United States EPA-600A-81 -003
Environmental Protection February 1981
Agency
>EPA Research and
Development
Results: (
Interlaboratory Comparison -
Acute Toxicity Tests
Using Estuarine Animals
Prepared for
Office of Pesticides
and Toxic Substances
Prepared by
Environmental Research
Laboratory
Gulf Breeze FL 32561
-------
EPA-600A-81-003
February 1981
RESULTS: INTERLABORATORY COMPARISON—ACUTE
TOXICITY TESTS USING ESTUARINE ANIMALS
by
Steven C. Schimmel
Environmental Research Laboratory
U.S. Environmental Protection Agency
Gulf Breeze, Florida 32561
my
•irborn Street
s 60604
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
GULF BREEZE, FLORIDA 32561
-------
INTRODUCTION
Under Section 4 of the Toxic Substances Control £ct (TSCA), the admini-
strator for the Environmental Protection Agency (EPA) can require environmental
effects testing of chemical substances if (1) the manufacture, use, distribu-
tion, or disposal of that substance may present an unreasonable risk of injury
to the environment; or (2) the substance will be produced in substantial quan-
tities and is expected to enter the environment, and there are insufficient
data to predict the effects of the chemical substance on the environment.
When the administrator issues a test rule for the performance of envi-
ronmental effects testing, he must also provide test standards to be used in
the development of test data. Before test standards are proposed, steps
should be taken by the Agency to insure that data developed by each test
standard are adequate and reliable.
This report summarizes the results of "round-robin" or precision tests to
validate proposed test standards and to determine the degree of variability
between data developed by different researchers using the same methodology.
The tests were performed by 2 of EPA's Office of Research and Development
laboratories and 4 laboratories under contract.
Contractors were instructed to follow the American Society for Testing and
Materials (ASTM) "Proposed Standard Practice for Conducting Basic Acute
Toxicity Tests with Fishes, Macroinvertebrates, and Amphibians" (Draft 6).
because EPA's Office of Toxic Substances (OTS) Fish Acute Toxicity Test
Standard and the Mysid Shrimp Static and Flow-through Acute Toxicity Test
Standard were not completed at the time of the contract award. Notebooks,
progress reports, and final reports of the contract laboratories were examined
to determine laboratory adherence to the ASTM test methods, which are similar
to the now completed OTS Test Standards.
The ASTM document specifies certain required test methods or conditions in
order for a test to be considered satisfactory. These conditions are always
associated with the word "must." For example, Section 11.1.1 states (in part)
that "for static tests at least 10 organisms must be exposed to each treat-
ment." Suggestions for good test practices are generally phrased in words such
as "should," rather than "must." For example, Section 10.3.1 states (in part):
"In any single test all fish should be from the same year class and the stan-
dard length of the longest fish should be no more than twice that of the
shortest fish." Each laboratory's final report was carefully scrutinized to
determine if the "musts" in the ASTM method were fulfilled. If not, an
attempt was made to determine if the test results were affected.
Test chemicals used in the "Round Robin" were silver nitrate and endo-
sulfan. Selection of these chemicals depended on a number of factors, including:
1
-------
(1) toxicity to selected species at or below water solubility; (2) chemical
type (an organic and an inorganic); (3) scarcity of toxicity data in the lit-
erature on tests with these chemicals and species; (4) ease of chemical anal-
ysis; and (5) relatively low mammalian toxicity.
The test species selected for the "Round Robin" were the copepod, Acartia
tonsa, the mysid shrimp, Mysidopsis bahia, and the sheepshead minnow,
Cyprinodon variegatus. One static toxicity test was required from each labor-
atory for each species exposed to both endosulfan and-silver nitrate. The LC50
values (concentration estimated to kill 50% of the test animals) required for
these static tests were to be based on nominal concentrations. In addition,
each laboratory was required to conduct flow-through tests on M. bahia and C.
variegatus, using endosulfan and silver nitrate. The LC50 values obtained from
the flow-through tests were to be based on both measured and nominal concen-
trations.
TOXICITY TEST RESULTS
Toxicity test data from each of the laboratories were analyzed at the
Environmental Research Laboratory, Gulf Breeze, using programs for probit
analysis, moving average and the binomial method. Results of statistical
analyses of all test data are listed in Tables 1, 2, and 3.
Acartia tonsa Toxicity Te<%sts
Eleven of the twelve A., tonsa toxicity tests produced data that were
amenable to statistical analysis (Tables 1 and 2). Lab 5 did not produce a
successful test with A., tonsa in an exposure to silver nitrate. This labora-
tory, as well as Lab 6, was totally inexperienced in handling and testing the
species. Eleven attempts made by Lab 5 to collect field populations of this
copepod and hold them in the laboratory for at least four days resulted in
inadequate control survival. An LC50 in the endosulfan test was obtained by
Lab 5 only when salinity was maintained at ambient (^8 °/oo) salinity con-
ditions. Any attempt to raise the salinity gradually over a long period of
time caused excessive (>15%) control mortality. .Lab 6 attempted at least 30
toxicity tests with field-captured animals; all were unsuccessful. After
obtaining lab-cultured individuals (M4 days old), two successful tests were
conducted in five trials. Of the six laboratories, three maintained contin-
uous A. tonsa cultures. No problems of control mortality were mentioned by any
of the three. Only one of the three laboratories (Lab 3) that tested
field-captured copepods conducted successful tests. The Environmental
Research Laboratory, Gulf Breeze wished to address the problems of high con-
trol mortality in field-captured A. tonsa by contracting Lab 3 (outside the
scope of the present contract) to conduct control survival experiments with
us. Two trials were attempted, using Lab 3's personnel and methods of col-
lection, holding, acclimation, and testing. Neither trial was successful.
The problem of excessive control mortality is two-fold: (1) The life
expectancy for _A. tonsa is approximately 30 days, and a 96-hour test is
approximately 1~3% of that life expectancy; and (2) The chance is great that a
significant number of field-collected animals are senescent and will die
-------
A
during a test, whether in the control or experimental group. Thus, the
problems of a long test period (relative to the life expectancy) and that of
senescence in field-collected stocks produce excessive control mortality. The
same argument might be made for field captured H. bahia, but in these tests
each laboratory used lab-cultured juveniles (<48 hours old).
We recommend that either of two changes be required when A. tonsa are
tested: (1) that the test duration be reduced to 48 hours; or (2) lab cul-
tures be maintained and adults M4 days old be used fqp toxicity tests.
(Adults are suggested because they are more visible, thus easier to handle.)
We believe that the second recommendation is more acceptable since in the
first, the possibility of using senescent adults still exists; therefore, the
likelihood for excessive control mortality still exists. Laboratory cultures
of A., tonsa are easily maintained. (Labs 1,2,4 and [now] Lab 6 maintain
cultures with little difficulty.)
Because of the problems associated with _A. tonsa, the copepods will not be
discussed further in the test results section.
Mysidopsis bahia and Cyprinodon variegatus Toxicity Tests
Sixty-six LC50 values of a required 72 (92%) were calculated in the
"Round-Robin" procedure (Tables 1, 2, and 3). The condition for an accep-
table LC50 value was that veither probit analysis, moving average or a binomial
LC50 value could be calculated. Two mysid tests were not acceptable. In the
first, the mysids exposed to endosulfan in a static test produced by Lab 4 did
not elicit mortality >50% (Table 2). In the second test (Lab 3; Table 3), the
measured concentration of endosulfan was not amenable for use in the calcula-
tion of an LC50, i.e., most concentrations were non-detectable or less than 15%
of nominal concentrations. The test was not repeated.
In order to estimate the variability of the LC50 values generated by each
laboratory for a test type (i.e., mysid, endosulfan, static test, nominal
concentration), we determined the ratio of the highest LC50 the lowest LC50
(H/L ratio) for. that particular test type. The H/L ratios for 11. bahia
exposed to silver nitrate was 2.2 in static tests and 1.9 in flow-through tests
based on nominal concentrations (Table 1); the H/L ratio for flowthrough tests
based on measured silver concentrations was 4.8 (Table 3). In the endosulfan
mysid tests, the H/L ratios were 6.1 for static tests and 5.2 for flow-through
tests based on nominal concentrations (Table 2). The H/L ratios for mysid
flow-through tests based on measured endosulfan concentrations was 3.4 (Table
3).
Thirty-two of the required 36 tests (89%) with C_. variegatus produced data
that were acceptable for the calculation of an LC50 (Tables 1, 2, and 3). Lab
5 failed to produce silver nitrate LC50 values for _C. variegatus in both static
and flow-through tests (Tables 1 and 3). The reason given for this failure was
that the solubility of silver nitrate in 28 °/oo seawater was approximately
2,000 to 3,000 ug/1. Precipitation of the test chemical occurred at or above
3,000 ug/1. Therefore, _C. variegatus placed in these nominal concentrations
may not have been actually exposed to those concentrations due to the precip-
itation resulting in mortality less than the 50%. The silver nitrate LC50
-------
values generated by other laboratories using this species ranged from 640 to
1,584 Mg/1 in static tests, and from 818 to 2,684 yg/1 in flow-through tests.
Lab 4 indicated that they observed precipitation of silver nitrate in the
sheepshead minnow static test, and no LC50 could be calculated. The H/L ratio
for the _C. variegatus silver nitrate static test was 2.5; that for the flow-
through tests based on nominal concentrations was 3.3 (Table 1). The H/L ratio
for the _C. variegatus silver nitrate tests based on measured concentration was
4.2 (Table 3).
Additional statistical analyses of the data in Tables 1, 2, and 3 were
attempted by the ERL, Gulf Breeze consultant Dr. Jerry Oglesby but were frus-
trated by the lack of intra-laboratory replication. Analyses of the slopes of
the mortality-concentration curves were made, but results were not amenable to
interpretation. More observations on the problems associated with these test
data and recommendations on the statistical design of future "Round-Robins"
will be presented in the discussion and conclusions section of this report.
Analysis of the Adherence of Laboratories to the ASTM
Method and Scope of Work.
Final reports from each laboratory involved in the acute "Round-Robin"
were carefully reviewed to determine if the "must" requirements of the ASTM
document and the Contract's Scope of Work were implemented. Tables 4, 5, 6,
and 7 address these requirements and their implementation.
Reports from each laboratory were reviewed to determine the quality of
seawater used in their toxicity tests. The ASTM method requires that diluent
water (water from marine or estuarine sources that is pumped or otherwise
delivered into the laboratory) be analyzed at least monthly for the following
parameters: salinity, temperature, pH, dissolved oxygen (D.O.), particulates,
total organic carbon (TOG) or chemical oxygen demand (COD), total organo-
chlorine pesticides plus PCB's. In addition, diluent water used in the
toxicity tests must be adjusted to levels specified in the Contract's Scope of
Work. The Scope of Work parameters and their requirements were: salinity,
28 °/oo +1.5 °/oo; temperature, 22° C + 0.5° C. Dissolved oxygen require-
ments for water at the start of the tests must be between 90 and 100% satura-
tion; TOG, <2 mg/1; COD, <5 mg/1; particulates, <20 mg/1. Laboratories varied
in how closely they followed the diluent and test water quality requirements of
the ASTM draft (Table 4). Most laboratories failed to measure particulates in
their diluent water, did not measure total organic carbon (TOG), or chemical
oxygen demand (COD), and only one laboratory (Lab 6) analyzed diluent water for
total organochlorine pesticides. Test water quality, however, was maintained at
the required values and measurements of salinity, temperature, pH, and D.O. were
made according to the ASTM method (Table 4). One exception was the A. tonsa
endosulfan test conducted by Lab 5. In that test, salinity was maintained at 8
°/oo rather than the required 28+1.5 °/oo. Acceptable control survival
could not be attained by that laboratory if the salinity was altered substan-
tially from ambient conditions.
It is our judgment that although the above deviations of ASTM and Scope
of Work were made, the tests 'concerned (other than the _A. tonsa study) should
-------
J
be considered valid. In all cases, diluent water was filtered before it was
used as test water, and D.O. values during the tests were acceptable.
The ASTM method defines the maximum amount of loading (grams of exposure
animal/liter of exposure water), carrier concentration in the test water for
static and flow-through tests, and minimum turnover rates of test water
required in aquaria used in flow-through tests. All laboratories met the
loading requirements in both static and flow-through tests (Table 5). Lab 3
used 1.0 ml carrier/1 seawater in both the _A. tonsa ar^d M. bahia endosulfan
tests; a maximum of 0.5 ml/1 is allowed. Lab 4 provided only two turnovers per
day in their silver nitrate and endosulfan flow-through tests; a minimum of
five turnovers per day is required. In addition, Lab 4 did not use a conven-
tional delivery apparatus, but mixed the toxicant and diluent water in Mariott
bottles and allowed the water to flow to the exposure chambers at a prescribed
rate. Although not technically in violation of the ASTM method, this procedure
is not recommended because of the potential for hydrolysis of test chemicals,
adsorption of hydrophobic compounds to the glass of the bottles, potential for
significant decrease in dissolved oxygen and increase in microflora that may
affect the test results. Lab 4 indicated that the measured endosulfan was
significantly below nominal in their flow-through tests; therefore, the de-
livery apparatus may have affected the measured concentration.
Test Animals
v
In contrast, six laboratories adhered more closely to the ASTM test method
and Scope of Work requirements of using: <2 day old mysids; 28-day-old
sheepshead minnows; 14-day acclimation time for sheepshead minnows; and accep-
table control mortality (_<15% for A., tonsa, <10% for M. bahia, and <5% for
C. variegatus) (Table 6). There were two exceptions: Lab 3 did not test
3uvenile M. bahia <2 days old, and Lab 2 did not acclimate £. variegatus for
the required 14-day period. Personnel at Lab 3 were questioned on the age
discrepancy and agreed to repeat the static studies with 2-day-old mysids.
The 96-hour LC50 values generated for silver nitrate with the younger mysids
(181 yg/1) compared favorably with those 6- to 8-days-old (203 yg/1); that for
the 2-day-old mysids exposed to endosulfan was 0.29 yg/1, compared to that of
the older animals (96-hr LC50 =0.24 yg/1). The flow-through tests were not
repeated with 2-day-old mysids, but should be a close estimate of those with
the older individuals. Lab 2 never complied with the necessary acclimation
time for sheepshead minnows (14 days for fish). Lab 2 purchased 21-day-old
fish from a supplier and tested the fish after seven days acclimation. Jus-
tification for the reduced acclimation period was that younger fish would be
adversely affected in shipment from the supplier. We disagree with the
contention, but do not believe that the inadequate acclimation time would
invalidate the test.
Each laboratory was evaluated to determine how it performed in analytical
chemistry relative to the requirements in the ASTM methods document (Table 7).
Criteria for silver nitrate and endosulfan were: validation of analytical
methods (including acceptable percentage recoveries), use of reagent blanks,
and percentage acceptable measured water concentrations.
-------
The ASTM method requires that the measured water concentrations "must be
no more than 30% higher or lower than the concentration calculated from the
composition of the stock solution and the calibration of the toxicant-delivery
system." Labs 4 and 5 were not in compliance with the validation of the ana-
lytical method and use of reagent blanks for both endosulfan and silver
nitrate (Table 7). In both cases, no mention was made of these requirements in
the final reports; therefore, it was assumed that the requirements were not
met. i
Performance of the laboratories in meeting the requirements for accep-
table water concentrations were varied (Table 7). Labs 1 and 2 performed
extremely well for both chemicals, whereas Labs 3 and 4 performed poorly. We
believe that the 30% ASTM requirement is somewhat arbitrary since some chemi-
cals (such as silver nitrate) are very water soluble in the parts per billion
range, where endosulfan is very insoluble. Poor performance in analysis can
result for endosulfan in most cases and may also cause problems for silver
nitrate at the parts per million level, which were encountered in the sheeps-
head minnow tests. Therefore, poor performance in the measured water cate-
gories in this "Round-Robin" may not be as serious as it appears, and the
problem may be that of too stringent requirements stipulated in the ASTM
method.
DISCUSSION AND CONCLUSIONS
• The results of the Acute Toxicity "Round-Robin" with estuarine animals
indicate that the mean H/L ratio for all mysid and sheepshead minnow tests
based on nominal concentrations was 3.5 (Tables 1 and 2); that for both species
based on measured concentrations was 4.0 (Table 3). Therefore, LC50 values
produced by different laboratories using these species and the ASTM method
should fall within a factor of 4.0. If the Acartia tonsa tests are included,
the mean H/L ratio for tests based on nominal concentrations would be 4.8. We
were somewhat disappointed in the results of this study. In acute tests at
ERL, Gulf Breeze, LC50 values in repeat tests seldom vary by more than a factor
of two; other laboratories indicate similar variability.
Several factors might explain the variability we report here in the
"Round-Robin." One factor is that the participating laboratories are widely
separated geographically. Labs 1, 2, and 4 are situated in New England and 3,
5, and 6 are located on the Gulf Coast. One might expect racial differences
in the species and this could translate to differing sensitivities. Also,
different solvents were used (e.g., acetone, triethylene glycol, and methanol)
in the endosulfan tests and different exposure apparatuses were used. More
important, laboratories had differing degrees of test experience with the
selected species and with flow-through tests. Labs 5 and 6 had never tested
A. tonsa before this "Round-Robin," and Labs 2, 4, and 5 are relatively new
Taboratories in the field of aquatic toxicology. All factors considered, the
H/L ratio of 4.0 derived for the M. bahia and C. variegatus studies is not
surprising. Additional experience accrued by the participants and other
facilities capable of conducting these tests will, undoubtedly, increase the
precision of the test.
-------
Based on the results of these studies and the probability of funding more
acute "Round Robin" tests in the future, we believe some changes are war-
ranted: (1) that a minimum of four laboratories participate and a minimum of
three replicate tests be required from each laboratory. In this way, intra-
laboratory, as well as inter-laboratory, variability can be estimated. Ob-
viously, funding of the "Round-Robin" was a major obstacle that prevented
replication. Over 50% of the allotted extramural monies were spent on the
four contract laboratories. Therefore, two replicates may have been gener-
ated, but would not provide the three replicates necessary for the desired
statistical analyses; and (2) only laboratory-cultured animals should be used.
When ML. bahia and _A. tonsa are used, only juveniles or young adults are
suitable for acute tests.
-------
Table 1. Results of silver nitrate static and flow-through Acute Toxicity Tests using estuarine animals.
Results are LC50 values (probit analysis based on nominal concentrations in yg/1); numbers in
parentheses are the 95% confidence intervals.
SILVER NITRATE
LABORATORY
Std.
Ratio
1
2
3
4
5
6
X
dev.
high/ low
Acartia tonsa
Static
30.9
(22.3-46.7)
66.0
(59-74)
35.8
(30.5-41.6)
23.5
(17.2-29.6)
c
36.4
(29.4-45.8)
38.5
16.2
2.8
Mysidopsis bahia
Static
264
(229-307)
251
(207-303)
203
(158-263)
248a
(219-283)
178
(163-190)
117
(98-140)
210
56
2.2
Flow- through
274
(239-326)
282
(2357,330)
168
(138-209)
325
(277-426)
248a
(219-282)
211
(178-289)
251
56
1.9
Cyprinodon variegatus
Static
1584
(1423-1835)
1182
(1028-1354)
640a
(360-1140)
Non-
Flow- through
1524
(1368-1753)
860a
(769-991)
818
(700-957)
1980
calculable15 (1868-2089)
c
1082
(1006-1169)
1122
388
2.5
c
2684a
(2258-3419.)
1573
788
3.3
aMoving average LC50 calculation
Mortality <50% in highest concentration
cUnable to complete the test
-------
Table 2. Results of endosulfan static and flow-through Acute Toxicity Tests using estuarine animals. Results
are LC50 values (probit analysis based on nominal concentrations in pg/1); numbers in parentheses are
the 95% confidence intervals.
ENDOSULFAN
Acartia tonsa
LABORATORY
1
Static
(0.
0.45
31-0.
a
70)
(0
0. 12a
2
3
4
5
(0.
(0.
(0.
(0.
11-0.
0.05
04-0.
0.28
18-0.
0.40
30-0.
14)
06)
41)
58)
(0
(0
Mysidopsis bahia
Static Flow-through
1.12
.85-1.52)
0.46
.38-0.56)
0.24
.16-0.54)
Non-
calculable0
(1
0.03a
6
~x
Std. dev.
Ratio high/low.
(0
.0-0.
0.24
0.25
15.
56)
(
1.47
.25-1.72)
0.73
.58-. 95)
0.84
0.5
6.1
1.77
(1.39-2.
0.34
(0^,27-0.
0.36
(0.28-0.
1.04
(0.82-1.
1.52
(1.24-1.
1.04
(0.73-1.
1.02
0.53
5.2
a
36) ,
40)
53)
33)
a
88)
49)
(2.
(2
(1.
(0
(2.
(3.
Cyprinodon variegatus
Static Flow-through
2.87
35-4.15)
2.7
.4-3.1)
1.4a
08-1.92)
1.2b
.8-2.0)
2.81
61-3.02)
3.45
22-3.69)
2.41
0.91
2.9
(1
(2
(0
(1.
(2.
(1.
1.61b
.0-2.0)
2.5
.3-2.7)
0.71a
.5-0.9)
1.4
31-1.55)
2.74a
56-2.90)
1.19
04-1.37)
1.69
0.78
3.8
aMoving average LC50 calculation
bBinoraial LC50 calculation
cMortality <50% in highest concentration
-------
Table 3. Results of flow-through Acute Toxicity Tests using estuarine animals. Test results are LC50 values
(probit analysis, measured concentrations in yg/1); the number in parentheses are the 95% confidence
intervals.
LABORATORY
Std.
Ratio
1
2
3
4
5
6
JC
Dev.
High/Low
SILVER
Mysidopsis bahia
256
(224-301)
300
(256-346)
86
(68-112)
313a
(267-377)
65a
(51.6-87.0)
132
(109-184)
192
111
4.8
NITRATE
Cyprinodon variegatus
1,356
(1,213-1,566)
898
(804-1,035)
441
(393-485)
l,510a
(1,413-1,615)
d
1,876
(1,692-2,023)
1,216
558
4.2
ENDOSULFAN
Mysidopsis bahia
12. 9a
(1.01-1.75)
0.38
,<0. 32-0. 44)
Non-calculable"
0.94
(0.82-1.10)
1.16a
(0.95-1.45)
0.75
(0.48-1.19)
0.94
0.36
3.4
Cyprinodon variegatus
1.15C
(0.72-1.42)
1.1
(1.09-1.12)
0.34C
(0.25-0.42)
0.60
(0.58-0.62)
0.88a
(0.82-0.93)
0.83
(0.70-1.03)
-*"*""
0.81
0.31
3.4
aMoving average LC50 calculation
^Measured concentrations were unacceptable
eBi«omial LC50 calculation
"Test unsuccessful
-------
Table 4. Performance of each laboratory in adhering to diluent and test water condition requirements speci-
fied in ASTM acute toxicity test procedures. A "Yes" designates compliance with the requirements;
"No" designates that the required observations were not made.
LABORATORY
I
2
3
4
5
6
1
2
3
4
5
6
aAll tests
Salinity pH
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes
Salinity
(28 +
1.5 °/oo)
Yes
Yes
Yes
Yes
Yesa
Yes
acceptable, except Acartia tonsa
DILUENT WATER QUALITY
D.O. Particulates
Yes No
Yes Yes
Yes f No
Yes No
Yes No
Yes Yes
TEST WATER CONDITIONS
Temperature
(22 ± 0.5 C) «0.8
Yes
Yes
Yes
Yes
Yes
Yes
endosulfan test, which was conducted
TOG
or
COD
No
Yes
No
No
No
No
PH
range)
Yes
Yes
Yes
Yes
Yes
Yes
at ambient 0
Total
Organochlorine
Pesticides
No
No
No
No
No
Yes
D.O. (>50%
saturation
Yes
.*, Yes
Yes
Yes
Yes
Yes
-8 °/oo)
salinity.
-------
Table 5. Performance of each laboratory in adhering to specific test condition requirements stipulated in the
ASTM acute toxicity test procedures. A "Yes" designates compliance with the requirements; "No"
indicates noncompliance.
LABORATORY STATIC TESTS
Loading
«0.8 g/1)
1 Y5/day)
Yes
Yes
Yes
Nob
Yes
Yes
(1 rag/1).
Accep-
table
Test
Apparatus
Yes
Yes
Yes
No°
Yes
Yes
"Inadequate turnover rate, ^2/day in all flow-through studies.
,—-
cAlthough not in violation of ASTM requirements, apparatus used is not recommended because chemical was
mixed with diluent water in a Mariotte bottle and then delivered.
-------
Table 6. Performance of each laboratory in adhering to test animal acclimation, age, and control mortality
requirements specified in ASTM acute toxicity test procedures and the contract Scope of Work. A
"Yes" designates compliance with the "must" requirements; "No" indicates noncompliance.
LABORATORY
1
2
3
4
5
6
MYSIDOPSIS BAHIA
Age (X2 days)
Yes
Yes
Noa
Yes
Yes
Yes
Control Mortality
Yes
Yes
Yes
Yes
Yes
Yes
CYPRINODON VARIEGATUS
Age Acclimation time
(28 days) (14 days)
Yes
Yes
r
Yes
Yes
Yes
Yes
Yes
Nob
Yes
Yes
Yes
Yes
Control Mortality
Yes
Yes
Yes
Yes
Yes
Yes
a6-8-day-old mysids used in flow-through test.
by-day acclimation used.
o tv> m er
sr (A.' u< ,,
_. ocij w,
" -"
yj' C: fl) V,
a •< •":
-------
Table 7. Performance of each laboratory in adhering to analytical chemistry
requirements in the ASTM acute toxicity test procedures and Scope
of Work. A "Yes" designates compliance with the requirements;
"No" indicates non-compliance.
LABORATORY
1
2
3
4
5
6
1
2
3
4
5
6
Validation of
Analytical Method
Yes
Yes
Yes
No
No
Yes
V
Yes
Yes
Yes
No
No
Yes
SILVER NITRATE
Reagent blank Tests having acceptable
used { measured water concen-
trations (%)a
Yes
Yes
Yes
No
No
Yes
ENDOSULFAN
Yes
Yes
Yes
No
No
Yes
100
100
0
0
0
40
80
67
0
0
40
50
a<30% variation.
14
------- |