982
result in over-regulation of compounds.
If data on water solubility, octanol-water partitioning,
biodegradability, or structure-activity relationships
suggest that a compound may substantially bioconcentrate in
-ish, a bioconcentration test with fathead minnows should be
performed to quantitate empirically the degree to which such
a compound may be accumulated in "fish tissue.
There are two methodologies used today to estimate
bioconcentration potential; the kinetic approach and the
s teady-s tate approach. Bishop and Maki (1980) and Hamelink
(1977) review both. Using the kinetic approach, Bishop,and
Maki(1980), Branson et al. (1975), Cember et al . (1978) and
Krzeminsky (1977) proposed the'use of first-order kinetic
expressions from relatively short (_<_ 5 days) fish exposure
and depuration periods to calculate uptake and depuration
rate constants. These rate constants are then used to
estimate the BCF at the time of apparent steady-state, and
the time to 50% elimination. The steady-state' method, in
more wide-spread use, exposes fish for a longer period of
time until steady-state in the tissue is experimentally
observed (Barrows et al. 1980, Bishop and Maki 1980, Veith
et al . 1979) and continues with a depuration phase until 50
or 95% elimination has been observed. The estimation of:
bioconcentration using the kinetic approach can not account
and adjust for changes in the rates of uptake and depuration
such as those observed by Barrows et al . (1980) and Melancon
and Lech (1979). The use of the kinetic approach also
requires access to a sophisticated computer system,
apparatus not readily available to many laboratories.
-------
ES-7
August, 1982
Although Bishop and Maki (1980) and Branson et al.
(1975) have shown excellent agreement between estimates of
bioconcentration factors for some compounds using both
approaches, we recommend a modified steady-state method far
determination of bioconcentration potential. The empirical
nature of the data, the relative ease with which the test
I
can be perform'ed, and the number of researchers and
laboratories that have performed such tests make this test
more appropriate at this time. As the data base for
comparisons of BCF's between the two methods grows, the
kinetic approach may become more useful and valuable. Under
] I ' !
the Toxic Substances Control Act, we are required to review
all test guidelines annually, and in the future we will
consider adopting the kinetic approach.
2. Test Substance Concentrations
Although virtually all researchers involved in
bioconcentration testing state that the exposure
concentration should be below toxic effect levels, there are
Eew data supporting this recommendation. Tests determining
bioconcentration factors with fathead minnows and,PCBs
(DeFoe et al. 1978), toxaphene (Mayer et al. 1977), three
chlorinated cyclodiene intermediates (Spehar et al. 1979),
and acrolein (Macek et al. 1976) showed that there was
little difference between BCF's calculated from different
exposure concentrations up to and including at least one
concentration that caused a reduction in survival or growth.
Mayer (1976) however performed a study with di-2-
ethylhexyl phthalate and fathead minnows and found that the
BCF's increased sixfold, from 155 to 886, as the exposure
concentration decreased from 62 to 1.9 ug/1. Bishop and
-------
ES-7
August,- 1982
Maki (1980) tested four compounds with bluegills at two
concentrations a factor of 10 apart. In the tests with DDT,
tetradecylheptaethoxylate (AE), and EDTA, similar BCF's were
observed at both concentrations tested. Elimination of
accumulated res idues however was cons iderably slower at the
lower EDTA concentration. In the test with LAS however, the
BCF at the low concentration of 0.064 mg/1 v-/as 26 0, more
than twice the BCF observed at 0.63 mg/1. Of 15 compounds
tested with bluegill at two concentrations by Macek et al.
(1975), four yielded BCF's at the lower concentration that
were | 2.2 to 6.2 times the BCF at the higher exoosure
: i • I • i; i i
level. For two other compounds, the reverse was true; the
BCFs at the higher concentration were 2.8 and 5.5 times
those at the lower level. As even these differences were
not great, available data does not •.vain-rant required testing
of all compounds at two exposure concentrations.
What would be most useful for the hazard and risk
assessment processes "would be the use of an exposure
concentration that approximates the expected or estimated
environmental concentration. One should take care, however,
that the selected concentration is at least three times its
detection limit in water and will allow quantification of
the residues in tissue. Test concentrations of 1-10 ug/1
would be appropriate for many compounds.
3. Duration of Test
The exposure period should be long enough to demonstrate
that steady-state has been reached in the fathead minnow
tissue- Most compounds will reach steady state within the
recommended 28 day maximum uptake period (Barrows et al.
1930, Bishop and Maki 1980, Macek et al. 1975, Veith et al.
-------
ES-7
August, 1982
1979). The depuration period should continue until at least
95% of the accumulated residues have been eliminated. This
will normally occur within the 14 day maximum depuration
period.
Most of the 33 chemicals tested by Barrows et al.
(1980), reached steady-state in 3-10 days, and 50%
depuration was usually reached in less than 1 day.
Consequently it is clear that the relatively long uptake and
depuration periods (_>_ 28 days uptake, 14 days depuration)
used by many researchers are usually not required.
Before starting a bioconcentra.tion test, an estimation ,
of the BCF and the time to steady-state should be made.
Kenaga and Goring (1930) present data and methods to
estimate the BCF. The two most commonly used factors for
predicting bioconcentration potential are water solubility
and octanol-water partitioning. Water solubility can be
determined empirically in the laboratory or, in some cases,
ta'ken from the literature (Chiou et al. 1977, Kenaga and
Goring 1980). Octanol-wa ter partition coefficient.--; can be
determined empirically, estimated by reverse-phase high
pressure liquid chroma tog rap hy according to Veith and Morris
(1978), calculated according to Leo et al. (1971) or taken
from the literature (Chiou et al . 1977, Hansch et al. 1972,
Kenaga and Goring 1980).
The time to steady-state (3 in hours) can be estimated
from the water solubility or the octanol-water partition
coefficient using the equations developed by ASTM (1980b):
-------
ES-7
August, 1982
S=3.0/antilog (0.431 log W-2.11) or
S=3.0/antilog (-0.414 log P + 0.122)
where
W=water solubility (rag/1) and
P=octanol-water partition coefficient
Presented below is a summary of data correlating various
exposure times to the corresponding estimates of partition
and BFC
Log P Log BGFa BCF
2
4
7
12
18
22
28
1,
8,
33,
120,
316,
524,
933,
585
710
113
226
228
807
254
3.
3.
4.
5.
5.
5.
5.
2
94
52
08
5
72
97
2
2
3
4
4
4
4
.0
.6
.1
,6
.0
.1
2
5
4
•;>
6
.37
1
10
14
23
105
446
387
4150
,000
,521
,686
aLog BCF was estimated using the equation of Veith et
al. (1979) where log BCF=0.85 log P - 0.70.
Based on the estimate of the time to steady state, one
of the following sampling schemes may be used to generate
appropriate data.
-------
ES-7
August, 1982
Time to Steady-State
Test Period/ Sa<4
Exposure 1
1
2
3
4
Depuration 1
6
1 ?b
1
34-14
Sampling
4b
1
3
7
10
12
14
1
2
6
> S15-21
Days
1
3
7
10
14
18
22
1
3
10
S>21
1
3
7
10
14
21
23
1
3
7
10
14
alength of estimated time to steady state in days.
hours .
B. Test Conditions
1. Test Species
a.' Selection
The most common fish species used to determine the
bioconcentration potential of compounds under flow-through
conditions have been the fathead minnow, Pimephales promelas
(Mayer 1976, Spehar et al. 1979, Veith et al. 1979); rainbow
trout, Salmo gairdneri (Blanchard et al . 1977, Branson et
al. 1975, Melancon and Lech 1979, Neely et al. 1974,
Reinert et al. 1974); and bluegill Lepomis macrochirus
(Barrows et al. 1980, Gonz et al. 1975, Macek et al . 1975).
The fathead minnow has been selected as the te;j t species
for use. It can be easily cultured in the laboratory (U.S.
EPA 1971), thus insuring an almost constant supply of
healthy fish of the proper size throughout the year. It has
-------
ES-7
August, 1982
been used extensively in life-cy.cle chronic toxicity tests
and early-life stage tests as summarized by Macek and
Sleight (1977) and iMcKim (1977). A definitive study on
measuring and estimating the bioconcentration factor of
chemicals in fish has also been performed using the fathead
minnow as the test species (Veith et al . 1979), Results of
this study clearly demonstrate the suitability of this
species. In tests with hexachlorobenzene and 1,2,4-
trichlorobenzene, the authors found that fathead minnows
accumulated these compounds to the same extent as green
sunfish (Lepora^s cyanellus ) and approximately twice as much
: I . ' i ' :
as rainbow trout.
In a separate set of tests with hexachlorobenzene, th-e
authors determined that the age and size of fathead minnows
had little effect on bioconcentration. Tests with newly
hatched fry, 30 and 90-day old juveniles, and approximately
180-day old adults yielded similar BCF's.
The source of the test fish was also found by Veith et
al. not to be a significant source of variation in the
bioconcentration of a PCB mixture (Aroclor 1016®). Tests
with three different fish populations from a laboratory
brood culture and with two populations from ponds yielded
s imilar BCF's . .
Studies by DeFoe et al. (1978) and Nebeker et al. (1974)
demonstrated that gravid fathead minnows bioconcentrated PCB
mixtures twice as much as males during laboratory tests.
This increase was due to the increased lip id content of the
females compared to the males. Consequently we recommend
that immature fathead minnows less than 120 days old be
used. Fish older than 120 days should not be used as by the
-------
ES-7
August, 1982
end of testing they may be sexually mature. The detection
limits of the analytical procedures used to measure tissue
residues may preclude the use of very small fish (eg. <30
days old) with compounds that do not bioconcentrate
appreciably.
2. Maintenance of Test Species
a. Acclimation
Brauhn and Schoettger (1975) found that fish that had
become accustomed to unres tricted-swimming in rearing ponds
underwent intense competition for food and swimming space
when placed in confined holding tanks. These authors
recommended that fish be maintained in conf ined holding
tanks with color backgrounds and light intensities similar
to those in the testing area to prevent additional stress
when transferred to test chambers.
Although the importance of acclimation to the test
temperature is still unclear, a maximum gradual change of
3°C/day is recommended at this time. Peterson and Anderson
(1969) concluded that complete acclimation to temperature,
based on changes in locomotor activity and oxygen
consumption, requires approximately two weeks before
metabolism is back to normal. They also determined that the
rate of change was more important than the amount of change.
3. Facilities
a. Dilution Water
A constant supply of good quality dilution water is
needed to maintain consistent experimental conditions during
testing. A change in water quality during a test may alter
the response of the test fish to the test solution.
Although there is substantial information on the effects of
10
-------
ES-7
August, 1982
pH (Folmar et al. 1979, Mauck et al. 1976, 1977, Menendez
1976, Mount 1973) and hardness (Carroll et al. 1979,
Holcombe and Andrew 1978, Howarth and Sprague 1978, Sauter
et al . 1976) on the acute and chronic toxicity of compounds,
there is no apparent data describing the effects of these
water quality parameters on bioconcentration potential.
A dependable source of clean surface or ground water
usually will provide water having greater consistency in its
chemical makeup than that from-a municipal water supply.
Municipal water may have originated from several sources
which differ in chemical makeup. Municipal wateij- , frequently
is also treated'chemically as part of a purification
process. Since the proportions in which waters from
different sources may be mixed, and since the chemical
treatment given water 'during the purification process may be
different from time to. time, the chemical makeup of
municipal water may vary considerably. Reconstituted water,
while theoretically 'more consistent from batch to batch than
either surface or ground water or municipal watar, may in
some instances lack trace minerals required by some species
of fish. Cairns (1969), however, performed several acute
toxicity tests on some compounds with both reconstituted
water and natural water and found that the data generated
from the tests in natural water were not consistent or
reproducible, whereas the results from the tests with
reconstituted water were consistent.
Fish culturists do not know all of the conditions
required to maintain fish health, nor do they know all of
the components and combinations of components in water that
adversely affect the health of fish (Brauhn and Schoettger
11
-------
ES-7
August t 1982
1975). Nevertheless, to avoid possible inconsistencies and
inaccuracies in test results, healthy £ish are needed for
use in bioconcentration tests. There is, therefore, a need
to determine that the dilution water, whatever its source,
is able to support the fathead minnow in a healthy condition
for the duration of the test period.
An appropriate way to make that determination is to
place young fathead minnows in the dilution water under
flow-through conditions for 60 days and observe their
behavior, growth and development. Ideally, those
observations should be made'bv an experienced biologist
familiar with certain stress 'reactions which are difficult
for an untrained observer to identify (Brauhn and Schoettger
1975).
Surface and ground water may vary considerably in, their
chemistry depending upon the season of the year and
precipitation patterns. Variations in the chemistry of
surface water may involve the quantity of particulate
matter, dissolved organic and inorganic chemicals, un-
ionized ammonia and various other contaminants. As an
indication of the uniformity OL the dilution water, it is
recommended in the guideline that certain substances be
quantified at least twice a year or more frequently if it is
suspected that the concentration of one or more of those
substances have changed significantly. The maximum
acceptable concentrations listed for these substances are
among those generally accepted as concentrations which do
not adversely affect freshwater fish (APHA 1975, ASTM
1930a) . Concentratioas in excess of the values cited in the
guideline may affect the data developed from the
12
-------
ES-7
August, 1982
bioconcentration test.
Recognizing that some variation in water chemistry is
normal in natural surface or ground waters, a 20 percent
fluctuation from month to month in water hardness,
alkalinity and conductivity and a variance of 0.4 pH units
is acceptable.
b. Construction materials
Due to the toxicity of many heavy metals at low
concentrations (U.S. EPA 1976) and the tendency of metal
pipe, galvanized sheeting, laboratory equipment, etc. to
leach metals into water, no metal other than stainless steel,
(preferably #316) should be used. For the same reason,
plasticized plastics should not be used due to the high
toxicity of the main component, di-^2-e thyl-hexyl phthalate,
(Mayer and Sanders 1973) which may leach Into aquaria
systems (Carmignani and Bennett 1976). To avoid any
possible stress from exposure to -low levels of metals,
phthalate esters, and other potential contaminants, #316
stainless steel, glass and perfluorocarbon plastics (e.g.
Teflon®) should be used whenever possible and economically
feasible. If other plastics should be used, conditioning
with a continuous flow of dilution water > 25°C should be
performed for a minimum of 48 hours.
c. ^Testing apparatus
The size and s'lopo of the test chambers are not
important as long as they accomodate the loading
requirements. The chambers should however, be large enough
and contain enough water such that the fish are not stressed
by crowding .
The following criteria presented by Hodson (1979) should
be considered when selecting or designing a toxicant
13
-------
ES-7
August, 1982
delivery system: 1) capable of stopping the delivery of the
toxicant if delivery of the dilution water stops, 2)
consistent in delivery amounts throughout the test period,
3) independent of electrical failure, 4) independent of
temperature and humidity fluctuations, 5) capable of
delivering small quantities, 6) easy to construct with few
moving parts and 7) easy to operate.
Any one of several types of toxicant delivery devices
can be used as long as it has been shown to be accurate and
reliable throughout the testing period. The greater the
variation in the quantity of test substance introduced, the
greater the spread1 of response 'values measured during
testing. Syringe injector systems (Barrows et al. 1980,
Spehar et al . J.979), metering pump systems (Veith et al .
1979) and modified proportional diluters (Macek et al. 1975,
Neely et al. 1974) have been used successfully.
The solubility of the test compound should also be taken
into account'when selecting a delivery system. If the
compound can be dissolved in water, a device capable of
delivering amounts of test solution greater than 1 ml will
probably be needed. If a carrier should be used, a system
capable of accurately delivering very small amounts (< 100
ul) will be required to minimize the carrier concentration
in the test solution.
d. Cleaning
Before use, the test system should be cleaned to remove
dust, dirt and other debris and residue that may remain from
the previous use of the system. Any of these substances may
affect the results of a test by sorption of test materials
or by exerting an adverse effect on test organisms. If any
14
-------
ES-7
August, 1982
test chambers or parts of the testing apparatus have
obviously absorbed test compound, those chambers or parts
should be discarded. New chambers should also be cleaned to
remove any chemical or dirt residues that have accumulated
during construction or storage. Detergent is used to remove
hydrophobia or lipid-like substances. Acetone is used for
the same purpose and to remove any detergent residues. It
is important to use pesticide-free acetone to prevent the
contamination of the chambers with pesticides which may-be
toxic to the test organisms or which might otherwise
influence the outcome of the test. Nitric acid'is used to
clea'n metal residues from the system. The final rinse with
water washes away the nitric acid. At the end of a test,
the system should be washed in preparation for the next
test.
Conditioning the test system with dilution water before
it is used allows an equilibrium to be established between
i
the chemicals in the water and the materials i of the test
system. Since a test system may sorb or react with
substances in the dilution water, allowing this equibibrium
to become established before the test begins lessens the
chances of additional changes in water chemistry occurring
during a test.
Even after extensive washing, new facilities still may
contain toxic residues. A good way to determine if toxic
residues remain is to test for their presence by maintaining
fathead minnows in those facilities for a period of time
equal to or exceeding the time required to complete a test.
e. Carriers
Carriers may be used to aid in the dissolution of test
15
-------
ES-7
August, 1982
compounds into dilution water only after significant efforts
to dissolve it in dilution water stocks or test solutions
have failed. Schoor (1975) believes that the use of a
carrier may interfere with the uptake of the test compound
by the test organism; if the carrier molecules affect the
adsorption of the test compound at the gill surface, there
will be a resultant change in the rate of tra'nsport into the
test organism. The author also states that the use of a
carrier may increase the concentration of a compound in the
test solution above solubility by creating a stable water
emuls ion. '
; i i i i i I
When a carrier is necessary, triethylene glycol (TEE),
dimethyl formamide (DMF) or acetone may be used. The
solvents should be tried in ' the order stated due to their
relative toxicity to fathead minnows as reported by Cardwell
et al . (manuscript, 1980). The minimum amount should be
used, and the concentration of TEG should not exceed 80
i
mg/1, the MATC (maximum acceptable toxicant concentration)
value. Concentrations of DMF and acetone should not exceed
5.0 mg/1, the MATC for DMF. Although there is no MATC value
for acetone, its acute toxicity is similar to that of DMF.
4. Environmental Conditions
a. Loading
The flow rate through a test chamber should be high
enough to maintain the dissolved oxygen concentration
greater than 60% of saturation (5.3 mg/1 at sea level and
22°C) , minimize buildup of ammonia, and limit to 30% the
loss of compound by adsorption onto the walls of the test
apparatus and by absorption by the fish.
16
-------
ES-7
August, 1982
Different researchers have used widely different flow
and loading rates. Analysis of the data reported by Veith
et al. (1979) indicated that they used a loading rate of
approximately 0.1 g/l/day and a turnover rate of 12. Spehar
et al. (1979) used a loading rate of 0.125 g/l/day and had
27 aquaria turnovers per day. Barrows et al. (1980) used a
loading rate that ranged from 0.4 to 1.0 g/l/day and a
considerably slower turnover rate of 6-7. With many of the
compounds studied by Barrows however, the compound
concentration in water dropped substantially below what it
was before the fish were introduced And usually took 1 to 3
days to recover to pre-test levels (Personnel communication,
Barrows). In a study by Blanchard et al. (1977) a loading
of 1.5 g/l/day and a turnover rate of 6 were not sufficient
to prevent loss of 14C-sec-butyl-4-chlorodiphenyloxide from
the test water. The concentration of the test substance
decreased more than 50% during the first 12 hours of
exposure and did not return to the expected concentration
until after 72 hours. Such a phenomenon did not occur in
the study by Veith (personal communication) where a higher
flow rate was used.
We recommend a maximum loading rate of 0.1 g/l/day and a
minimum turnover rate of 6. An even lower loading may be
needed if the compound is suspected to readily degrade/ is
highly volatile or is expected to accumulate quickly and
substantially in fish.
b. Dissolved Oxygen (See Loading)
c. Temperature
Since fish are poikilothermic, most biochemical
activities are affected by the water temperature to which
17
-------
ES-7
August, 1982
they are exposed. Although there are some exceptions,
Prosser (1973) states that there is approximately a 2 fold
increase in fish metabolism for each 10°C rise in water
tempe rature.
During 95-hour studies with rainbow trout and methyl
mercuric chloride HgCl, MacLeod and Pessah (1973) found that
accumulation of mercury increased with temperature and that
this increase was due to an increase in metabolic rate.
Re inert et al . (1974) exposed rainbow trout to HgCl and
DDT separately, at 5, 10 and 15°C and found that
accumulation of mercury increased 78% between 5 and 15°C
: i ;
and DDT accumulation increased l'40% over the same
temperature range. They stated that this increase was not
due to the intrinsic nature of the chemicals but due to the
increased metabolism of the fish.
•In the study by Veith et al. (1979), fathead minnows
were exposed to Aroclor 1254® at 5, 10, 15, 20 and 25°C.
The log BCF's increased substantially between 5,'10 and 15°C
and slightly between 20 and 25°C. There was little
difference between log BCF's at 15 and 20°C.
Although researchers have performed apparently
successful tests at 16°C (Barrows et al. 1980), 20°C (Macek
et al . 1975) and 25°C (Veith et al . 1979), and there are
some indications that greater BCF's will be generated at
increased temperatures, we recommend testing at 22 ± 1°C.
As testing at 25°C may induce sexual maturation (U.S. EPA
1971), the test temperature should be less than 25°C. A
test temperature of 22°C is also consistent with the test
temperature recommended in a similar TSCA test guideline for
acute toxicity tests with fathead minnows. Having identical
18
-------
ES-7
August, 1982
test temperatures will limit the number of separate fish
populations required to be held and acclimated at the
testing laboratory.
d . Light
Although many researchers have used a 16 hour light - 3
hour dark photoperiod (Neely et al 1974, Spehar et al.
1979, Veith et al . 1979) and an ASTM task group has
recommended its use in Draft 10 of the Proposed Practice for
Conducting ' Bioconcentration Tests with Fishes and Saltwater
Bivalve Molluscs (1980b), there is no scientific
justification given for its use.
To retard gamete maturation, a photoperiod of 12 hours
light-12 hours dark with a 15-30 minute transition period is
recommended.
C. Report ing
An estimate of the time to steady state, the steady-
otnte 3CF, and the time to 50% or 95% elimination should be
made for each compound tested. If. steady-state has not been
observed during the maximum 28 day exposure period or if 95%
elimination has not been achieve during 14 days depuration,
data generated during these tests should be used to estimate
these values. The 3IOFAC program developed by Blau and A-^in
(1978) uses nonlinear regression techniques to estimate the
uptake and depuration rate constant, the steady-state BCF,
the time to reach 90% of steady-state, the time to reach 50%
elimination, and the variability associated with each
es t imate.
III. Economic Aspects
The Agency awarded a contract to Enviro Control,
Inc. to provide us with an estimate of the cost for
19
-------
ES-7
August, 1982
performing a bioconcentration test using the fathead minnow
according to this Guideline. Enviro Control supplied us
with two estimates; a protocol estimate and a laboratory
survey estimate.
The protocol estimate was $8,274. This estimate was
prepared by identifying the major tasks needed to do a test
and estimating the hours to accomplish each task.
Appropriate hourly rates were then applied to yield a total
direct labor charge. An estimated average overhead rate of'
115%, other direct costs of $300, a general and
administrative rate of 10%, and a fee of 20% were then added
to the direct labor charge to yield the final estimate.
Enviro Control estimated that differences in salaries,
equipment, overhead costs and other factors between
laboratories could result in as much as 50% variation from'
this estimate. Consequently, they estimated that test costs
could range from $4,137 to $12,411.
The laboratory survey estimate was $10,938; the mean of
the estimates received from four laboratories. The
estimates ranged from $6,000 to $15,750 and were based on
the costs to perform the test according to this Guideline.
Enviro Control listed the following as possible sources of
variation in the cost estimates:
o understanding the Guideline
o overhead rates
o salary rates
o in-house expertise
o worker productivity and efficiency
o degree of automation
o accuracy of protocol costing procedures
20
-------
ES-7
August, 1982
IV. REFERENCES
APHA. 1975. American Public Association, Ameriocan
Water Works Association, and Water-Pollution Control
Federation. Standard Methods for Examination of Water
and Wastewater, 14th ed. New York: American Public
Health Association
ASTM. 1930a. American Society for Testing and
Materials. Standard Practice for Conducting Basic Acute
Toxicity Tests With Fishes, Macroinve cte'orates, and
Amphibians. E729-80.
ASTM. 1980b. American Society for Testing and
Materials. Proposed Standard Practice for Conducting
Bioconcentration Tests With Fishes and Saltwater Bivalve
Molluscs -. ' Draft No. 10, August 22,^ 1980.
.Barrows ME, PetrocelLi SR, Macek KJ. 1980.
Bioconcentration and elimination of selected water
pollutants by bluegill sunfish ( Lepomis macroch ira s ) .
In: Hague R, ed. Dynamics, Exposure and Hazard
.Assessment of Toxic Chemicals. Ann Arbor, Michigan:
Arbor Science Pub., Inc.
Bishop WE, Make AW. 1980. A critical- comparison of two
biooO'vr>5_atrat ion test methods. In: Eaton ,jn,. Parrish
PR, Hend ricks AC, eds . Aquatic Toxicology. ASTM STP
707. American Society for Testing and Materials: pp.
61-73.
Carmignani GM, Bennett JP- 1976. Leaching of plastics
used in closed aquaculture system. Aquaculture 7:89-91.
Carroll JJ, Ells S J, Oliver W3. 1979. Influences of
hardness constituents on the acute toxicity of cadmium
to brook trout (Salvelinus fontinalis). Bull. Environm.
Contam. Toxicol. 22:575-581.
Cember H, Curtis EH, Blaylock BG. 1978. Mercury
bioconcentration in fish: temperature and concentration
effects. Environm. Pollution 17:311-319./
21
-------
ES-7
August, 1982
Chiou CT, Freed VH, Schmeddiny DW, Kohnert RL. 1977.
Partition coefficient and bioaccumulation of selected
organic chemicals. Environ. Sci. Techno. 11:475-478
DeFoe DL, Veith GD, Carlson RW. 1978. Effects of
Aroclor@ 1243 and 1260 on the fathead minnow (Pimephales
promelas). J. Fish. Res. Board Can. 35:997-1002.
Folmar LC, Sanders HO, Julin AM. 1979. Toxicity of the
herbicide glyphosate and several of its formulati.oas to
fish and aquatic invertebrates. Arch. Environm. Contam.
Toxicol. 8:269-278.
Blanchard FA, Takahashi IT, Alexander HC, Bartlett Ea.
1977. Uptake, clearance and bioconcentration of 14C-
Sec-4 chlorodiphenyl oxide in rainbow trout. In: Mayer
FL, Hanelink JL,i eds. Aquatic Toxicology and Hazard!
Evaluation. ASTM STP 634. American Society for Testing
and Materials: pp. 162-177.
Blau GE, Agin GL. 1978. A users manual for BI OF AC; A
computer program for characterizing the ratio of uptake
and clearance of chemicals in aquatic organisms. Dow
Chemical Co. Midland, Michigan.
Branson DR, BlauGE, Alexander HC, Neely WB. 1975.
Bioconce-i trration of 2,2,4,4, - tetrachlorobiphenyl in
rainbow trout as measured by an accelerated test. Tran.
Am. Fish. Soc. 4:735-792.
Brauhn JL, Schoettger RA. 1975. Aquisition and culture
of research fish: rainbow trout, fathead minnow, channel
catfish and bluegill. Corvallis, Oregon: U.S.
Environmental Protection Agency. EPA-660/3-75-011.
Cairns J. 1969. Fish bioassay - reproducibil i ty and
rating. Revista be Biologia. 7:7-13.
Cardwell RP, Foreman DG, Payne TR, Wilbur D J. 1930.
Acute and chronic toxicity of four organic chemicals to
fish. Manuscript.
22
-------
ES-7
August, 1982
Ganz CR. , Schulze J, Stensby PS, Lyman PL,. Macek KJ.
1975. Accumulation and elimination studies of four
detergent fluorescent whitening agents in bluegill
(Lepomis macrochirus). Environment. Sci. Techno.
8:738-744
Hamelink JL. 1977. Current bioconcentration test
methods and theory. In: Mayer FL, Hamelink JL. eds .
Aquatic Toxicology and Hazard Evaluation. ASTM STP
634. American Society for Testing and Materials. oo.
149-161.
Hansch C, Leo A, Nikaitani D J. 1972. On the additive-
cohstitutive character oC partition coefficients. Org.
Chem. 37:3090-3092.
Hodson RV. 1979. Metering device for toxicants used in
bioassays with aquatic organisms. Prog. Fish Cult.
41(3) :129-1,3^..
Holcomb GW. And rev/ RN. 1978. The acute toxicity of
zinc to rainbow and brook trout. Duluth, Minnesota:
U.S. Environmental Protection Agency. EPA-6 00/3-78-0 94 .
Howarth RS, Sprague JB. 1978. Copper lethality to
rainbow trout in waters of various hardness and pH.
Water Research 12:455-462.
Kenaga EE, Goring CAI. 1980. Relationship between
water solubility, soil-sorption, octanol-water
pactitioning, and concentration of chemicals in biota.
In: Eaton JG, Parrish PR, Hendricks AC, eds. Aquatic
Toxicology. ASTM STP 707. American Society for Testing
and Materials. pp. 78-115.
Krzeminski SF, Gilbert JT, Ritts JA. 1977. A
pharmacokinetic model for predicting pesticide residues
in fish. Arch. Environ. Contarn. Toxicol. 5(2):157-166.
Leo A, Hansch C, Elkins D. 1971. Partition
coefficients and their uses. Chem. Reviews 7:525-616.
Macek KJ, Barows ME, Kras ny RF, Sleight BH III. 1975.
Bioconcentration of 14C=pesticides by bluegill sunfish
during continuous aqueous exposure. In: Veith GD,
Konasewich DE, eds. Sypos ium on Structure-activity
Correlation in Studies of Toxicity and Bioc~>-"icentration
with Aquatic Organisms. Windsor Ontario: IJC, Great
Lakes Research Advisory 3rd: pp. 119-141.
23
-------
Augus t,
ES-7
1982
Maced KJ, Lindberg M, Sauter S, Buxton KS, Costa PA.
1976. Toxicity of four pesticides to water fleas and
fathead minnows. Duluth, Minnesota:
Protection Agency. EPA 600/3-76-099.
U.S. Environmental
Macek KJ, Sleight SH III. 1977- Utility of toxicity
t:escs with embryos and fry of fish in evaluating hazards
associated with the chronic toxicity of chemicals to
fishes. In: Mayer FL, Hamelink JL, eds. Aquatic
Toxicology and Hazard Evaluation. ASTM STP 634.
American Society for Testing and Materials
137-146,
MacLeod JC, Pessah E. 1973.. Temperature effects on
mercury accumulation,, toxicity and metabolic rate in
rainbow trout (Salmo gairdneri). J. Fish. Res. Board
Can. 30:435-492";
Mayer FL.
pnthalate
Fish. Res
1976. Residue dynamics of di-2-ethylheyl
in fathead minnows'' ( Pimephalesm oromelas ) .
Board Can. 33:2610-2613.
Mayer FL, Sanders HO. 1973.
esters in aquatic organisms.
Perspective 3:153-157,
Toxicity of phthalic acid
Environmental Health
Mayer FL, Mehrle PM Jr., Dwyer WP. 1977. Toxap'nene:
chronic toxicity to fathead minnows and channel
catfish. Duluth, Minnesota: U.S. Environmental
Protection Agency. EPA-600/3-77-069.
Mauck WL, Olson LE, Marking LL. 1976. Toxicity of
natural pyrethriny and five pyrethroids to fish. Arch,
Environ. Contamin. Toxicol. 4:18-29.
Mauck WL, Olson LE, Hogan JW. 1977. Effects of water
quality on deactivation and toxicity of mexacarbate
(Zectron©) to fish. Arch, Environ. Contam. Toxicol.
6;385-393.
McKim JM. 1977. Evaluation of tests with early life
stages of fish for predicting long-term toxicity. J.
Fish. Res. Board Can. 34:1148-1154.
24
-------
ES-7
August, 1982
Melancon MJ Jr., Lech JJ. 1979. Uptake,
b iotrans forma tion, disposition, and elimination of 2-
methylnaphthalene and naphthalene in several fish
species. In: Marking LL, KimecLe RA, eds . Aquatic
Toxicology. ASTM STP 667. American Society of Testing
and Materials: pp. 5-22.
Menendez, R. 1976. Chronic effects of reduced pH on
brook trout (Salvelinus fontinalis ) . J. Fish. Res.
"Octrd. Can. 33:118-123.
Mount DI. 1973. Chronic effect of low pH on fatm-.-. 1
minnow survival, growth and reproduction. Wat or-
Research 7-987-993.
Nebeker AV, Puglisi FA, DeFoe DL. 1974. Effect of
polychlorinated byphenyl compounds on survival and
reproduction of the fathead minnow and flagf ish. Trans.
Am. Fish. Soc. 103:562-568. '
Neely WB, Branson DR, Blau GF. 1974. Partition
coefficient to measure bioconcentration potential of
organic chemicals in fish. Environ. Sci. Technol,
8:1113-1115.
Peterson RH, Anderson JM. 1969. Influence of
temperature change on spontaneous locomotor activity and
oxygen consumption of atlantic salmon (Salmo solar)
acclimated to two temperatures. J. Fish. Res. Board
Can. 26:93-109.
Prosser CL. 1973. Comparative Animal Physiology, 3rd
ed. Philadelphia: W.B. Saunders Co.
Reinert RE, Stone LJ, Willford W J. 1974. Effects of
temperature on accummulation of methylmercuric chloride
and p,p DDT by rainbow trout (Salmo gairdneri). J. Fish.
Res. Board Can. 31:1649-1652.
Sauter S, Buxton KS, Macek KJ, Petrocelli SR. 1976.
Effects of exposure to heavy metals on selected
freshwater fish. Duluth, Minnesota; U.S. Environmental
Protection Agency. EPA-600/3-76-105.
25
-------
REPORT DOCUMENTATION i._REPORT NO.
PAGE EPA 560/6-82-002 Part_J-
4. Title and Subtitle
Environmental Effects Test Guidelines
3. Recipient's Accession No
PB82-232992
1 5. Report Date
August, 1982
I 6.
7. Author(s)'
8. Performing Organization Rept. No.
9. Performing Organization Name and Address
i 10. Proiect/Task/Work Unit No.
Office of Pesticides and Toxic Substances
Office of Toxic Substances (TS-792)
United States Environmental Protection Agency
401 M Street, S.W.
Washington, D.C. 20460
, 11. Contract(C) or Grant(G) No.
I (G)
12. Sponsoring Organization Name and Address
13. Type of Report & Period Covered
Annual Report
15. Supplementary Notes
16. Abstract (Limit: 200 words)
These documents consitute a set of 21 environmental effects test guidelines (and,
in some cases, support documents) that may be cited as methodologies to be used
in chemical specific test rules promulgated under Section 4(a) of the Toxic
Substances Control Act (TSCA). These guidelines cover testing for invertebrate
toxicity, aquatic vertebrate toxicity, avian toxicity, phytotoxicity, and
bioconcentration. The guidelines will be published in loose leaf form and
updates will be made available as changes are dictated by experience and/or
advances in the state-of-the-art.
17. Document Analysis a. Descriptors
b. Identifiers/O^ien-Ended Terms
... COSATI Field/Group
18. Availability Statement
Release unlimited
\ 19. Security Class (This Report)
Unclassified
21. No. of Pages
492-
20. Security Class (This Page)
Unclassified
22, Price
(See ANSI-Z39.18)
See /nsfructions on Reverse
OPTIONAL FORM 272 (4-77)
(Formerly NTIS-35)
Oeoartment of Commerce
------- |