United States       Prevention, Pesticides     EPA 712-C-96-113
          Environmental Protection    and Toxic Substances     April 1996
          Agency         (7101)
&EPA    Ecological Effects Test
          Guidelines
          OPPTS 850.1000
          Special Considerations
          for Conducting Aquatic
          Laboratory Studies
                'Public Draft"

-------
                           INTRODUCTION
     This guideline is one of a series of test guidelines that have been
developed by the Office of Prevention, Pesticides and Toxic Substances,
United States Environmental Protection Agency for use in the testing of
pesticides and toxic substances, and the development of test data that must
be submitted to the Agency for review under Federal regulations.

     The Office of Prevention,  Pesticides and Toxic Substances (OPPTS)
has  developed this guideline through  a  process of harmonization that
blended the testing  guidance and requirements that existed in the Office
of Pollution Prevention and Toxics  (OPPT) and appeared in Title 40,
Chapter I,  Subchapter R of the Code of Federal Regulations  (CFR), the
Office of Pesticide Programs (OPP) which appeared in publications of the
National Technical  Information Service (NTIS) and the guidelines pub-
lished by the Organization for Economic Cooperation and Development
(OECD).

     The purpose of harmonizing these guidelines into a single set of
OPPTS  guidelines is to minimize variations among the testing procedures
that must be performed to meet the data requirements of the U. S. Environ-
mental Protection Agency under the Toxic  Substances Control Act (15
U.S.C. 2601) and the Federal Insecticide,  Fungicide and Rodenticide Act
(7U.S.C. I36,etseq.).

     Public Draft Access Information: This draft guideline is part of a
series of related harmonized guidelines that  need to  be considered as a
unit. For copies: These guidelines are available electronically from the
EPA Public Access  Gopher (gopher.epa.gov) under the heading "Environ-
mental Test Methods and Guidelines" or in paper by contacting the OPP
Public    Docket    at    (703)    305-5805    or     by    e-mail:
guidelines@epamail.epa.gov.

     To Submit Comments: Interested persons are invited to submit com-
ments. By mail: Public Docket and Freedom of Information Section, Office
of Pesticide Programs, Field Operations Division (7506C), Environmental
Protection Agency,  401  M  St.  SW.,  Washington, DC 20460. In  person:
bring to: Rm. 1132, Crystal Mall #2, 1921 Jefferson Davis Highway, Ar-
lington, VA. Comments may also be submitted  electronically by  sending
electronic mail (e-mail) to: guidelines@epamail.epa.gov.

     Final  Guideline Release: This guideline is available  from the U.S.
Government Printing Office, Washington, DC 20402 on The Federal Bul-
letin  Board.   By  modem   dial   202-512-1387,   telnet   and  ftp:
fedbbs.access.gpo.gov (IP 162.140.64.19),  or  call 202-512-0135 for disks
or paper copies.  This  guideline is also available electronically in ASCII
and PDF (portable document format) from the EPA Public Access  Gopher
(gopher.epa.gov) under the heading  "Environmental Test Methods and
Guidelines."

-------
OPPTS 850.1000   Special considerations for conducting aquatic lab-
oratory studies.
     (a) Scope—(1) Applicability. This guideline is intended to meet test-
ing requirements of the Federal Insecticide, Fungicide, and Rodenticide
Act  (FIFRA) (7 U.S.C.  136,  et seq.)  and the Toxic Substances Control
Act (TSCA) (15 U.S.C. 2601).

     (2) [Reserved]

     (b) Introduction. (1)  This  guideline provides additional information
on how to design and conduct  aquatic laboratory studies with emphasis
on the importance of adequate  characterization of the test  material and
proper understanding of how  the material behaves under test conditions.
This guideline also attempts to interpret those areas that need to be defined
and set limits for designing and conducting laboratory studies.

     (2) Agency guidance for performing aquatic testing sets forth a rea-
sonable position and approach for testing criteria, limits, and standards.
However, standards are  set with the recognition that certain problems will
arise and provisions must be made to accommodate unavoidable problems.
This document provides for exceptions, while at the same time maintaining
a high level of scientific integrity so that testing will provide information
that  is scientifically defensible and protective  of the environment, while
taking into consideration the chemistry of the test material.

     (c) General considerations. (1) Note that for aquatic toxicity testing,
the solubility and stability of the test material must be known for the con-
ditions under which it will be tested and chemical analysis  of the batch
test material must be performed. Determining  the solubility  and stability
of the test material  in the mixture or test solution is  an important part
of these studies.

     (2) The behavior of a test material should be based on experiments
which are conducted under the same conditions as those occurring during
the test. These include but are not limited to:

     (i) Test solution characteristics (salt or freshwater).

     (ii) Temperature, pH, conductivity, lighting.

     (iii) With test organisms in place.

     (iv) Use of the same test containers.

     (v) Use of the same  flow-through systems where appropriate.

     (3) All chemistry methods used in preliminary trials, in range-finding
tests, in establishing percent purity of batches of test material, or in meas-
uring concentrations in test containers must be submitted with the  study.
The  documentation must include a complete description of the method so

-------
that a bench chemist can determine the necessary equipment and perform
the analysis. It must  also include  the raw data,  standards, and  chro-
matograms  from a representative analysis using the method.  This rep-
resentative analysis must be conducted with the specific media for which
it will be used during the  test—i.e.,  analysis should be performed  under
test conditions.  The actual limit of detection (LOD) and limit of quantifica-
tion (LOQ) must be identified.

    (d) Definitions.

    EEC is the effective environmental concentration.

    LOD is the limit of detection below which the qualitative presence
of the material is uncertain.

    LOEC is the lowest-observable-effect-concentration.

    LOQ is the  limit of quantification  below  which  the  quantitative
amount of the material is uncertain relative the amount.

    Measured concentration is an analytically derived measure above the
LOQ.

    NOEC  is the no-observed-effect-concentration.

    Nominal concentration is, for aquatic tests, the  nominal test  level,
which is the concentration that would exist if all test material added to
the test solution was completely dissolved and  did not  dissipate in any
way.

    Recommended means  that the procedure or test is preferred in order
to avoid problems, but it is not required.  If the  recommended procedure
or test is not performed, the study will not necessarily be rejected.

    Solubility is defined as the amount of chemical  retained in the super-
natant of a conventionally centrifuged sample of test medium.

    (e) Stability.  (1) A test material is considered to  be  stable under test
conditions if, under those  conditions,  it does not degrade, volatilize, dis-
sipate, precipitate,  sorb to test container walls, or otherwise decline to con-
centrations less than 70 percent of the day-0 measured concentration dur-
ing the study period. If it  is expected to decline to less than 70 percent
of the day-0 measured concentration during the study period, either static
renewal or  flow-through design  is needed to try to ensure that the test
concentration is maintained at levels greater than or equal to 70 percent.
The only exception is testing with  algae  and diatoms,  which  cannot  be
tested in static  renewal or flow-through systems (see discussion in para-
graph (m) of this guideline  on testing with algae and diatoms).

    (2) Static renewal is one method to ensure relatively continuous con-
centrations when the test material is not stable under test conditions.  At

-------
a minimum, the renewal cycle should be based on the stability of the test
material under test conditions. The time to renewal (renewal cycle) should
be shorter than the time it takes for the concentration of the test material
to decline to < 70 percent. (The renewal cycle may be shorter than required
by stability characteristics of the test material because of other factors,
such as dissolved oxygen, feeding, etc.)

     (f) Sample storage. If samples of growth medium, stock solutions,
or test solutions collected for chemical analysis cannot be analyzed imme-
diately, they should be  handled and stored appropriately to minimize loss
of the test material. Loss could be caused by such processes as microbial
degradation, hydrolysis, oxidation, photolysis, reduction, sorption, or vola-
tilization.  Stability  determination under storage conditions, whether it re-
fers to storing the test material before testing or storing samples awaiting
analysis, is required by GLP regulation.

     (g) Preliminary trials. (1) The  Agency recommends preliminary test-
ing  for problem chemicals. The information about stability and solubility
of problem chemicals should be developed under test conditions. This in-
formation can be gained while doing the  currently required range-finding
studies. A list of recommended preliminary tests is as follows:

     (i) Stability trials  should be conducted under test conditions. These
trials must be documented and submitted  to the Agency for review with
the study to which they  apply.

     (ii) Solubility trials should be conducted under  test conditions. These
trials must be documented and submitted with the  study to  the Agency
for  review.  Surfactants and charged polymers will be self-dispersing in
water and should be tested at or below their dispersability limits.

     (iii) If solubility is a problem (<100 ppm), trials should be conducted
under test conditions using various solvents that are most likely to be ef-
fective and that  are widely recognized as  being nontoxic and other means
to ensure that the appropriate methods are used during the laboratory tests
to enhance solubility. Once a solvent is chosen based upon more simplistic,
comparative evaluations, the decision should be confirmed in the prelimi-
nary trials with only that solvent.

     (iv)  Chemical analysis methods as  detailed  in paragraph (1)  of this
guideline.

     (v) Stability of the test material in the  samples to  be collected for
chemical analyses should be determined during the laboratory studies. This
includes determining whether and how samples can be stored  for future
analysis.

-------
     (2) Laboratory studies must be designed taking into account this pre-
liminary information. This means the trials described are to be conducted
before the definitive laboratory studies are initiated.

     (h) Toxicity tests  with poorly soluble materials. (1) Existing OPP
guidelines  for  aquatic toxicity tests require that chemicals be tested  up
to a maximum dissolved concentration of 100 ppm (milligrams per liter)
for pesticides or 1,000 ppm for industrial chemicals in an effort to  obtain
an LC50 or EC50. This amount of test material is considered to represent
a conservative  measure  of the most bioavailable fraction, which may  in-
clude some colloidal material not removed by centrifugation in addition
to the truly dissolved fraction.

     (2)  Applicants  must demonstrate the technique  used to maximize
chemical dissolution in the test media under standard conditions. Consider-
ation of the  optimum technique  should include use of nontoxic solvents,
saturation (solubility) columns, sonication,  minor adjustments to environ-
mental conditions (i.e., temperature, pH, etc.),  as appropriate.  Minor ad-
justments should not extend outside the recommended range of conditions
for the specific test organism.

     (3) Current policy  allows chemicals that are poorly soluble (solubility
<100 ppm) or dispersible in water to be tested up to the  maximum water
solubility or dispersibility limit  obtainable for the  given test  conditions
employed, provided that certain prerequisites apply:

     (i) Concentrations  of test chemical in test media are measured at ap-
propriate intervals and from  appropriate test chambers of all  test levels
are determined from centrifuged supernatant or other appropriate  separa-
tion (e.g., filtrate). Self-dispersing industrial chemicals (e.g.,  surfactants,
detergents, or charged [polymers) should be sampled directly.

     (ii) Testing is also performed with a more soluble formulation e.g.,
emulsifiable  concentrate, if one exists (in addition to testing with the tech-
nical-grade material). Testing  with a more  soluble  formulation will not,
however, be required if it does not provide a twofold increase in solubility.

     (4) Studies that involve radical changes in environmental test condi-
tions outside the recommended range of values for temperature, salinity,
pH, etc., will be considered on a case-by-case basis.

     (i) Methods  for solubility enhancement—(1) Saturator columns.
The use of saturation columns as an aid in the dissolution of test material
and in confirming maximum solubility is  recommended but  not required
for nonvolatile test chemicals with test media  solubilities of 10 ppm or
less. Methods  for using these columns in aquatic  toxicity tests  can  be
adapted from the methods established  for their use in determining water
solubility under OECD's Column Elution Method (see  OPPTS 830.7840).
Saturator columns may be considered to generate test  solutions for static

-------
studies  or for flow-through studies. Furthermore, saturator  columns  for
these studies need only be considered if conventional techniques for dosing
the water  do not result in water concentrations within twice the  stated
solubility of the compound.

     (2) Emulsifiers and formulation  testing.  Testing with a more solu-
ble formulation, if one  exists and  which  may contain emulsifiers, dis-
persants, solubilizing agents, etc., is  required for all active ingredients sub-
ject to aquatic  organism testing and having a  water solubility less than
100 ppm and less than an EC/LC50. A defined EC/LC50 provides a great-
ly improved basis for risk assessment.

     (3) Effect of temperature.  Solubility is  a  function of temperature
and is especially sensitive at the limits of solubility. Generally, below satu-
ration, increases of as much as  10  °C may affect the  solubility up to a
factor of 2. However if test  solutions are close to saturation, small changes
in temperature may result in supersaturated solutions. In addition, control
of temperature is important  because  of its well-known effects on the actual
toxicity of the compound.

     (4) Centrifugation. Conventional  centrifugation is required for all
test media where undissolved test material, precipitate, flocculant, or col-
loidal suspension except for surfactants or charged polymers) are observed
in the test  chambers or where the  solubility and (hence bioavailability)
are in question. Filtration may be used instead of centrifugation if the ana-
lytical method is validated over a range of acceptable concentrations.

     (j) Measurement at initiation  and termination of testing—(1) Ini-
tial analysis, (i) Analysis at the 0-hour: A 30-min interval is generally
required between the addition of the test substance  and the introduction
of the test organisms. 0-hour measurement should be made when test orga-
nisms are added. Industry will have to justify an exception from the 30-
min requirement for adding test organisms if the characteristics of the test
material and test system require a longer equilibration time. If preliminary
trials have been performed, this delay should be  predictable.

     (ii) In flow-through tests, the study should  be conducted with knowl-
edge  of the time it will take for the  test  material to  reach equilibrium
or  steady-state in the test container. Initiation  of the test and  scheduling
of the sampling times must be based on this information. In some  cases,
a flow-through system may have to be run for an extended-time pretest
in an attempt to achieve equilibrium or steady-state conditions. If equi-
librium or steady-state cannot be achieved, and/or it appears that the meas-
ured concentrations will be  substantially below (< 70 percent) nominal,  the
study report should reflect that the laboratory was aware of this problem.
The study report should clearly identify the problem,  indicate the steps
taken to mitigate it and justify the study design and dosing levels.  How-
ever, if sufficient analytical methods are available and acceptable toxicity

-------
data are produced, additional testing and evaluation with the sole objective
of obtaining initial measured concentrations greater than 70 percent of
nominal will not be required.

     (2) Analysis  at test termination. Where indicated,  measurement at
test termination is considered necessary to determine if the test organisms
were exposed to the test material throughout the entire study and at what
levels. A  significant change in  test concentration during the last part of
the study  may substantially alter the results.  For example, if the test con-
centration dropped dramatically during the last few days of a study, the
effects that may have been caused by such exposure may not occur. The
EC50 or  LC50  developed from that study would be misleading if it is
called a "96-hLC50".

     (k) Replicates and concentration measurement. (1) Average  con-
centrations of replicates are used in regression  analysis. When  replicate
test  containers and measurement of test concentration  are required,  each
replicate in  each test concentration must be analyzed separately because
the responses in each replicate  are viewed as independent and it is nec-
essary to  know  what the concentrations were so variation can be deter-
mined. Exceptions to this occur when:

     (i) Replicate treatment containers under static tests or static renewal
conditions are filled from a bulk  preparation.  In this  case, only samples
from the bulk supply for each test level must be analyzed.

     (ii) A "splitter" is used in  a flow-through test to  feed more  than
one  replicate.  In this case, only  samples from  one  replicate per treatment
level require analysis. It is recommended that samples be collected from
all replicates and be stored in case anomalous concentrations  are measured
in the one that is analyzed. Analyzing the other replicates may shed light
on the cause and extent of the anomalous measurements.

     (2) Replicates receiving flow from a splitter should  be sampled and
analyzed  alternately. In  other words,  if there  are  two replicates (A  and
B), replicate A  should  be analyzed in the first  week and replicate  B in
the second week, etc.

     (3) To the extent possible,  variability in  measured concentrations
should be minimized. The  goal for limiting variability of measurements
between replicates of the same  concentration,  and over time in the same
concentration, is maintaining the ratio of the highest concentration to the
lowest concentration  at 1.5:1 or  less. Generally,  variability above  this
amount is not acceptable.

     (4) An important factor in  considering the limits of variability is the
avoidance of overlapping  mean test concentrations between test levels.
High variability puts into  question the reliability of the environmental
chemistry method and/or the concentrations on  which to base statistical

-------
analysis and toxicological conclusions. If variability beyond the 1.5:1 ratio
occurs, an exception to it should be justified.

     (5) This justification  should clearly state the problem, explain why
it occurred, provide  scientific justification, and identify all measures taken
to mitigate the problem. The justification also should include the fully de-
veloped chemistry method, including the  documentation necessary for a
bench chemist to review and evaluate it.

     (6) For cases in which variability problems are suspected,  preliminary
trials are strongly recommended. If it becomes clear that high variability
cannot be  avoided,  an exception  should be justified. Any justification
should be provided in advance. Agency scientists will decide on the valid-
ity of the rationale for the exception, and may recommend other methods
to reduce potential variability.

     (1) Use of chemical  analysis to confirm exposure in aquatic test-
ing—(1) Acute static tests. Except for acute  aquatic algae  and  diatom
studies  (which can  only be conducted  as static tests), acute static tests
may be conducted only if, among other things, the test material has been
shown to be stable under the test conditions, as defined in paragraph  (e)
of this guideline. (Other factors not addressed in this guideline may pre-
clude conducting a static test even if the test material is stable under test
conditions. These include,  but are not limited to, problems in maintaining
dissolved oxygen levels, feeding requirements,  and concern for bacterial/
microbial contaminants.) In an  acute static test with a test material that
is stable and readily soluble at the treatment levels, measurements of each
test concentration are not absolutely required. However:

     (i) For static tests, the concentration of toxicant should be measured
at the beginning and end of the test in all test  chambers. Further, measure-
ment of the toxicant's degradation products is desirable, but not required.

     (ii) The study may be rejected if the following  occurs:

     (A) The test material was not stable under test  conditions.

     (B) Precipitates formed.

     (C) Solubility was likely to have been a  problem at the levels tested.

     (iii) If the recommended chemical measurements were made to verify
exposure levels, the  study  may not be rejected. Whether the study design
was  modified in a scientifically defensible attempt to accommodate these
chemical characteristics will also be considered.

     (iv) If variability is  expected to be  a problem, it is recommended
that  measurements  of test  concentrations be made at each test  level at
0-hour, 48-h and, for tests longer than 48 h, at test termination. Replicate

-------
test containers should be measured separately, except as explained under
paragraph (k) of this guideline.

     (2) Acute static renewal. Refer to the general discussion of replicates
under paragraph (k) of this guideline. If a static renewal test is conducted,
each test chamber must be sampled for chemical  analysis  at the 0-hour,
at the end of the first (or longest) cycle,  and at test termination. It is rec-
ommended that measurements be made at the end of each renewal cycle
acute flow-through.

     (3) Acute flow-through. If a flow-through test is conducted, each
test concentration must be measured at the 0-hour and at test termination.
It is recommended that for 96-h tests, an intermediate measurement be
made at 48-h to verify midtest exposure if variability  is expected to be
a problem. (All  acute aquatic algae and diatom tests must be  conducted
as static. Flow-through and static renewal systems are not recommended
for  these tests, since they are conducted  with microscopic  organisms that
cannot be protected from loss when renewing or draining water from the
test containers.  Static tests  for Lemna gibba can be conducted,  regardless
of stability.)

     (4) Chronic static renewal. Refer to the general discussion of rep-
licates under paragraph (k) of this guideline. Concentrations must be meas-
ured at each test  level at  0-hour,  at the end  of the last  renewal cycle
(at test termination), and at the beginning and end of an intervening cycle
at least once per week. The longest cycle in a  sequence should be used
if variable-cycle periods are employed.

     (5) Chronic flow-through. Refer to the general discussion of rep-
licates under paragraph (k) of this guideline. In each concentration, meas-
ure  at 0-hour, every 7 days,  and at test termination. At the beginning of
a study, the exact flow of the system and  water output at each  splitter
must be documented. In addition, system flow must be metered and mon-
itored visually or  mechanically  on a  daily basis (every 24 h), and it is
recommended that the system flow be metered and monitored twice a day
(approximately every 12 h). Measurement of test concentration is required
each time metering fluctuation or malfunction is detected or observed. A
record of the regular inspections must be maintained and  provided with
the  study report.

     (m) Measured concentrations versus nominal concentrations. This
section describes acceptable limits of deviation of measured from nominal
concentrations.

     (1) Test endpoints are used as if the organisms were exposed to the
test material at the statistically developed value (LC50 or  EC50) for the
entire test duration. One aspect of the risk assessment is to  compare con-
cern levels based on the LC50 is to initial immediate concentrations. How-
ever, field conditions may exist in which concentrations that may be of

                                  8

-------
acute concern may last longer or occur frequently enough to be comparable
to the 48-h, 96-h  or  120-h test duration.  Even though a pesticide may
degrade rapidly under one environmental condition (in water, for example),
the possibility of repeated exposure needs to be considered.  Repeated ex-
posure from reservoirs of the active ingredient, occurring in environmental
compartments where persistence is greater,  may occur. The Agency takes
these eventualities into account in order to  generate risk assessments that
adequately address hazard to the aquatic ecosystem.

     (2) Presumably, a safer chemical one  that may degrade rapidly, has
low solubility,  and is  used at low rates.  While these characteristics may
result in lower exposure levels in the field,  the  risk they represent can
only be determined if  the actual toxicity  of the pesticide is known or the
level below which the pesticide is not likely to result in 50 percent mortal-
ity (i.e., an LC50 > X-concentration situation.) When potentially low, real-
istic exposure levels are calculated and  used for risk assessments, it is
imperative that the  actual toxicity of the pesticide at those levels be  deter-
mined. If the test is conducted using  nominal concentrations, the results
could reflect a higher  apparent effect  concentration (e.g. LC50,  EC50, or
NOEL). As a result, potential risk may be missed because the comparison
would be between a  low "realistic"  exposure and a high nominal  test
level that was not the true toxicity level. A  risk assessment based on such
a  comparison and  data  would be  faulty and  could not be scientifically
defended.

     (3) Pesticide chemicals that are used at very low levels tend to have
high biological activity.  For this reason,  it  is imperative that the toxicity
data developed for these pesticides be accurate and scientifically defen-
sible.

     (4) Measured concentrations are used when they are available because
they indicate what the  exposure was in the test chambers. When measured
concentrations are indicated, they are considered necessary because:

     (i) There are concerns that the actual concentrations to which the test
organisms are exposed may differ from  "nominal."  This variation may
be  due to chemical characteristics, test conditions, or mechanical appara-
tus.

     (ii)  Measured concentrations confirm  that the test system was de-
signed appropriately and is operating acceptably. Characteristics that make
testing difficult (low solubility, short half-life, high binding potential, etc.)
must be  accounted for in the exposure estimates.  They are not a reason
for developing misleading toxicity values from laboratory tests.

     (5) Measurement of test concentrations is not performed just to  deter-
mine if the technician knows how to  mix the test solution once. Among
other things, it also ensures that the test solution was mixed correctly each

-------
time.  It corroborates the precision of the technician or mechanics of the
test system.

    (6) If test levels  are not measured,  the  nominal values are used to
calculate the LC50, EC50, NOEC and LOEC. If the test material has de-
graded or has become unavailable because of insolubility or sorption, the
pesticide may be characterized as less toxic than it really is. For example,
if based on nominal test levels, the LC50 is  5 ppm, the pesticide would
be considered moderately toxic. No higher tier testing  would be required
and that value (5 ppm) would be  the basis for developing  concern levels
with which to compare EECs. But if, in reality, the concentrations to which
the organism was actually  exposed were only between 0.1 and 1 ppm,
the LC50 may well be closer 0.5  ppm.  For  pesticides this would result
in labeling, and  could trigger higher tier tests.  More importantly, it would
yield  substantially lower concern levels with  which to  compare exposure
levels.

    (7) When a laboratory test design has been specifically modified to
accommodate the instability of test material or other factors  likely to cause
variability  in test concentrations, and the  design is  judged adequate based
on sufficient preliminary information, the  study will not be  rejected solely
on the grounds that measured concentrations varied by  more than 30 per-
cent of the nominal concentration. (This assumes that the preliminary sta-
bility  tests were conducted under test  conditions  essentially identical to
the actual test conditions.) An increase in measured test concentration of
more  than  30 percent from  the nominal concentration during the test will
generally not result in rejection,  provided that the following  conditions
are met:

    (i) A reasonable and scientific explanation is given, and the variability
of results produced by the chemical analysis method is  adequately charac-
terized.

    (ii) All test containers exhibit a similar (but not necessarily identical)
shift.  (If concentrations in some containers go up  substantially (>30 per-
cent)  and  test concentrations in other  containers  go down substantially
(>30 percent), they will not  be considered to have exhibited  a similar shift.
The most important criterion is that test levels must not experience a shift
in "order." That is, the highest test level should remain highest, the next
should remain second, etc. If orders are shifted, the test may be rejected,
since  regression analysis would not yield  statistically sound median lethal
concentrations and confidence limits.)

    (iii) The variability of the measured concentrations is acceptable.

    (iv) A statistically valid endpoint can be  derived from the measured
concentrations (either an LC50, EC50, or that  the LC50 or EC50 is greater
than 100 ppm).

                                  10

-------
     (v) The preliminary  stability information is provided with complete
documentation and description of methods used to derive such information.

     (8) In some cases, high variability cannot be avoided because the test
concentrations are approaching  the limit of detection or because  of un-
avoidable binding of the test material to the chemical analysis apparatus.
When the ratio of the highest concentration to the lowest measured con-
centration is expected to vary by more than 1.5, the registrant is strongly
advised to justify an exception to this requirement in advance of conduct-
ing the aquatic laboratory studies. This exception justification should con-
sist of:

     (i)  Documentation  of the preliminary  trials  indicating this problem.

     (ii) The specific steps that will be taken to reduce the variation.

     (iii) The fully developed chemical analysis method.

     (iv) The raw data, standards, and chromatogram from a representative
analysis using the method. For  each chemistry method, the actual mini-
mum detection level and level of quantification must be identified.

     (9) The Agency will  decide on each exception justification on a case-
by-case basis. However, if a series of aquatic tests  are to be conducted
with one chemical and it  is anticipated that these limits will be exceeded,
one exception justification may  cover more than one study. The Agency
will then exercise judgment in evaluating  studies  with test materials that
are difficult to measure.

     (10)  Conducting flow-through or  static renewal tests  with  aquatic
algae is not feasible with the current state of the  practice. Therefore, the
following is recommended for a test material that, based on preliminary
stability testing, is expected to degrade to less than 70 percent of the nomi-
nal concentration.  The  study should be  conducted normally, with  con-
centrations measured at 0-hour and at test termination. Although it is unde-
sirable to allow the concentrations to decline throughout the study, the
problem  may be unavoidable. In this case, the LC50 regression analysis
is  based  on the mean measured concentration. If the concentration is ex-
pected to decline to  less than the minimum detection level before  the end
of the study, then it is recommended that interim  chemical measurements
be made to determine the decline rate.

     (11) For purposes of consistency, the aquatic test with a vascular plant
(Lemna gibba) need not be done using a  flow-through or static renewal
system with the sole purpose of maintaining test concentrations. There may
be other reasons for conducting a static renewal study.
                                  11

-------