OWEP-82-OOl
IN SJIU ACUTE/CHRONIC TOXICOLOGICAL
MONITORING OF INDUSTRIAL EFFLUENTS
FOR THE NPDES BIOMONITORING PROGRAM
USING FISH AND AMPHIBIAN EMBRYO-LARVAL
STAGES AS TEST ORGANISMS
SEPTEMBER 1981
WESLEY J. BIRGE
JEFFREY A. BLACK
OFFICE OF WATER ENFORCEMENT AND PERMITS
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
0WEP-82-0Q1
IN SITU ACUTE/CHRONIC TOXICOLOGICAL MONITORING OF
INDUSTRIAL EFFLUENTS FOR THE NPDES BIOMONITORING PROGRAM
USING FISH AND AMPHIBIAN EMBRYO-LARVAL STAGES AS TEST ORGANISMS
September 1981
Final Report
EPA Contract No. 68-01-5052
Directive of Work No. 14
JRB Associates Subcontract No. 2800-03-218
Wesley J. Birge
Jeffrey A. Black
University of Kentucky
Lexington, Kentucky 40506
Project Manager, DOW 14
Stephen L. Bugbee (First Year)
William F. Brandes tSecond Year)
JRB Representative
Eddy J. Forman
-------
NOTICE
This report has been reviewed through the Office
of Water Enforcement and Permits, EPA, and approved.
Approval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names or
commercial products constitute indorsement or
recommendation for use.
-------
TABLE OF CONTENTS
Page
LIST OF TABLES iii
LIST OF FIGURES v
ACKNOWLEDGMENTS vi
ABSTRACT vi i
INTRODUCTION AND OBJECTIVES 1
RATIONALE AND SCIENTIFIC BACKGROUND 2
GENERAL WORK PLAN 5
EMBRYO-LARVAL BIOMONITORING SYSTEMS AND TEST PROCEDURES 12
Static, static-renewal, and flow-through systems 12
Animal species and duration of exposure 14
Test responses and expression of data 16
INITIAL PERFORMANCE EVALUATIONS 17
Tests with reference toxicant 17
Tests with coal-ash effluent 19
Initial tests with industrial effluents 20
RESULTS OF ON-SITE BIOMONITORING OF MUNICIPAL AND
INDUSTRIAL EFFLUENTS 32
ANALYSIS OF RESULTS AND EVALUATION OF EFFLUENT
BIOMONITORING PROCEDURES 66
Reliability and sensitivity of alternative
embryo-larval test systems 66
Test organisms and responses 67
Comparison of on-site and laboratory testing
of industrial effluents 70
Comparison of acute and embryo-larval effluent
biomonitoring 71
APPLICATIONS OF EFFLUENT BIOMONITORING WITH EMBRYO-LARVAL STAGES ... 77
Characterization of effluents 77
Use and legal defensibility of effluent toxicity data 79
Characterization of receiving waters 84
Use of embryo-larval toxicity testing in the
evaluation of effluent treatability 87
Recommendations for future work 89
SUMMARY 91
Development of embryo-larval test systems 91
Application of test systems to effluent biomonitoring 92
i
-------
TABLE OF CONTENTS - continued.
Page
APPENDIX - TEST PROCEDURES FOR EMBRYO-LARVAL BIOMONITORING 95
Introduction 96
Description and Operation of Embryo-Larval
Biomonitoring Systems 96
Effluent Sampling Procedures, Dilution Water,
and Test Conditions 103
Selection and Handling of Test Organisms 105
A. Fish 106
B. Amphibians 108
Test Responses, Expression of Data, and
Statistical Procedures 110
Performance Evaluations and Personnel Requirements 112
Cost Analysis for Effluent Monitoring Using
Embryo-Larval Toxicity Tests 113
Equipment Inventory 116
REFERENCES 118
-------
LIST OF TABLES
Table Number Page
1 Toxicity tests and biomonitoring studies performed 9
2 Effects of phenol on embryo-larval stages of Xenopus
laevis as determined in flow-through, static-renewal,
and static tests 22
3 Effects of coal-ash effluent on embryo-larval stages
of the bluegill sunfish as determined in flow-through,
static-renewal, and static tests 23
4 Effects of coal-ash effluent on newly hatched larvae
of the bullfrog as determined in flow-through, static-
renewal, and static tests 24
5 Effects of coal-ash effluent on embryo-larval stages
of the rainbow trout as determined in flow-through,
static-renewal, and static tests 25
6 Selected chemical characteristics of coal-ash effluent
used in flow-through, static-renewal, and static
embryo-larval tests 26
7 Initial 12-hour static-renewal embryo-larval tests on
composite and grab samples of industrial effluents
analyzed in the laboratory 27
8 LC50 and LC^ values for composite and grab samples of
industrial effluents analyzed in laboratory
embryo-larval tests 30
9 General characteristics of composite and grab samples
of industrial effluents analyzed in 12-hour static-
renewal embryo-larval tests conducted in the laboratory ... 31
10 On-site toxicity tests on primary and secondary sewage
treatment plant effluents using newly hatched larvae
of the channel catfish 42
11 On-site embryo-larval biomonitoring of a chemical
manufacturing plant final effluent . 43
12 On-site embryo-larval biomonitoring of a synthetic
rubber plant final effluent 46
13 On-site embryo-larval biomonitoring of the chlorinated
final effluent from a secondary sewage treatment plant
receiving tannery waste 49
14 On-site study at a tannery-secondary sewage treatment
plant complex involving embryo-larval biomonitoring of
raw tannery waste, unchlorinated effluent from
secondary treatment, and chlorinated final effluent
from secondary treatment 52
iii
-------
LIST OF TABLES - continued.
Table Number Page
15 Chemical and toxicological characteristics of
three effluents from a tannery-secondary sewage
treatment plant complex 56
16 On-site embryo-larval biomonitoring of a
metal plating plant final effluent 57
17 On-site embryo-larval biomonitoring of final
effluent and effluent components from a metal
plating plant using embryo-larval stages of the
fathead minnow 60
18 Chemical and toxicological characteristics of
effluent components and final effluent from a
metal plating plant 62
19 General characteristics of final effluent, effluent
components, and dilution water used in embryo-larval
biomonitoring experiments 63
20 Comparison of effluent toxicity determined in
on-site and laboratory static-renewal tests using
fish and amphibian enbryo-larval stages 72
21 Comparative evaluations of effluent toxicity as
determined in flow-through acute and embryo-larval
tests with the fathead minnow 76
22 LC50 and LCj values for final NPDES effluents
tested on site 78
23 Life-cycle MATC values compared with embryo-larval
LC^'s for organic compounds 82
24 MATC's compared with LCi values determined in
static-renewal tests with rainbow trout
embryo-larval stages 83
25 Candidate species for use in embryo-larval
toxicity tests on industrial and municipal
effluents 109
iv
-------
LIST OF FIGURES
Page
Figure 1 Static-renewal test system 13
Figure 2 Mobile laboratory and test systems (I) 101
Figure 3 Mobile laboratory and test systems (II) 102
Figure 4 Design of the flow-through effluent
test system 104
v
-------
ACKNOWLEDGMENTS
We should like to gratefully acknowledge the use of certain facilities
and the helpful advice and counsel received from William Peltier, Ron Wei don,
and others at the Region IV EPA Laboratory, Athens, Georgia. In addition,
we are deeply indebted to personnel and graduate students in our laboratory
who contributed to this study or to the preparation of the manuscript,
including William E. McDonnell, Albert G. Westerman, Barbara A. Ramey,
Stephen Ballard, Donald M. Bruser, Richard Cassidy, William H. Benson,
Paul C. Francis, Bobbie Welch, and Pamela Robertson. We should also like
to express our gratitude to Eddy Forman, JRB Coordinator, and to William
Brandes and Stephen Bugbee, Office of Water Enforcement and Permits, EPA,
who served as project managers. We also are appreciative of help received
from Robert Logan (Kentucky Division of Water Quality) whose efforts were
instrumental in obtaining certain effluent test samples.
vi
-------
ABSTRACT
The Clean Water Act has necessitated further requirements and revisions
in the industrial permit program maintained under the National Pollutant Dis-
charge Elimination System (NPDES). It is now evident that in many cases the
identification and control of toxic substances cannot be accomplished solely
on the basis of an effluent guideline approach, emphasizing the need for
reliable and economical procedures with which to quantify directly the net
toxicity of complex effluents and estimate acute and chronic effects on
aquatic biota. Therefore, the major objective of this investigation was
to develop and evaluate fish and amphibian embryo-larval test procedures
for the toxicological characterization of municipal and industrial effluents.
In both laboratory and on-site studies, flow-through, static-renewal,
and static tests were conducted simultaneously on a broad array of effluents.
The principal test organisms included the African clawed frog (Xenopus
laevis), bluegill sunfish (Lepomis macrochirus), channel catfish (Ictalurus
punctatus). fathead minnow (Pimephales promelas), and rainbow trout (Salmo
qairdneri). Except with the trout, exposure was initiated at fertilization
and usually maintained through 4 days posthatching, giving exposure periods
of 6 to 9 days depending upon the species. In most studies with the trout,
exposure was started either at fertilization or at the eyed-egg stage and
continued for 9 days. Combining frequencies for mortality and gross terato-
genesis, probit analysis was used to calculate LCcq and LC-| values with
95% confidence limits. The LC-. was defined as the toxicity threshold and
was expressed as the effluent Concentration (% by volume) required to produce
1% control-adjusted impairment (I.e., mortality, terata) of test populations.
Flow-through tests consistently provided the most reliable evaluations of
effluent toxicity and usually gave the lowest LCcq values. However, LC5q's
determined in flow-through and static-renewal tests usually differed only
by a factor of 2 or less, and the more economical static-renewal procedure
was considered adequate for most routine toxicological screening. Static
tests generally proved more variable and less sensitive and were not rec-
ommended. Concerning the different animal species, optimum results were
obtained with the bluegill sunfish, channel catfish, fathead minnow, and
rainbow trout, but numerous other species should prove satisfactory for
embryo-larval effluent testing.
The majority of effluents were collected from chemical, rubber, and
plastics manufacturing plants, metal plating plants, and sewage treatment
plants. Of 19 industrial and municipal effluents or effluent components
studied, 18 exhibited moderate to high toxicity in embryo-larval tests.
Ten of these were final NPDES effluents which entered receiving waters and,
taking the most sensitive test in each case, the LCcq values were 0.3%,
6.4%, 6.6%, and 21.6% for 4 major effluents analyzes on site and 0.04%,
9.4%, 29.3%, 39.2%, 43.0%, and 100% for the remaining 6 NPDES effluents
vi i
-------
which were tested only in the laboratory. Corresponding LC, values
ranged from 0.001% to 2.63% for the 6 NPDES effluents analyzed in the
laboratory and 0.001% to 0.8% for those analyzed on-site. During these
4 on-site studies, conducted using a mobile laboratory, effluent samples
also were collected and transported to the laboratory for simultaneous
testing. In all instances, effluent toxicity was substantially less when
measured in the laboratory. Also in the 4 major field biomonitoring
studies, both fish acute and embryo-larval tests were performed con-
currently on the NPDES effluents. The acute toxicity determinations were
conducted by the EPA region IV biomonitoring team and consisted of 96-hr
flow-through tests with the fathead minnow. Embryo-larval tests gave more
reliable detection and better quantification of effluent toxicity. The
LCcq and LC-. values for the 4 effluents, determined in flow-through tests
witn the fathead minnow, ranged from 0.3% to 29.4% and 0.001% to 2.8%,
respectively, and narrow 95% confidence intervals were observed in all
cases. In the acute tests, it was not possible to determine LCcq values
for 3 of the 4 effluents. The acute LCcq for the most toxic effluent was
8.0% and this was about 27 times the emBryo-larval LCcq (0.3%), and it
differed from the embryo-larval LC-j value (0.001%) by more than 3 orders
of magnitude.
On the basis of these and other results, it was concluded that embryo-
larval tests provided a sensitive, reliable, and economical means of
quantifying the toxicity of complex effluents. Test results (J_«e.«» LC-.)
also can be used to calculate dilution factors required to preclude signifi-
cant mortality and teratogenesis of sensitive reproductive stages. Such
dilution factors, together with transport-fate data and other essential
information, should prove useful in assessing the impact of an effluent
upon its receiving system. In 3 on-site studies, tests also were conducted
on effluent components or effluents at different stages of treatment to
determine the utility of embryo-larval biomonitoring for identifying toxic
effluent fractions and for determining the effectiveness of waste treatment
processes. Results indicated that more accurate and definitive assess-
ments were possible with toxicity data than with chemical parameters.
In view of results of this and earlier studies, it also was proposed
that embryo-larval tests can be used to estimate effluent concentrations
that produce chronic effects on aquatic biota and which result in long-
term ecological degradation. In addition, embryo-larval test procedures
developed for effluent biomonitoring appeared equally suitable for direct
evaluations on receiving waters. In determinations made under actual
environmental conditions, net effects of important variables, including
toxic interactions and other factors which affect the toxicity and bio-
availability of effluent contaminants, are directly reflected in test
responses.
viii
-------
1
INTRODUCTION AND OBJECTIVES
The Federal Water Pollution Control Act, as amended in 1977 by the
Clean Water Act (PL 95-217), provides statutory limitations on toxic
discharges which may endanger water quality and environmental health.
Under section 101(a)(3), it is stated that the national goal is to pro-
hibit discharges of toxic pollutants in toxic amounts. Compliance with
the Clean Water Act has necessitated further requirements and revisions
in the industrial permit program maintained under the National Pollutant
Discharge Elimination System (NPDES). Recent experiences gained in
implementing and enforcing NPDES regulations have reaffirmed the fact
that in many cases identification and control of toxic substances cannot
be accomplished solely on the basis of an effluent guideline approach,
and the Environmental Protection Agency's Office of Water Enforcement
and Permits has stressed the need for biomonitoring to provide supple-
mental toxicity data. This emphasizes the necessity for reliable and
scientifically valid procedures with which to obtain direct quantifica-
tion of the net toxicity of complex effluents and to estimate acute and
chronic effects on aquatic biota of receiving waters. Accordingly, this
project was undertaken to design test systems and develop methods for in
situ (on-site) embryo-larval toxicological monitoring of industrial wastes,
and to evaluate such test systems for toxicity screening as applied in
the NPDES program.
The principal objectives of this study included 1) development of
enbryo-1arval test systems using fish and amphibian species for on-site
-------
2
toxicological evaluations of complex effluents; 2) application of embryo-
larval test systems to on-site biomonitoring of municipal and industrial
effluents; 3) comparisons of sensitivity and reliability of embryo-larval
and acute effluent biomonitoring; 4) evaluation of embryo-larval toxicity
tests for use under the NPDES program; and 5) a description of effluent
biomonitoring procedures using sensitive life-cycle stages of fish and
amphibians. Further work involved determining the reliability of labora-
tory testing of composite effluent samples, using direct on-site biomonitor-
ing as a basis for comparison.
RATIONALE AND SCIENTIFIC BACKGROUND
Present effluent guidelines are based largely on general water quality
characteristics (^.£., pH, BOD, suspended solids) and criteria for specific
elements and compounds. However, many chemical, physical, and biological
variables can affect the net toxicity of a complex effluent as well as its
impact upon receiving waters and aquatic biota. In addition, toxicant
criteria and other guidelines developed largely from laboratory tests are
prone to some imprecision when applied under actual field conditions. This
problem is further compounded by the fact that many municipal and industrial
effluents contain large suites of contaminants which simultaneously enter
receiving waters. Thus, using the present guideline approach, it is
difficult 1) to quantify effluent toxicity accurately, 2) to assess the
extent to which effluent toxicity varies with ambient conditions, and 3)
to provide reliable estimates of environmental impact.
-------
3
On-site biomonitoring has shown promise as a reliable means of quan-
tifying biological effects of complex effluents (1, 2). Such technology,
if properly applied, can provide direct toxicological readout
mortality, teratogenesis) for combined effects of full suites of toxicants
contained in effluents and receiving waters. Net effects of toxic inter-
actions and other variables are directly reflected in test responses. As
the latter are determined under actual field conditions (£.£., ambient
toxicant concentrations, pH, hardness, suspended solids), this minimizes
the need to extrapolate from laboratory data. In addition to evaluating
whole effluents, direct toxicological monitoring may be conducted simul-
taneously on effluent dilutions, permitting determination of dilution
ratios required to reduce or preclude toxic effects. On-site toxicological
monitoring also may be applied to effluent treatability, providing a
quantitative basis for identifying toxic effluent fractions and/or
evaluating effectiveness of waste treatment procedures (3).
Peltier (2), in an in-depth study, established the basis for applying
acute toxicity testing to effluent biomonitoring, particularly under the
NPDES program. In addition to developing a detailed test protocol for
acute tests with fish and certain invertebrates, he clearly demonstrated
the advantages of and indeed the necessity for direct toxicological
characterization of complex effluents. He stated:
Since it is not economically feasible to determine the toxicity
of each of the thousands of potentially toxic substances in
complex effluents or to conduct exhaustive chemical analyses of
effluents, the most direct and cost-effective approach to the
measurement of the toxicity of effluents is to conduct a bioassay
with aquatic organisms representative of indigenous organisms in
receiving waters.
-------
4
As noted above, the purpose of this study was to investigate the use
of embryo-larval testing to provide improved detection and quantification
of effluent toxicity and to consider the use of such testing to estimate
chronic toxicity. Several recent investigations lend support to this
prospect. Using data from a substantial number of studies conducted with
a broad selection of organic and inorganic toxicants, McKim (4) estimated
maximum acceptable toxicant concentrations (MATC's) from embryo-larval
responses observed through 30 to 90 days posthatching. A high correlation
was established between these estimated MATC's and those determined in life-
cycle tests. However, 30- to 90-day tests still exceed cost limitations
required for successful implementation of on-site effluent biomonitoring
under the NPDES program. More recently, Birge, _et_ al_. (5-7) conducted
short-term embryo-larval tests on numerous inorganic and organic toxicants.
Exposure was initiated at or soon after fertilization and continued through
4 to 8 days posthatching. Dose-response data were subjected to log probit
analysis (8), and LCj values, defined as toxicant concentrations which
produced 1% control-adjusted impairment in test populations, compared
favorably with MATC's estimated or determined in 30- to 90-day embryo-larval
and life-cycle tests (Tables 23, 24). As a number of fish and amphibian
species suitable for testing have egg hatching times in the range of 2 to
5 days (.§.£., bluegill sunfish, fathead minnow, Xenopus laevis), it appears
plausible that embryo-larval tests of 6 to 9 days may constitute a sensitive,
reliable, and cost-efficient means of screening effluents for toxicity.
Despite the short duration of such tests, fertilized eggs, all embryonic
-------
5
stages, and early larvae are subjected to exposure. Due to the great
complexity of animal development, which involves gene expression and many
synthetic processes regulated by sensitive enzyme systems, embryos and
larvae are subject to a broad spectrum of toxicant-induced responses. For
example, aquatic toxicants may affect biochemical, physiological, and other
phenomena associated with 1) fertilization, 2) gene expression and cellular
differentiation, 3) proliferation and growth, 4) systemic functions, and
5) the initial accommodation to a free-living existence. Due to their
sensitivity and simple culture requirements, eggs and larvae of aquatic
organisms are particularly suitable for use in toxicity testing (9, 10).
GENERAL WORK PLAN
This two-year investigation was initiated June 19, 1979, and the
various studies conducted are summarized in Table 1. The initial task was
to adapt embryo-larval procedures used in toxicity testing for application
to effluent biomonitoring. Using a reference toxicant (i..e.., phenol), coal-
ash effluents, and effluents or process waters from selected industries,
laboratory studies were undertaken to compare economy of operation and
reliability of static, static-renewal, and flow-through test procedures
and to modify conventional test systems for optimum use with the broad
array of complex effluents likely to be encountered under the NPDES program.
Specific tasks during the first year of the project were as follows.
o Perform laboratory embryo-larval tests on effluents (l.£., 24-hr
composites) and a reference toxicant (j_..e., phenol).
-------
6
o Compare applicability and reliability of embryo-larval static,
static-renewal, and flow-through procedures for testing complex
effluents.
o Modify test systems for practical application to field testing.
o Consider effects of experimental conditions (.e.j},., pH, DO) on
toxicological monitoring and define test parameters which insure
reliable and reproducible results.
o Evaluate and compare sensitivity of alternative test animals (_e.£.,
fish, amphibians) and select species suitable for use in effluent
testing.
o Design a mobile laboratory for on-site biomonitoring, including
installation of embryo-larval test systems and necessary supporting
faci1ities.
During the second year of the project, five on-site studies were
conducted. The first involved a waste treatment plant in Lexington,
Kentucky. Seven-day tests were performed simultaneously on two different
sewage effluents. One was taken after primary treatment and the other was
collected after secondary treatment but prior to chlorination. This initial
field study was conducted 1) to evaluate and perfect set-up and testing
procedures, 2) to develop adequate standards for quality assurance, and
3) to determine work loads, staff assignments, and management policies.
In addition, selection of this monitoring site was intended to demonstrate
the use of embryo-larval testing as a means of evaluating the effectiveness
of waste treatment procedures.
During the remainder of the project, four principal on-site studies
were conducted outside the Lexington area. These involved major industries
selected in consultation with JRB Associates and EPA, and included the
following:
-------
7
1. Chemical Manufacturing Plant
2. Synthetic Rubber Plant
3. Metal Plating Plant
4. Tannery-Secondary Sewage Treatment Plant Complex
Under conditions agreed upon for this investigation, identification of
specific industries was not to be disclosed. In each of the four cases,
project requirements involved testing the final NPDES effluent. However,
at two sites, tests also were conducted on various effluent components or
on effluents at various levels of treatment. This concluding phase of
the investigation involved several major tasks, including 1) final design
and evaluation of on-site embryo-larval biomonitoring procedures for
detecting and quantifying effluent toxicity, 2) use of on-site embryo-
larval testing to determine effectiveness of waste treatment, 3) comparison
of acute and embryo-larval biomonitoring as applied to complex effluents,
and 4) comparison of on-site biomonitoring with laboratory testing of
effluent samples.
For each NPDES effluent, embryo-larval testing was performed with
static, static-renewal, and flow-through procedures, using the same animal
species fathead minnow, Xenopus laevis). In addition, four test
organisms were compared for sensitivity using one or more test systems
(!•£•» static-renewal, flow-through). During on-site biomonitoring,
effluent samples also were collected and transported to the laboratory
(University of Kentucky) for simultaneous testing. Furthermore, studies
at each of the last four sites were planned to coincide with visits by
-------
8
the EPA Region IV biomonitoring unit. Using the same effluent and dilution
water sources, the EPA team performed on-site acute toxicity tests with
the fathead minnow, and this formed the basis of comparison between acute
and embryo-larval biomonitoring.
-------
Table 1. Toxicity tests and biomonitoring studies performed.
Test* Study Site Test System Test Organism
Initial Laboratory Studies (First Year)
Phenol
U.K.
Lab2
Flow-through
Xenopus laevis
U.K.
Lab
Static-renewal
Xenopus laevis
U.K.
Lab
Static
Xenopus laevis
Coal-Ash Effluent
U.K.
Lab
Flow-through
Rainbow Trout, Bluegill
Sunfish,
Bullfrog
U.K.
Lab
Static-renewal
Rainbow Trout, Bluegill
Sunfish,
Bullfrog
U.K.
Lab
Static
Rainbow Trout, Bluegill
Sunfish,
Bullfrog
Chemical Manufacturing Plant #1
Undiluted process water
U.K.
Lab
Static-renewal
Bluegill Sunfish
Chemical Manufacturing Plant #1
Diluted process water
(final effluent)
U.K.
Lab
Static-renewal
Bluegill Sunfish
Chemical Manufacturing Plant #2
Cooling water and storm runoff
(final effluent)
U.K.
Lab
Static-renewal
Channel Catfish
Chemical Manufacturing Plant #3
Undiluted process water
(final effluent)
U.K.
Lab
Static-renewal
Channel Catfish
Synthetic Rubber Plant #1
Undiluted process water3
(final effluent)
U.K.
Lab
Static-renewal
Bluegill Sunfish
Synthetic Rubber Plant #2
Cooling water and storm runoff
(final effluent)
U.K.
Lab
Static-renewal
Channel Catfish
Plastics Manufacturing Plant
Cooling water
(final effluent)
U.K.
Lab
Static-renewal
Channel Catfish
-------
Table 1 - continued.
Test*
Study Site Test System
Biomonitorinq Studies (Second Year)
Lexington Sewage Treatment Plant
Primary effluent
Unchlorinated secondary
effluent
Chemical Manufacturing Plant
Diluted process water
(final effluent)
Synthetic Rubber Plant
Undiluted process water^
(final effluent)
Tannery-Sewage Treatment Plant
Chlorinated, secondary effluent
(final effluent)
On Site
On Site
On Site
On Site
On Site
On Si te
U.K. Lab
On Site
On Site
On Site
U.K. Lab
On Site
On Site
On Site
U.K. Lab
Static-renewal
Static-renewal
Flow-through
Flow-through
Static-renewal
Static
Static-renewal
Flow-through
Static-renewal
Static
Static-renewal
Flow-through
Static-renewal
Static
Static-renewal
Test Organism
Channel Catfish
Channel Catfish
Channel Catfish
xenopus laevis, Fathead Minnow
xenopus laevis, Fathead Minnow, Rainbow Trout
xenopus laevis. Fathead Minnow
Xenopus laevis
xenopus laevis, Fathead Minnow
xenopus laevis, Fathead Minnow, Rainbow Trout
Xisnopusr laevis
Rainbow Trout
xenopus laevis, Fathead Minnow
xenopus laevis, Fathead Minnow, Rainbow Trout
Xenopus laevis
Rainbow Trout
-------
Table 1 - continued.
Test*
Study Site
Test System
Test Organism
Tannery-Sewage Treatment Plant (cont.)
Unchlorinated secondary
effl uent
On Site
Static-renewal
Xenopus
laevis,
Fathead
Minnow, Rainbow Trout
Tannery effluent
On Site
Static-renewal
Xenopus
laevis,
Fathead
Minnow, Rainbow Trout
Metal Plating Plant
Mixed waste
(final effluent)
On Site
On Site
Flow-through
Static-renewal
Xenopus
Xenopus
laevis,
laevis,
Fathead
Fathead
Minnow
Minnow
On Site
Static
Xenopus
laevis,
Fathead
Minnow
U.K. Lab
Static-renewal
Xenopus
laevis
Component 1
(sludge-bed filtrate)
On Site
Static-renewal
Fathead
Minnow
Component 2
(process water)
On Site
Static-renewal
Fathead
Minnow
Component 3
(brazing water)
On Site
Static-renewal
Fathead
Minnow
Component 4
On Site
Static-renewal
Fathead
Minnow
(cooling water)
Final effluents include only those discharges which were subject to NPDES regulations.
2
Laboratory of aquatic toxicology and ecology (W.J. Birge), School of Biological Sciences,
University of Kentucky.
3
The NPDES discharge contained undiluted process water mixed with certain other wastes (e.c[., sanitary discharge).
-------
12
EMBRYO-LARVAL BIOMONITORING SYSTEMS AND TEST PROCEDURES
Static, static-renewal« and flow-through systems. As noted above,
procedures for effluent biomonitoring using fish and amphibian embryo-larval
stages were modified from those previously developed for conventional tox-
icity testing (1, 5, 7, 9, 11). Deep petri dishes served as exposure
chambers in all tests. These covered Pyrex dishes were modified for use
in static-renewal and flow-through systems. In static-renewal tests,
effluent and dilution water were changed at regular 12- or 24-hr intervals.
A combination inlet-outlet tube was attached to the lower wall of the
exposure chamber, approximately 8 mm above the bottom of the dish, to
permit solution changes without disturbing test organisms. Generally 12
exposure chambers were used in each static-renewal test, permitting dupli-
cate dishes for controls and each of five effluent concentrations (.e .jj,.,
100%, 50%, 10%, 1%, 0.1%). The static-renewal exposure chamber and the
system used for effluent changes were described in an earlier publication
(12) and are illustrated further in Figures 1 and 2. The chamber holds
400 mL of test solution, and approximately 50 mL remain when emptied to
the level of the combination inlet-outlet. The renewal interval was 12 hrs,
except in tests with phenol (reference toxicant) and effluents from the
last four biomonitoring sites (Table 1). During these studies, solutions
were changed at regular 24-hr intervals. Control populations were main-
tained with the same dilution water source and renewal interval as specified
for the effluents. Static tests were performed using the same procedures
-------
EMBRYO-LARVAL
TEST CHAMBER
Three-Way
Valve
EFFLUENT AND
DILUTION WATER
AIR
PUMP
Rapid Disconnect
u>
Figure 1. Static-renewal test system. Test chambers (400 mL capacity),
modified from deep petri dishes, are equipped with a combination
inlet-outlet tube positioned approximately 8 mm above the bottom
of the dish and constructed of 4-mm I.D. glass tubing. A 3-way
valve fitted to the glass tube makes it possible to change effluent
and dilution water and to maintain aeration without direct handling
of the exposure chambers. Effluent and dilution water are changed
at regular 12-hr intervals. Solutions are added to the exposure
chamber, using a 3" Pyrex funnel attached to the 3-way valve by a
short section of tubing. Exposure chambers are emptied to the
level of the combination inlet-outlet using an additional line
attached to the 3-way valve. Between solution changes, this outlet
line is used to provide continuous, moderate aeration to the exposure
chambers if required to maintain dissolved oxygen at a minimum of
60% saturation.
WASTE
RESERVOIR
-------
14
as described for static-renewal testing, except that standard deep petri
dishes served as exposure chambers and solution changes were omitted.
Inlet and outlet tubes (10 mm I.D. Pyrex) were annealed to petri
dishes used as exposure chambers in the flow-through tests. The inlet was
positioned approximately 7 mm above the bottom of the dish, and the outlet
was attached on the opposite side, just below the shoulder. Volume to the
level of the outlet was approximately 300 mL, and flow rate was maintained
at 200 mL/hr, giving a retention time of 1.5 hr. Flow rate was monitored
using either timed volumetric measurements or Gilmont no. 12 flow meters.
Control water, ful 1-strength effluent, and 50% effluent were delivered to
respective exposure chambers using variable speed, multiple-channel peri-
staltic pumps. Effluent concentrations below 50% were obtained using
a serial diluter system previously designed by Freeman and Birge (unpublished
observations). The diluter was provided with standpipes which were adjusted
as necessary to maintain constant dilution ratios. The latter were determined
daily, using timed volumetric measurements on effluents and dilution water.
Overall fluctuation in exposure concentration usually was within 5% and
seldom exceeded 10%. Test chamber solutions were monitored daily for pH,
alkalinity, hardness, conductivity, dissolved oxygen, and temperature.
Test procedures and the diluter system are further described in the
Appendix. In the studies that follow, only those toxic discharges which
were subject to NPDES regulations are designated as final effluents.
Animal species and duration of exposure. During the first year of
the project, all testing was conducted in the laboratory using a reference
-------
15
toxicant (i.,e., phenol), coal-ash effluents, and various industrial effluents
(composite and grab samples). Animal species used in these embryo-larval
investigations included the African clawed frog (Xenopus laevis), bluegill
sunfish (Lepomis macrochirus), channel catfish (Ictalurus punctatus), and
rainbow trout (Salmo qairdneri). Exposure was initiated within 12 hrs of
fertilization and continued through 4 to 8 days posthatching. Average
hatching times for the warmwater species (20-22°C) were 2 days for Xenopus,
3 days for bluegill, and 5 days for catfish. For the rainbow trout, the
hatching period was 23 days (12.5°C). In addition, a limited number of
96-hr tests were initiated using 1-day-old larvae of the bullfrog (Rana
catesbeiana).
During the second year, the initial on-site study was conducted at the
Lexington Sewage Treatment Plant, and effluents from primary and secondary
treatment were subjected to 7-day tests initiated with 1-day-old larvae of
the channel catfish. During the four additional on-site biomonitoring
investigations conducted in the second year, including the correlated
laboratory studies, effluent toxicity tests were performed using embryos
and larvae of Xenopus, fathead minnow, and rainbow trout. Egg exposure
for warmwater species was initiated 8 to 12 hrs after fertilization,
except during the last two on-site biomonitoring studies (i_.,e., Tannery-
Sewage Treatment Plant Complex, Metal Plating Plant) in which exposure of
fathead minnow eggs was delayed up to 24 hrs after spawning. However, as
eggs were transported in an ice-packed container, thereby slowing develop-
mental time, the exposure period to hatching did not vary significantly
-------
16
from that observed in other tests with this species. In addition, trans-
port conditions did not appear to affect egg viability, as the survival of
control organisms ranged from 80% to 92% in flow-through and static-renewal
tests. Exposure to effluents usually was maintained through 4 days post-
hatching, except in a few on-site tests which were delayed due to the
unavailability of eggs. In such cases, treatment was curtailed 3 days
after hatching due to time constraints imposed on field investigations.
In embryo-larval tests conducted with the warmwater species, temperature
was maintained at 20-22°C and average hatching times were 2 days for Xenopus
and 4.5 days for fathead minnows. In studies with the rainbow trout, tempera-
ture was maintained at 11.5-12.5°C and both fresh eggs and eyed eggs were
used as test organisms. In the case of fresh eggs, exposure was initiated
30 min to 2 hr after fertilization and maintained for 9 days. Eyed eggs
were exposed through 4 days posthatching and average hatching time was
5 days, giving an exposure period of 9 days.
Test responses and expression of data. Test responses included
frequencies of egg hatchability, embryo-larval survival, and teratogenesis.
Determinations of teratic organisms were limited to gross defects considered
likely to preclude survival, as discussed previously by Birge, et al_. (11,
13). Defects most commonly encountered were acute kyphosis, lordosis,
scoliosis, and other gross anomalies of the vertebral column. Percent
hatchability was based on all organisms, normal and aberrant, which lived
to complete the hatching process and frequencies of teratic organisms ob-
served among experimental and control animals were expressed as percentages
-------
17
of newly hatched populations. Depending upon test conditions and the
supply of eggs, sample size (organisms per dish) usually ranged from 50
to 100 and was never less than 25. In determining percent survival,
teratic organisms were counted as lethals, except when tests were
terminated prior to hatching , fresh trout eggs). Taking accumu-
lative dose-response data at the end of the exposure period, log probit
analysis (8) was used to determine LC^g values (percent effluent by
volume) with 95% confidence limits. In addition, LC-j values, defined
as concentrations producing 1% control-adjusted impairment of test
populations, were calculated for most effluents. Mortality consistently
was the predominant test response. Moreover, as teratogenesis was based
solely upon severe defects which preclude survival or normal reproduction
and as it is particularly important to quantify effluent toxicity in terms
of organismal impairments that affect population density and viability, it
was deemed more appropriate to express combined test responses as lethal
concentrations (LC) rather than as effective concentrations (EC).
INITIAL PERFORMANCE EVALUATIONS
Tests with reference toxicant. Phenol initially was selected as a
reference toxicant with which to compare reliability and sensitivity of
flow-through, static-renewal, and static test procedures. Embryo-larval
stages of Xenopus laevis were used as test organisms and exposure was
maintained from fertilization through 8 days posthatching, giving a
treatment time of 10 days. In the flow-through study, good regulation
-------
18
of phenol concentrations was obtained and the LC^q was 7.5 mg/L (Table 2).
Though the duration of exposure was somewhat longer, this value was within
the range of LCgg's reported in a previous investigation with six other
amphibian species (14). When embryo-larval responses were analyzed 4 days
after hatching in phenol tests conducted at a water hardness of 100 mg/L
as CaCOg. LC^'s varied from 0.04 to 0.23 mg/L when determined with more
sensitive amphibian species (.e.jj,., Rana pi pi ens, Rana catesbeiana) and LC&q
values ranged from 2.45 to 9.87 mg/L in tests with more tolerant species
(l.£., Bufo fowleri, Rana palustris). In toxicity tests conducted through
8 days posthatching at a water hardness of 50 mg/L, LC^g values were 0.54
and 1.19 mg/L when determined with embryo-larval stages of the rainbow
trout and goldfish, respectively (9). This is in general agreement with
the on-site effluent studies reported below, in which Xenopus embryos and
early larvae usually were more tolerant than similar stages of the fathead
minnow and rainbow trout.
Compared to results obtained in the flow-through test, there was
approximately a two-fold reduction in toxicity when phenol was administered
using static-renewal procedures, and no significant effects on test organisms
were observed in the static test (Table 2). Though accurate mean phenol
exposures could not be determined in static and static-renewal tests,
analyses performed on water samples from the exposure chambers revealed
substantial reductions in phenol concentration with time. Analyses
performed during the static-renewal study indicated that actual phenol
concentrations were within 5% of nominal concentrations when water samples
-------
19
were collected at the beginning of the 24-hr renewal intervals. However,
phenol concentrations usually dropped to 50% or less of nominal values when
determinations were performed at the end of the 24-hr renewal intervals. By
comparison, phenol retention at the conclusion of the static test generally
was less than 10%. As biomass volume was not excessive (< 0.5 cc per exposure
chamber), tissue uptake likely was not the principal factor affecting phenol
exposure concentrations.
Tests with coal-ash effluent. The three embryo-larval test procedures
were further evaluated in studies with a complex coal-ash effluent. The
latter was produced in the laboratory, using procedures previously described
by Birge (1). This involved flow-through washing of a 50-kg sample of
precipitator-collected fly ash from a coal-fired power plant. The rate of
flow was 1 L/hr and retention time in the fly-ash leaching chamber was 42
hrs. Effluent toxicity was evaluated using flow-through, static-renewal,
and static tests with embryonic and larval stages of each of three organisms
(j_.£., rainbow trout, bluegill sunfish, bullfrog). The results are presented
in Tables 3-5. Selected chemical characteristics of the influent water and
the resulting effluent are given in Table 6. The same influent water source
also was used to perform effluent dilutions and to maintain control organisms.
Coal-ash effluents are known to contain detectable concentrations of many
toxic metals, and trace metal concentrations decrease with leaching time
(1). Due to their predictable leaching patterns and high concentrations in
ash effluents, aluminum and zinc were selected as reference toxicants with
which to compare exposures in the three test systems.
-------
20
The LC,jq values did not differ significantly when determined using
flow-through or static-renewal procedures with bluegill embryo-larval stages
or early bullfrog larvae (Tables 3, 4), and the median lethal concentrations
varied by only a factor of two in tests with the trout (Table 5). However,
toxicity of the coal-ash effluent was appreciably greater in static tests
and this presumably was due to higher concentrations of toxic metals, as
reflected by the values obtained for aluminum and zinc (Table 6). Several
factors undoubtedly contributed to this situation. For example, concentra-
tions of coal-ash metals (J_.£., A1, Zn) were much more stable in aqueous
test systems than was phenol. In addition, as noted above, metal concen-
trations decreased with leaching time and, therefore, the effluent became
progressively less toxic during the course of testing. While this diminu-
tion in toxicity was reflected in the flow-through and static-renewal tests,
mean toxicity over the duration of the exposure period was overestimated in
the static system due to the one-time sampling of the ash effluent.
Initial tests with industrial effluents. The first application of
embryo-larval toxicity testing to actual industrial effluents was undertaken
in the laboratory. Composite and grab samples of seven different effluents
from six major industries, including the final NPDES effluent from each, were
evaluated using static-renewal procedures with developmental stages of the
bluegill sunfish and channel catfish (Table 7). In all cases, tests were
performed with undiluted effluent, as well as with dilutions ranging down
to 0.01% effluent. Good delineation of dose-response data was observed
by 4 days posthatching in all but one case, and LC^q values ranged from
-------
21
0.04% for process water from Chemical Manufacturing Plant No. 3 to 43% for
process water from Synthetic Rubber Plant No. 1. Lowest toxicity was
observed for cooling water/storm runoff from Chemical Manufacturing Plant
No. 2. Though the median lethal concentration could not be determined, it
was estimated to be about 100% effluent.
The LC-j values determined for the seven effluent sources ranged from
0.001% to 2.63%. The LC-| was defined as the percent effluent concentration
which produced 1% control-adjusted impairment in test populations, and this
value can be used to estimate the effluent dilution required to preclude
appreciable toxicity to embryos and early larvae. The LC^q and LC-j values
are summarized in Table 8 and general effluent characteristics are given in
Table 9. No unusual variations in effluent characteristics were noted except
for high conductivity readings obtained for four of the seven effluents.
An extremely high conductivity of 75,000 umhos/cm was observed for the most
toxic effluent (process water, Chemical Manufacturing Plant No. 3). However,
the next highest conductivity (4,420 pmhos/cm) was recorded for the second
least toxic effluent (Synthetic Rubber Plant No. 1), indicating that there
was no consistent correlation between conductivity and toxicity.
-------
22
Table 2. Effects of phenol on embryo-larval stages of xenopus laevis
as determined in flow-through, static-renewal, and
static tests.
Phenol
Concentration (mg/L)
Percent Survival at
8 Days Posthatchingl
Test System
Nominal
Actual
(mean ± S.E.)
Flow-through
10.0
1.00
0.100
0.010
0.0010
12.6 ± 0.6
1.30 ± 0.06
0.112 ± 0.012
0.0096 ± 0.0011
0.0014 ± 0.0005
42
73
95
98
98
LC50 (mg/L)
(95% confidence limits)
7.5
(4.7 - 14.2)
Static-renewal
10.0
1.0
0.1
0.01
0.001
-
56
80
86
98
100
LC50 (mg/L)
(95% confidence limits)
18.0
(8.1 - 57.9)
Static
10.0
1.0
0.1
0.01
0.001
-
93
97
97
99
94
LC50 fag/L)
ND2
^Average hatching time was 2 days, giving an exposure period of 10 days.
Survival values were control-adjusted in this initial test.
2
Not determined.
-------
23
Table 3. Effects of coal-ash effluent on embryo-larval stages of the
bluegill sunfish as determined in flow-through, static-renewal,
and static tests.
Test System
Ash Effluent
Concentration
(Percent)
Percent ,
Hatchability
Percent Survival at
4 Days Posthatching
Flow-through
100
39(11)
33
10
76(1)
73
1
94
94
0.1
95
94
0.01
98
97
Control
99
98
LC50 (% Effluent) 43-2
(95% confidence limits) (30.4 - 66.2)
Static-renewal 100 46(10) 42
10 70(1) 68
1 86 82
0.1 94 94
0.01 98 97
Control 97 97
LC50 (% Effluent) 52.3
(95% confidence limits) (29.1 - 100)
Static 100 34(14) 24
10 64(3) 59
1 82 78
0.1 91 88
0.01 95 92
Control 96 94
LC50 (% Effluent) 19.4
(95% confidence limits) (11.7 - 33.2)
Egg hatchability was based on all animals, normal and aberrant, which
completed hatching. Frequencies of teratic survivors appearing in
hatched populations were expressed parenthetically.
-------
24
Table 4. Effects of coal-ash effluent on newly hatched
larvae of the bullfrog as determined in
flow-through, static-renewal, and static tests.
Test System
Ash Effluent
Concentration
(Percent)
Percent Survival
after 96 Hours
F1ow-through
100
43
10
75
1
85
0.1
91
0.01
98
Control
98
LC50 (% Effluent) 83.5
(95% confidence limits) (45.8 - 100)
Static-renewal 100 49
10 70
1 86
0.1 95
0.01 98
Control 98
LC50 (% Effluent) 95.1
(95% confidence limits) (51.2 - 100)
Stati c 100 33
10 65
1 84
0.1 94
0.01 97
Control 96
LC50 (% Effluent) 31.0
(95% confidence limits) (20.1 - 50.5)
-------
25
Table 5. Effects of coal-ash effluent on embryo-larval stages of the
rainbow trout as determined in flow-through, static-renewal,
and static tests.
Ash Effluent
Percent .
Percent Survival at
Test System Concentration
(Percent)
Hatchability
4 Days Posthatching
Flow-through 100
23(8)
20
10
69(2)
67
1
84(1)
83
0.1
92
92
0.01
93
93
Control
94(1)
93
LC50 (% Effluent)
25.2
(95% confidence limits)
(20.3 - 31.1)
Static-renewal 100
41(15)
35
10
83(5)
78
1
91(3)
87
0.1
94(1)
93
0.01
95(1)
94
Control
97
97
LC50 (% Effluent)
53.4
(95% confidence limits)
(40.9 - 72.3)
Static 100
18(40)
11
10
58(10)
52
1
85(5)
81
0.1
92(2)
90
0.01
94(1)
93
Control
97
97
LC50 (% Effluent)
10.1
(95% confidence limits)
(8.0 - 12.6)
Egg hatchability was based on all animals, normal and aberrant, which
completed hatching. Frequencies of teratic survivors appearing in
hatched populations were expressed parenthetically. Exposure was
initiated at fertilization and continued through 4 days posthatching.
-------
Table 6. Selected chemical characteristics of coal-ash effluent used in flow-through,
static-renewal, and static embryo-larval tests.
Effluent
Characteristics
1,2
Test System
Flow-through Static-renewal
Static
Dilution
Water
pH
7.7
+
0.1
7.9
+
0.1
7.8 ± 0.1
7.8
±
0.1
Alkalinity (mg/L as CaCOj)
58
+
2
61
+
2
60 ± 1
65
±
2
Hardness (mg/L as CaCO^)
75
+
7
70
+
6
75 ± 3
74
±
6
Conductivity (umhos/cm)
145
+
4
144
±
5
141 ± 2
138
±
3
Aluminum (yg/L)
20
±
10
20
+
10
120 ± 50
10
+
10
Zinc (pg/L)
11.0
+
1.9
14.0
+
2.9
36.0 ± 10.8
5.0
+
1.8
1
Chemical characteristics expressed as mean ± standard error In tests with bluegill sunfish.
?
"Aluminum and zinc were selected as reference toxicants.
-------
Table 7. Initial 12-hour static-renewal embryo-larval tests on composite and grab samples of
industrial effluents analyzed in the laboratory.
Test
Species
Effluent
Percent ,
Hatchability
Percent Survival
Effluent Source
Concentration
(Percent)
Hatching
4 Days
Posthatching
Chemical Manufacturing Plant #1
Undiluted process water
Bluegill
Sunfish
100
50
10
5
1
0.1
0.01
Control
20(59)
55(12)
79(2)
90
95
98
98
99
8
48
77
90
95
98
98
99
0
22
63
82
91
97
97
98
LC50 (% Effluent)
(95% confidence limits)
16.0
(13.1 - 19.2)
Chemical Manufacturing Plant #1
Diluted process water
(final effluent)
Bluegil1
Sunfish
100
50
10
5
1
53(14)
74(5)
90(1)
96
98(1)
45
71
89
96
97
0
47
79
89
95
0.1 98 98 95
0.01 99 99 98
Control 99 99 98
LC50 (% Effluent)
(95% confidence limits)
29.3
(24.1 - 35.1)
-------
Table 7 - continued.
Effluent Source
Test
Species
Effluent
Concentration
(Percent)
Percent ,
Hatchability
Percent
Hatching
Survival
4 Days
Posthatching
Chemical Manufacturing Plant §2
Channel
100
59(7)
55
45
Cooling water and storm runoff
Catfish
50
84(2)
82
78
(final effluent)
10
93
93
91
5
97
97
94
1
96
96
93
0.5
97
97
96
0.1
100
100
98
0.01
100
100
98
Control
98
98
97
LC50 (% Effluent)
-vlOO
Chemical Manufacturing Plant #3
Channel
100
0
0
0
Undiluted process water
Catfish
50
0
0
0
(final effluent)
10
0
0
0
5
0
0
0
1
0
0
0
0.5
23
23
0
0.1
76
76
33
0.01
94
94
78
Control
98
98
98
LC50 (% Effluent)
0.04
(95% confidence limits)
(0.03 - 0.05)
Synthetic Rubber Plant #1
Undiluted process water
(final effluent)
Bluegill
Sunfish
100
50
10
5
1
0.1
0.01
Control
45(6)
65(3)
83(1)
91(1)
93
97
100
98
42
63
82
90
93
97
100
98
20
54
79
90
92
96
99
97
LC50 {% Effluent)
confidence limits)
43.0
(34.1 - 55.
-------
Table 7 - continued.
Test
Species
Effluent
Percent .
Hatchability
Percent Survival
Effluent Source
Concentration
(Percent)
Hatching
4 Days
Posthatching
Synthetic Rubber Plant #2
Cooling water and storm runoff
(final effluent)
Channel
Catfish
100
50
10
5
1
0.5
0.1
0.01
Control
34(14)
57
70
86
94
98
96
97
98
29
57
70
86
94
98
96
97
98
0
17
46
72
88
96
94
95
97
LC50 (35 Effluent)
(95% confidence limits)
9.4
(6.7 - 13.0)
Plastics Manufacturing Plant
Cooling water
(final effluent)
Channel
Catfish
100
50
10
5
1
0.5
0.1
0.01
Control
40(10)
68
86
98
99
98
100
98
99
36
68
86
98
99
98
100
98
99
19
53
76
92
97
96
98
98
99
LC50 (% Effluent)
(95% confidence limits)
39 2
(28.6 - 55.4)
*Egg hatchability was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearing in hatched populations were expressed parenthetically.
-------
Table 8. LC50 and LCj values for composite and grab samples of Industrial effluents analyzed in
laboratory embryo-larval tests.
Test Species (% ^ 95* Confidence
Chemical Manufacturing Plant if3
Undiluted process water Channel Catfish
(final effluent)
Synthetic Rubber Plant #2
Cooling water and storm runoff Channel Catfish
(final effluent)
Chemical Manufacturing Plant #1
Undiluted process water
Chemical Manufacturing Plant #1
Diluted process water
(final effluent)
Plastics Manufacturing Plant
Cooling water
(final effluent)
Synthetic Rubber Plant #1
Undiluted process water
(final effluent)
Bluegill Sunfish
Blueglll Sunfish
Channel Catfish
Chemical Manufacturing Plant #2
Cooling water and storm runoff Channel Catfish
(final effluent)
0.04
9.4
16.0
29.3
39.2
Bluegill Sunfish 43.0
0.03 - 0.05
6.7 - 13.0
13.1 - 19.2
24.1 - 35.1
28.5 - 55.4
34.1 - 55.0
0.001 0.0006 - 0.003
^100
0.34
1.17
2.63
1.27
1.43
2.53
0.12 - 0.67
0.65 - 1.80
1.44 - 4.03
0.36 - 2.67
0.56 - 2.65
0.33 - 6.20
U>
o
-------
Table 9. General characteristics of composite and grab samples of industrial effluent analyzed in
12-hour static-renewal embryo-larval tests conducted in the laboratory.*
Effluent Source
Test Species
pH
Alkalinity
(mg/L as
CaC03)
Hardness
(mg/L as
CaC03)
Conductivity
(iimhos/cm)
Dissolved
Oxygen
(mg/L)
Dilution Water
Chemical Manufacturing Plant #1
Undiluted process water
Chemical Manufacturing Plant #1
Diluted process water
(final effluent)
Chemical Manufacturing Plant #2
Cooling water & storm runoff
(final effluent)
Chemical Manufacturing Plant #3
Undiluted process water
(final effluent)
Synthetic Rubber Plant #1
Undiluted process water
(final effluent)
Synthetic Rubber Plant #2
Cooling water & storm runoff
(final effluent)
Plastics Manufacturing Plant
Cooling water
(final effluent)
Bluegill Sunfish
Bluegill Sunfish
7.4 ± 0.2
8.1 ± 0.2
8.1 ± 0.2
66 ± 3
82 ± 7
183 ± 3
218 ± 12 234 ± 25 1440 ± 25
250 ± 30 211 ± 19
Channel Catfish 7.7 ± 0.04 178 ± 2
Channel Catfish 8.0 ± 0.5
109 ± 10
360 ± 12
196 ± 6
Channel Catfish 8.7 ± 0.04 197 ± 1
227 ± 41 1760 ± 10
Channel Catfish 7.7 ± 0.04
77 ± 1
144 ± 26
166 ± 4
7.6 ± 0.2
6.7 ± 0.2
7.0 ± 0.2
7.7 ± 0.2
to
287 ± 13 75000 ± 557 7.9 ± 0.2
Bluegill Sunfish 7.9 ± 0.2 114 ±4 251 ± 26 4420 ± 58 7.1 ± 0.2
7.5 ± 0.3
7.6 ± 0.1
Chemical characteristics expressed as mean ± standard error.
-------
32
RESULTS OF ON-SITE BIOMONITORING OF MUNICIPAL AND INDUSTRIAL EFFLUENTS
During the field investigations, on-site embryo-larval biomonitoring
experiments were performed at five different test locations, including
1) Lexington Sewage Treatment Plant, 2) Chemical Manufacturing Plant,
3) Synthetic Rubber Plant, 4) Tannery-Secondary Sewage Treatment Plant
Complex, and 5) Metal Plating Plant. Biomonitoring was conducted in a
mobile laboratory designed to accommodate flow-through, static-renewal,
and static test systems, as described in the Appendix. Personnel from the
EPA Region IV Laboratory were on site at the last four locations to perform
acute effluent toxicity tests. These joint investigations provided an
opportunity for direct comparisons of acute and embryo-larval biomonitoring
data. The on-site biomonitoring experiments were performed using animal
species and test procedures given above. Specific tests performed at the
different field sites are summarized in Table 1. Though our principal
responsibility was to evaluate final effluents designated under the NPDES
program, additional tests were performed on individual effluent components
(_i_.£., Metal Plating Plant) and effluents at different stages of treatment
(!•£•> Tannery-Secondary Sewage Treatment Plant Complex). Specifications
on effluent production and treatment were obtained through the EPA Region
IV Laboratory and were provided by plant officials.
The mobile laboratory was moved to each biomonitoring site 2 to 3 days
prior to the actual onset of testing to allow adequate time for the assembly
of exposure systems and stabilization of test conditions (,e.
-------
33
temperature). In analyzing dose-response data, log probit analysis (8) was
applied to combined frequencies of embryo-larval mortality and teratogenesis
to determine control-adjusted LC^q values. These values represented expo-
sure concentrations, expressed as percent effluent by volume, that produced
50% impairment of test populations.
During the period of August 1-12, 1980, the first on-site biomonitoring
experiment was conducted at a Lexington sewage treatment plant. The plant
received domestic sewage which was passed through a grit chamber to primary
settling tanks with sludge collectors. Secondary treatment was achieved in
an aerated activated sludge system, followed by clarification in secondary
settling tanks. Effluent then received tertiary treatment in a lagoon
system prior to chlorination and discharge. The mean wastewater flow at
the time of study was approximately 7.61 million gallons per day. Toxicity
tests on sewage effluents were conducted using larvae of the channel catfish.
Treatment was initiated 1 day after hatching and continued for 7 days.
Secondary sewage effluent (unchlorinated) was administered to test organisms
in both the flow-through and static-renewal systems. In addition, primary
sewage effluent was tested using static-renewal procedures. Carbon-fi1tered
tap water (11), transported from the University of Kentucky laboratory to
the biomonitoring site, was used for effluent dilutions and maintenance of
control animals.
Solutions in all test chambers were monitored daily for general test
parameters, using procedures described above. Temperature was regulated
at 22.0 ± 0.5°C, and dissolved oxygen, as observed in other tests, was
-------
34
near saturation. Other water quality characteristics are summarized in
Table 19, and dose-response data are given in Table 10. The secondary
effluent was not highly toxic in either flow-through or static-renewal
tests, producing only 21% to 40% mortality at the 100% concentration.
However, toxicity was substantially greater for sewage effluent which
received only primary treatment. For example, ful1-strength and half-
strength primary effluent concentrations produced 100% and 53% mortality,
respectively, and the LC50 was 50.2%. Based on these data and earlier
observations (3), it appears that biomonitoring with fish developmental
stages affords a useful means of evaluating the effectiveness of waste
treatment. Results of this investigation also were used to revise and
perfect field monitoring procedures.
The second on-site biomonitoring experiment was performed at a major
chemical manufacturing plant in north-central Kentucky during the period
of September 6-17, 1980. The plant produced formaldehyde, urea-formaldehyde
resins, and phenol-formaldehyde resins. Wastewater was routed through two
equalization ponds, two anaerobic ponds, and a clarifier. Prior to dis-
charge, the effluent reportedly was diluted 50:1 with well water. The
mean wastewater flow at the time of study was approximately 2.8 million
gallons per day.
Biomonitoring experiments performed on site included flow-through
tests with the fathead minnow and Xenopus, static-renewal tests with the
fathead minnow, Xenopus, and the rainbow trout, and static tests with the
fathead minnow and Xenopus (Table 1). The effluent was pumped continuously
-------
35
from the NPDES sampling station to the mobile laboratory. Dilution water
for the embryo-larval tests was collected periodically from a deep well
located on the premises. Solutions in all flow-through and static-renewal
test chambers were monitored daily for alkalinity, conductivity, dissolved
oxygen, hardness, pH, and temperature. Measurements for the static system
were recorded daily, except for alkalinity and hardness which were monitored
at the beginning and end of the exposure period. Temperature was regulated
at 20.5 ± 0.5°C in tests with the fathead minnow and Xenopus and at 12.5 ±
0.5°C with the trout. Other test conditions are summarized in Table 19.
Dose-response data and LC^q values for these tests are summarized in
Table 11. Undiluted effluent proved to be highly toxic to embryo-1arval
stages. For example, in flow-through tests with the fathead minnow, exposure
to 100% effluent reduced survival to 4% at 3 days posthatching, compared to
93% survival in control populations. Full-strength effluent was somewhat
less toxic to Xenopus stages, as survival was 35% when determined in the
flow system. In static-renewal tests with the trout, early embryos were
highly sensitive to the effluent, as survival frequencies after 9 days of
exposure were 64%, 44%, 16%, and 0% at treatment levels of 1%, 10%, 50%,
and 100%, respectively. When exposure was initiated at the eyed-egg stage,
the trout was substantially more tolerant.
As determined in flow-through tests, effluent LC^q values were 29.4%
with the fathead minnow and 63.8% with Xenopus (Table 11). In static-renewal
tests with these species, LC50's were 48.9% and xl00%, respectively. The
lowest LC5q value was 6.6%, calculated in static-renewal tests with early
-------
36
trout embryos. Due to high survival frequencies at most effluent exposure
concentrations, LCgg's could not be determined in other tests.
The third on-site bionionitoring experiment was conducted at a major
synthetic rubber plant in north-central Kentucky during the period of
September 19 through October 2, 1980. The plant produced three polymers,
including polybutadiene rubber, styrene-butadiene rubber, and polybutadiene-
acrylic acid-acrylonitrile rubber. The first two polymers were used primarily
in the tire industry and the last was a proprietary product. Wastewater from
the plant was from four sources, including sanitary effluents, surface runoff,
lime slurry from the well water softening process, and process water. The
latter contained latex solids, rubber particles, unreacted monomers, sodium
chloride, and toluene. In the treatment procedure, the pH of process waste-
water was adjusted to 7.5, and the lime slurry and an anionic polyelectrolyte
polymer were added just before primary clarification. The clarified wastewater
was then mixed with the sanitary discharge, and the resulting effluent entered
an extended aeration pond. After a 30-hr retention time, the final wastewater
was discharged from the plant, and the mean flow was approximately 1.13 million
gallons per day.
Biomonitoring experiments performed on site included flow-through tests
with the fathead minnow and Xenopus, static-renewal tests with the fathead
minnow, Xenopus, and the rainbow trout, and static tests with Xenopus (Table
1). The effluent was continuously pumped from the NPDES sampling station
to the mobile laboratory. Dilution water for the embryo-larval tests was
collected periodically from a deep well located on the premises. General
-------
37
test parameters were monitored as described above and results are summarized
in Table 19. Temperature was maintained at 20.5 ± 0.5°C in tests with the
fathead minnow and Xenopus and at 12.5 ± 0.5°C with the trout.
Test responses are summarized in Table 12. Undiluted effluent produced
substantial frequencies of mortality. In flow-through tests, exposure to
100% effluent through 3-4 days posthatching reduced survival to 13% and 21%
for fathead minnows and Xenopus, respectively, compared to 90% to 93% survival
for control organisms. As administered in static-renewal tests, undiluted
effluent reduced survival of early trout embryos to 0% after 9 days. The
LC50 values determined in flow-through tests with fathead minnows and
Xenopus were 8.3% and 22.9%, respectively (Table 12). Using static-renewal
procedures with these species, the effluent LC^g's were 15.6% and -x.100%.
By comparison, the LC^q value was 6.4% when early trout embryos were exposed
in the static-renewal system. The survival of trout eyed eggs was considerably
higher, with an LC5q of 48.4%.
The fourth on-site biomonitoring experiment was performed at a
secondary sewage treatment plant in southeastern Kentucky during the period
of November 12-24, 1980. The plant received approximately 25% of its waste-
load from an adjacent tannery, and the designated NPDES effluent was the
chlorinated outflow from secondary treatment. Domestic sewage passed
through a grit chamber into a primary clarifier, at which point settleable
solids were removed. Tannery waste entered a second primary clarifier and
was mixed with the clarified domestic wastewater. Subsequently, the com-
bined wastewater was subjected to treatment in an aerated activated sludge
-------
38
system, followed by secondary clarification and chlorination. Approximately
1.1 million gallons of outfall per day entered the receiving water.
Biomonitoring experiments on the final chlorinated effluent included
flow-through tests with the fathead minnow and Xenopus, static-renewal tests
with the fathead minnow, Xenopus, and the rainbow trout, and a static test
with Xenopus. Final chlorinated effluent was continuously pumped from the
NPDES sampling station to the mobile laboratory. Dilution water for all
tests was taken 5 miles upstream from the treatment plant discharge.
General test parameters were monitored as described above, and results
are summarized in Table 19. In tests with the fathead minnow and Xenopus,
temperature was maintained at 22.0 ± 1.0°C. For the trout experiments,
temperature was regulated at 11.5 ± 0.5°C.
Results of the toxicity tests are summarized in Table 13. Except for
6% survival observed in the static test with Xenopus, 50% effluent always
produced complete mortality. The fathead minnow and Xenopus were about
equally sensitive to the final effluent, as reflected by LC5Q values of
0.3% and 0.4% derived in flow-through tests. The LCgg values were 0.5%,
0.7%, and 0.9% when determined in static-renewal tests with early trout
embryos, fathead minnow embryo-larval stages, and Xenopus embryo-larval
stages, respectively.
The Tannery-Secondary Sewage Treatment Plant Complex afforded an
unusual opportunity to monitor simultaneously a highly toxic untreated
effluent, as well as the final treated NPDES effluent. Raw tannery wastes
and unchlorinated and chlorinated effluents from the secondary sewage
-------
39
plant were tested using static-renewal procedures to determine the
applicability of on-site biomonitoring for evaluating the effectiveness
of waste treatment. These tests were repeated with three animal species,
and the results are presented in Table 14 and summarized in Table 15. As
the tannery wastes were reported to contain substantial concentrations of
metallic toxicants, the different effluent sources and the upstream
dilution water were analyzed for cadmium, chromium, copper, iron, and
zinc, and these data are reported in Table 15, together with general
water quality parameters.
Taking data for all three animal species, LCgg's ranged from 0.08% to
0.4% for untreated tannery wastes, 0.9% to 1.6% for unchlorinated secondary
effluent, and 0.5% to 0.9% for the final chlorinated secondary effluent
(Tables 14, 15). Chlorination resulted in a moderate increase in toxicity.
However, it was apparent from these data that the overloaded sewage treatment
plant provided little improvement in effluent quality. As a matter of fact,
the slight reduction in toxicity, as well as the decreases observed for
certain metals and general physicochemical parameters (£.£., conductivity,
hardness, alkalinity) may have resulted as much from dilution by domestic
waste input as from the treatment process. As noted above, the tannery
effluent comprised approximately one-fourth of the treatment plant waste-
load. Despite reductions in the concentrations of certain metals (.e.jj,.,
Cd, Cr, Cu), significant metal residues remained in the final effluent,
and this undoubtedly contributed to the high toxicity. The impacted
receiving water was a third-order stream of moderate size.
-------
40
The last on-site biomonitoring experiment was performed at a metal
plating plant in northeastern Kentucky during the period of April 7-18,
1981. The final NPDES effluent from this plant was comprised of five
separate components, including wastewater from the brazing-quencher
operation, sludge-bed filtrate, cooling water, chemically-treated waste-
water from the plating process, and surface runoff. Flows from these
discharges combined to form a tributary on which the NPDES sampling site
was situated. At the time of testing, wastewater flow in the tributary
was estimated to be approximately 155,000 gallons per day. This was
somewhat above the average flow of about 108,000 gallons per day, due
to local precipitation and increased surface runoff. Of this total
tributary flow, approximately 43% was cooling water, 24% was plating
process water, 3% was sludge-bed filtrate, less than 1% was brazing
water, and about 30% was surface runoff.
On-site biomonitoring experiments conducted on the final effluent
included flow-through, static-renewal, and static tests with the fathead
minnow and Xenopus. The final effluent was continuously pumped from the
NPDES sampling station to the mobile laboratory. Dilution water for all
tests was taken one-half mile upstream from the outfall, and the receiving
water was a second-order stream which entered a major river approximately
one-half mile below the test site. All tests were monitored as described
above. Temperature averaged 22.0 ± 0.5°C, and other physicochemical
parameters of the test water are summarized in Table 19.
Results of toxicity tests conducted on the final effluent are given
-------
41
in Table 16. As determined in the flow-through system, undiluted effluent
reduced survival frequencies to 6% and 46% for the fathead minnow and
Xenopus, respectively, compared to control survival of 77% to 80%. Effluent
LCgo values were 21.6% for the fathead minnow and -vl00% for Xenopus.
Toxicity was significantly less when measured in static-renewal tests.
For example, the effluent LC5q was 44.7% when the fathead minnow was
used. In static-renewal and static tests with Xenopus, LC^g's could not
be calculated.
As noted above, the final effluent was a mixture of four separate
discharges, supplemented by surface runoff. Additional tests were
performed in order to evaluate the comparative toxicity of the individual
effluent components (Table 17). In Table 18, these data are summarized
together with analyses of selected metals (i.e., Cd, Cr, Cu, Fe, Zn) and
general physicochemical characteristics. Components 1 and 2 were the most
toxic, with LC^q values of 0.01% and 0.05%, respectively, and they comprised
about 27% of the final effluent. The LC^q's for Components 3 and 4 were
23.6% and 25.4% and these collectively comprised about 44% of the final
effluent, with the remainder coming from surface runoff. The latter was
above normal due to local rainfall at the time of testing, and this likely
accounted for the lower toxicity of the final effluent, for which the LC^g
was 44.7%. With some exceptions, overall metal concentrations and conduc-
tivity were higher in the more toxic components. However, judging from
data summarized in Table 18, it would be difficult to estimate the toxicity
of the effluent components accurately on the basis of selected chemical
criteria, whereas on-site biomonitoring provided a direct means of measuring
toxicity.
-------
42
Table 10. On-site toxicity tests on primary and secondary sewage treatment ^ ^
plant effluents using newly hatched larvae of the channel catfish. '
Effluent Source
Test System
Effluent
Concentration
(Percent)
Percent Survival
after 7 Days
After Primary
Treatment
Static-renewal
100
50
10
1
0.1
Control
0
47
100
100
100
100
LC50 (% Effluent)
(95% confidence limits)
50.2
(46.5 - 53.5)
After Secondary
Treatment, Unchlorinated
Static-renewal
100
50
10
1
0.1
Control
60
100
99
100
100
99
LC50 (% Effluent)
ND3
After Secondary
Treatment, Unchlorinated
Flow-through
100
50
10
1
0.1
Control
79
97
100
100
96
97
LC50 {% Effluent)
ND
Initial on-site biomonitoring test conducted at Lexington, Kentucky.
2
Tests were initiated with 1-day-old larvae and continued through 8 days
posthatching.
3
Not determined.
-------
Table 11. On-site embryo-larval biomonitoring of a chemical manufacturing plant final effluent.
Test Species
Test System
Effluent
Concentration
(Percent)
Percent Hatchability* 2
or Embryonic Survival '
Percent
Hatching
Survival
4 Days
Posthatching
Xenopus laevis
Flow-through
100
63(8)
58
35
50
71(9)
66
53
10
88(4)
85
78
1
92(4)
88
83
0.1
98(1)
97
91
Control
98
98
96
LC50 {% Effluent)
63.8
(95% confidence 1
imits)
(39.8 - 100)
Xenopus laevis
Static-renewal
100
79(11)
70
46
50
86(6)
82
73
10
94(4)
90
81
1
97(1)
96
87
0.1
95
95
93
Control
97
97
97
LC50 (% Effluent)
^100
Xenopus laevis
Static
100
83(12)
73
71
50
87(7)
81
79
10
92(4)
88
86
1
96(2)
94
92
0.1
98
98
98
Control
96(2)
94
94
LC50 (% Effluent)
ND3
-------
Table 11 - continued.
Test Species
Test System
Effluent
Concentration
(Percent)
Percent Hatchability j 2
or Embryonic Survival
Percent Survival
Hatching PosJhatching
Fathead Minnow
Flow-through
100
18(41)
10
4
50
51 13)
45
38
10
76(9)
70
69
1
82(4)
79
79
0.1
90
90
90
Control
93
93
93
LC50 (% Effluent)
29.4
(95% confidence limits)
(22.1 - 36.4)
Fathead Minnow
Static-renewal
100
26(32)
18
5
50
60(7)
56
54
10
82(1)
81
79
1
88
88
86
0.1
93
93
93
Control
96
96
96
LC50 U Effluent)
48.9
(95% confidence 1
imits)
(38.3 - 60.2)
Fathead Minnow
Static
100
72(9)
66
66
50
77(5)
73
73
10
80
80
80
1
92
92
92
0.1
97(3)
94
94
Control
89
89
89
LC50 {% Effluent)
ND 3
-------
Table 11 - continued.
Test Species
Test System
Effluent
Concentration
(Percent)
Percent Hatchability. ~
or Embryonic Survival '
Percent
Hatching
Survival
4 Days
Posthatching
Rainbow Trout
Static-renewal
100
88
88
81
(eyed)
50
88(2)
87
84
10
91(3)
88
88
1
92(3)
89
88
0.1
99(1)
97
96
Control
99(1)
98
96
LC50 U Effluent) ND3
Rainbow Trout Static-renewal 100
50
10
1
0.1
Control
LC50 {% Effluent) 6.6
(95% confidence limits) (1.9 - 13.5)
*Egg hatchability was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearinq in hatched populations were expressed parenthetically.
2
Trout tests were conducted using fresh eggs, unless specified otherwise.
3
Not determined.
0
16
44
64
80
86
4*
cn
-------
Table 12. On-site embryo-larval biomonitoring of a synthetic rubber plant final effluent.
Test Species Test System
Effluent
Concentration
(Percent)
Percent Hatchability j ^
or Embryonic Survival '
Percent
Hatching
Survival
3 Days
Posthatching
xenopus laevis Flow-through
100
77(24)
58
21
50
85(20)
67
45
10
90(15)
76
57
1
93(9)
85
78
0.1
99(6)
93
89
Control
100(3)
97
93
LC50 (% Effluent)
(95% confidence limits)
22.9
(13.4 - 36.9)
xenopus laevls Static-renewal 100 85(31) 58 48
50 90(23) 70 67
10 95(17) 78 78
1 96(7) 89 85
0.1 97(5) 93 91
Control 100(3) 97 93
LC50 {% Effluent) 'v-lOO
xenopus laevis Static^ 100 88(24) 67 9
50 87(22) 67 58
10 94(12) 83 77
1 96(4) 93 89
0.1 98(2) 96 95
Control 98(2) 97 97
LC50 (% Effluent)
(95% confidence limits)
39 3
(27.9 - 50.6)
-------
Table 12 - continued.
Test Species Test System
Effluent
Concentration
(Percent)
Percent Hatchability 9
or Embryonic Survival '
Percent
Hatching
Survival
3-4 Days
Posthatching^
Fathead Minnow Flow-through
100
50
10
1
0.1
Control
35(22)
52(12)
65(5)
77(4)
85(2)
94(1)
27
46
61
74
83
92
13
20
50
67
79
90
LC50 (% Effluent)
(95% confidence limits)
(4.68-313.2)
Fathead Minnow Static-renewal
100
50
10
1
0.1
Control
49(15)
57(15)
70(4)
77(1)
85
94
42
48
67
76
85
94
12
34
59
70
83
91
LC50 (% Effluent)
(95% confidence limits)
15.6
(8.5 - 24.7)
Rainbow Trout Static-renewal
(eyed)
100
50
10
1
0.1
Control
85(11)
92(4)
96(3)
99(3)
99
99
75
88
93
96
99
99
37
53
75
89
93
97
LC50 (% Effluent)
(95% confidence limits)
48.4
(25.5 - 100)
-------
Table 12 - continued.
Test Species
Test System
Effluent
Concentration
(Percent)
Percent Hatchability , «
or Embryonic Survival '
Percent Survival
Hatching posJha?ching
Rainbow Trout
Static-Renewal
100
0
• —
50
14
-
10
56
-
1
62
-
0.1
85
-
Control
93
-
LC50 {% Effluent) 6.4
(95% confidence limits) (3.3 - 10.3)
*Egg hatchability was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearing in hatched populations were expressed parenthetically.
2
Trout tests were conducted using fresh eggs, unless otherwise specified.
3
Fungal contamination was detected at the 100% effluent exposure concentration and probably
contributed to mortality.
4
Tests with fathead minnovsand eyed trout stages were continued through 3 and 4 days posthatching,
respectively.
-------
Table 13. On-site embryo-larval biomonitoring of the chlorinated final effluent from a secondary sewage
treatment plant receiving tannery waste.
Test Species Test System
Effluent
Concentration
(Percent)
Percent Hatchability ^ 2
or Embryonic Survival *
Percent
Hatching
Survival
4 Days
Posthatching
xenopus laevis Flow-through
100
50
10
1
0.1
Control
0
0
24(12)
35(2)
80
95
0
0
21
34
80
95
0
0
9
24
78
94
LC50 (% Effluent)
(95% confidence limits)
0.4
(0.3 - 0.6)
Xenopus laevis Static-renewal
100
50
10
1
0.1
Control
0
2(100)
23(20)
48(17)
85
91
0
0
18
40
85
91
0
0
15
39
80
91
LC50 (% Effluent)
(95% confidence limits)
0.9
(0.5 - 1.5)
tfenopus laevis Static
100
50
10
1
0.1
Control
0
35(83)
39(12)
49
81
96
0
6
35
49
81
96
0
6
27
46
81
96
LC50 (X Effluent)
(95% confidence limits)
1.2
(0.7 - 2.0)
-------
Table 13 - continued.
Test Species Test System I
Effluent
Concentration
(Percent)
Percent Hatchability . ?
or Embryonic Survival '
Percent Survival
Hatching posthauhing
Fathead Minnow Flow-through
100
—
-
_
50
0
0
0
10
30(39)
18
7
1
62(11)
55
33
0.1
77(6)
72
61
0.01
83
83
81
Control
91
91
91
LC50 (% Effluent)
0.3
(95% confidence limits)
(0.2 - 0.6)
Fathead Minnow Static-renewal
100
0
0
0
50
24(100)
0
0
10
28(29)
20
16
1
52
52
44
0.1
80
80
76
Control
96
96
96
LC50 (* Effluent)
(95% confidence limits)
0.7
(0.3 - 1.5)
Rainbow Trout Static-renewal 100
50
10
1
0.1
Control
0
0
4
47
67
97
LC50 (% Effluent)
(95% confidence limits)
0.5
(0.3 - 0.8)
-------
Table 13 - continued.
*Egg hatchabllity was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearing in hatched populations were expressed parenthetically.
2
Trout tests were conducted using fresh eggs.
-------
Table 14. On-site study at a tannery-secondary sewage treatment plant complex Involving embryo-larval
biomonitoring of raw tannery waste, unchlorinated effluent from secondary treatment, and
chlorinated final effluent from secondary treatment.1*2
Effluent
Percent Hatchability ^ a
or Embryonic Survival *
Percent
Survival
Effluent Source
Test Species
Concentration
(Percent)
Hatching
4 Days
Posthatching
Tannery
Xenopus laevis
100
50
10
1
0.1
Control
0
0
0
42
73
90
0
0
0
42
73
90
0
0
0
34
63
90
LC50 (% Effluent)
(95% confidence limits)
0.4
(0.2 - 0.7)
Tannery
Fathead Minnow
100
50
10
1
0.1
Control
0
0
8(100)
44(18)
92
92
0
0
0
36
92
92
0
0
0
20
68
92
LC50 {% Effluent)
(95% confidence limits)
0.3
(0.1 - 0.5)
Tannery
Rainbow Trout
100
50
10
1
0.1
Control
0
0
0
7
43
97
-
-
LC50 (% Effluent)
(95% confidence limits)
0.08
(0.03 - 0.13)
-------
Table 14 - continued.
Effluent
Percent Hatchability 3 *
or Embryonic Survival '
Percent
Survival
Effluent Source
Test Species
Concentration
(Percent)
Hatching
4 Days
Posthatching
After Secondary
Treatment, Unchlorinated
Xenopus laevis
100
50
10
1
0.1
Control
0
20(38)
34(13)
58(13)
80
90
0
13
29
51
80
90
0
0
21
51
78
90
LC50 (X Effluent)
(95% confidence limits)
1.5
(0.7 - 2.6)
After Secondary
Treatment, Unchlorinated
Fathead Minnow
100
50
10
1
0.1
Control
0
24(100)
56(14)
72
92
92
0
0
48
72
92
92
0
0
28
52
76
92
LC50 (% Effluent)
(95% confidence limits)
1.6
(0.4 - 3.5)
After Secondary
Treatment, Unchlorinated
Rainbow Trout
100
50
10
1
0.1
Control
0
0
24
50
73
97
-
-
LC50 (% Effluent)
(95% confidence limits)
0.9
(0.5 - 1.4)
-------
Table 14 - continued.
Effluent
Percent HatchabiHty 3 4
or Embryonic Survival '
Percent
Survival
Effluent Source
Test Species
Concentration
(Percent)
Hatching
4 Days
Posthatchlng
After Secondary
Treatment, Chlorinated
(final effluent)
Xenopus laevis
100
50
10
1
0.1
Control
0
2(100)
23(20)
48(17)
85
91
0
0
18
40
85
91
0
0
15
39
80
91
LC50 (% Effluent)
(95% confidence limits)
0.9
(0.5 - 1.5)
After Secondary
Treatment, Chlorinated
(final effluent)
Fathead Minnow
100
50
10
1
0.1
Control
0
24(100)
28(29)
52
80
96
0
0
20
52
80
96
0
0
16
44
76
96
LC50 {% Effluent)
(95% confidence limits)
0.7
(0.3 - 1.5)
After Secondary
Treatment, Chlorinated
(final effluent)
Rainbow Trout
100
50
10
1
0.1
Control
0
0
4
47
67
97
-
-
LC50 (% Effluent)
(95% confidence limits)
0.5
(0.3 - 0.8)
-------
Table 14 - continued.
Untreated tannery effluent constituted approximately 25% of the wasteload of the
sewage treatment plant.
2
Tests were performed using static-renewal procedures.
Egg hatchabllity was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratlc survivors appearing in hatched populations were expressed parenthetically.
4
Trout tests were conducted using fresh eggs.
Ol
OI
-------
Table 15. Chemical and toxicological characteristics of three effluents from a tannery-secondary
sewage treatment plant complex.
Effluent Source
Effluent Characteristics
1,2
Tannery ^
STP
(unchlorlnated)
Final STP
(chlorinated)
Dilution
Water
PH
8.5 ± 0.1
7.8 ± 0.1
7.5 ± 0.1
6.9 ± 0.1
Alkalinity (mg/L as CaCO^)
473 ± 31
207 ± 6
203 ± 4
13 ± 1
Hardness (mg/L as CaCO^)
1317 ± 58
406 ± 23
385 ± 19
16 ± 1
Conductivity (ymhos/cm)
5829 ± 99
1744 ± 97
1746 ± 111
18 ± 1
Cadmium (yg/L)
21 ± 9
-
5 ± 1
n.d.^
Chromium (yg/L)
856 ± 134
-
118 ± 3
n.d.
Copper (yg/L)
29 ± 2
-
16 ± 1
6 ± 5
Iron (yg/L)
66 ± 7
-
80 ± 3
n.d.
Zinc (yg/L)
53 ± 8
-
122 ± 16
2 ± 2
LCjjq (Rainbow trout)
0.08%
0.9%
0.5%
-
LCjjq (Fathead minnow)
0.3%
1.6%
0.7%
-
LC,jq (xenopus laevis)
0.4%
1.5%
0.9%
-
cn
cn
i
Chemical characteristics expressed as mean ± standard error.
"LC50 values were determined during on-site biomonitoring, using embryo-larval static-renewal
procedures.
3
Untreated tannery effluent constituted approximately 25% of the wasteload of the secondary
sewage treatment plant.
*Not detected.
-------
Table 16. On-site embryo-larval biomonitoring of a metal plating plant final effluent.
Effluent
Percent
Survival
1
Test Species Test System
Concentration
Percent Hatchability
(Percent)
Hatching
4 Days
Posthatching
Xenopus laevis Flow-through^
100
62(2)
61
46
50
82(9)
75
68
10
86(3)
83
76
1
98
98
89
0.1
91(2)
89
72
Control
93
93
77
LC50 (% Effluent)
M00
Xenopus laevis Static-renewal
100
74(H)
64
64
50
85(5)
78
78
10
86(6)
80
70
1
87(1)
86
86
0.1
93
93
92
Control
98(1)
97
94
LC50 (* Effluent)
ND^
Xenopus laevis Static
100
84(14)
73
70
50
87
87
80
10
88
88
81
1
93
93
88
0.1
94
94
89
Control
98
98
89
LC50 (% Effluent)
-------
Table 16 - continued.
Test Species Test System
Effluent
Concentration
(Percent)
Percent Hatchabillty*
Percent
Hatching
Survival
4 Days
Posthatching
Fathead Minnow Flow-through
100
50
10
1
0.1
Control
10(2)
37(17)
60(3)
73(1)
82
80
8
31
58
72
82
80
6
31
53
72
82
80
LC50 (% Effluent)
(95% confidence limits)
21.6
(13.9 - 29.5)
Fathead Minnow Static-renewal
100
50
10
1
0.1
Control
21
53
73
81
87
87
21
53
73
81
87
87
13
43
72
79
83
86
LC50 {% Effluent)
(95% confidence limits)
44.7
(32.5 - 54.6)
Fathead Minnow Static^
100
50
10
1
0.1
Control
23(33)
37(11)
40(9)
15
33
36
15
33
36
87
75(3)
87
73
85
71
LC50 (% Effluent)
(95% confidence limits)
10.6
(0.00 - 29.3)
-------
Table 16 - continued.
*Egg hatchabllity was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearinq in hatched populations were expressed parenthetically.
2
Fungal contamination was detected in two flow-through chambers used for xsnopus (i.e., 0.1X,
control) and in most static chambers used for the fathead minnow, and this probably contributed
to mortality.
3
Not determined.
cn
vo
-------
Table 17. On-site embryo-larval biomonitorlng of final effluent and effluent components
from a metal plating plant using embryo-larval stages of the fathead minnow.1
Final Effluent and
Effluent Components
Effluent
Concentration
(Percent)
Percent
Hatchability^
Percent Survival
Hatching Pos^1ng
Component 1
10
1(100)
0
0
(Sludge-bed filtrate)
5
17(16)
14
14
1
34
34
34
0.1
-
-
0.01
50
50
50
Control
87
87
86
LC50 (% Effluent)
0.01
(95% confidence limits)
(0.0005 - 0.04)
Component 2
10
3
3
0
(Process Water)
5
11
11
5
1
16(23)
13
11
0.1
34(7)
32
32
0.01
67
67
63
Control
87
87
86
LC50 (* Effluent)
0.05
(95% confidence limits)
(0.02 - 0.10)
Component 3
100
51(14)
43
4
(Brazing Water)
50
44(12)
39
37
10
57
57
57
1
84
84
80
0.1
85
85
83
Control
86
86
85
LC50 (% Effluent)
(95% confidence limits)
23.6
(14.1 - 33.2)
-------
Table 17 - continued.
Final Effluent and
Effluent Components
Effluent
Concentration
(Percent)
Percent
Hatchability2
Percent Survival
Hatching PosJha?ching
Component 4
100
47
47
31
(Cooling Water)
50
30
30
27
10
62
62
58
1
73
73
73
0.1
76
76
73
Control
87
87
86
LC50 (* Effluent)
25.4
(95? confidence limits)
(9.0 - 63.0)
Final Effluent
100
21
21
13
50
53
53
43
10
73
73
72
1
81
81
79
0.1
87
87
83
Control
87
87
86
LC50 (% Effluent)
(95% confidence limits)
44.7
(32.5 - 54.6)
Tests were conducted using static-renewal procedures.
2
Egg hatchability was based on all animals, normal and aberrant, which completed hatching.
Frequencies of teratic survivors appearing in hatched populations were expressed
parenthetically.
-------
Table 18. Chemical and toxicological characteristics of effluent components and final effluent from
a metal plating plant.
Effluent Characteristics*'^
Effluent Components3
Final
Dilution
1
2
3
4
Effluent
Water
4
Percent Final Effluent
3
24
<1
43
-
PH
9.2
+
0.1
9.2 ± 0.2
7.8 ± 0.1
7.2 ± 0.1
8.1 ± 0.1
8.1 ± 0.1
Alkalinity (mg/L as CaCO^)
1844
+
204
157 ± 9
86 ± 5
310 ± 26
273 ± 10
260 ± 7
Hardness (mg/L as CaCO^)
148
±
13
86 ± 6
163 ± 3
438 ± 7
281 ± 11
272 ± 5
Conductivity (pmhos/cm)
12209
+
1966
866 ± 37
189 ± 3
509 ± 58
716 ± 42
229 ± 5
Cadmium (ug/L)
31
+
5
9 ± 2
8 ± 1
58 ± 46
6 ± 1
n.d.^
Chromium (pg/L)
741
±
282
484 ± 205
6 ± 3
585 ± 48
296 ± 135
13
Copper (pg/L)
239
+
44
57 ± 12
14 ± 2
11 ± 1
23 ± 5
n.d.
Iron (ug/L)
377
+
66
425 ± 157
44 + 21
36 ± 5
292 ± 146
31
Zinc (pg/L)
2488
+
585
893 ± 223
1173 ± 139
74 ± 8
379 ± 115
n.d.
LCjjq (Fathead minnow)
0.
,01%
0.05%
23.6%
25.4%
44.7%
-
Chemical characteristics expressed as mean ± standard error.
2
LC50 values were determined during on-site biomonitoring, using embryo-larval static-renewal procedures.
3
The final NPDES effluent was comprised of four separate components, including sludge-bed filtrate (1),
process water (2), brazing process water (3), and cooling water (4). The final effluent was further
diluted with surface runoff at time of testing.
^Remaining component of final effluent was from surface runoff.
^Not detected.
-------
Table 19. General characteristics of final effluent, effluent components, and dilution water used in
embryo-larval biomonitoring experiments.*
Test Site
Test System
Test
Species
pH
Alkalinity Hardness Dissolved
(mg/L as (mg/L as 0x^en
CaC03) CaC03) (urns/cm) (mg/L)
Lexington Sewage Treatment Plant
Dilution water
Primary effluent
Unchlorinated secondary
effluent
Static-renewal
(on site)
Static-renewal
(on site)
Flow-through
(on site)
Chemical Manufacturing Plant
Dilution water
Diluted process water
(final effluent)
Flow-through
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Static
(on site)
Static-renewal
(laboratory)
Catfish
Catfish
Catfish
Xenopus &
F. minnow
Xenopus &
F. minnow
Trout
Afenopus &
F. minnow
Xenopus
7.1 ± 0.1
7.6 ± 0.1
7.6 ± 0.04
7.6 ± 0.03
7.6 ± 0.1
7.6 ± 0.1
7.5 ± 0.1
7.5 ± 0.2
7.9 ± 0.1
7.7 ± 0.3
61 ± 2 105 ± 8
124 ± 64 102 ± 10
173 ± 15 130 ± 8
181 ± 3
312 ± 2
133 ± 9
276 ± 14 327 ± 6
304 ±6 387 ± 6
252 ± 11 338 ± 11
288 ± 13 358 ± 14
352 ± 28
233 ± 34 368 ± 7
186 ± 1 8.3 ± 0.1
8.0 ± 0.4
308 ±9 8.0 ± 0.2
312 ±3 8.4 ± 0.2
308 ± 3 8.0 ± 0.1
402 ± 7 7.8 ± 0.1
359 ± 7
8.1 ± 0.1
337 ± 11 10.0 ± 0.2
419 +3 8.1 ± 0.4
373 + 32 7.9 ± 0.2
-------
Table 19 - continued.
Test Site
Test System
Test
Species
Synthetic Rubber Plant
Dilution water
Undiluted process water
(final effluent)
Flow-through
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Static
(on site)
Static-renewal
(laboratory)
Xenopus &
F. minnow
Xenopus &
F. minnow
Trout
xenopus
Trout
Tannery-Sewage Treatment Plant
Dilution water
Chlorinated secondary
effluent
(final effluent)
Flow-through
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Static
(on site)
Static-renewal
(laboratory)
Xenopus &
F. minnow
Xenopus &
F. minnow
Trout
Xenopus
Trout
Alkalinity Hardness Dissolved
(mg/L as («g/L as °Wn
CaC03) CaC03) (pfflhos/cm) (l«L)
pH
7.6 ± 0.1 409 ± 34 445 ± 19 476+6 8.0 ±0.1
7.7 ± 0.1 134 ± 3 296 ± 9 3962 ± 143 8.3 ±0.1
7.7 ± 0.1 140 ± 3 294 ± 5 4138 ± 193 7.9 ±0.1
7.7 ± 0.1 174 ± 17 298 ± 14 3204 ± 290 10.4 ±0.1
7.8 ± 0.2 134 ± 4 297 ± 31 4986 ± 60 8.1 + 0.1 en
7.2 ± 0.1 154 ± 4 298 ± 4 4323 ± 121 8.8 ±0.1
6.9 ±0.1 13 ± 1 16 ± 1 18 ± 1 8.0 ± 0.1
7.5 ± 0.1 203 ±4 385 ± 19 1746 ± 111 7.9 ± 0.1
8.1 ± 0.1 203 ± 4 317 ± 19 1490 ± 38 7.4 ±0.1
8.0 ± 0.1 211 ± 1 - 1317 ± 93 8.4 ± 0.1
8.0 ± 0.1 203 ±1 291 ± 9 1397 ±3 7.3 ± 0.1
8.2 ± 0.1 248 ± 9 281 ± 16 1210 ± 65 9.0 ±0.1
-------
Table 19 - continued.
Test Site
Test System
Test
Species
PH
Alkalinity
(mg/L as
CaC03)
Hardness
(mg/L as
CaC03)
Conductivity
(pmhos/cm)
Dissolved
Oxygen
(mg/L)
Unchlorinated secondary
effluent
Tannery effluent
Static-renewal
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Xenopus &
F. minnow
Trout
Xenopus &
F. minnow
Trout
7.8 ± 0.1
8.1 ± 0.1
8.5 ± 0.1
8.4 ± 0.2
207 ± 6
210 ± 1
503 ± 16
406 ± 23 1744 ± 97
5417 ± 60
7.5 ± 0.1
1333 ± 133 8.7 ± 0.1
473 ± 31 1317 ± 58 5829 ± 99 5.7 ±0.1
8.6 ± 0.1
Metal Plating Plant
Dilution water
Mixed waste
(final effluent)
Component 1
(sludge-bed filtrate)
Component 2
(process water)
Component 3
(brazing water)
Component 4
(cooling water)
Flow-through
(on site)
Static-renewal
(on site)
Static
(on site)
Static-renewal
(laboratory)
Static-renewal
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Static-renewal
(on site)
Xenopus &
F. minnow
Xenopus &
F. minnow
Xenopus &
F. minnow
Xenopus
8.1 ± 0.1
8.1 ± 0.1
8.1 ± 0.1
8.0 ± 0.3
8.0 ± 0.1
260 ±7 272 ± 5 229 ± 5 8.7 ± 0.03
263 ± 12 265 ± 14 731 ± 58 7.8 ± 0.1
273 ± 10 281 ± 11 716 ± 42 8.0 ± 0.1
228 ± 36 302 ± 54 553 ± 8
240 ±8 307 ± 4
F. minnow 9.2 ± 0.2
F. minnow 7.8 ±0.1
F. minnow 7.2 ± 0.1
157 ± 9
86 ± 6
86 ± 5 163 ± 3
310 ± 26 438 ± 7
189 ± 3
509 ± 58
8.3 ± 0.1
765 ± 112 8.5 ± 0.4
F. minnow 9.2 ± 0.1 1844 ± 204 148 ± 13 12209 ± 1966 7.6 ± 0.2
866 ± 37 8.0 ± 0.2
8.0 ± 0.2
7.7 ± 0.3
1
Chemical characteristics expressed as mean ± standard error.
-------
66
ANALYSIS OF RESULTS AND
EVALUATION OF EFFLUENT BIOMONITORING PROCEDURES
On the basis of observations presented above, it is apparent that
embryo-larval tests afford an effective means of detecting and quantifying
effluent toxicity. It is also evident that biomonitoring results may vary
significantly with the different test systems and animal species selected
for use. Therefore, it is important to standardize test procedures within
limits necessary to insure reliable results and yet afford reasonable
economy of operation.
Rel iabil it.y and sensitivity of alternative embryo-1 arval test systems.
Ten comparative evaluations were made in which flow-through, static-renewal,
and static procedures were all used to test selected effluents or the
reference toxicant d.,e., phenol). In each case, the three tests were
conducted using the same exposure period, animal species, dilution water
and, in so far as possible, general test parameters (.e.jj., temperature,
dissolved oxygen, water hardness). In most instances, flow-through tests
provided the greatest resolution of dose-response data and the lowest LC^q
values. When toxicity appeared greater in static tests, such results were
due to effluent sampling during periods of peak toxicity, as in the coal-ash
study, or to extrinsic factors (e.ji., fungal contamination). Though such
problems were encountered infrequently, static test conditions were more
susceptible to deterioration (Tables 12, 16).
In twelve instances, flow-through and static-renewal tests were
-------
67
conducted simultaneously. Based on 95% confidence intervals, animal
responses did not vary significantly for the two different methods in
eight of the twelve cases. The LC^q values usually were within a factor
of about two or less, and seldom differed by more than a factor of 2.5.
Considering these data and the greater economy of operation, static-renewal
procedures appear applicable for routine toxicological screening of
effluents. However, when precise quantification of effluent toxicity is
required, the flow-through test is recommended. Based on the above
considerations, static procedures appear less suitable for use in
effluent biomonitoring with embryo-larval stages. Compared to static-
renewal procedures, the small gain in economy does not offset the reduced
sensitivity and greater variability of results observed in static tests.
Furthermore, as the composition of many effluents varies significantly with
time, static tests provide a less representative measure of toxicity.
Test organisms and responses. Of those organisms used in on-site
effluent biomonitoring, early trout embryos (fertilization through 9 days)
and embryo-larval stages of the fathead minnow (fertilization through 3 to
4 days posthatching) consistently were the most sensitive, based on LC5q
values. Results obtained with these species gave the most reliable data
for detection and quantification of effluent toxicity. Eyed trout eggs
carried through 4 days posthatching and embryo-larval stages of Xenopus
were appreciably more tolerant. For example, using data obtained in on-
site static-renewal tests, the LC5q values for the Chemical Manufacturing
Plant effluent were 6.6%, 48.9%, and -vl00% when determined with early
-------
68
trout embryos, fathead minnow embryo-larval stages, and Xenopus embryo-
larval stages, respectively (Table 11). An LC^g could not be determined
with eyed trout eggs. Using the same procedures, LC5q values for the
Synthetic Rubber Plant effluent were 6.4%, 15.6%, 48.4%, and -vl00% when
determined with early trout embryos, fathead minnow embryo-larval stages,
eyed trout eggs, and Xenopus embryo-larval stages, respectively (Table 12).
In instances when static-renewal procedures are the method of choice for
testing industrial effluents, Xenopus stages and eyed trout eggs probably
should not be used as the principal test organisms. However, differences
in sensitivity between Xenopus and the fathead minnow decreased somewhat
when flow-through procedures were used. For example, taking on-site
effluent toxicity data for the Tannery-Sewage Treatment Plant Complex,
Synthetic Rubber Plant, Metal Plating Plant, and Chemical Manufacturing
Plant, the LC^q values were 0.3%, 8.3%, 21.6%, and 29.4% when determined
with embryo-larval stages of the fathead minnow and 0.4%, 22.9%, -*.100%,
and 63.8% when determined with embryo-larval stages of Xenopus (Tables
11, 12, 13, 16).
In the event that embryo-larval toxicity testing is applied to
effluent biomonitoring under the NPDES program, it will become necessary
to standardize procedures sufficiently to provide reasonable quality
assurance. However, some flexibility should be maintained regarding
the selection of animal species. An abundant supply of viable eggs is
essential for a reliable testing program, and this requirement can be
met more effectively and economically by drawing upon a reasonable
-------
69
complement of species.
In addition to animal species evaluated during field studies,
embryo-larval stages of the channel catfish and bluegill sunfish were
used in effluent toxicity tests conducted in the laboratory (Table 8),
and good results were obtained with both species. Therefore, on the
basis of field and laboratory observations reported above, organisms
considered suitable for either flow-through or static-renewal effluent
testing included embryo-larval stages of the bluegill sunfish, channel
catfish, and the fathead minnow, and fertilized eggs of the rainbow
trout. Based on previous results obtained in conventional toxicity testing
with embryo-larval stages (5, 7, 9, 11, 14), a number of additional warm-
water species should prove suitable for use, and a list of preferred and
optional species is included in the Appendix.
Despite the greater tolerance of its embryo-larval stages, Xenopus
may be useful as an optional test species. The reproductive biology and
embryology of this species have been well-documented, due to the fact
that it has served as one of the principal experimental organisms in the
field of developmental biology. Parental stocks are easily maintained,
spawning can be induced with hormonal injections, and a typical spawn
contains from 1,000 to 2,000 eggs. Eggs can be taken soon after fertili-
zation and are easily handled in test systems. When testing is required
at locations which lack fish culturing facilities, it may prove more
practical and economical to maintain a Xenopus colony.
Both embryo-larval mortality and teratogenesis proved to be important
-------
70
and easily discernible test responses. Though mortality was the predominant
response, most effluents evaluated at concentrations of 10% to 10056 in on-
site studies produced significant frequencies of teratic organisms (Tables
11-14, 16, 17), and for the more toxic effluents or effluent components,
teratogenesis often was significant at concentrations ranging down to
0.1% to 1% (Tables 13, 14, 17). Due to the limited duration of exposure,
no consideration was given to growth. Though some attention was given to
retarded development, delayed hatching, and minor anomalies, no additional
responses with clearly identifiable endpoints were observed which could be
applied consistently to effluent testing. As embryo-larval mortality and
gross teratogenesis are sensitive, irreversible responses, their occurrence
can be expected to impact upon reproductive success and population density
and, therefore, they constitute the most feasible criteria to use in short-
term tests with early life stages of fish and amphibians.
Comparison of on-site and laboratory testing of industrial effluents.
One aspect of this investigation was to explore the possibility of using
embryo-larval stages in laboratory tests to screen industrial effluents
for toxicity. Therefore, during the four major on-site studies, 24-hr
composite samples of the final NPDES effluents designated for testing
were transported to the laboratory for simultaneous analyses. Two to
three 20-liter aliquots of each effluent were collected using a standard
composite sampler (ISC0) equipped with an ice bath. Upon collection, each
Pyrex container was stoppered, packed in ice, and transported to the
laboratory for immediate use. Due to the heavy volume of testing conducted
-------
71
during the on-site studies, it was necessary to perform this correlated
field/laboratory investigation using static-renewal procedures and test
organisms for which there was the greatest availability. Consequently,
three of the four tests were conducted using either eyed trout eggs or
Xenopus embryo-larval stages, and this resulted in less sensitivity
than would have been desired. However, it is evident from the dose-
response data that effluent toxicity consistently was less when measured
in the laboratory (Table 20). Results given for the Tannery-Secondary
Sewage Treatment Plant Complex were particularly significant. These
tests were performed using fertilized trout eggs, and the effluent LC^q
values were 0.5% and 32.1% as determined in the field and laboratory,
respectively. This difference was between one and two orders of magnitude,
considering the narrow 95% confidence intervals. Though LC^q values could
not be determined for most of the remaining tests, a survey of dose-response
data indicated less deviation between on-site and laboratory determinations,
but disparities could have been greater had more sensitive test organisms
been used. Considering results given in Table 20, it is apparent that
effluent toxicity can be quantified more accurately in on-site tests.
However, based on these results and those obtained for seven industrial
effluents evaluated during the first year of the project (Table 8), tests
conducted in the laboratory using embryo-larval stages of sensitive
aquatic species could prove useful in preliminary effluent screening.
Comparison of acute and embryo-larval effluent biomonitorinq. In each
of the four major biomonitoring studies, the EPA Region IV team performed
-------
72
Table 20. Comparison of effluent toxicity determined in on-site and laboratory
static-renewal tests using fish and amphibian embryo-larval stages.
NPDES
Effluent Source^
Test Species
Effluent
Concentration
(Percent)
Percent
On Site
O
Survival
Laboratory
Chemical Manufacturing
Plant
Xenopus
laevis
100
50
10
1
0.1
Control
46
73
81
87
93
97
72
72
78
84
87
85
LC50 (% Effluent)
^100
ND3
Synthetic Rubber Plant
Rainbow
(eyed)
Trout
100
50
10
1
0.1
Control
37
53
75
89
93
97
93
85
90
94
96
95
LC50 (% Effluent)
(95% confidence limits)
48.4
(25.5 - 100)
ND
Metal Plating Plant
Xenopus
laevis
100
50
10
1
0.1
Control
64
78
70
86
92
94
83
80
86
90
83
88
LC50 {% Effluent)
ND
ND
Tannery-Sewage Treatment
Plant
Rainbow
Trout
100
50
10
1
0.1
Control
0
0
4
47
67
97
0
41
78
90
92
94
LC50 (% Effluent)
(95% confidence limits)
0.5
(0.3 - 0.8)
32.1
(25.5 - 38.4)
Final NPDES effluents specified for testing.
2
Percent survival was determined 4 days after hatching for xenopus and eyed stages
of the rainbow trout, and after 9 days of exposure when testing was initiated with
freshly fertilized trout eggs.
3
Not determined.
-------
73
96-hr acute tests on the designated NPDES effluents, using the fathead minnow
in a flow-through system. One objective of this joint investigation was to
evaluate effluent toxicity using both acute and embryo-larval test procedures.
Precisely the same effluent sources and dilution waters were used in these
comparisons and the results are summarized in Table 21. Embryo-larval tests
gave more reliable detection and better quantification of effluent toxicity.
In addition, it was possible to use dose-response data from embryo-larval
tests to calculate LC^ values. The LC-j was defined as the toxicity thresh-
old for embryo-larval mortality and/or teratogenesis. This provided an
additional reference point with which to evaluate effluent toxicity and,
as discussed below (p. 79), such values may be used to estimate effluent
concentrations which produce chronic effects on aquatic biota.
In view of data presented in Table 21, it was not possible in the EPA
acute tests to determine LC^q values in three of the four cases studied,
and this precluded an adequate quantitative comparison between fish acute
and embryo-larval test responses for the selected effluents. We were able
to make such a comparison only in the case of the most toxic effluent,
which originated from the Tannery-Sewage Treatment Plant Complex. In this
instance, the acute LC^q of 8.0% effluent was about 27 times the embryo-
larval LCjq (0.3%) and differed from the embryo-larval LC^ value (0.001%)
by more than three orders of magnitude. However, it is likely that the
quantitative difference between fish acute and embryo-larval test responses
will narrow with decreasing effluent toxicity. Considering results for
the Metal Plating Plant and most other effluents of equal or lesser toxicity
-------
74
summarized in Tables 8 and 21, it is estimated that fish acute LCgg values
probably would not differ by more than one and two orders of magnitude from
fish embryo-larval LCjq and LC-j values, respectively, but further study will
be required to establish such correlations.
It is obvious from results in Table 21 that fish acute tests cannot
be used consistently to quantify the effects of effluents of intermediate
or moderate toxicity. This is due to the fact that it is not possible to
determine LCgg values without extending exposure ranges by concentrating
toxic substances beyond levels present in 100% effluent. As it is not
practical or plausible to concentrate effluents for such purposes, it
appears that fish acute tests, which have been used extensively in assess-
ments of specific toxicants, will prove less useful in the characterization
of complex effluents.
In the development of freshwater criteria for specific elements or
compounds, principal reliance traditionally has been placed on MATC's
developed in chronic life-cycle studies {4, 7). Due to the great time and
cost of life-cycle tests, reliable MATC's are available for a relatively
small number of priority toxicants. In order to facilitate the use of
toxicity data in hazard assessment, application factors have been employed
to estimate MATC's or comparable values from acute LC^g's (16). However,
it does not appear that application factors will prove as useful in the
characterization of complex effluents. This assumption is based on several
considerations, including 1) the inability to determine fish acute LC5Q values
for many effluents and 2) the difficulty of establishing a reliable application
-------
75
factor for a complex toxicant mixture, the composition of which may
fluctuate significantly with time. Furthermore, it is evident that MATC's
can be estimated more accurately and about as economically by using short-
term embryo-larval tests, as discussed below (p. 79). The LC-j also can be
used to calculate the dilution factor required to preclude significant
mortality and teratogenesis of sensitive reproductive stages. Such
dilution factors, together with transport-fate data and other essential
information, should prove useful in assessing the impact of an effluent
on its receiving system.
-------
77
APPLICATIONS OF EFFLUENT BIOMONITORING WITH EMBRYO-LARVAL STAGES
Characterization of effluents. Using LC^q's determined in flow-through
and static-renewal tests, moderate to high toxicity was observed with all but
one of the 19 industrial and municipal effluents and effluent components
studied (Tables 8, 10, 14, 17, 22). Only the secondary treatment effluent
from the Lexington Sewage Treatment Plant failed to exert significant toxicity
(Table 10). Using the most sensitive test in each case, the LC^q's for the
18 toxic effluents ranged from 0.04% to 0.9% for six; 6.4% to 16.0% for four;
21.6% to 29.3% for four; and 39.2% to 100% for four (Tables 8, 10, 14, 17, 22).
Out of these, ten were final NPDES effluents which entered receiving waters
and the LC^q values were 0.3%, 6.4%, 6.6%, and 21.6% for the four major
effluents analyzed on site (Table 22) and 0.04%, 9.4%, 29.3%, 39.2%, 43.0%,
and %100% for the effluent samples tested in the laboratory (Table 8). It
is likely that LC^q's for the last six effluents would have been lower if
determined in on-site tests.
The LC-j values ranged from 0.001% to 2.63% for the six final effluents
analyzed in the laboratory (Table 8) and, taking data for the most sensitive
animal species (j_..e.i fathead minnow or trout), the LC-|'s varied from 0.001%
to 0.8% for the four major NPDES effluents analyzed on site (Table 22).
Considering data presented above, it is apparent that toxicity tests with
embryo-larval stages provided a sensitive and reliable means of evaluating
the toxicity of complex effluents.
-------
Table 22. LC^q and LCj values for final NPDES effluents tested on site.
NPDES Effluent Source
Test System
Test Species1
(% Efffuent)
95% Confidence
Limits
LCi
(« Effluent)
95% Confidence
L1mi ts
Chemical Manufacturing
Flow-through
Fathead minnow
29.4
22.1 - 36.4
2.8
1.1 - 5.1
Plant
Static-renewal
Rainbow trout
6.6
1.9 - 13.5
0.08
0.001 - 0.47
Synthetic Rubber Plant
Flow-through
Fathead minnow
8.3
4.6 - 13.2
0.02
0.002 - 0.08
Static-renewal
Rainbow trout
6.4
3.3 - 10.3
0.07
0.01 - 0.24
Tannery-Sewage Treatment
Flow-through
Fathead minnow
0.3
0.2 - 0.6
0.001
0.0002 - 0.006
Plant
Static-renewal
Rainbow trout
0.5
0.3 - 0.8
0.004
0.001 - 0.01
Metal Plating Plant
Flow-through
Fathead minnow
21.6
13.9 - 29.5
0.8
0.1 - 2.0
^Embryo-larval tests for the fathead minnow were maintained through 3-4 days posthatching; trout were exposed
from fertilization through 9 days of development.
-------
79
Use and legal defensibil ity of effluent toxicity data. On the basis of
the above results, embryo-larval testing appears highly suitable for use
under the NPDES program. Permit users could apply on-site testing 1) for
periodic monitoring of their effluent discharges, 2) to design and/or evaluate
effectiveness of waste treatment systems, and 3) to provide additional docu-
mentation in permit applications. Regulatory agencies at State and Federal
levels could employ embryo-larval testing, or data obtained by such means,
1n revising NPDES program requirements, in permit issuance, in granting
special dispensations, in determining compliance, and in the documentation
of violations considered likely to result in litigation. As the Clean Water
Act specifically addresses the discharge of toxic pollutants in toxic amounts,
direct toxicological measurements should have greater relevance in legal
actions than chemical criteria based on laboratory investigations.
Such effluent toxicity testing is authorized under CWA sections 308 and
402 (35-37). Even in cases where promulgated BAT or BCT effluent guidelines
exist, toxicity-based permit limits could be observed on a case-by-case basis
under CWA section 402(a)(1). Biological monitoring is defined under CWA
section 502 and, as reviewed by Weber (38), some form of biological monitoring
is stated or implied in at least 19 sections of the amended Federal Water
Pollution Control Act.
While embryo-larval effluent biomonitoring provides the means for
sensitive detection and reliable quantification of toxic discharges, such
testing also may prove useful in estimating levels of effluent toxicity which
produce chronic effects on aquatic biota. In recent studies by Birge, et al.
(5, 6), comparisons were made between LC-| values and maximum acceptable
-------
80
toxicant concentrations (MATC) determined for a substantial number of organic
and inorganic toxicants. The LC-j values were determined in embryo-larval
tests which extended from fertilization through 4 days posthatching, using
essentially the same procedures as given above for effluent testing. The
MATC's were estimated in 30- to 90-day embryo-larval tests or determined in
chronic life-cycle studies, as reported in the literature. The MATC is
generally defined as the highest toxicant concentration which has no adverse
effect on test organisms (7). However, this value cannot be fixed precisely
in most life-cycle studies and, therefore, it usually is taken to fall within
the range between the highest no-effect level and the lowest concentration
which produces statistically significant responses. Though comparisons for
particular toxicants were complicated somewhat by differences in test
procedures, water characteristics (.§..2.., hardness), and the use of different
animal species, the LC^ values generally fell within or near the MATC ranges
(Tables 23, 24; ref. 5, 6). Correlations were particularly close in cases
where toxicity determinations were made using the same animal species
(¦£•£•» Table 24; chromium, lead, silver). When different species were used
for the same toxicant, variations between LC-j's and MATC's were no greater
than variations among the different MATC's (,e.£., Table 24; mercury, zinc).
Comparisons of LC-|'s with the MATC for 2,4-D were particularly interesting,
indicating the rainbow trout to be more sensitive than the fathead minnow,
and the latter to be more sensitive than the largemouth bass. This order of
species sensitivity was not inconsistent with that frequently observed in
tests with other toxicants (7, 11).
-------
81
These findings further support the premise that probit LC-| values
determined in short-term embryo-larval tests usually approximate the levels
of sensitivity which can be statistically verified in chronic life-cycle
studies. Limitations involved in long-term investigations frequently
curtail use of sufficient replicate exposures and the precise control of
test variables required to provide statistical differentiation of low-level
test responses. On the other hand, short-term embryo-larval tests can be
conducted with greater precision and the dose-response relationship usually
can be characterized more adequately, thus permitting reliable probit
analysis of test responses. These factors tend to nullify the differences
in sensitivity between chronic life-cycle studies and short-term embryo-
larval tests. Probit analysis further permits a best-fit determination for
the entire dose-response, and discrete lethal concentrations (_e.£., LCgo»
LC-jq, LC-|) with confidence limits can be determined at any point on the
curve. This option should become more important if and when toxicity is
integrated more quantitatively with other factors (,e.£., transport-fate
phenomena, assimilative capacity, reproductive resiliancy) also germane to
the development of freshwater criteria and numerical effluent limitations.
Many MATC's, particularly if 95% confidence limits are applied to the upper
and lower values, reflect rather broad concentration ranges. Such data
will prove less useful if the hazard assessment process is further modified
to integrate toxicity with important environmental and biological factors
which also affect the magnitude of impact on aquatic biota. As chronic
studies are not generally applicable to effluent testing, due to time and
cost constraints and various technical problems, effluent biomonitoring
-------
Table 23. Life-cycle MATC values compared with embryo-larval LCj's for organic compounds.*
Organic
Compound
Water
(mg/L
50
LCi
Hardness
as CaC03)
200
Embryo-Larval
Test Species?
MATC
Water Hardness
(mg/L as CaC03)
Life-Cycle liie-Qycle
Test Species *est3
Atrazine (pg/L)
29.0
77.2
Rainbow trout
65 - 120
35.7
Brook trout
pic (15)
2,4-D (mg/L)
0.03
0.02
Rainbow trout
0.3 - 1.5
111 - 192
Fathead minnow
pic (16)
13.1
3.2
Largemouth bass
8.2
8.9
Goldfish
Malathion {pg/L)
141
440
Goldfish
200 - 580
111 - 192
Fathead minnow
pic (16)
NTA (mg/L)
16.9
20.2
Rainbow trout
54 - 114
34.0 - 45.2
Fathead minnow
pic (17)
28.5
30.1
Goldfish
138
131
Channel catfish
PCB (pg/L)
(Capacitor 21)
0.5
0.9
Largemouth bass
(A1254) 1.8-4.6
44 - 46
Fathead minnow
clc (18,15
-
1.0
Rainbow trout
(A1242) 5.4-15.0
3.5
1.3
Redear sunfish
(A1248) 1.1-3.0
(A1260) 2.1-4.0
Modified from a previous study by Birge, et al. (5 ).
9
Tests were conducted using a flow-through system and organisms were exposed from fertilization through
4 days posthatching.
MATC's taken from partial (pic) and complete (clc) life-cycle tests. References are given parenthetically.
-------
83
Table 24. MATC's compared with LCi values determined in static-renewal
tests with rainbow trout embryo-larval stages.*
2
Element
LCi3
(yg/L)
MATC
(yg/L)
Species
Test4
Ref.
Cadmi um
8.0
1.7
.
3.4
brook trout
clc
20
3.0
-
6.5
flagfish
el
4
3.8
-
11.7
brown trout
el
21
4.1
-
12.5
coho salmon
el
21
7.4
-
16.9
flagfish
clc
4
8.1
-
16.0
flagfish
el
22
Chromium
21.5
51
-
105
rainbow trout
el
23
200
-
350
brook trout
clc
24
Copper
3.4
3.0
-
5.0
brook trout
el
23
5.0
-
8.0
brook trout
el
23
9.4
-
17.4
brook trout
clc
25
Lead
10.3
4.1
-
7.6
rainbow trout
pic
26
7.2
-
14.6
rainbow trout
pic
26
31.3
-
62.5
flagfish
clc
4
58
-
119
brook trout
clc
27
71
-
146
rainbow trout
el
23
Mercury
0.2
0.07
-
0.13
fathead minnow
clc
4
0.17
-
0.33
flagfish
pic
4
0.29
-
0.93
brook trout
clc
28
Si 1ver
0.1
0.09
-
0.17
rainbow trout
pic
29
Zinc
216
30
-
180
fathead minnow
pic
30
139
-
267
flagfish
el
22
532
•
1368
brook trout
pic
4
Modified from a previous study by Birge, et ai. (6).
2
Administered in static-renewal tests from fertilization through 4 days
posthatching.
3
Determined with the probit method of Finney (8), rather than the procedure
of Daum (31) used in earlier investigations (1, 3, 10).
4
MATC's were estimated from 30- to 90-day embryo-larval tests (el) or
determined in partial (pic) and complete (clc) life-cycle studies.
-------
84
with embryo-larval stages should provide an economical and reliable means
of estimating chronic effects under the NPDES program.
In determining the LC-j, it is important to achieve an adequate deline-
ation of test responses. Sharp truncations of or internal discontinuites
within the dose-response may skew or preclude the calculation of LC-| values.
In this initial investigation, due to the requirements of comparing three
test systems and evaluating several alternative animal species, it generally
was not possible to repeat tests or to use an extended selection of exposure
concentrations. Therefore, it usually was necessary to compromise on five
effluent exposure concentrations spaced to cover an extended dose-response
range. Despite this limitation, adequate characterization of test responses
usually was obtained, particularly with the more sensitive animal species
(jL.,e., fathead minnow, rainbow trout). In future investigations, however,
it may be advisable to use an initial "range-finding" test to gauge effluent
toxicity, permitting a more precise selection of an exponential series of
exposure concentrations for use in final testing (2). While probit analysis
is recommended, embryo-laryal test responses alternatively can be analyzed
using standard statistical procedures.
Characterization of receiving waters. Though it is possible to define
effluent toxicity, it is often a complex matter to estimate the impact of
toxic outfall upon receiving waters. Dilution, sediment and water character-
istics, structure and density of the biomass, and various other factors may
grossly alter effects. Toxic properties and pharmacodynamics of the contam-
inants, their propensity for persistence, and possible interactions may
-------
85
further complicate hazard assessment. Consequently, when receiving waters
are affected by industrial or municipal effluents, which frequently contain
complex mixtures of toxicants, reliable impact assessments may be difficult
to achieve using present effluent guidelines. One of the most important
applications of on-site biomonitoring may be in the direct evaluation of
receiving systems. Toxicity tests performed simultaneously on the effluent,
the mixing zone, and contiguous water (,e.£., upstream, downstream) could
provide comparative data useful in estimating acute and chronic effects on
aquatic biota. Using a mobile laboratory, such as that described in the
Appendix, and taking representative effluent and water samples at 12- to
24-hr intervals, toxicity tests could be performed on six or more sites
during a single operation. As such studies would be conducted under actual
environmental conditions, net effects of important variables — toxic
interactions and other factors which affect the toxicity and bioavailability
of effluent contaminants—would be reflected directly in test responses.
Such results, together with appropriate chemical monitoring data, should
provide more precise evaluations on the environmental impact of industrial
and municipal wastes. Correlated effluent/receiving water studies, involving
toxicological and chemical monitoring, as well as biotic surveys, should also
provide essential baseline information required 1) to quantify the extent to
which transport-fate phenomena and other characteristics of the receiving system
affect bioavailability and toxicity of effluent contaminants and 2) to determine
an accurate means of extrapolating from effluent toxicity determinations to
probable effects in receiving waters. It is of interest to note that Ladd (39)
-------
86
has stressed the need to apply biomonitoring not just to effluents but
also to receiving streams, whereby all the interactions affecting toxicity
would be considered. In addressing the potential of effluent biomonitoring,
Mount (40) recently stressed the fact that the use of an organism is
implicit in characterizing toxicity, and that toxicity is a property of
wastes and should be regulated within limits necessary to preclude harmful
impact upon receiving waters. He further stated (p. 5):
Some of the advantages of biomonitoring, as opposed to only chemical
monitoring, are that it probably more closely approximates the
receiving water conditions than nearly anything we can do, and it
does consider the interactions that may occur between the components
of a waste stream.
In a previous investigation by Birge, et £l_. (32), embryo-larval tests
were conducted on water samples from 11 different streams and rivers which
were selected to represent various stages of ecological degradation. Test
results correlated closely with independent ecological parameters (.e.^.,
species diversity, density) used to estimate environmental impact. The
principal ecological criterion was the retention of fish species diversity
over a period of 10 to 20 years, during which time most of the streams were
.impacted to varying extents by agricultural and industrial development
and/or urbanization. At the time of study, two heavily impacted streams
had lost all of their original 15 indigenous fish species, and species
retention varied from 13% to 100% for the remaining nine streams and rivers.
A high correlation (r = 0.98) was obtained when percent egg hatchability was
compared with fish species retention. Those streams for which retention
was 80% or more supported generally healthy and diverse aquatic fauna.
-------
87
Hatching success also correlated with changes in diversity and density of
macroinvertebrate populations in five instances where such data were
available. On the basis of these results, it was concluded that bio-
monitoring studies using early life stages of fish and amphibians provide
an accurate means by which to evaluate quality of receiving waters and
estimate prospects for long-term ecological degradation. Thus, this study
provided significant field validation concerning the use of short-term
embryo-larval tests for estimating chronic effects of environmental toxicants
on aquatic biota.
Use of embryo-larval toxicity testing in the evaluation of effluent
treatability. In three of five on-site studies, consideration was given to
effluent components or effluents at different stages of treatment. The
objective was to determine the utility of biomonitoring with embryo-larval
stages in identifying toxic effluent fractions and in determining effectiveness
of waste treatment processes. The initial testing was performed at one of
several sewage treatment plants in Lexington, Kentucky. Using newly hatched
larvae of the channel catfish as test organisms, an LCg0 of 50.2% was obtained
for the effluent from primary treatment and no appreciable toxicity was
observed for secondary effluent (Table 10). This particular plant had
established a good performance record, and no apparent impact had been
reported for the receiving waters, a fourth-order stream of relatively small
size.
During the on-site investigation conducted at the Tannery-Secondary
Sewage Treatment Plant Complex in southeastern Kentucky, static-renewal
tests using three different animal species were conducted simultaneously on
-------
88
raw tannery waste, unchlorinated secondary effluent, and the chlorinated
final secondary effluent which entered a third-order stream of moderate size.
The tannery effluent which provided approximately 25% of the wasteload of
the treatment plant was highly toxic, as judged by LC^q values of 0.08%, 0.3%,
and 0.4% with rainbow trout {fresh eggs), the fathead minnow, and Xenopus.
respectively (Table 15). The LC^q values ranged from 0.9% to 1.6% and 0.5%
to 0.9% for unchlorinated and chlorinated secondary effluents, and the 95%
confidence intervals were uniformly small (Table 14). Though values obtained
for a number of physicochemical parameters indicated some improvement in
water quality after treatment, a more accurate and definitive assessment
was possible using toxicity data. It was especially significant that results
with the three different animal species were highly consistent concerning
differences observed in toxicity among the three effluent sources (Table 15).
In this particular case, the receiving water had been heavily impacted.
The results discussed above clearly indicate the value of comparative
toxicity data taken before and after effluent treatment. However, in
designing and monitoring waste treatment systems, it is also necessary to
characterize different effluent components and identify toxic fractions
which contribute to the final discharge. During the on-site study conducted
at a metal plating plant in northeastern Kentucky (Tables 17, 18), embryo-
larval tests were used to differentiate among four separate components. As
summarized in Table 18, LC^q values determined in static-renewal tests with
the fathead minnow were 0.01%, 0.05%, 23.6%, 25.4%, and 44.7% for Components
1, 2, 3, 4, and the final effluent, respectively. In addition to the four
-------
89
specific plant effluents, surface runoff contributed at least 30% to the
final effluent. Due to local precipitation at the time of testing, the
contribution from surface runoff was difficult to quantify and could have
been greater. This dilution undoubtedly accounted in substantial measure
for the lower toxicity of the final effluent. Considering data summarized
in Table 18, it appeared that the quality of the different effluent components
and the final effluent could be compared more precisely and economically
using toxicity data than by exhaustive analyses of multiple physicochemical
parameters. Though the latter are important in monitoring waste components,
the complexity encountered in the analysis of such data often impedes accurate
estimates of toxicity. In this and other investigations, it was possible
to obtain direct quantification of toxicity using simple and inexpensive
test procedures (j_.,e., static-renewal).
It was concluded, therefore, that on-site biomonitoring was highly
desirable, if not critical, for comparative assessments on effluent fractions
and for judging impact potential of untreated and treated effluents. It
is also important to note that static-renewal procedures were judged suitable
for on-site testing, and that this generally could be accomplished 1) with
low to moderate cost, 2) without need for sophisticated instrumentation,
3) with minimum space requirements, and 4) without need for highly technical
personnel. The benefits of such testing to industry should far outweigh
investments in time and materials.
Recommendations for future work. In view of the potential of toxico-
logical biomonitoring in hazard assessment under the NPDES program, a
-------
90
number of recommendations should be made concerning further work, as follows:
1) Application of on-site biomonitoring to receiving waters. This
should include comparative toxicological and chemical monitoring of the
effluent outfall, the mixing zone, and the contiguous waters. In addition,
coordinated field studies should be carried out on the impacted waters,
involving a) faunistic surveys (e.£., fish, macroinvertebrates, benthic
invertebrates), b) toxicant residues in fish tissues, and c) histopathological
analyses of fish tissues. Transport-fate phenomena and other important
characteristics of the receiving system should be taken into account in
correlating results and in assessing the reliability of biomonitoring as
a means of quantifying effluent toxicity and estimating impact on receiving
waters.
2) Extension of baseline on-site embryo-larval biomonitoring studies
to include a larger, representative selection of industrial and municipal
effluents and waste treatment systems.
3) Further perfection and standardization of embryo-1arval biomonitoring
procedures, including computer programs for data analysis and impact assess-
ment.
4) Incorporation of additional test parameters for on-site biomonitoring
(£.£., short-term Daphnia test, tissue residue analyses).
5) Establishment on a trial basis of a laboratory screening program for
municipal and industrial effluents, working with regulatory agencies and
selected industries. Composite effluent samples would be analyzed a) to
develop a data base on effluent toxicity, b) to assist in the development of
-------
91
waste treatment procedures, and c) to assess the effectiveness of laboratory
screening.
6) Develop a short-course training program, including laboratory and
field experience, to familiarize potential users with biomonitoring techniques.
7) Extend biomonitoring techniques to Include the screening of important
health-related effects (e,.£., teratogenesis, mutagenesis, carcinogenesis).
8) Conduct an international symposium on biomonitoring as applied to
aquatic hazard assessment.
SUMMARY
The principal objectives of this study included 1) development of
embryo-larval test systems using fish and amphibian species for on-site
toxicological evaluations of complex effluents; 2) application of embryo-
larval test systems to on-site biomonitoring of selected municipal and
industrial effluents; 3) comparisons of sensitivity and reliability of
acute and embryo-larval effluent biomonitoring; 4) evaluation of embryo-
larval tests for characterization of effluents under the NPDES program; and
5) a description of procedures for effluent testing with fish and amphibian
embryo-larval stages. Further work involved determining the reliability of
laboratory testing of composite effluent samples, using direct on-site
biomonitoring as a basis for comparison.
Development of embryo-larval test systems. Of four test organisms
used in on-site effluent biomonitoring, early trout embryos (fertilization
through 9 days) and embryo-larval stages of the fathead minnow {fertilization
-------
92
through 3 to 4 days posthatching) consistently were the most sensitive, based
on LCgQ values (percent effluent). Results obtained with these species gave
the most reliable data for on-site detection and quantification of effluent
toxicity. Eyed trout eggs carried through 4 days posthatching and embryo-
larval stages of Xenopus were appreciably more tolerant. Ten comparative
evaluations were made in which flow-through, static-renewal, and static
tests were simultaneously used to evaluate selected effluents and a reference
toxicant (j,..e., phenol). In each case, the three tests were conducted using
the same exposure period, animal species, dilution water and, in so far as
possible, general test parameters (.e.jl., temperature, dissolved oxygen,
water hardness). In most instances, flow-through tests provided the greatest
resolution of dose-response data and the lowest LC^q values. Results obtained
with flow-through and static-renewal tests often did not vary significantly,
and the LC^g values usually were within a factor of about two or less, and
seldom differed by more than a factor of 2.5. Static procedures proved less
suitable for use in effluent biomonitoring with embryo-larval stages.
Appl ication of test systems to effluent biomonitoring. Moderate to
high toxicity was observed with all but one of 19 industrial and municipal
effluents and effluent components studied. Using the most sensitive test
in each case, the LC^g's for the 18 toxic effluents ranged from 0.04% to
0.9% for six; 6.4% to 16.0% for four; 21.6% to 29.3% for four; and 39.2%
to -vl00% for four. Out of these, ten were final effluents released to
receiving waters and the LC5Q values were 0.3%, 6.4%, 6.6%, and 21.6% for
four major effluents analyzed on site and 0.04%, 9.4%, 29.3%, 39.2%, 43.0%,
-------
93
and -vl00% for other effluents tested in the laboratory.
During the major on-site studies, grab and composite samples from
NPDES effluents designated for testing were transported to the labora-
tory for simultaneous analyses. Based on the results, it was concluded
that effluent toxicity can be quantified more accurately in on-site tests.
However, tests conducted in the laboratory using embryo-larval stages of
sensitive aquatic species could prove useful in preliminary effluent
screening.
In each of four on-site biomonitoring studies, 96-hr acute tests were
performed with the fathead minnow using a flow-through system. By comparison,
embryo-larval tests gave more reliable detection and much better quantifi-
cation of effluent toxicity. For example, in studies with the most toxic
effluent, LC^q values obtained with the fathead minnow were 8.0% and 0.3%
in acute and embryo-larval tests, respectively. In the remaining three tests,
effluent LC^g values ranged from 8.3% to 29.4% when determined with embryo-
larval stages, but fish acute LC^q values could not be determined.
In three on-site studies, consideration also was given to effluent
components or effluents at different stages of treatment. The objective was
to determine the utility of biomonitoring with embryo-larval stages for iden-
tifying toxic effluent fractions and for determining the effectiveness of
waste treatment processes. Results indicated that more accurate and defini-
tive assessments were possible with toxicity data than with physicochemical
parameters. It was concluded, therefore, that on-site biomonitoring was
highly desirable, if not critical, for comparative assessments on the impact
-------
94
potential of untreated and treated effluents and effluent components.
On the basis of results obtained, it was evident that embryo-larval
effluent biomonitoring provided sensitive detection and reliable quantifi-
cation of toxic discharges, and that such testing also may prove useful in
estimating levels of effluent toxicity which produce chronic effects on
aquatic biota. Permit users could apply on-site testing 1) for periodic
monitoring of effluent discharges, 2) to design and/or evaluate effectiveness
of waste treatment systems, and 3) to provide additional documentation in
permit applications. Regulatory agencies at State and Federal levels could
employ embryo-larval testing, or data obtained by such means, in revising
NPDES program requirements, in permit issuance, in granting special dispen-
sations, in determining compliance, and in the documentation of violations
considered likely to result in litigation.
-------
95
APPENDIX
TEST PROCEDURES FOR EMBRYO-LARVAL BIOMONITORING
-------
96
Introduction
The purpose of this Appendix is to provide recommended procedures for
the performance of effluent toxicity tests using embryo-larval stages of fish
and amphibians. The methods given here were modified from those previously
developed for more conventional laboratory studies (1,5, 7, 9, 11). In
addition to procedures discussed below, detailed methods on acute testing
and extensive background information pertinent to effluent monitoring have
been presented in comprehensive publications by Peltier (2) and Weber and
Peltier (33).
Description and Operation of
Embryo-Larval Biomonitoring Systems
Two test systems have been found suitable for embryo-larval biomonitoring
of municipal and industrial effluents. The flow-through procedure usually is
recommended for definitive tests. However, the static-renewal system provides
a cost-effective and reliable means for performing preliminary screening of
effluent samples. Generally 12 exposure chambers are used in each static-
renewal test, permitting duplicate dishes for controls and each of five
effluent concentrations. This assembly of 12 test units requires less than
4 square feet of bench space. Solutions usually are renewed every 12 or 24
hrs, but the interval can be modified to accommodate special needs. For
example, if an effluent contains highly volatile components, toxicity may be
characterized more accurately by using a shorter renewal period (.e.jj,., 6-8 hrs).
-------
97
The static-renewal test system is illustrated in Figures 1 and 2.3.
The exposure chamber is a Pyrex deep petri dish (400 mL), modified by the
addition of a glass inlet/outlet tube (4 mm I.D.). This tube, attached to
a 3-way valve, allows 1) gradual flow of effluent solution into the exposure
chamber, 2) siphoning of spent test water from the chamber, and 3) moderate,
continuous aeration of test water if required. Effluent and effluent
dilutions are delivered by gravity flow to the exposure chambers using
3" Pyrex funnels and silicone or latex tubing. During the solution renewal
step, a small volume of test water (50 mL) is retained in the exposure
chamber.
The flow-through system used for embryo-larval biomonitoring is illus-
trated in Figures 3.2 and 4. Inlet and outlet tubes (10 mm I.D. Pyrex)
are annealed to deep petri dishes used as exposure chambers. The inlet
is positioned approximately 7 mm above the bottom of the dish and the
outlet is attached to the opposite side, just below the shoulder. Teflon
or stainless steel screening is placed at the ends of the influent and
effluent tubes to prevent loss of test organisms. Flow rate through each
exposure chamber is set at 200 mL/hr and monitored by timed volumetric
measurements or Gilmont no. 12 flow meters. Retention time for the 300-mL
chamber is 1.5 hrs.
Full-strength effluent is continuously pumped from the NPDES sampling
station to an overflow-equipped effluent reservoir situated inside the mobile
laboratory. Incoming effluent is filtered through glass wool and delivered
to a serial diluter by a peristaltic pump (Brinkmann model 131900). Dilution
-------
98
water, held in a 30-gallon Nalgene tank, is administered to the diluter by
gravity flow, which is regulated using a float valve. The purpose of the
glass wool filter is to remove only coarse particulates which might produce
malfunctioning of the dilution system, and this should not significantly
alter effluent toxicity. Considering flow rate, as well as retention time
in the effluent reservoir and diluter, elution of soluble toxicants from
such particulates should occur prior to the exposure of test organisms.
The diluter, as seen in Figure 3.2, is constructed of 4.7 mm Plexiglas
and has overall dimensions of 40 cm (1) by 10 cm (w) by 45 cm (h). The
structure is divided into several compartments, including an upper head
box (15 cm deep) and four dilution (mixing) chambers which are 10 cm
square and 10 cm deep. The length of the box can be increased if more than
four dilution chambers are desired. The head box receives dilution water
which is distributed to the dilution chambers via adjustable standpipes.
Standpipes are fabricated from 15-cm lengths of 3-mm O.D. glass tubing and
fitted with size 00 rubber stoppers. Each standpipe is fire-polished to
deliver approximately 30 ± 1 mL/min when the intake is positioned at mid-
depth in the head box. Vertical adjustment of the standpipes permits precise
regulation of dilution water flow.
Each dilution chamber is provided with three staggered baffles to
insure thorough mixing of effluent and dilution water. Overflow notches
cut in the front panel maintain a constant working volume of 850 mL. Given
a total flow rate of 33 mL/min (1:10 dilution ratio) for combined effluent
and dilution water, retention time in the mixing chamber is approximately
-------
99
26 min. A peristaltic pump delivers ful1-strength effluent to the first
mixing chamber through 1.6-mm I.D. silicone tubing, which enters the
diluter through the rear panel. A second pump channel extracts diluted
effluent from the first mixing chamber and delivers it to the next mixing
chamber, where further dilution is accomplished by standpipe flow. This
process is repeated for all subsequent dilutions. Effluent concentrations
(!•£•> 10%, 1%, 0.1%, 0.01%) are carried by peristaltic pump channels from
each dilution chamber to duplicate embryo-larval exposure chambers.
Dilution ratios are determined daily, using timed volumetric measurements
of effluents (peristaltic pump flow) and dilution water (standpipe flow).
The 100% and 50% effluent concentrations are provided directly to exposure
chambers by peristaltic pump flow. Dilution water for control populations
is pumped directly from the diluter head box or the dilution water reservoir
(Figure 3.2). The flow system requires approximately 800 mL/hr of undiluted
effluent and 8 L/hr of dilution water to provide continuous flow to two
replicate exposure chambers for control water, full-strength effluent, and
four effluent dilutions. This diluter system was developed in a previous
study by Freeman and Birge (unpublished observations), and details con-
cerning its design and construction currently are being submitted for
publication. In tests reported above, overall errors in flow rates and
dilution ratios usually were within 5% and seldom exceeded 10%.
The flow-through system, as described above, provided a reliable
means of quantifying the toxicity of the various effluents tested to date.
However, should an effluent prove particularly difficult to test, due to
-------
100
highly volatile or insoluble components, the system can be modified to
minimize such problems by employing procedures previously described by
Birge, et jil_. (9).
The test systems described above are housed in a mobile laboratory
especially designed for embryo-larval biomonitoring (Figure 2.1). Dimensions
of the laboratory are 14 feet in length, 7 feet 10 inches in width, and 7
feet 6 inches in height. The unit is constructed with an aluminum exterior,
3 inches of Styrofoam insulation in the walls and ceiling, and one-half inch
plywood interior walls. The two-inch thick wooden floor is overlaid with
three-quarter inch marine plywood and heavy-duty vinyl floor covering.
Four inches of Styrofoam insulation are laid beneath the floor and retained
in place with a 12-gauge galvanized metal subfloor. Due to the heavy
insulation, the high reflectivity of the aluminum exterior, and the lack
of windows, a 5000-BTU air conditioner is adequate to maintain temperature
down to 17°C even when ambient temperature reaches 35°C. A 1500-watt
portable heater provides adequate temperature regulation during cold
weather. When lower temperatures are required for testing coldwater
species (e.£., trout; 12 ± 1°C), up to 14 exposure chambers can be con-
tained in a 4.8 cubic foot table-top refrigerator (Figures 2.2, 2.4).
As shown in Figure 3.1, stand-up-height kitchen cabinets with Formica
tops provide ample space for test systems and monitoring equipment. Adjust-
able shelves are used for additional equipment, peristaltic pumps, reagents,
and labware. If a local electrical hookup is not available, power is
supplied to the laboratory by a diesel generator. The latter provides
current to five 20-amp circuits and approximately 24 outlets.
-------
i'tfsfc
-------
102
Figure 3
3.1 Interior view of the front end of the mobile laboratory. Furnishings
include adjustable shelving and stand-up-height kitchen cabinets with
Formica bench tops. Other facilities include an air conditioner (A),
a dilution water reservoir (B), a flow-through effluent test system
with embryo-larval exposure chambers and related equipment (C), a
flow-through effluent reservoir contained in a stainless steel sink (D),
a dilution water storage tank (E), and general laboratory apparatus
[e.g.., pH meter, dissolved oxygen meter). Due to the heavy insulation
of the mobile laboratory, a 5000 BTU air conditioner is adequate to
regulate temperature down to 17°C even when ambient temperature reaches
35°C. Dilution water and effluent reservoirs are filled and the flow
system is set in operation at least 12 to 24 hrs prior to the onset of
testing (introduction of test organisms). During this period, flow
rates and dilutions ratios are monitored and adjusted. This time
interval also permits effluent and dilution water to reach room temper-
ature which is regulated in the optimum range for selected test organisms.
Effluent is pumped into the mobile laboratory from an external source
using submersible pumps. Approximately 800 mL/hr of effluent are
required to supply a series of 10 exposure chambers (duplicates of
five effluent concentrations). Due to the low volume of flow,which
can be regulated with a hose clamp,and the standing time in the
overflow-equipped effluent reservoir (approximately 45-75 min),
undiluted effluent normally reaches the selected test temperature
before entering the flow-through system. However, if necessary,
further temperature adjustment can be achieved by regulating effluent
flow rate, by lengthening the effluent lines, or by varying retention time
in the effluent reservoir. After optimum conditions have been reached,
test organisms (e.g., fish or amphibian eggs) are placed in the
exposure chamber.
3.2 Enlarged view of flow-through effluent test system. A thirty-gallon
Nalgene tank serves as the dilution water reservoir (A). The latter
is connected by a float valve (B) to a four-stage serial diluter (C)
which receives a continuous flow of full-strength effluent provided by
a peristaltic pump (D). The same peristaltic pump is used to supply
100% and 50% effluent directly to embryo-larval test chambers (E).
Lower dilutions (e.g., 10%, 1%, 0.1%, 0.01%) are conveyed from the
diluter to exposure chambers using an additional peristaltic pump (F).
Flow rates through peristaltic pump lines are monitored using Gilmont
no. 12 flow meters (G).
-------
-------
103
Effluent Sampling Procedures, Dilution Water, and
Test Conditions
As noted above, effluent used 1n the flow-through system is pumped
continuously from the NPDES sampling station to an effluent reservoir in the
mobile laboratory. Approximately 800 mL/hr of effluent are required to supply
a series of ten exposure chambers (duplicates of five exposure concentrations).
Due to the low volume of flow, which can be regulated with a hose clamp, and
the retention time in the effluent reservoir (approximately 45-75 min), un-
diluted effluent normally reaches the selected test temperature (.e.jj., 22 ± 1°C
for warmwater species) before entering the flow-through system. If necessary,
additional temperature control can be achieved by adjusting effluent flow
rate, by lengthening effluent reservoir inlet or outlet lines, or by varying
retention time in the effluent reservoir. For static-renewal tests, samples
are taken from the effluent reservoir every 12 or 24 hrs, diluted to the
required effluent concentrations, and administered to the test organisms.
A 12-hr renewal interval is recommended. Dilution water is usually collected
from an appropriate source near the test site (e.£., deep wells; receiving
water upstream from the plant discharge) and stored in 25-gallon tanks in
the mobile laboratory. However, when dilution water of acceptable quality
cannot be obtained locally, the alternative sources given by Peltier (2) or
the reconstituted water described by Birge, et al_. (7, 11) are recommended.
When tests are conducted with coldwater species, dilution water and
effluent must be cooled prior to testing. For flow-through tests, a
-------
Input
Over flow -
EFFLUENT
RESERVOIR
Slondptpe-
DILUTER-
PUMP NO.I
f \ S~\
III
. 1 • 1
I III
M-
11 11
i'i
i i i
PUMP NO 2
DILUTION
WATER
-Head Box
'Flow Meter
100%
10%
1%
0.1 %
001%
0%
(control)
o
EFFLUENT EXPOSURE CHAMBERS
Figure 4. Design of the flow-through effluent test system.
Effluent from a flow-through reservoir is delivered to a four-stage serial diluter by a peristaltic pump. Dilution water,
held in a 30-gallon Nalgene reservoir, is conveyed to the diluter head box by gravity flow and distributed to the dilution
chambers (i.e., 1-4) via adjustable standpipes. Full-strength effluent is delivered by a peristaltic pump to the first
mixing chamber (i.e., 1) and combined with dilution water to effect the desired effluent concentration. A second pump
channel extracts diluted effluent from the first mixing chamber and delivers it to the next chamber (i.e., 2) where it is
further diluted by standpipe flow, and this process is repeated for all subsequent dilutions. A second peristaltic pump
is used to supply effluent dilutions (e.£., 1056, 1%, 0.1%, 0.01%) to corresponding exposure chambers, and 100% effluent
is pumped directly from the effluent reservoir. Dilution water for control populations is supplied from either the
dilution *er reservoir or the diluter head box. Flow rat >re monitored using calibrated flow meters or t volu-
metric Hi rements. This system will accommodate 12 or mt jxposure chambers, permitting replicate tests.
-------
105
refrigerated water bath or immersion cooler may be required. However,
static-renewal exposure chambers and a one-day supply of replacement
solution (t4 liters) can be maintained in a table-top refrigerator.
Continuous aeration should be supplied to the dilution water reservoir
and this usually is sufficient to maintain dissolved oxygen at or close
to saturation in the exposure chambers. If necessary, aeration should
be supplied directly to the exposure chambers to prevent dissolved oxygen
from falling below 60% saturation. In the latter case, air flow should
be maintained at a low rate to minimize loss of volatile effluent com-
ponents.
Selection and Handling of Test Organisms
Requirements for the selection of animal test species should include good
general health, an uncontaminated native habitat, adequate seasonal avail-
ability of viable eggs, ease of transport and handling, and sufficient
sensitivity to toxicant stress. Though not essential, it is desirable to
select organisms that are indigenous to the effluent receiving stream or
similar waters. Biomonitoring experiments can be initiated either with
eggs spawned at the test site or with fresh spawns collected from field
or laboratory populations. Eggs generally can be transported in ice-
packed containers to the test location by air freight or automobile. It
is recommended that tests be initiated as soon after fertilization as
possible. For warmwater species, tests should be continued through 4
days posthatching. With coldwater fish such as trout, a testing period
spanning the first 8 to 9 days of embryonic development has proved
-------
106
adequate for evaluating effluent toxicity. Fish and amphibian species found
suitable for use in short-term embryo-larval toxicity tests are included in
Table 25, and characteristics of several of these species are discussed
below.
A. Fish
1. Fathead minnow (Pimephales promelas). Eggs and larvae of the fat-
head minnow are used extensively in toxicity testing and are relatively
sensitive to many contaminants. The egg is approximately 1 to 3 mm in
diameter, and each female produces 50 to 60 eggs. Eggs can be obtained
throughout the year from established laboratory cultures. Although arti-
ficial spawning methods can be employed, lower egg viability usually
results. The developmental period to hatching is 4 to 5 days at 22°C.
Because this organism is somewhat susceptible to fungal contamination, it
is recommended that dead embryos be removed from exposure chambers daily.
2. Rainbow trout (Salmo qairdneri). Embryo-larval stages of this
species also have been used widely in toxicity testing and usually are
highly sensitive to aquatic contaminants. The egg is 6 to 8 mm in diameter,
and each female produces between 1,000 and 3,500 eggs. The adults can be
spawned in the laboratory or in the field, permitting initiation of tests
immediately after fertilization. Although developmental time to hatching
is relatively long (23 days at 12.5°C), an exposure period spanning the first
8 to 9 days of development has proved adequate for reliable assessments of
effluent toxicity. Depending upon strain and geographic location, gravid
adults are available 9 months of the year (August-April). It should be
-------
107
noted that eyed-egg stages are considerably more tolerant than early
embryos.
3. Channel catfish (Ictalurus punctatus). Embryo-larval stages of
the catfish are quite sensitive to many aquatic contaminants. The egg is
approximately 4 to 8 mm in diameter, and each female produces from 6,000
to 15,000 eggs. The developmental period to hatching is 5 to 6 days at
22°C. While an ideal test species 1n many respects, catfish have a short
spawning season (June - July) and the eggs are susceptible to fungal con-
tami nation.
4. Largemouth bass (Micropterus salmoides). Compared to embryos and
larvae of the above three species, developmental stages of the largemouth
bass usually are somewhat less sensitive to aquatic toxicants. The egg is
1 to 2 mm in diameter, and each female produces approximately 2,000 to
10,000 eggs. The developmental time to hatching is 3 to 4 days at 22°C
and spawning, depending upon temperature, generally occurs from March
through June.
5. Bluegill sunfish (Lepomis macrochirus). Embryos and larvae of this
species usually exhibit sensitivity similar to that observed with life-
stages of the largemouth bass. The egg is small (0.75-1.5 mm in diameter),
and each female produces approximately 600 to 2,000 eggs. The developmental
period to hatching is 2 to 3 days at 22°C, and the spawning season ranges
from April through July.
6. Goldfish (Carassius auratus). Goldfish generally are similar to
bass and bluegill in terms of embryo-larval sensitivity to aquatic toxi-
-------
108
cants. The egg is 1 to 2 mm in diameter, and each female produces between
5,000 and 20,000 eggs. As spawning can be induced by hormonal injection,
eggs are available for testing over much of the year (February-October).
The developmental period to hatching is approximately 3 days at 22°C.
B. Amphibians
1. African clawed frog (Xenopus laevis). Embryos and larvae of the
African clawed frog, a species recently introduced to the United States,
appear to exhibit sensitivity somewhat less than that observed for most
other amphibian species (,e.£., leopard frog). The eggs measure 1.2 to 1.7
mm in diameter, and each female produces between 1,000 and 2,000 eggs. By
use of hormonal injections, eggs are available for testing nearly 12 months
of the year. The developmental period to hatching is approximately 2 days
at 22°C. The reproductive history and embryology of Xenopus laevis is
especially well known, as it is one of the major experimental organisms
used in developmental biology. Compared to fish species, parental stocks
can be maintained with simpler facilities and lower cost.
2. American toad (Bufo americanus). Compared to many other amphibian
species, embryos and larvae of the American toad are more tolerant to
aquatic toxicants. The eggs are approximately 1.0 to 1.4 mm in diameter,
and each female produces from 2,000 to 4,000 eggs. The spawning season
spans approximately 5 months (April-August), and developmental time to
hatching is 2 to 3 days at 22°C.
3. Bullfrog (Rana catesbeiana). Embryos and larvae of the bullfrog
appear to be comparable in sensitivity to developmental stages of
-------
109
Table 25. Candidate species for use in embryo-larval toxicity tests
on industrial and municipal effluents.1
PREFERRED SPECIES
Fish
Bluegill sunfish (Lepomis aacrochirus)
Channel catfish (ictalurus punctatus)
Fathead minnow (Pimephales promelas)
Goldfish (Carassius auratus)
Largemouth bass (Micropterus salmoides)
Rainbow trout (salmo gairdneri)
fresh eggs
Redear sunfish (Lepomls microlophus)
Amphibians
Bullfrog {l&na catesbeiana)
Leopard frog (ife/ia pipiens)
OPTIONAL SPECIES
Fi sh
Rainbow trout (Salmo gairdneri)
eyed eggs
Amphibians
African clawed frog (xenopus laevis)
American toad (Bufo americanus)
Fowler's toad (Bufo fowleri)
Narrow-mouthed toad (g astrophryrte carolinensis)
Species were restricted to those used in our effluent testing program.
Many other aquatic species should prove suitable for this purpose.
-------
no
the bass, goldfish, and bluegill sunfish. The egg measures 1.2 to 1.7 mm
in diameter, and each female produces from 2,000 to 10,000 eggs. The
spawning season is from May through August, and developmental time to
hatching is 4 to 5 days at 22°C.
4. Leopard frog (Rana pipiens). Embryo-larval stages of the leopard
frog have been used extensively in experimental biology and aquatic toxicity
tests and are quite sensitive to many contaminants. The egg averages 1.6 mm
in diameter, and each female produces from 2,000 to 4,000 eggs. Gravid
adults can be obtained from biological supply houses or field populations.
Animals can be spawned over a 7-month period (November-May), using hormonal
injection. Developmental time to hatching is 5 to 6 days at 22°C. This
species is especially applicable to effluent testing due to its broad
geographic distribution and other characteristics.
5. Narrow-mouthed toad (Gastrophryne carolinensis). The narrow-mouthed
toad consistently has been one of the most sensitive species tested in our
laboratory. The egg is 1.0 to 1.2 mm in diameter and each female produces
from 100 to 200 eggs. The spawning season is quite long (April-November),
and developmental time to hatching is approximately 3 to 4 days at 22°C.
Limitations associated with the use of this species involve its rather narrow
geographic distribution and low egg production.
Test Responses,
Expression of Data, and Statistical Procedures
In the effluent toxicity test systems described above, organisms are
exposed to full-strength effluent, several effluent dilutions, and dilution
-------
Ill
(control) water. Eggs are examined daily to gauge extent of development
and to remove'dead animals. Sample size generally ranges from 50 to 100
organisms per exposure chamber. Two species may be tested in the same
exposure chamber provided they are separated by a Teflon or stainless steel
screen. Only eggs of high viability should be used, and control survival
should average about 80% or more.
Test responses include frequencies of egg hatchability, embryo-larval
survival, and teratogenesis. Determinations of teratic organisms are
limited to gross defects considered likely to preclude survival, as
discussed previously by Birge et al_. (11, 13). Defects most commonly
encountered are acute lordosis, scoliosis, and other gross anomalies of
the vertebral column. Percent hatchability is based on all organisms,
normal and aberrant, which live to complete the hatching process. In
determining percent survival, teratic organisms are counted as lethals,
except when tests are terminated prior to hatching (!•.§.•» fresh trout
eggs). Taking accumulative dose-response data at the end of the exposure
period, log probit analysis (8) is used to determine LC^g values (percent
effluent by volume) with 95% confidence limits. In addition, LC^ values,
defined as concentrations producing 1% control-adjusted impairment of test
populations, are calculated to estimate effluent dilution factors required
to preclude toxic effects to embryo-larval stages and to estimate effluent
concentrations likely to produce chronic effects. In determining the LC^,
it is important to develop an adequate delineation of test responses.
Sharp truncations of or internal discontinuities within the dose-response
-------
112
may skew or preclude the calculation of LC^ values. For this reason, it
is advisable to use an initial "range-finding" test to estimate effluent
toxicity, permitting a more precise selection of an exponential series of
exposure concentrations for use in final testing (2).
Performance Evaluation and Personnel Requirements
Reference toxicants may be used to evaluate and standardize performance
of embryo-larval effluent biomonitoring systems. Such evaluations usually
are required to assess the precision and reproducibility of the flow-through
procedure. A known amount of toxicant Ce.jj,.. zinc, phenol) is added to
the effluent reservoir which supplies the serial diluter, and test solutions
from the exposure chambers are analyzed for toxicant concentrations. Accuracy
of the test system can be evaluated by comparing calculated nominal concen-
trations with actual toxicant determinations.
During the performance of effluent toxicity tests, flow rates from the
diluter head box and peristaltic pumps should be monitored regularly to
determine the accuracy of dilution ratios. Solutions in the exposure
chamber should be analyzed daily for general water quality characteristics.
Temperature, dissolved oxygen, specific conductivity, and pH may be deter-
mined using a YSI telethermometer with thermocouple (model 42SC), YSI oxygen
meter (model 54A), YSI conductivity meter (model 33), and a Corning pH meter
(model 610). Hardness and alkalinity measurements are accomplished using
the EDTA and methyl orange titrimetric procedures described in Standard
Methods (34). In addition to the above parameters, it may be desirable to
-------
113
analyze test solutions for one or more of the principal effluent toxicants.
General parameters of test water should be determined either in exposure
chambers (,e.£., temperature, DO) or soon after sample collection (£.£.,
alkalinity, hardness). Special attention should be given to procedures
used in the collection and preservation of effluents. Effluent samples
to be used for toxicity evaluations should be stored without residual air
space in tightly capped containers (,e.£.» Teflon, Pyrex, stainless steel),
and tests should be performed as soon as possible to minimize storage time.
If tests are not to be initiated soon after collection, effluent samples
should be refrigerated. When chemical analyses are to be performed, handling
and preservation procedures may vary according to the characteristics of
the effluent or the specific toxicants selected for analysis, and reference
should be made to Standard Methods (34) or other sources (2, 33).
Embryo-larval tests described above can be performed without the use
of highly technical personnel. Though a minimum of two years of study or
experience in biology, chemistry, or related disciplines is preferred,
individuals of high aptitude but less formal background may prove satis-
factory, given adequate training.
Cost Analysis for Effluent Monitoring Using
Embryo-Larval Toxicity Tests
Estimates were subject to some imprecision due to the fact that the
work performed was largely experimental. The cost of a mobile laboratory
as described above ranges from $10,000 to $12,500, depending upon optional
-------
114
features (.e.jl.. exterior lighting, electrical service, plumbing, construction
materials). For extensive field work, a diesel-powered generator also is
required. The cost varies with amperage output and different models, ranging
from approximately $5,500 to $8,500 for a suitable unit. Minimum equipment
necessary to perform flow-through studies, either in the laboratory or on
site, totals approximately $8,500 (e.£.> diluter, peristaltic pumps, pH
meter, telethermometer, dissolved oxygen meter, titration apparatus, exposure
chambers, general labware), whereas that required for static-renewal testing
costs about $3,000 (e.j£., pH meter, titration apparatus, dissolved oxygen
meter, telethermometer, and labware). Once obtained, these facilities can
be reused extensively with minimal maintenance.
On-site testing, as conducted in this study using a mobile laboratory,
requires a two-man crew for a period of 12 to 13 days. Cost of food, lodging,
and local travel averages $125 per day. Expendable supplies total about
$300 for each flow-through test and $200 when the static-renewal system is
used.
Effluent testing conducted by industry personnel, excluding space
requirements, electrical supply, and cost of test organisms, would require
approximately $200 in expendable supplies and 32 hrs of labor for a series
of static-renewal tests performed over an 8- to 10-day period. Labor
would more than double with the use of flow-through procedures. Exclusive
of permanent equipment and space requirements, a set of static-renewal
tests likely could be completed for about $500 (labor and expendable
supplies), and the cost probably would increase to $1,000 to $1,200 for
-------
115
flow-through tests. These are minimal estimates and could increase
depending upon wage scale, efficiency of personnel, and conditions at
particular testing sites Ce.,2.* access to effluent, availability of test
organisms).
-------
116
Equipment Inventory
A. Test System and Monitoring Equipment
Nalgene rectangular tank (30 gallon)
Nalgene round tank (25 gallon)
Peristaltic pump (Brinkmann model 131900 or model IP-12)
Submersible pump (Little Giant model 2E-NDVR)
Refrigerators (Norcold model DE250, 2.5 cu. ft.; Gerald model GR54,
4.8 cu. ft.)
pH Meter with probe (Corning model 610)
Conductivity meter with probe (YSI model 33)
Oxygen meter with probe (YSI model 54A)
Telethermometer (YSI model 42SC)
Air pumps (Hagen Optima)
Diluter
Pyrex test chambers with covers - flow design
Pyrex test chambers with covers - static and static-renewal design
Long-stem 3" Pyrex funnels with covers
Liquid flow meters (Gilmont no. 12)
B. Test System and Monitoring Supplies
Silicone tubing, assorted diameters
Tubing connectors
Latex rubber tubing
Air valves
Pyrex carboys (5 gallon)
Nalgene carboys (5 gallon)
Pyrex Erlenmeyer flasks
Pyrex graduated cylinders
Pyrex funnels
Burettes with clamps and stands
Disposable beakers (50, 100, 250, 1000 mL)
Stainless steel screening
Pasteur pipettes and bulbs
Pyrex disposable pipettes
Thermometers
C. Chemicals and Reagents
Distilled water
Reagents for hardness determinations
Reagents for alkalinity determinations
Standards for pH meter calibrations
Disinfectant
-------
117
D. Safety Equipment
First aid kit
Fire extinguishers
Hard hats
Safety glasses
Organic vapor respirators and refill cartridges
Rubber gloves
Latex gloves
Flashlight and batteries
E. Miscellaneous Supplies and Equipment
Nalgene wash bottles
Pyrex disposable screw-cap test tubes
Glass tubing
Hose clamps
Egg spawning supplies
Data collection supplies (stopwatch, calculator)
Tape (masking, electrical, duct)
Aluminum foil
Kimwipes
Towels
Waders
F. Laboratory Maintenance Equipment
Air conditioner (5000 BTU)
Space heater (1500 watts)
3-ton hydraulic jacks and jack stands
Electrical extension cords
Gas and diesel tanks
Tool kit and assorted hand tools
Diesel generator (60 amp)
-------
118
REFERENCES
1. Birge, W.J. 1978. Aquatic toxicology of trace elements of coal and
fly ash. hi Energy and Environmental Stress in Aquatic Systems,
J.H. Thorp and J.W. Gibbons, eds., DOE Symposium Series (CONF-771114),
Washington, D.C. pp. 219-240.
2. Peltier, W. 1978. Methods for Measuring the Acute Toxicity of Effluents
to Aquatic Organisms. EPA-600/4-78-012, U.S. Environmental Protection
Agency, Cincinnati, Ohio. 52 p.
3. Birge, W.J., J.E. Hudson, J.A. Black, and A.G. Westerman. 1978. Embryo-
larval bioassays on inorganic coal elements and in situ biomonitoring
of coal-waste effluents, hi Surface Mining and Fish/Wildlife Needs in
The Eastern United States, D.E. Samuel, J.R. Stauffer, C.H. Hocutt, and
W.T. Mason, eds., Publication no. FWS/OBS-78/81, U.S. Department of the
Interior, Washington, D.C. pp. 97-104.
4. McKim, J.M. 1977. Evaluation of tests with early life stages of fish
for predicting long-term toxicity. 0. Fish. Res. Board Can., 34(8):
1148-1154.
5. Birge, W.J., J.A. Black, and D.M. Bruser. 1979. Toxicity of Organic
Chemicals to Embryo-larval Stages of Fish. EPA-560/11-79-007, U.S.
Environmental Protection Agency, Washington, D.C. 60 p.
6. Birge, W.J., J.A. Black, A.G. Westerman, and J.E. Hudson. 1980. Aquatic
toxicity tests on inorganic elements occurring in oil shale. l£ Oil
Shale Symposium: Sampling, Analysis and Quality Assurance, March 1979,
C. Gale, ed., EPA-600/9-80-022, U.S. Environmental Protection Agency,
Cincinnati, Ohio. pp. 519-534.
7. Birge, W.J., J.A. Black, and B.A. Ramey. 1981. The reproductive
toxicology of aquatic contaminants. Hazard Assessment of Chemicals -
Current Developments, J. Saxena and F. Fisher, eds., Academic Press,
New York. pp. 59-115.
8. Finney, D.J. 1971. Probit Analysis, Third Edition. Cambridge Press,
New York. 333 p.
9. Birge, W.J., J.A. Black, J.E. Hudson, and D.M. Bruser. 1979. Embryo-
larval toxicity tests with organic compounds. Jii Aquatic Toxicology,
L.L. Marking and R.A. Kimerle, eds., Special Technical Publication 667,
American Society for Testing and Materials, Philadelphia, Pa. pp. 131-147.
10. Birge, W.J., J.A. Black, and A.G. Westerman. 1979. Evaluation of
aquatic pollutants using fish and amphibian eggs as bioassay organisms.
In Animals as Monitors of Environmental Pollutants, National Academy
of Sciences, Washington, D.C. pp. 108-118.
-------
11
12
13
14
15
16
17
18
19
20
21
22
119
Birge, W.J., J.A. Black, A.G. Westerman, and J.E. Hudson. 1979. The
effects of mercury on reproduction of fish and amphibians.
Biogeochemistry of Mercury in the Environment, J.O. Nriagu, ed.,
Elsevier/North Holland Biomedical Press, Amsterdam, pp. 629-655.
Birge, W.J. and J.J. Just. 1975. Bioassay Procedures Using Developmental
Stages as Test Organisms. Research Report No. 84, U.S. Department of
the Interior, Washington, D.C. 36 p.
Birge, W.J. and J.A. Black. 1977. Sensitivity of Vertebrate Embryos
to Boron Compounds. EPA-560/1-76-008, U.S. Environmental Protection
Agency, Washington, D.C. 62 p.
Birge, W.J., J.A. Black, and R.A. Kuehne. 1980. Effects of Organic
Compounds on Amphibian Reproduction. Research Report No. 121, U.S.
Department of the Interior, Washington, D.C. 39 p.
Macek, K.J., K.S. Buxton, S. Sauter, S. Gnilka, and J.W. Dean. 1976.
Chronic Toxicity of Atrazine to Selected Aquatic Invertebrates and
Fishes. EPA-600/3-76-047, U.S. Environmental Protection Agency, Duluth,
Minn. 50 p.
Mount, D.I. and C.E. Stephan. 1967. A method for establishing acceptable
toxicant limits for fish - malathion and the butoxyethanol ester of 2,4-D.
Trans. Am. Fish. Soc., 96: 185-193.
Arthur, J.W., A.W. Lemke, V.R. Mattson, and B.J. Halligan. 1974. Toxicity
of sodium nitrilotriacetate (NTA) to the fathead minnow and an amphipod
in soft water. Water Res., 8: 187-193.
Nebeker, A.V., F.A. Puglisi, and D.L. DeFoe. 1974. Effect of poly-
chlorinated biphenyl compounds on survival and reproduction of the
fathead minnow and flagfish. Trans. Am. Fish. Soc., 103: 562-568.
DeFoe, D.L., G.D. Veith, and R.W. Carlson. 1978. Effects of Aroclor
1248 and 1260 on the fathead minnow (Pimephales promelas). J. Res.
Board Can., 35(7): 997-1002.
Benoit, D.A., E.N. Leonard, G.M. Christensen, and J.T. Fiandt. 1976.
Toxic effects of cadmium on three generations of brook trout (Salve!inus
fontinalis). Trans. Am. Fish. Soc., 105: 550-560.
Eaton, J.G., J.M. McKim, and G.W. Holcombe. 1978. Metal toxicity to
embryos and larvae of seven freshwater fish species - I. Cadmium.
Bull. Environ. Contam. Toxicol., 19: 95-103.
Spehar, R.L. 1976. Cadmium and zinc toxicity to flagfish, Jordanella
floridae. J. Fish. Res. Board Can., 33: 1939-1945.
-------
120
23. Sauter, S., K.S. Buxton, K.J. Macek, and S.R. Petrocel1i. 1976. Effects
of Exposure to Heavy Metals on Selected Freshwater Fish, Toxicity of
Copper, Cadmium, Chromium and Lead to Eggs and Fry of Seven Fish Species.
EPA-600/3-76-105, U.S. Environmental Protection Agency, Duluth, Minn.
75 p.
24. Benoit, D.A. 1976. Toxic effects of hexavalent chromium on brook trout
(Salvelinus fontinalis) and rainbow trout (Salmo gairdneri). Water Res.,
10: 497-500.
25. McKim, J.M. and D.A. Benoit. 1974. Duration of toxicity tests for
establishing "no effect" concentrations for copper with brook trout
(Salvelinus fontinalis). J. Fish. Res. Board Can., 31: 449-452.
26. Davies, P.H., J.P. Goettl, J.R. Sinley, and N.F. Smith. 1976. Acute
and chronic toxicity of lead to rainbow trout Sal mo gairdneri, in hard
and soft water. Water Res., 10: 199-206.
27. Holcombe, G.W., D.A. Benoit, E.N. Leonard, and J.M. McKim. 1976. Long-
term effects of lead exposure on three generations of brook trout
(Salvelinus fontinalis). J. Fish. Res. Board Can., 33: 1731-1741.
28. McKim, J.M., G.F. Olson, G.W. Holcombe, and E.P. Hunt. 1976. Long-term
effects of methylmereuric chloride on three generations of brook trout
(Salvelinus fontinalis): toxicity, accumulation, distribution, and
elimination. J. Fish. Res. Board Can., 30: 485-492.
29. Davies, P.H., J.P. Goettl, Jr., and J.R. Sinley. 1978. Toxicity of
silver to rainbow trout (Salmo gairdneri). Water Res., 12: 113-117.
30. Brungs, W.A. 1969. Chronic toxicity of zinc to the fathead minnow
(Pimephales promelas Rafinesque). Trans. Am. Fish. Soc., 98: 272-279.
31. Daum, R.J. 1969. A revision of two computer programs for probit analysis.
Bull. Entom. Soc. Am., 16: 10-15.
32. Birge, W.J., J.A. Black, A.G. Westerman, P.C. Francis, and J.E. Hudson.
1977. Embryopathic Effects of Waterborne and Sediment-Accumulated
Cadmium, Mercury, and Zinc on Reproduction and Survival of Fish and
Amphibian Populations in Kentucky. Research Report No. 100, U.S.
Department of the Interior, Washington, D.C. 28 p.
33. Weber, C.I. and W. Peltier. 1980. Draft - Use of Effluent Toxicity
Testing in the Second Round of NPDES Permit Issuance. U.S. Environmental
Protection Agency, Washington, D.C.
34. American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1975. Standard Methods for the
Examination of Water and Wastewater, 14th Edition. American Public Health
Association, Washington, D.C. 1193 p.
-------
121
35. Newton, B. and R. Brandes. 1981. Use of effluent toxicity testing in
the second round of NPDES permit issuance. Draft policy statement,
Industrial Permits Branch, Office of Water Enforcement, U.S. Environ-
mental Protection Agency, Washington, D.C. 46 p.
36. Environmental Protection Agency. 1979. Interim NPDES Compliance Bio-
monitoring Inspection Manual. MCD-62, U.S. Environmental Protection
Agency, Washington, D.C.
37. Environmental Protection Agency. 1979. Draft Application Forms for
EPA Programs. Fed. Reg., 44(116): 34346-34416.
38. Weber, C.I. 1980. Federal and state biomonitoring programs, ^n
Biological Monitoring for Environmental Effects, D.L. Worf, ed.,
Lexington Books, Lexington, Mass. pp. 25-52.
39. Ladd, E.C. 1977. An industrial view of biological monitoring,
Biological Monitoring of Water and Effluent Quality, J. Cairns, Jr.,
K.L. Dickson, and G.F. Westlake, eds., Special Technical Publication
607, American Society for Testing and Materials, Philadelphia, Pa.
pp. 24-29.
40. Mount, D.I. 1980. Biomonitoring of effluents in perspective. In
Proceedings of the Seminar on Biological Monitoring and Its Use in the
NPDES Permit Program, EPA-600/9-80-026, Office of Research and Develop-
ment, U.S. Environmental Protection Agency, Cincinnati, Ohio. pp. 3-7.
------- |