(Revised January, 1972)

Proposed Bioassay Procedures are established by the approval of both
the Committee on Aquatic Bioassays and the Director of the National
Water Quality Laboratory. The main reasons for establishing than are:
(1) to permit direct comparison of test results, (2) to encourage
the use of the best procedures available, and (3) to encourage
uniformity. These procedures should be used by National Water Quality
Laboratory personnel whenever possible, unless there is a good reason
for using some other procedure.
Proposed Bioassay Procedures consider the basic elements that are
believed to be important in obtaining reliable and reproducible
results in laboratory bioassays. An attempt has been made to adopt
the best acceptable procedures based on current evidence and opinion,
although it is recognized that alternative procedures may be adequate.
Improvements in the procedures are bing considered and tested, and
revisions will be made when necessary. Comments and suggestions are
Committee on Aquatic Bioassays
Director, National Water Quality Laboratory

Proposed Bioassay Procedure for
Fathead Minnow Pimephales promelas Rafinesque Chronic Tests
(Revised January, 1972)
A. Physical system
1.	Diluter: Proportional diluters (Mount and Brungs, 1967) should
be employed for all long-term exposures. Check the operation
of the diluter daily, either directly or indirectly, through
measurement of toxicant concentrations. A minimum of five
toxicant concentrations and one control should be used for
each test with a dilution factor of not less than 0.30. An
automatically triggered emergency aeration and alarm system
must be installed to alert staff in case of diluter or water
supply failure.
2.	Toxicant mixing: A container to promote mixing of toxicant
bearing and w-cell water should be used between diluter and
tanks for each concentration. Separate delivery tubes
should run from this container to each duplicate tank.
Check at least once every month to see that the intended
amounts of water are going to each duplicate tank or chamber.
3.	Tank: Two arrangements of test tanks (glass, or stainless
steel with glass ends) can be utilized:
a.	Duplicate spawning tanks measuring 1 x 1 x 3 ft. long
with a one sq. ft. portion at one end screened off
and divided in half for the progeny. Test water is
to be delivered separately to the larval and spawning
chambers of each tank, with about one-third the water
volume going to the former chamber as to the latter.
b.	Duplicate spawning tanks measuring 1 x 1 x 2 ft. long
with a separate duplicate progeny tank for each
spawning tank. The larval tank for each spawning
tank should be a minimum of 1 cu. ft. dimensionally
and divided to form two separate larval chambers with
separate standpipes, or separate 1/2 sq. ft. tanks
may be used. Test water is to be supplied by delivery
tubes from the mixing cells described in Step 2 above.
Test water depth in tanks and chambers for both a & b
above should be 6 inches.
4.	Flow rate: The flow rate to each chamber (larval or adult)
should be equal to 6 to 10 tank volumes/24 hr.

5.	Aeration: Total dissolved oxygen levels should never be allowed
to drop below 60% of saturation, and flow rates must be Increased
If oxygen levels do drop below 60%. As a first alternative flow
rates in the spawning tanks can be increased above those specified
in A.4. Aerate with oil free air only if testing a non-
volatile toxic agent and then only as a last resort to maintain
dissolved oxygen at 602 of saturation.
6.	Cleaning: All adult tanks, and larvae tanks and chambers
after larvae swim-up, oust be siphoned a minimum of 2
times weekly and brushed or scraped when algal or fungus
growth becomes excessive.
7.	Spawning substrate; Use spawning substrates made from
inverted cement and asbestos halved, 3-inch ID drain tile,
or the equivalent, each of these being 3 inches long.
8.	Egg cup: Egg incubation cups are made from either 3-inch
sections of 2-inch OD (1 1/2-inch ID) diameter polyethylene
water hose or 4-oz., 2-inch OD round glass jars with the
bottoms cut off. One end of the jar or hose sections is
covered with stainless steel or nylon screen (with a minimum
of 40 meshes per inch). Cups are oscillated in the test
water by means of a rocker arm apparatus driven by a 2 r.p.m.
electric motor (Mount, 1968). The vertical-travel distance
of the cups should be 1 to 1 1/2 inches.
9.	Photoperiod: The photoperiods to be used (Appendix A)
simulate the dawn to dusk times of Evansville, Indiana.
Adjustments in day-length are to be made on the first
and fifteenth day of every Evansville test month. The
table is arranged so that adjustments need be made only
in the dusk times. Regardless of the actual date that
the experiment is started, the Evansville test photo-
period should be adjusted so that the mean or estimated
hatching date of the fish used to start the experiment
corresponds to the Evansville test day-length for
December first. Also, the dawn and dusk times listed
in the table need not correspond to the actual times
where the experiment is being conducted. To illustrate
these points, an experiment started with 5-day-old
larvae in Duluth, Minnesota, on August 28 (actual
date), would require use of a December 5 Evansville
test photoperiod, and the lights could go on anytime
on that day just so long as they remained on for 10
hours and 45 minutes. Ten days later (Sept. 7 actual
date, Dec. 15 Evansville test date) the day-length

would be changed to 10 hours and 30 minutes. Gradual
changes In light Intensity at dawn and dusk (Drummond
and Dawson, 1970), 1 desired, should be Included
within the day-lengths shown, and should not last for
more than 1/2 hour from full on to full off and vice
10.	Temperature: Temperature should not deviate instantaneously
from 25 C by more than 2 C and should not remain outside the
range of 2k to 26 C for more than 48 hours at a time.
Temperature should be recorded continuously.
11.	Disturbance: Adults and larvae should be shielded from
disturbances such as people continually walking past the
chambers, or from extraneous lights that might alter the
Intended photoperiod.
12.	Toxic materials: Construction materials which contact the
diluent water should not contain leachable substances and
should not sorb significant amounts of substances from the
water. It is best to avoid such problems when possible, rather
than to try to overcome them through chemical measurements.
Even though what is significant will vary from test to test
and most materials sorb and/or leach small amounts of at least
some substances, some generalizations are possible. Stainless
steel is probably the preferred construction material. Glass
absorbs some trace organics significantly. Rubber should not
be used. Plastic containing fillers, additives, stabilizers,
plasticizers, etc., should not be used. Teflon, nylon, and
their equivalents should not contain leachable materials and
should not sorb significant amounts of most substances. Un-
plasticized polyethylene and polypropylene should not contain
leachable substances, but may sorb very signficant amounts of
trace organic compounds.
The water used should be from a well or spring if at all
possible, or alternatively from a surface water source. Only
as a last resort should water from a chlorinated municipal
water supply be used. If it is thought that the water supply
could be conceivably contaminated with fish pathogens, the
water should be passed through an ultraviolet or similar
sterilizer immediately before it enters the test system.
B. Biological system
1. Test animals: If possible, use stocks of fathead minnows from
the National Water Quality Laboratory in Duluth, Minnesota or
the Fish Toxicology Laboratory in Newtown, Ohio. Groups of
starting fish should contain a mixture of approximately equal
numbers of eggs or larvae from at least three different females.

Set aside enough eggs or larvae at the start of the test to
supply an adequate number of fish for the acute mortality
bloassays used in determining application factors.
2.	Beginning test; In beginning the test, distribute 40 to
50 eggs or 1 to 5-day-old larvae per duplicate tank using
a random assignment. All acute mortality tests should be
conducted when the fish are 2 to 3 months old. If eggs
or 1 to 5-day-old larvae are not available, fish up to
30 days of age may be used to start the test. If fish
between 20 and 60 days old are used, the exposure should
be designated a partial chronic test. Extra test animals
may be added at the beginning so that fish can be removed
periodically for special examinations (see B.12.) or for
residue analysis (see C.4.).
3.	Food: Feed the fish a frozen trout pellet (e.g., Oregon
Hoist). A minimum of once dally fish should be fed ad
libitum the largest pellet they will take. Diets should
be supplemented weekly with live or frozen-live food
(e.g., Daphnla, chopped earthworms, fresh or frozen brine
shrimp, etc.). Larvae should be fed a fine trout starter
a minimum of 2 times daily, ad libitum; one feeding each
day of live young zooplankton from mixed cultures of
small copepods, rotifers, and protozoans is highly
recommended. Live food is especially important when
larvae are just beginning to feed, or about 8 to 10 days
after egg deposition. Each batch of food should be
checked for pesticides (including DDT, TDE, dieldrin,
lindane, methoxychlor, endrin, aldrin, BHC, chlordane,
toxaphene, 2,4-D, and PCBs), and the kinds and amounts
should be reported to the project officer or recorded.
4.	Disease: Handle disease outbreaks according to their
nature, with all tanks receiving the same treatment
whether there seems to be sick fish in all of them or
not. The frequency of treatment should be held to a
5.	Measuring fish: Measure total lengths of all starting fish
at 30 and 60 days by the photographic method used by McKim
and Benolt (1971). Larvae or juveniles are transferred
to a glass box containing 1 inch of test water. Fish
should be moved to and from this box in a water-filled
container, rather than by netting them. The glass box
is placed on a translucent millimeter grid over a
fluorescent light platform to provide background
Illumination. Photos are then taken of the fish over

the millimeter grid and are enlarged Into 8 by 10 inch
prints. The length of each fish is subsequently
determined by comparing it to the grid. Keep lengths of
discarded fish separate from those of fish that are to be
6.	Thinning; When the starting fish are sixty (+ 1 or 2) days
old, impartially reduce the number of surviving fish in
each tank to 15. Obviously injured or crippled Individuals
may be discarded before the selection so long as the number
is not reduced below 15; be sure to record the number of
deformed fish discarded from each tank. As a last resort in
obtaining 15 fish per tank, 1 or 2 fish may be selected for
transfer from one duplicate to the other. Place five spawning
tiles in each duplicate tank, separated fairly widely to reduce
interactions between male fish guarding them. One should
also be able to look under tiles from the end of the tanks.
During the spawning period, sexually maturing males must be
removed at weekly intervals so there are no more than four
per tank. An effort should be made not to remove those
males having well established territories under tiles where
recent spawnings have occurred.
7.	Removing eggs: Remove eggs from spawning tiles starting at
12:00 noon Evansville test time (Appendix A) each day.
As indicated in Step A.9., the test time need not correspond
to the actual time where the test is being conducted. Eggs
are loosened from the spawning tiles and at the same time
separated from one another by lightly placing a finger on
the egg mass and moving it in a circular pattern with
increasing pressure until the eggs being to roll. The
groups of eggs should then be washed into separate,
appropriately marked containers and subsequently handled
(counted, selected for incubation, or discarded) as soon as
possible after all eggs have been removed and the spawning
tiles put back into the test tanks. All egg batches must
be checked initially for different stages of development.
If it is determined that there is more than one distinct
stage of development present, then each stage must be
considered as one spawning and handled separately as
described in Step B.8.
8.	Egg incubation and larval selection; Impartially select
50 unbroken eggs from spawnings of 50 eggs or more and
place them in an egg incubator cup for determining
viability and hatchability. Count the remaining eggs and
discard them. Viability and hatchability determinations
must be made on each spawning (>49 eggs) until the number
of spawnings (>49 eggs) in each duplicate tank equals the

number of females in that tank. Subsequently, only eggs
from every third spawning (>49 eggs) and none of those
obtained on weekends need be set up to determine hatch-
ability; however, weekend spawns must still be removed from
tiles and the eggs counted. If unforseen problems are
encountered in determining egg viability and hatchability,
additional spawnings should be sampled before switching to
the setting up of eggs from every third spawning. Every
day record the live and dead eggs in the incubator cups,
remove the dead ones, and clean the cup screens. Total
numbers of eggs accounted for should always add up to
within two of 50 or the entire batch is to be discarded.
When larvae begin to hatch, generally after 4 to 6 days,
they should not be handled again or removed from the egg-
cups until all have hatched. Then, if enough are still
alive, 40 of these are eligible to be transferred
immediately to a larval test chamber. Those individuals
selected out to bring the number kept to 40 should be
chosen impartially. Entire egg-cup-groups not used for
8urvival and growth studies should be counted and
9. Progeny transfer: Additional important information on
hatchability and larval survival is to be gained by
transferring control eggs immediately after spawning to
concentrations where spawning is reduced or absent, or
to where an affect is seen on survival of eggs or larvae,
and by transferring eggs from these concentrations to
the control tanks. One larval chamber in, or corresponding
to, each adult tank should always be reserved for eggs
produced in that tank.
10. Larval exposure: From early spawnings in each duplicate
tank, use the larvae hatched in the egg incubator cups
(Step B.8. above) for 30 or 60 day growth and survival
exposures in the larval chambers. Flan ahead in setting
up eggs for hatchability so that a new group of larvae is
ready to be tested for 30 or 60 days as soon as possible
after the previously tested group comes out of the larval
chambers. Record mortalities, and measure total lengths
of larvae at 30 and, if they are kept, 60 days post-
hatch. At the time the larval test is terminated they
should also be weighed. No fish (larvae, juveniles, or
adults) should be fed within 24 hr's. of when they are to
be weighed.

11.	Parental termination: Parental fish testing should be
terminated when, during the receding day-length photo-
period, a one week period passes in which no spawning
occurs in any of the tanks. Measure total lengths and
weights of parental fish; check, sex and condition of
gonads. The gonads of most parental fish will have
begun to regress from the spawning condition, and thus
the differences between the sexes will be less distinct
now than previously. Males and females that are readily
distinguishable from one another because of their
external characteristics should be selected initially for
determining how to differentiate between testes and
ovaries. One of the more obvious external characteristics
of females that have spawned is an extended, transparent
anal canal (urinogenital papilla). The gonads of both
sexes will be located Just ventral to the kidneys. The
ovaries of the females at this time will appear transparent,
but perhaps containing some yellow pigment, coarsely
granular, and larger than testes. The testes of males
will appear as slender, slightly milkly, and very finely
granular strands. Fish must not be frozen before making
these examinations. Use discarded fish for tissue residue
analysis (skin, bone, muscle, gill, brain, liver, kidney,
GI tract, and gonad should be considered) and physiological
measurements of toxicant related effects.
12.	Special examinations: Fish and eggs obtained from the test
should be considered for physiological, biochemical, histological
and other examinations which may indicate certain toxicant
related effects.
13.	Necessary data; Data that must be reported for each tank of a
chronic test are:
a.	Number and individual total length of normal and deformed
fish at 30 and 60 days; total length, weight and number
of either sex, both normal and deformed, at end of test.
b.	Mortality during the test.
c.	Number of spawns and eggs.
d.	Hatchability.
e.	Fry survival, growth, and deformities.

C. Chemical sy9tem
1.	Preparing a stock solution; If a toxicant cannot be introduced
Into the test water as is, a stock solution should be prepared
by dissolving the toxicant in water or an organic solvent.
Acetone has been the most widely used solvent, but dimethylformanide
(DMF) and triethylene glycol may be preferred in many cases.
If none of these solvents are acceptable, other water-miscible
solvents such as methanol, ethanol, isopropanol, acetonitrile,
dimethylacetamide (DMAC), 2-ethoxyethanol, glyme (dimethylether
of ethylene glycol, diglyme (dimethyl ether of diethylene glycol)
and propylene glycol should be considered. However, dimethyl
sulfoxide (DMSO) should not be used if at all possible because
of its biological properties.
Problems of rate of solubilization or solubility limit should be
solved by mechanical means if at all possible. Solvents, or as
a last resort, surfactants, can be used for this purpose, only
after they have been proven to be necessary in the actual test
system. The suggested surfactant is p-tert-octylphenoxynonaethoxy-
ethanol (p-1, 1, 3, 3-tetramethylbutylphenoxynonaethoxyethanol,
0PE,q) (Triton X-100, a product of the Rohm and Haas Company, or
The use of solvents, surfactants, or other additives should be
avoided whenever possible. If an additive is necessary, reagent
grade or better should be used. The amount of an additive used
should be kept to a minimum, but the calculated concentration of
a solvent to which any test organisms are exposed must never exceed
one one-thousandth of the 96-hr. TL50 for test species under the
test conditions and must never exceed one gram per liter of water.
The calculated concentration of surfactant or other additive to
which any test organisms are exposed must never exceed one-twentieth
of the concentration of the toxicant and must never exceed one-tenth
gram per liter of water. If any additive is used, two sets of
controls must be used, one exposed to no additives and one exposed
to the highest level of additives to which any other organisms
in the test are exposed.
2.	Measurement of toxicant concentration: As a minimum the
concentration of toxicant must be measured in one tank at each
toxicant concentration every week for each set of duplicate
tanks, alternating tanks at each concentration from week to
week. Water samples should be taken about midway between the
top and bottom and the sides of the tank and should not include
any surface scum or material stirred up from the bottom or sides
of the tank. Equivolume daily grab samples can be composited
for a week if it has been shown that the results of the analysis
are not affected by storage of the sample.

Enough grouped grab samples should be analyzed periodically
throughout the test to determine whether or not the concentration
of toxicant Is reasonably constant from day to day In one tank
and from one tank to its duplicate. If not, enough samples must
be analyzed weekly throughout the test to show the variability
of the toxicant concentration.
3.	Measurement of other variables: Temperature must be recorded
continuously (see A.10.).
Dissolved oxygen must be measured in the tanks daily, at least
five days a week on an alternating basis, so that each tank is
analyzed once each week. However, if the toxicant or an additive
causes a depression In dissolved oxygen, the toxicant concentration
with the lowest dissolved oxygen concentration must be analyzed
daily in addition to the above requirement.
A control and one test concentration must be analyzed weekly for
pH, alkalinity, hardness, acidity, and conductance or more often,
if necessary, to show the variability in the test water. However,
If any of these characteristics are affected by the toxicant
the tanks must be analyzed for that characteristic daily, at
least five days a week, on an alternating basis so that each
tank is analyzed once every other week.
At a minimum, the test water must be analyzed at the beginning
and near the middle of the test for calcium, magnesium, sodium,
potassium, chloride, sulfate, total solids, and total dissolved
4.	Residue analysis: When possible and deemed necessary, mature
fish, and possibly eggs, larvae, and juveniles, obtained from
the test, should be analyzed for toxicant residues. For fish,
muscle should be analyzed, and gill, blood, brain, liver, bone,
kidney, GI tract, gonad, and skin should be considered for
analysis. Analyses of whole organisms may be done in addition
to, but should not be done in place of, analyses of individual
tissues, especially muscle.
5.	Methods; When they will provide the desired information with
acceptable precision and accuracy, methods described in Methods
for Chemical Analysis of Water and Wastes (EPA, 1971) should be
used unless there is another method which requires much less time
and can provide the desired information with the same or better
precision and accuracy- At a minimum, accuracy should be measured
using the method of known additions for all analytical methods
for toxicants. If available, reference samples should be
analyzed periodically for each analytical method.

D.	Statistics
1.	Duplicates; Use true duplicates for each level of toxic agent,
i.e., no water connections between duplicate tanks.
2.	Distribution of tanks: The tanks should be assigned to locations
by stratified random assignment {random assignment of one tank
for each level of toxic agent in a row followed by random assign-
ment of the second tank for each level of toxic agent in another
ot an extension of the same tow).
3.	Distribution of test organisms: The test organisms should be
assigned to tanks by stratified random assignment (random assignment
of one test organism to each tank, random assignment of a second
test organism to each tank, etc.). At time of thinning (B.4.) the
choice of males and females must also be made randomly.
E.	Miscellaneous
1.	Additional information: All routine bioassay flow through methods
not covered in this procedure (e.g., physical and chemical
determinations, handling of fish) should be followed as
described in Standard Methods for tie Examination of Water and
Wastewater, (American Public Health Association, 1971), or
information requested from appropriate persons at Duluth or
New town.
2.	Acknowledgments: These procedures for the fathead minnow
were compiled by John Eaton for the Committee on Aquatic
Bioassays. The participating members of this committee are:
Robert Andrew, John Arthur, Duane Benoit, Gerald Bouck,
William Brungs, Gary Chapman, John Eaton, John Hale,
Kenneth Hokanson, James McKim, Quentin Pickering, Wesley
Smith, Charles Stephan, and James Tucker.
3.	References: For additional information concerning flow
through bioassays with fathead minnows, the following
references are listed:
American Public Health Association. 1971. Standard
methods for the examination of water and wastewater.
13th ed. APHA. New York.
Brungs, William A. 1969. Chronic toxicity of zinc to the
fathead minnow, Pimephales promelas Rafinesque. Trans. Amer.
Fish. Soc., 98(2): 272-279.

Carlson, Dale R. 1967. Fathead minnow, Plmephales promelas
Raflnesque, in the Des Moines River, Boone County, Iowa, and
the Skunk River drainage, Hamilton and Story Counties, Iowa.
Iowa State Journal of Science, 41(3): 363-374.
Drummond, Robert A., and Walter F. Dawson. 1970. An
inexpensive method for simulating Diel patterns of lighting
in the laboratory. Trans. Amer. Fish. Soc., 99(2): 434-435.
Isaak, Daniel. 1961. The ecological life history of the
fathead minnow, Plmephales promelas (Rafinesque). Ph.D.
Thesis, Library, Univ. of Minnesota.
Markus, Henry C. 1934. Life history of the fathead minnow
(Plmephales promelas). Copeia, (3): 116-122.
McKim, J. M., and D. A. Benoit. 1971. Effect of long-term
exposures to copper on survival, reproduction, and growth
of brook trout Salvelinus fontinalis (Mitchill). J. Fish.
Res. Bd. Canada, 28: 655-662.
Mount, Donald I. 1968. Chronic toxicity of copper to
fathead minnows (Plmephales promelas, Rafinesque). Water
Research, 2: 215-223.
Mount, Donald I., and William Brungs. 1967. A simplified
dosing apparatus for fish toxicology studies. Water Research,
1: 21-29.
Mount, Donald I., and Charles E. Stephan. 1967. A method
for establishing acceptable toxicant limits for fish 
malathion and the butoxyethanol ester of 2,4-D. Trans.
Amer. Fish. Soc., 96(2): 185-193.
Mount, Donald I., and Charles E. Stephan. 1969. Chronic
toxicity of copper to the fathead minnow (Plmephales promelas)
in soft water. J. Fish. Res. Bd. Canada, 26(9): 2449-2457.
Mount, Donald I., and Richard E. Warner. 1965. A serial-
dilution apparatus for continuous delivery of various
concentrations of materials in water. PHS Publ. No. 999-
WP-23. 16 pp.
Pickering, Quentin H., and Thomas 0. Thatcher. 1970. The
chronic toxicity of linear alkylate sulfonate (LAS) to
Pimephales promelas, Rafinesque. Jour. Water Poll. Cont.
Fed., 42(2): 243-254.

Pickering, Quentln H., and William N. Vigor. 1965. The
acute toxicity of zinc to eggs and fry of the fathead
minnow. Progressive Fish-Culturist, 27(3); 153-157.
Venna, Prabha. 1969. Normal stages in the development
of Cyprinus carpio var. communis L. Acta biol. Acad. Sci.
Hung., 21(2): 207-218.
Approved by the Committee
on Aquatic Bioassays
Approved by the Director, NWQL

Appendix A
Test (Evansville, Indiana) Photoperiod
For Fathead Minnow Full Chronic
Dawn to Dusk
6:00 - 4:45)
6:00 - 4:30)
6:00 - 4:30)
6:00 - 4:45)
6:00 - 5:15)
6:00 - 5:45)
6:00 - 6:15)
6:00 - 7:00)
6:00 - 7:30)
6:00 - 8:15)
Day-length (hour and minute)
5-month pre-
spawning growth
6:00 - 8:45)
6:00 - 9:15)
6:00 - 9:00)
6:00 - 8:30)
4-month spawning
OCT. 1
post spawning period