U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
PB-252 488
ACUTE TOXICITY OF SELECTED TOXICANTS TO SIX SPECIES
OF FISH
CHEMICO PROCESS PLANTS COMPANY
PREPARED FOR
ENVIRONMENTAL RESEARCH LABORATORY
MARCH 1976
-------
KEEP UP TO DATE
Between the time you ordered this report—
which is only one of the hundreds of thou-
sands in the NTIS information collection avail-
able to you—and the time you are reading
this message, several new reports relevant to
your interests probably have entered the col-
lection.
Subscribe to the Weekly Government
Abstracts series that will bring you sum-
maries of new reports as soon as they are
received by NTIS from the originators of the
research. The WGA's are an NTIS weekly
newsletter service covering the most recent
research findings in 25 areas of industrial,
technological, and sociological interest—
invaluable information for executives and
professionals who must keep up to date.
The executive and professional informa-
tion service provided by NTIS in the Weekly
Government Abstracts newsletters will give
you thorough and comprehensive coverage
of government-conducted or sponsored re-
search activities. And you'll get this impor-
tant information within two weeks of the time
it's released by originating agencies.
WGA newsletters are computer produced
and electronically photocomposed to slash
the time gap between the release of a report
and its availability. You can learn about
technical innovations immediately—and use
them in the most meaningful and productive
ways possible for your organization. Please
request NTIS-PR-205/PCW for more infor-
mation.
The weekly newsletter series will keep you
current. But learn what you have missed in
the past by ordering a computer NTISearch
of all the research reports in your area of
interest, dating as far back as 1964, if you
wish. Please request NTIS-PR-186/PCN for
more information.
WRITE: Managing Editor
5285 Port Royal Road
Springfield, VA 22161
Keep Up To Date With SRIM
SRIM (Selected Research in Microfiche)
provides you with regular, automatic distri-
bution of the complete texts of NTIS research
reports only in the subject areas you select.
SRIM covers almost all Government re-
search reports by subject area and/or the
originating Federal or local government
agency. You may subscribe by any category
or subcategory of our WGA (Weekly Govern-
ment Abstracts) or Government Reports
Announcements and Index categories, or to
the reports issued by a particular agency
such as the Department of Defense, Federal
Energy Administration, or Environmental
Protection Agency. Other options that will
give you greater selectivity are available on
request.
The cost of SRIM service is only 450
domestic (60£ foreign) for each complete
microfiched report. Your SRIM service begins
as soon as your order is received and proc-
essed and you will receive biweekly ship-
ments thereafter. If you wish, your service
will be backdated to furnish you microfiche
of reports issued earlier.
Because of contractual arrangements with
several Special Technology Groups, not all
NTIS reports are distributed in the SRIM
program. You will receive a notice in your
microfiche shipments identifying the excep-
tionally priced reports not available through
SRIM.
A deposit account with NTIS is required
before this service can be initiated. If you
have specific questions concerning this serv-
ice, please call (703) 451-1558, or write NTIS,
attention SRIM Product Manager.
This information product distributed by
US. DEPARTMENT OF COMMERCE
National Technical Information Service
5285 Port Royal Road
Springfield, Virginia 22161
-------
EPA-600/3-76-008
March 1976
PB 252 488
Ecological Research Series
I*'
TOXICANT^ TO
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
^*H*T^ff "*
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials. Problems are assessed for their long- and short-term
influences. Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects. This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/3-76-008
2.
4. TITLE AND SUBTITLE
Acute Toxicity of Selected Toxicants to Six
Species of Fish
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
March 1976 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Rick D. Cardwell, Dallas G. Foreman,
Thomas Ri Payne., Doris J. Wilbur
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Chemico Process Plants Company-Envirogenics Systems
9200 East Flair Drive
El Monte, California 91734 ,
10. PROGRAM ELEMENT NO.
1BA021 (ROAP/Task No. 16AAE/05)
11. CONTRACT/KRXHXNO.
68-01-0748
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA-ORD
19. SUPPLEMENTARY NOTES
16. ABSTRACT
The relationship between median lethal concentration and exposure time was determined
for five chemicals and up to six species of freshwater fish in a flow-through system.
The lowest median lethal concentrations found were 0.114 mg/1 for sodium cyanide,
0.118 mg/1 for sodium pentachlorophenate, 2.9 mg/1 for selenium dioxide, 18.0 mg/1
for sodium arsenite, 25.4 mg/1 for beryllium sulfate, and greater than 100 mg/1 for
lead chloride.
Toxicity curves relating median lethal concentration to exposure time were of three
types. One curve, resembling a rectangular hyperbola, characterized the toxicity of
sodium cyanide, while another curve, sigmoid in shape, characterized the toxicity of
selenium dioxide. Both types of curves were observed in toxicity tests with sodium
pentachlorophenate, sodium arsenite and beryllium sulfate. Linear toxicity curves
were recorded for some fish species exposed to selenium dioxide, sodium arsenite
and beryllium sulfate, but these were usually encountered when exposure times were
less than 96 hr.
This report was submitted in fulfillment of Contract Number 68-01-0748 by the Chemico
Process Plants Company-Envirogenics Systems under sponsorship of the Environmental
Protection Agency. Work was completed as of December, 1973.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Bioassay, Fishes, Sodium cyanide,
Fresh water, Metals, Halohydro-
carbons, Water pollution
Acute toxicity, Sodium penta-
chlorophenate, Selenium dioxide,
Sodium arsenite, Beryllium
sulfate, Lead chloride, Blue-
gill, Channel catfish, Brook
trout, Flagfish, Goldfish,
Fathead minnow PRICES SUM
6F
ECT TO CHANGE
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OE_PA
20. SECURITY CLASS (This page)
UNCLASSIFIED
EPA Form 2220-1 (9-73)
ft U. 5. 60rtMUKIIT FIINTMG OFFICE: 1976-657-695/5391 Region No. 5-11
-------
EPA-600/3-76-008
March 1976
ACUTE TOXICITY OF SELECTED
TOXICANTS TO SIX SPECIES OF,FISH
Rick D. Cardwell
Dallas G. Foreman
Thomas R. Payne
Doris J. Wilbur
Chemico Process Plants Company-Envirogenics Systems
El Monte, California 91734
Contract No. 68-01-0748
Project Officer
Charles E. Stephan
Environmental Research Laboratory
Duluth, Minnesota 55804
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL RESEARCH LABORATORY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental
Research Laboratory, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the U.S.
Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation
for use.
11
-------
ABSTRACT
The relationship between median lethal concentration and
exposure time was determined for five chemicals and up to six
species of freshwater fish in a flow-through system. The lowest
median lethal concentrations found were 0. 114 mg/$, for sodium
cyanide, 0. 118 mg/J? for sodium pentachlorophenate, ;2. 9 mg/jfc
for selenium dioxide, 18.0 mg/4 for sodium arsenite, 25.4 mg/^
for beryllium sulfate, and greater than 100 mg /A for lead
chloride.
Toxicity curves relating median lethal concentration to
exposure time were of three types. One curve, resembling a
rectangular hyperbola, characterized the toxic it y of sodium
cyanide, while another curve, sigmoid in shape, characterized
the toxicity of selenium dioxide. Both types of curves were
observed in toxicity tests with sodium pentachlorophenate, sodium
arsenite and beryllium sulfate. Linear toxicity curves were
recorded for some fish species exposed to selenium dioxide,
sodium arsenite and beryllium sulfate, but these were usually
encountered when exposure times were less than 96 hr.
This report was submitted in fulfillment of Contract Number
68-01-0748 by the Chemico Process Plants Company-Envirogenics
Systems under sponsorship of the Environmental Protection Agency.
Work was completed as of December, 1973.
ill
-------
CONTENTS
Page
ABSTRACT ii
LIST OF FIGURES iv
LIST OF TABLES v
ACKNOWLEDGEMENTS vi
Sections
I. Conclusions I
II. Recommendations 2
III. Introduction 4
IV. Materials and Methods 7
V. Results 19
VI. Discussion 45
VII. References 63
VIII. Appendix 71
Preceding page blank
-------
FIGURES
No. Page
Relationship Between Median Lethal 21
Concentration (LC50) of Sodium
Pentachlorophenate and Exposure Time
for Four Species of Freshwater Fish.
Relationship Between Median Lethal 25
Concentration (LC50) of Sodium Cyanide
as CN~ and Exposure Time for Five Species
of Freshwater Fish.
Relationship Between Median Lethal 31
Concentration (LC50) of Selenium Dioxide
and Exposure Time for Six Species of
Freshwater Fish.
Relationship Between Median Lethal 36
Concentration (LC50) of Sodium Arsenite
and Exposure Time for Six Species of
Freshwater Fish.
Changes in the Concentration of Dissolved 39
Beryllium (as BeSC»4) Spiked into 10,6 of
Diluent Water at a Level of 89 mg/X BeSO4-
Relationship Between Median Lethal 43
Concentration (LC50) of Beryllium Sulfate
(BeSO4) and Exposure Time for Three
Species of Freshwater Fish.
vi
-------
TABLES
No. Page
1 Chemicals Used In Acute Toxicity 8
Tests
2 Representative Quality of Source 13
Water
3 Water Duality During Toxicity Tests 20
of Sodium Pentachlorophenate
4 Water Quality During Toxicity Tests 23
of Sodium Cyanide
5 Lowest Levels of pH and Total Alkalinity 27
Encountered in Toxicity Tests of
Selenium Dioxide (SsCO
6 Water Quality During Toxicity Tests 28
of Selenium Dioxide
7 Highest Levels of pH and Total Alkalinity 33
Encountered in Toxicity Tests of Sodium
Arsenite (NaAsCX,)
8 Water Quality During Toxicity Tests 34
of Sodium Arsenite
9 Minimum Levels of pH and Total 40
Alkalinity Observed in Toxicity Tests
of Beryllium Sulfate (BeSCO
10 Water Quality During Toxicity Tests 41
of Beryllium Sulfate
vu
-------
ACKNOWLEDGEMENTS
Mr. Charles E. Stephan provided many helpful suggestions
during the course of the research and critically reviewed the
manuscript. Dr. Keith H. Sweeny also provided valuable advice
on analytical methodology and aided in making the initial chemical
analyses. Mr. R. Stankiewicz provided much technical assistance
in computer programming and analysis. Mr. W. E. -Wright per-
formed the majority of the water quality analyses and contributed
substantially to the monitoring of the toxicity tests.
viii
-------
SECTION I
CONCLUSIONS
1. Sodium pentachlorophenate was the most toxic chemical
studied, producing some mortality at concentrations less than
100 i-ig/l, while lead chloride, even at concentrations well above
the solubility limit ( ~2 mg/ty, was not toxic in the'moderately
hard, somewhat alkaline diluent water. The order of toxicity was
sodium pentachlorophenate, sodium cyanide, selenium dioxide,
sodium arsenite, beryllium sulfate, and lead chloride.
2. The order of sensitivity of the test species varied with
each toxicant. Brook trout and fathead minnow were generally
the most sensitive and goldfish and bluegill the least sensitive.
Channel catfish and flagfish were of intermediate sensitivity.
3. Upon prolonged exposure of the various fish species, two
main types of toxicity curves, hyperbolic and sigmoid, were
observed. The toxicity curve resembling a rectangular hyperbola
was found exclusively in toxicity tests of sodium cyanide, but
infrequently in toxicity tests of the other compounds. It denoted
that acute toxicity ceased at certain toxicant concentrations and
exposure times, and was believed to reflect one basic mode of
toxicant action. The sigmoid toxicity curve predominated in
toxicity tests of all compounds except cyanide and was observed
exclusively in tests of selenium dioxide conducted for 336 hr.
All species of fish exhibited both types of toxicity curves.
4. Because fish exposed to high concentrations of beryllium
produced substantial amounts of mucus, it was postulated that the
specimens exposed to this chemical perished mainly from the
"mucus coagulation syndrome. " Enhanced mucus production was
not seen in fish exposed to lead chloride, selenium dioxide, sodium
arsenite, sodium cyanide or sodium pentachlorophenate at the
concentrations employed.
1
-------
SECTION II
RECOMMENDATIONS
1. The ultimate shapes of the toxicity curves and the mechan-
isms of toxic action of some of the chemicals are in doubt and should
be further1 explored.
2. Several approaches for assessing the mechanism of toxicant
action should be considered. First, the physiological condition of
the test specimens should be established at the onset of toxicant
exposure and monitored throughout the toxicity test. The para-
meters should be selected on the basis of how accurately they assess
the state and functional integrity of the organism and how readily they
will be accepted and used by aquatic toxicologists and water pollution
biologists. Suggested parameters include hematocrit and hemoglobin
for determining the functional state of the blood, specifically the
presence of any anemia; total lipid, or more specifically saturated
and mono-unsaturated fatty acids, for evaluating the fish's capacity
to withstand a 14 day or more period of fasting; and possibly serum
a- amino nitrogen, to determine the state of amino acid and protein
metabolism. Although many variables could be evaluated, in order
td be used widely, each must reflect the broadest range of physio-
logical activity possible without requiring excessive investment of
time and money. The results can be used to compare L.C50 values
from toxicity tests performed with different stocks of a fish species
in order to determine whether variations in condition may have
influenced the results. From the monitoring program, one could
determine whether or not the fish became severely weakened in the
later stages of exposure as a result of the stresses arising from
fasting, handling and confinement.-
3. For defining toxicity curves, toxicity tests should be con-
ducted until median lethal thresholds are reached rather than be
terminated after predetermined times.
-------
4. It would be informative to deter mine whether transient exposures
of fish to these chemicals can cause latent mortality or delayed
physiological dysfunction. Furthermore, the relationship between
toxicant concentration and exposure time and the severity of the
delayed functional effects should be explored.
5. The relative amounts of hydrolysis or chemical degradation
products arising from introduction of the parent compound into the
diluent water should be determined and correlated with relative uptake
by and distribution in the organisms. Median lethal concentrations
should be based on the form of the chemical found to be exerting the
predominant toxicity. Chemical studies should be incorporated into
the toxicity testing experimental design to resolve these questions.
-------
SECTION III
INTRODUCTION
Water quality criteria for the protection of biotic communities
in aquatic environments are 'based primarily on bioassays or toxicity
tests, which determine, as a first order approximation, the effect
of a given chemical or substance on a given population of organisms.
Acute toxicity tests are the first step in the experimental protocol
for determining the highest concentration of a chemical or chemicals
which will not affect long-term survival, growth or reproduction
of a species, termed the Maximum Acceptable Toxicant Concentration
or MATC (Mount and Stephaa ). As research has accumulated on the
acute effects of chemicals on aquatic organisms, it has become appar-
ent that there is a need for standardizing the experimental conditions,
for characterizing and defining the levels and states of the toxicant,
the water quality variables, and the conditions of the test fish, and for
2 23
providing more refined analysis of the data (Sprague ). As Sprague '
has noted, a thorough search of the literature or of compendiums such
4
as McKee a.id Wolf will often show that the toxicity of a particular
compound to various aquatic organisms varies up to a thousand-fold.
To partially ameliorate the problem, committees have formulated a
series of standard testing procedures (Committee on Methods for
te
7
5 6
Toxicity Tests with Aquatic Organisms ' ), to improve on the note-
worthy standard procedures developed earlier by Doudoroff et al.
and adopted as the standard bioassay method in the United States by
o
the American Public Health Association .
Many of the experimental protocols and conditions suggested in
the revised procedures have been incorporated into these studies
of the acute effects of sodium pentachlorophenate, sodium cyanide,
selenium dioxide, sodium arsenite, beryllium sulfate and lead
chloride on six species of freshwater fish: fathead minnow
(Pimephales promelas Rafinesque), goldfish [ Carassius auratus
4
-------
(Linnaeus) ], flagfish ( Jordanella floridae Goode and Bean), blue-
gill (L-epomis macrpchirus Rafinesque), channel catfish [ictalurus
gunctatus (Rafinesque) 1 and brook trout [ Salvelinus fontinalis
(Mitehill) ]. Toxicity tests of these compounds have been conducted
previously with at least one of the test species, but with at least
two of the chemicals studied, beryllium and selenium, little was
known of their effects on fish. Pomellee" added a beryllium
sulfate-tartaric acid complex to aquaria containing goldfish, "minnows"
and snails for 12 days to give a total beryllium ion (Be ) concentra-
tion of 28. 5 mg/Xand did not observe any identifiable toxic responses
by the test animals. During the course of the present study, Slonim
and Slonim and Slonim published two papers detailing principally
the interaction of water hardness and beryllium sulfate toxicity
to guppies (Poecilia reticulata Peters). Ninety-six hour median
______ __________
lethal concentrations (LC50) ranged from 0. 16 to 27.0 mg/£ Be
depending on hardness. According to the Ohio River Sanitation
1 O /I
Commission , cited by McKee and Wolf , 10 to 100 mg/Ji concentra-
tions of sodium selenite were "toxic" (nature of the toxic response
undefined) to goldfish in 98 to 144 hr and 8 to 19. 5 hr, respectively,
in hard water. Weir and Hine calculated an LC50 of 12. 0 mg/i
elemental selenium (Se) from goldfish mortalities arising from a
combination of 48 hr toxicant exposure and 168 hr (216 hr total)
confinement in uncontaminated, reconstituted deionized water.
A great deal of toxicological information is available con-
cerning the effects of arsenic, cyanide and pentachlorophenate com-
pounds on aquatic organisms. Upon examination of the compendium
14
of Becker and Thatcher , which summarizes many of the toxic
limits that have been established for these compounds, the difference
between the lowest and the highest LC50 values for fish exposed
to cyanide was eighteen-fold, but only four-fold for invertebrates.
Concentration of toxicant lethal to 50% of the test specimens.
-------
For fish exposed to sodium arsenite, the highest 48-hr LC50
estimate was 5. 5-fold greater than the lowest LC50 recorded.
For invertebrates in general, the magnitude of the difference
between 48-hr LC50 estimates was 44-fold. Although less directly
comparable data were available for pentachlorophenol, the "no
effect" levels were uniformly less than 1 mg/& for fish, except
for one observation that 5 mg/X sodium pentachlorophenate was
non-toxic to the green sunfish (Lepomis cyanellus Rafinesque).
Although it is apparent that toxicological information is available,
it is necessary to re-evaluate the toxicity relationships in terms
of the recently revised aquatic toxicological methods. Accordingly,
toxicity tests were conducted using intermittent-flow proportional
diluters, with water of known quality, and fish of known age and size.
Toxicant concentrations were measured several times during the
course of each test to determine the levels of chemical to which
the fish were actually being exposed.
Another objective of the tests was to determine the relation-
ship between concentrations found to kill 50% of the test specimens,
termed the median lethal concentration or LC50, and various exposure
times. Representated as toxicity curves, these relationships provide
valuable information on the type of toxic action, serve to identify
if and when the acute toxicity ceases (known as a median lethal
jj5 o i 5
threshold ' }, and permit interpolation of LC50 values for various
exposure periods for comparison with other estimates.
Median lethal concentrations derived from these tests can
be used in the design of long-term chronic toxicity tests and in cal-
culation of the "application factor" (Mount and Stephan ). The
application factor, derived by dividing the MATC by the LG50 for
48 to 336 hr exposure periods, is used to designate probable maximum
acceptable toxicant concentrations for species amenable only to acute
toxicity testing.
Concentration at which acute toxicity ceases to 50% of the test specimens.
-------
SECTION IV
MATERIALS AND METHODS
Acute toxicity tests of selenium dioxide (SeO,), sodium
arsenite (NaAsO^), sodium cyanide (NaCN), sodium pentachlorophenate
(NaPCP), beryllium sulfate (BeSO4), and lead chloride (PbCl2) were
conducted with up to six species of freshwater fish. The chemical
formulae for these compounds, some of their characteristics, and
the commercial suppliers are given in Table 1.
TEST FISH
The species utilized were bluegill, channel catfish, fathead
minnow, brook trout, flagfish, and Ozark-strain goldfish. All fish were
obtained from commercial dealers within the State of California except
minnows and flagfish, which were produced in the laboratory's culture
unit, brook trout, which were obtained from the California Department
of Fish and Game, and goldfish.
The test specimens were juvenile fish except brook trout, which
were adults1. In some experiments, fathead minnows and flagfish less
than 24-hr old at the onset of testing were used. The size, age and
density of the fish in the test chambers in each toxicity test are given
in Appendix Tables 1-5.
Fish obtained from commercial dealers were quarantined for
control and elimination of pathogens and acclimated to laboratory con-
ditions for at least one month subsequent to their arrival. With the
exception of the disease-free stock of fathead minnows and flagfish
produced by the culture unit, all species apparently harbored aeromonad
bacteria since low grade infections of bacterial hemorrhagic septicemia
were occasionally observed within a week after the fish's arrival or as
-------
Table 1 . CHEMICALS USED IN ACUTE TOXICITY TESTS
00
Chemical
common
name
Sodium
arsenite
Beryllium
sulfate
Lead
chloride
Selenium
dioxide
Sodium
cyanide
Pentachloro-
phenol
Source
J. T. Baker
Research Organic/
Inorganic Chem.
Corp.
J. T. Baker
Research Organic/
Inorganic Chem.
Corp.
J. T. Baker
Aldrich
Chemical Co.
Purity Formula
A.Rf NaAsO2
99. 5% BeSO.. 4H_O
T: £*
A. R. Pbd2
99. 9% Se02
A. R NaCN
99 + % C,C1,OH
(Gold
Label)
Water
solubility,
g/100 ixui
Soluble
43. 7830°C
0. 9920°C
38. 4U°C
48.010°C
Insoluble0
Probable
major
hydrolysis
product
H3AsO3
(BeOH)n + nH+
n
Pb(C03)2 • Pb(OH)2
HSeO 3
HCN
C6ci5o-
Analytical reagent
bSolubility at least 100 g/100 mjtdistilled water
cSoluble as the sodium salt.
-------
the water temperatures were being increased to the acclimation
temperature of 25°C (15° C for brook trout). The disease was treated
by adding sufficient oxytetracycline - HC1 ("TM-50'1 Pfizer) to the
food to gi.ve a concentration of 0. 44% active ingredient (A. I. ), or
75 mg A.I. /kg fish/day. Since immune fish can act as carriers and
transmit the disease to susceptible individuals at a later date (Snieszko
and Bullock ), oxytetracycline treatment was continued for several
weeks after evidence of disease had abated. According to Patterson ,
rainbow trbut (Salmo gairdneri Richardson) force-fed sufficient "TM-50"
in tablets to give doses over 1, 000 mg/kg fish did not incur mortality
or external pathology in a ten day post-treatment period. Depuration
of TM-50" from renal and somatic tissues of rainbow trout, and brown
trout (Salmo trutta Linnaeus) was usually complete in 14 days and from
18
hepatic tissues within 28 days. Herman also concluded that rainbow
trout could be fed more than 1, 000 mg oxytetracycline/kg in a single
dose without apparent ill effects. In reality, fish will not usually con-
sume TM-50-treated food dosed above approximately 75 mg A.I. /kg
fish/day. Because the aeromonad bacteria are free-living facultative
pathogens which usually cause pathogenesis in fish when they have been
stressed or held under unsanitary conditions (also a stresser), the hold-
ing tanks were treated weekly with 2 mg/^ "Roccal" (National Laboratories,
Mont vale, New Jersey) to reduce bacterial buildup.
Other diseases positively identified and for which therapy was in-
stituted included bacterial gill disease and external protozoan infections.
Upon arrival at the laboratory, brook trout had bacterial gill disease,
which was eliminated over a period of two weeks with daily, one hour
baths of 10 mg A.I. /i soluble oxytetracycline - HC1. The two-hr LC50
for the soluble form of oxytetracycline HC1 is greater than 500 ppm
its 19
(Herman ). Upon arrival, channel catfish carried the protozoans,
Trichodina sp. and Ichthyopthirius multifilus, and the external
-------
monogenetic trematode, Gyrodactylus sp. All parasites were con-
trolled over a three week period with daily one-hour baths of 50 to
100 mg/£ formalin (17 to 33 mg/£formaldehyde) accompanied with
artificial aeration. The maximum formalin concentration of 100 mg/j£
was selected from results of a static acute toxicity test conducted with
the newly-arrived fish. The tanks were also artificially aerated
during formalin treatment since it was observed that the 35 to 50 mm
fish became stressed when treated without aeration. Maintenance of
high dissolved oxygen concentrations during static formalin treatments
is important since formaldehyde lowers the oxygen uptake of fish
(Wedemeyer ).
During acclimation to a standard photoperiod of 12-hr light,
12-hr dark, appropriate water temperatures, and laboratory conditions
in general, brook trout, goldfish, bluegill and channel catfish were
fed twice daily with a dry pelleted ration (Moore-Clark Company,
Salt Lake City, Utah) at a rate of 2% of their body weight/day (trout)
or ad libitum (goldfish, bluegill and catfish). Juvenile fathead minnows
and flagfish were fed twice daily a combination of brine shrimp nauplii
(Artemia salina) and dry trout starter (Moore-Clark).
TOXICITY TESTING CONDITIONS
All test species were acclimated to the above conditions for
at least one month prior to introduction into the test chambers. For
specific acclimation to toxicity test conditions, fish were introduced
into the chambers 72 hr prior to toxicant introduction. They were
fasted during this period as well as during toxicant exposure, which
lasted up to 336 hr (408 hr total). The order of introduction of test
specimens and the position of the test containers were selected with
random number tables. Ten specimens were tested in each container
except for brook trout, where the large size of the fish necessitated
use of five individuals per container.
10
-------
All toxic ity tests were conducted with an intermittent-flow test
system consisting of a two-liter proportional diluter (Mount and
21
Brungs ) and twelve glass test chambers (30. 5 x 30. 5 x 30. 5 cm)
containing ZOA of test solution in the toxicity tests of adult trout and
16.5.6 in tests of the other species. Light intensity from fluorescent
lamps (Sylvania "Gro-Lux" and Durotest "Optima") averaged 1019 lux
(94 fc). Toxicity tests were conducted with two replicates, each
including one control and five toxicant concentrations. For all
toxicants except sodium cyanide, the diluter delivered sufficient water
to replace six tank volumes per day, assuring 90% molecular displace-
ment in nine hours. Because test concentrations of sodium cyanide
were observed to undergo a temporal decline, the rate of toxicant
introduction was doubled to compensate for the loss. Although water
flow into the tanks maintained dissolved oxygen concentrations above
70% of air saturation in most toxicity tests, it was not adequate in
those utilizing brook .trout, necessitating artificial aeration. Toxicant
concentrations were successively diluted by a factor of 0. 75.
As mortalities occurred, and at the end of the toxicity test, fish
were measured for total length to the nearest millimeter and for wet
body weight to the nearest gram or milligram, depending on size,
after excess moisture had been removed with toweling. All fish were
measured from each tank of one of the replicates. The data were later
pooled if visual examination of the data indicated no size differences
between treatments. Length and weight measurements were not taken
prior to toxicity testing since it was believed that the stresses associated
with handling and anesthesia (Houston, et al. ' Wedemeyer )
would be far more significant in terms of their influence on the LC50
than the changes in body weights during the course of the test.
11
-------
Water quality in the test chambers was measured 24 hr prior to,
and at least twice during toxicant addition for comparison of the effects
of the toxicant's presence on water quality, for checking the water's
suitability for uncompromised fish survival, and for estimating water
quality variation. Seven variables were determined using standard methods
recommended by the Environmental Protection Agency or the. American
Q
Public Health Association . Except for rare instances, measurements
were made on samples collected less than 4 to 6 hr earlier. Water
temperatures were monitored with thermistors connected to a 12-channel
recorder (Honeywell, Inc. , Philadelphia, Pa.). Dissolved oxygen
(D. O. ) concentrations were determined with the azide modification of
the Winkler method or with an oxygen meter (Model 54, Yellow Springs
Instrument Company, Yellow Springs, Ohio), pre-calibrated with
the Winkler method. In toxicity tests of sodium arsenite, dissolved
oxygen was measured with the meter because the toxicant interfered with
the Winkler method. Percentage saturation of the water was estimated
with a nomograph using the known water temperatures and D. O. concen-
trations. A glass electrode was used to determine pH, while total
alkalinity was determined by electrometric titration with 0. 02 N reagent
sulfuric acid to a pH of 4. 5. Acidity was also determined electrometri-
cally by titration with carbon dioxide-free 0. 05 N sodium hydroxide to a
pH of 8. 3. Measurement of total hardness was derived by titration with
disodium ethylenediaminetetraacetic acid using Eriochrome Black T
indicator. Specific conductance was measured with a conductivity bridge
(Beckman Instruments, Inc., Cedar Grove, New Jersey) using a cell
with a constant of 1. 0.
A number of ions were also determined by a commercial laboratory
to gain a more specific description of the water's composition (Table 2).
Calcium, magnesium, potassium, sodium, chloride and sulfate ions
were determined every four months over the preceding year, while
ammonia was measured biannually. Other substances were determined
once.
12
-------
Table 2. REPRESENTATIVE QUALITY OF SOURCE WATER
Variable
Calcium
Magnesium
Potassium
Sodium
Chloride
Sulfate
Sulfide
Nitrate
Nitrite
Unit
mg/t
mg/t
mg/t
mg/4
mg/t
mg/.e
mg/t
mg/Jt
mg/l
Concen-
tration
31.1
13. 1
2.0
15.4
11.3
8.6
< 0.002
4.65
0.005
Variable
Ammonia
Phenol
Fluoride
Cyanide
Iron
Copper
Zinc
Cadmium
Chromium
Unit
mg/t
NH3-N
mg/t,
mg/t
mg/4
mg/A
mg/t
mg/t
mg/t
mg/t
Concen-
tration
0. 16
0.001
0.96
0. 0005
0.001
0.005
0.001
0.010
0.025
13
-------
TOXICANT ANALYSIS
During each toxicity test, samples of water were collected for
toxicant analysis an average of three different times to determine the
levels of chemical to which the fish were exposed and to define variation-in
the toxicant concentrations during the course of the toxicity test. Water
samples were collected at mid-depth in the container, filtered through
Whatman No. 1 paper and stored at approximately 5 C until analysis.
Prerequisite to these determinations, analyses were made to describe
the accuracy and reproducibility of the methods applied.
All metals, namely arsenic, beryllium, lead and selenium,
were assayed on an atomic absorption spectrophotometer (Model 303,
Perkin-Elmer Corp. ) using sample preparation and analytical methods
25 26
specified by EPA or the Perkin-Elmer Corporation . To verify
the accuracy and reproducibility of the methods, known amounts of the
metals were spiked into laboratory water which had been filtered through
Whatman No. 1 paper. The results showed the methods employed
were highly reproducible and that samples spiked into water taken from
tanks containing control fish had concentrations in agreement with stan-
dard curves. The coefficients of variation for selenium, arsenic, beryl-
lium, and lead were 1. 1, 5.3, 0.6, and 1. 1%, respectively (Appendix
Tables 6, 7, 8 and 9).
Sodium cyanide was measured colorimetrically using a
colorimeter (Spectronic 20, Bausch and Lomb) and the pyridine-
pyrazalone method (EPA ). The analytical method was checked for
reproducibility and accuracy using the same protocol as described for
the metals. The coefficient of variation, 3. 1%, indicated low error of
measurement (Appendix Table 10).
14
-------
Insoluble pentachlorophenol was converted to water soluble
sodium pentachlorophenate by dissolving the former in one molar
sodium hydroxide for toxicity testing. Concentrations of sodium
pentachlorophenate were subsequently analyzed by gas-liquid chroma-
27
tography using the acetylation procedure of Rudling . The repro-
ducibility and accuracy of the method was checked with the afore-
mentioned protocol and showed a low coefficient of variation of 5. 7%
for five replicates spiked into filtered water taken from tanks containing
control fish (Appendix Table 11).
Stock toxicant solutions were prepared in deionized water as
needed and dispensed from Marietta bottles using a funnel dosing
apparatus developed by Mount and Brungs . Usually 2 mS, of stock
solution were dispensed per cycle of the proportional diluter, although
larger volumes were required for administering lead chloride, which
had a solubility in distilled water of approximately 9. 9 mg/^at 20° C
28
(Weast and Selby ). Stock solutions of sodium cyanide were prepared
daily because it was found that concentrations declined in the stock
solutions, presumedly due to volatilization of hydrocyanic acid formed
by hydrolysis (CN~ t H-Orz^HCN + OH"). Inthe tests with cyanide,
it was also necessary to augment the flow of water through the test
chambers, decreasing the 90% molecular displacement time to approxi-
mately five hours, to compensate for cyanide loss through volatilization
of HCN.
Concentrations of the chemicals measured in the chambers during
the course of the toxicity tests are given in Appendix Tables 12 through
16 for NaPCP, NaCN, SeO2> NaAsO2 and BeSO4.
15
-------
STATISTICAL ANALYSIS
Simple summary statistics were utilized to analyze length-
weight, water quality, and toxicant concentration data. Except were
noted, results are represented as means + one standard deviation.
Concentration-percent mortality data generated from the toxicity
tests were analyzed with logarithmic-probability (log-probit) methods
29
using either the manual procedure of Litchfield and Wilcoxon or the
computer program of Dixon . The log-probit method was selected
because it is a more objective approach than the graphical interpolation
method, offers a test of the regression line's goodness-of-fit, and
provides the statistics necessary for calculating 95% confidence limits
for median lethal concentrations (LC50) and for comparing differences
between two LC50 values. The LC50 is the concentration of toxicant
killing 50% of the test specimens in a specified period of time. In
most cases, concentration-percent mortality data from toxic ity tests
which were conducted in duplicate were pooled for statistical analysis.
Data were not pooled from tests of sodium cyanide using brook trout
since the tests were performed at different times under slightly dif-
ferent water quality conditions and toxicant concentrations.
For homogeneous data, upper and lower 95% confidence limits
for LC50 values determined from the computer program were calculated
as (LC50) (f) and LC50/f, respectively, where f = antilogarithm of the
1. 96 a (N'/2)~*, a is the standard deviation of the logarithm of the
population tolerance frequency distribution, and N' is the number of
test animals expected to have perished within the percent mortality
29
interval of 16 to 84% (Litchfield and Witoxon ). An equivalent equa-
tion is + 1.96 S. E. log LC50 (Bliss ). For heterogeneous data, i.e.,
where Chi-square analysis of the fitted line indicated lack of goodness-
of-fit, the equation f = (Student's t-value) ( a ) ( x3 /n ) was employed
(Bliss ). The logarithms of the median lethal concentrations were
plotted against the logarithms of the exposure times to give a toxic ity
curve (Sprague ). '
16
-------
The accuracy of LC50 estimates and their 95% confidence limits
29
generated by the Litchfield and Wilcoxon and computer program
30
(Dixon ) methods were compared with similar calculations made by
eight other aquatic toxicology laboratories using standard concentration-
percent response data (Appendix Table 17) supplied by the Committee
on Methods for Toxicity Tests with Aquatic Organisms. Our L.C50
estimates were in agreement with the average LiCSO computed by the
other laboratories (Appendix Table 18) using both methods of analysis.
In our laboratory, 95% confidence limits were usually narrower with
29
the computer method than with that of Litchfield and Wilcoxon
Median lethal times for measured toxicant concentrations were
calculated in some cases. Data were analyzed in the same manner as
for the calculation of LC50 values and were plotted on the same toxicity
curve, with the exception that 95% confidence limits were determined for
the independent variable, time, rather than the LC50.
Control mortality occurred in less than five percent of the toxicity
tests and was less than ten percent in all cases. Median lethal concen-
trations were corrected for control mortality, where applicable, using
Dixon's computer program.
In this program, it was considered more important to define the
relationship between median lethal toxicant concentration and exposure
time, i. e. , define the shape of the toxicity curve, than it was to determine
JLC50 values for specified exposure times (e.g. , 24, 48, 96 hr). Since
many workers in aquatic toxicology base their toxicity data on pre-
determined exposure times, such values were estimated using simple
linear regression (Steel and Torrie ) where applicable.
17
-------
All statistics were initially calculated for the undissociated
or parent compound used in the toxicity tests. Although it was known
that several of the parent chemicals were capable of hydrolysis or
other chemical reactions upon their introduction into the diluent water
(Table 1), determination of the exact forms of the chemicals, par-
ticularly those which were biologically significant in terms of their
relative toxicity, was beyond the scope of this project. However,
33 34
as Lee and Black, et al have recently emphasized, such know-
ledge is essential for assessing the toxicity of chemicals to aquatic
organisms. Since it is well known (Doudoroff, Leduc and Schneider )
that within the pH range of 6 to 8 cyanide toxicity is due mainly to
hydrocyanic acid (HCN), which was measured in the pyridine-pyrazalone
method along with cyanide ion as total cyanide, LC50 estimates and 95%
confidence limits for this chemical only were also calculated as cyanide
ion (CN").
18
-------
SECTION V
RESULTS
TOXICITY TESTS OF SODIUM PENTACHLOROPHENATE
Water Quality
The sodium salt of pentachlorophenol was soluble at all levels
employed for toxicity testing and did not change any of the seven
water quality variables monitored. In general, mean dissolved oxygen
concentrations ranged from 6. 1 _+ 0. 5 to 7. 4 jj- 0. 3 mg/A and were
greater than 70% of air saturation in all toxicity tests. Mean pH ranged
from 7. 59 _+ 0. 08 to 7. 94 j- 0. 16; total alkalinity, from 153 H- 4 mg/X
to 167 jf 2 mg/4 CaCO3; acidity, from 5. 9 jf 0. 9 to 7. 5 _t 6. 0 mg/4
CaCO3; total hardness, from 145 jh 6 to 156 _+ 11 mg/jfc CaCO,;
and specific conductance, from 367 -f 14 to 379 +_ 21 |J mhos/cm
(Table 3).
Toxicity
Sodium pentachlorophenate (NaPCP) produced total mortality
of all fish at concentrations greater than 1. 0 mg/<6. The descending
order of species sensitivity to NaPCP was brook trout, fathead
minnow, goldfish and bluegill.
Toxicity curves relating median lethal concentration of sodium
pentachlorophenate to exposure time for the four species indicated
that the halogenated phenol had a fairly narrow range of acute lethality
and had a toxic action which varied with respect to species (Figure 1).
Two types of toxicity curves were obtained: for brook trout, acute
lethality ceased at a NaPCP concentration of 0. 118 mg/jgand an
exposure time of 220 hr; but for fathead minnow, goldfish and bluegill,
mortalities did not cease at test concentrations as low as 0. 08 mg/l
and toxicant exposure times as long as 406 hr. The changes in the slopes
of the toxicity curves for fathead minnows, goldfish and bluegill occurred
19
-------
Table 3 . WATER QUALITY DURING TOXICITY TESTS
OF SODIUM PENTACHLOROPHENATE
to
o
Dissolved oxygen
Species
Fathead b
minnow
Blue gill
Brook b
trout
Goldfish
Water
temperature,
C mg/J&
24. 9a
+0.3
24.9
+0.3
15.4
+0.6
25.1
+0.2
7.4
+0.3
6.7
+0.3
8.4
+0.9
6.1
+0.5
>
% pH
saturation
88. 1
+3. 1
80.2
+3.2
83.1
+9.4
72.8
+6.6
7.83
+0.05
7.94
+0. 16
7.89
+0.06
7.59
+0.08
Total Total
alkalinity, hardness
mg/jfc Ca.CO-3
162
+ 4
153
+ 4
161
+ 4
167
+ 2
156
+11
145
+ 6
147
+ 8
148
+ 4
Specific
' conductance,
(jnnhos/cm
379
+21
367
+14
377
+13
367
+ 8
aMeans ^ 1 standard deviation are given
bAcidity averaged 5. 9 +0.9, 7.5 + 6.0, and 6. 4 + 2.0 mg/1 CaCOj in toxicity tests
using fathead minnow, "bluegill ancTbrook trout, respectively.
-------
1.0
w
H
H
?
k
O
rt
O
J
a
u
<;
H
2
W
8
w
b
O
o
m
U
0.6
0.4
0.1
1
1 T
T
QBROOK TROUT
^FATHEAD MINNOW
GOLDFISH
BLUEGILL
100
200
2 6 10 20 60
EXPOSURE TIME, hr
Fig. 1. Relationship between median lethal concentration (LC50) of
sodium pentachlorophenate and exposure time for four species
of freshwater fish.
600
-------
after approximately 96 hr, suggesting a bimodal type of toxicity
since fish died at a faster rate upon prolonged exposure to low con-
centrations than at lesser exposures to intermediate concentrations.
Changes in slope of the toxicity curves for very short exposures to
high concentrations basically reflect the time required for a specimen
to expire.
The range at which sodium pentachlorophenate was acutely
lethal was much narrower than that characterizing the four metals
tested. For brook trout, the 24-hr LC50 was 0. 315 mg/A, while
the median lethal threshold(219-hr LC50) was 0. 118 mg/Ji. The
difference between the 24 and 219-hr median lethal concentrations
was only 63% (Appendix Table 19). A similar relationship existed
for the three other species, where the differences between 24 and
336 hr LC50 estimates were approximately 67, 49 and 41% for fat-
head minnow, goldfish and bluegill, respectively. The differences
in sensitivity between minnow, goldfish and bluegill to sodium penta-
chlorophenate were slight, as indicated by the fact that the 336-hr
LC50 estimates were 0. 153 mg/Jiior fathead minnows, 0. 189 mg/Ji
for goldfish, and 0. 215 mg/Xfor bluegill (Appendix Tables 20 to
22 and Figure 1).
TOXICITY TESTS OF SODIUM CYANIDE
Water Quality
Sodium cyanide was soluble in all proportions used for toxicity
testing and did not alter the water quality variables measured.
Dissolved oxygen concentrations were above 78% of air saturation
in all tests, ranging from 6. 6 4- 0. 2 to 8. 5 j+ 0. 3 mg/Ji. The diluent
water was moderately alkaline, as shown by pH readings of
7. 64 j- 0. 08 to 7. 85 jf 0. 10 mg/£and total alkalinities of 162 J- 3
to 170 jh 6 mg/1 CaCO,, and of intermediate hardness (148 J- 2
to 155 _+ 10 mg/Ji CaCO-j) and conductivity (376 jh 28 to 391 +_15 Mmhos/
cm) (Table 4).
22
-------
Table 4 . WATER QUALITY DURING TOXICITY
TESTS OF SODIUM CYANIDE
to
CO
Water Dissolved oxygen,
Species
Fathead
minnow
Goldfish
Brook
trout
Bluegill
Channel
catfish
tempera-
ture,
o
C
25. 3a
+0.2
25.0
±0. 1
15.4
+0.5
25.4
+0.2
25.2
+ 0.2
mg/4
6.9
±0.1
6.6
+0.2
8.5
+0.3
6.7
+0.2
7.2
+0.3
%
satura-
tion
82.5
+1.7
78.4
+1.8
83.7
+2.2
.80.8
I 2.1
86.4
+3.6
PH
7.64
+0. 08
7.74
+0.08
7.83
+0.03
,7.68
-0. 13
7.85
+0. 10
Total
alkalinity,
mg//CaCO3
169
± 6
168
+ 5
170
+ 5
VI
162
+ 3
Total
hardness,
mg/j£CaCO3
154
+13
152
+ 4
155
+10
155
±16
148
+ 2
Specific
conductance,
(amhos/cm
391
+15
382
+11
376
+28
384
±14
384
+ 3
Means + 1 standard deviation are given.
-------
Toxicity
Sodium cyanide was the second most toxic chemical to the fish
species studied, being almost as acutely lethal as sodium pentachloro-
phenate. The order of species sensitivity was somewhat different than
that generally found for the other chemicals, notably in the substantial
sensitivity of bluegill. Fathead minnows were the most sensitive,
followed by bluegill, brook trout, channel catfish and goldfish.
For all species, sodium cyanide toxicity ostensibly ceased after
approximately 100 hr toxicant exposure, indicating a rapid type of toxi-
cant action. In one of the brook trout toxicity tests, but not in the other,
there was a slight shift in the slope of the toxicity curve after 200 hr.
The significance of the shift is difficult to assess owing to the proximity
of the last three LC50 estimates (Figure 2). A true median lethal thres-
hold was approached but not reached in the toxicity test of goldfish.
The lowest concentration of sodium cyanide found to produce
acute lethality was in the toxicity test using fathead minnows, where a
median lethal threshold of 0. 114 mg/j£ was obtained in 192 hr (Appendix
Table 23). Median lethal thresholds of 0. 116 mg/£ and 0. 126 mg/£ CN~
were obtained for bluegill and brook trout exposed for 168 and 288 hr,
respectively (Appendix Tables 24 and 25). Toxicant exposures were not
sufficiently long to facilitate estimation of a median lethal threshold for
channel catfish, although one .was being approached in the region of 0. 16 mg/£
CN after 30 hr (Appendix Table 26). A similar pattern was evident for
the toxicity test using goldfish, where exposure for 336 hr to sodium cyanide
concentrations as low as 0. 156 mg/jj CN. did not result in definition of
the. median lethal threshold. The 336-hr LC50 was 0. 261 mg/£ CN~
(Appendix Table 27).
24
-------
2.0
to
01
w
Q
O
g
0
3
o
w
o
m
O
1.0
0.6
0.1
I I I
A FATHEAD MINNOW
§3 BLUEGILL
O BROOK TROUT
A CHANNEL CATFISH
O GOLDFISH
10 20 60
EXPOSURE TIME, hr
100
200
600
Fig. 2. Relationship between median lethal concentration (LC50) of sodium cyanide
as GN~ and exposure time for five species of freshwater fish.
-------
As was also observed in the toxicity tests of sodium pentachloro-
phenate, the range within which sodium cyanide produced acute
mortality was very narrow. For the most sensitive species, the
fathead minnow, the difference between the 24-hr L.C50 estimate and
the median lethal threshold was only 20%. Although goldfish were
much less sensitive than minnows, the difference was almost equivalent
to that for minnows, 21%.
TOXICITY TESTS OF SELENIUM DIOXIDE
Water Quality_
Selenium dioxide was soluble in water at all concentrations
tested. The presence of the chemical, however, was found to affect
water quality, reducing pH and total alkalinity while increasing acidity,
probably because of conversion of selenium dioxide to biselenite ion
and selenous acid (SeO2 f H^Q ^ H + HSeO,"). The greatest
alteration in water quality occurred in one test using goldfish, where
pH and total alkalinity were reduced to 6. 58 and 97 mg/X CaCO,
respectively, at a measured concentration of 114 mg/4 SeO7 (Table 5).
£»
In all cases, the changes in water quality were greater at higher
toxicant concentrations. Water quality for the tests of each species
is given in Table 6. Because of the toxicant's influence on pH, alkalinity,
and acidity, means and standard deviations for these three variables
were derived from measurements made in chambers containing control
fish, while summary statistics for the other variables, i. e., dissolved
oxygen,total hardness, and specific conductance, were derived from
measurements made in all treatments, including controls.
26
-------
Table 5, LOWEST LEVELS OF pH AND TOTAL ALKALINITY
ENCOUNTERED IN TOXICITY TESTS OF SELENIUM DIOXIDE (SeO2)
Species
Measured
concentration
of SeO-,
mg/jf
Minimum level observed
PH
Total
alkalinity,
mg/JL CaCO3
Fathead minnow 54
Channel catfish 63
Brook trout 102
Flagfish 38
Goldfish 114
Bluegill 133
. 7.43
7.53
7.45
7.57
6.58
6.80
143
120
112
149
97
110
27
-------
Table 6. WATER QUALITY DURING TOXICITY TESTS
OF SELENIUM DIOXIDE
to
00
Species
Fathead
minnow
b
Flagfish
Blue gill
.b
Channel
catfish
Goldfish
Brook
trout
Water
temperature
°C
24. 7a
4-0.4
24.5
+0. 4
24.9
+0.4
24.9
+0.3
25.4
+0. 3
15.5
+0.6
Dissolved oxygen,
mg/4
7. 1
+0.6
7.8
+0.3
6.8
+0.4
6.7
+0. 9
6.8
+0.5
8.1
+;0.4
%
saturation
83.9
±7.8
92.6
+3.6
81.0
+;5.2
80.3
+10.6
81.3
+5.4
79.9
+ 3.8
PH
7.80
+0. 09
7. 90
+0.04
7.75
+0.05
7. 93
+0. 09
7.63
+0. 10
7.80
+0.07
Total
alkalinity,
mg/£
167
+ 4
164
+ 4
169
± 4
140
± 3
159
+ 5
157
+11
Total
hardness,
CaCO3
151
+ 9
152
13
150
± 6
140
± 3
148
+ 8
148
+10
Specific
conductance
(jmhos/cm
351
•+16
374
± 5
357
+23
u
375
+18
383
+20
Means +_ 1 standard deviation are given.
3Acidityaveraged 3. 7 _+ 2. 4, 4. 6 _+ 0. 9, and 7.5+; 3, 1 mg/4 Ca CO, in toxicity tests of
flagfish, catfish and trout, respectively.
-------
As in the tests of the other compounds, dissolved oxygen concen-
trations were greater than 70% of air saturation, and there was little
variation in pH or hardness of the diluent water entering the system.
Average pH readings from one test to the next ranged from 7. 63 + 0. 1
to 7. 93 +_ 0. 9, while total alkalinity varied from 140 +_ 3 mg/jfc CaCO,
in the toxicity test of channel catfish to 169 +_ 4 mg/£ CaCO, in that of
bluegill. Acidity was less than 10 mg/£ CaCO_. Variation in mean total
hardness and specific conductivities in all tests was less than ten percent
(Table 6).
External Pathology Produced in Fish
V
Exposure to acutely toxic concentrations of selenium dioxide
produced exteinal hemorrhaging or lesions or both within 48 hr of
the fish's exposure to the toxicant in four of the species tested:
brook trout, channel catfish, goldfish and fathead minnow. Similar
signs may have occurred in flagfish juveniles, but they would have been
obscure because of the small size of the specimens. The incidence of
cutaneous lesions appeared to increase at higher concentrations of toxicant.
Control fish did not exhibit external evidence of pathology or behavior which
appeared to deviate from the state judged to be normal. Channel catfish,
goldfish, and fathead minnows had hemorrhages only, whereas brook
trout had marked discoloration and lesions throughout the body, and
necrosis and sloughing of epithelial tissue in the cranial region, par-
ticularly in the snout. Tissue destruction was far more severe and
dramatic in brook trout. Although it is possible that a disease organism
was responsible for these observations, the rapidity of the onset of
pathological signs, the graded nature of the response, and the occurrence
of the pathology in fathead minnows, a species which had no previous
history of disease in our laboratory, contradicts a microbial etiology,
although the possibility of one cannot be excluded.
29
-------
Toxic ity
Selenium dioxide proved to be less toxic than sodium pentachloro-
phenate and sodium cyanide, but more toxic than sodium arsenite,
beryllium sulfate and lead chloride. The order of descending sen-
sitivity was fathead minnow, brook trout, channel catfish, flagfish,
goldfish and bluegill.
No acute median lethal thresholds were manifested for any of
the test species, even for those exposed for 336 hr. In fact, substantial
mortality resulted when fish had been exposed for more than 96 to
120 hr to low concentrations of selenium dioxide. Inspection of the
toxicity curve indicated that the rate of mortality was higher and that
the slope had changed (Figure 3). At shorter exposures, the relation-
ship between LC50 arid time was linear and the slopes of the toxicity
curves were similar for all species except flagfish, where the slope
of the line was less. None of the concentrations tested was high enough
to demonstrate the point at which the toxicity curve became asymptotic
to the LC50 coordinate (ordinate). The median lethal concentrations
calculated for various exposure periods, their 95% confidence limits,
and statistical information on the nature of the concentration-percent
mortality curves for given exposure periods, are detailed in Appendix
Tables 28 to 33 for fathead minnow, brook trout, channel catfish,
flagfish, goldfish and bluegill.
The range in which selenium dioxide was acutely lethal to fish
was uniformly broad. For example, the 24-hr L.C50 of 77. 3 mg/jfc
SeO, determined for juvenile bluegill, the least sensitive of the test
species, was more than four times its 336-hr LC50 of 17. 6 mg/X
SeO2. Similarly, the 24. 5-hr LC50 of 24. 3 mg/j6 SeCX for juvenile
fathead minnow, the most sensitive species, was more than eight times
its 168-hr LC50 of 2.9 mg/4SeO2.
30
-------
00
s
W
O
R
o
i— i
Q
W
o
o
m
U
200
100
60
20
10
6
I T
FATHEAD MINNOW
QBROOK TROUT
A CHANNEL CATFISH
D FLAGFISH
O GOLDFISH
0 BLUEGILL
I
1 T
I
10 20 60
EXPOSURE TIME, hr
100
200
600
Fig. 3. Relationship between median lethal concentration(LC50) of
selenium dioxide and exposure time for six species of freshwater fish.
-------
Similar patterns were evident for the other species. Sensitivity to
selenium dioxide was also affected by size of the test fish. The 96-hr
LC50 of 2. 9 mg/lSeO, estimated for 5 mm fathead minnow fry less
than 24 hr old was 40% that of the 96-hr JLC50 of 7. 3 mg/t SeO2
determined for fathead minnows two to three months older and four to
five times longer.
TOXICITY TESTS OF SODIUM ARSENITE
Water Quality
Sodium arsenite was soluble in the laboratory water at all con-
centrations tested. The presence of the chemical affected water quality,
but in a manner opposite that observed for selenium dioxide. Because
of hydrolysis of sodium arsenite to the very weakly acidic arsenous
acid, with concomitant liberation of hydroxyl ions (e. g. AsO, + 2H~O
,_—> H-AsO, t OH"), there were increases in pH and total alkalinity
and reductions in acidity on addition of sodium arsenite. As shown in
Table 7, the highest levels of pH and total alkalinity encountered were
approximately 8.9 and 256 mg/^CaCO,, respectively, at a concentration
of 272 mg/£ NaAsO2 in one test of goldfish. However, lesser toxicant
concentrations raised the levels of these variables, particularly pH, to
similar extents. Average water quality measured in the various tests
is summarized in Table 8. Because of the toxicant's effect on pH, total
alkalinity, and acidity, values for these three variables were necessarily
derived from measurements made in uncontaminated (control) water.
Water temperatures were maintained at 15 C for tests using brook
trout and at 25° C for the reitiaining species. Dissolved oxygen con-
centrations were greater than 70% of air saturation in all toxicity tests.
Average pH readings ranged from 7. 61 ji- 0. 08 in the tests using goldfish
to 7. 98 + 0. 21 in those using channel catfish, while the range in mean
total alkalinity was 140 jf 8 mg/4 CaCO? in the tests of channel catfish to
168 J- 4 mg/£ CaCO, in those employing brook trout. Variation in total
hardness was low (range of means 140 to 152 mg/lCaCO?), as was that
for specific conductance (range of means 367 to 388 |a mhos/cm).
32
-------
Table 7. HIGHEST LEVELS OF pH
AND TOTAL ALKALINITY ENCOUNTERED
IN TOXICITY TESTS OF SODIUM ARSENITE (NaAsO )
Species
Measured
NaAsO2
cone.,
mg/J&
pH
Total
alkalinity,
mg/l CaCO-
Bluegill
Goldfish
Channel
catfish
Fathead
minnow
Brook
trout
67.9
272.0
72.4
82. 1
84.7
8. 30
8.90
8.92
8.80
8.70
185
256
169
206
215
33
-------
Table 8 . WATER QUALITY DURING TOXICITY TESTS OF SODIUM ARSENITE
CO
Species
Goldfish
Fathead b
minnow
Brook c
trout
Bluegill
Channel j
catfish
Water
temperature,
°c
25. la
±0.4
25. 0
±0. 3
15. 1
±0.6
24.9
±0. 1
24.9
±0. 3
Dissolved
oxygen,
mg/^ saturation
6.7
±0. 5
6.9
±0. 3
7. 5
±0.2
6.8
±0.2
6.5
±0.4
80. 3
±5.8
82. 1
±3.0
73.2
±3.5
81. 1
±1.8
77.1
±5.1
pH
7.61
±0.08
7. 77
±0.13
7. 75
±0. 10
7.82
±0.06
7.98
±0.21
Total
alkalinity,
159
±5
166
±5
168
± 4
166
±4
140
±8
Total
hardness*
148
±7
149
±9
152
±3
147
±15
140
±4
Specific
conductance*
>^ M mhos /cm
373
±17
379
±24
388
±15
378
±23
367
±9
Means ± 1 standard deviation are given
Acidity averaged 5.4 ±0.4 mg/4
Acidity averaged 5. 5 ± 2. 1 mg/X
Acidity averaged 5.4 ± 3.4 mg/4
-------
External Pathology
Although evidence of cutaneous hemorrhaging was evident in
certain individuals of each species exposed to acutely toxic ..oncentra-
tions of sodium arsenite, none of the species except brook trout
developed hemorrhages and lesions on the scale noted for selenium
dioxide. Essentially, all brook trout developed patches of marked dis-
coloration or a mottled appearance within 12 to 24 hr after toxicant
introduction. Trout exposed for longer periods developed extensive
tissue destruction in the cranial regions, although lesions of the skin
were also evident in posterior areas.
Toxic ity
Sodium arsenite was the fourth most toxic chemical to the fresh-
water fish tested, ranking just behind selenium dioxide. The order
of species sensitivity was very similar to that observed for selenium
dioxide except that adult brook trout were more sensitive than juvenile
fathead minnows. Bluegills were the least sensitive species. The
minimum concentration of sodium arsenite found to be acutely lethal
was in the tests of adult brook trout, where the LC50 for 262 hr exposure
was 18. 0 mg/jfc NaAsO? or 10. 4 mg/4 elemental arsenic.
Toxicity curves relating median lethal concentration to exposure
time are given in Figure 4 for the six test species. Two types of curves
were evident. For brook trout and goldfish, median lethal thresholds
were reached, indicating a unimodal type of sodium arsenite acute
toxicity. In contrast, median lethal thresholds were not evident for
fathead minnows and bluegill, and probably not for channel catfish, and
two modes of toxicity were suggested by the curve. In the latter three
species, prolonged (i. e. > 96 hr) exposure to the lower concentrations
produced substantial mortality and a shift in the slope of the toxicity curve.
There were insufficient observations to distinguish whether the toxicity
curve for flagfish fry followed the pattern of minnows and bluegill.
35
-------
GO
0>
300
200
bfi
8
w
H
2
w
CO
Q
O
CO
«'^
O
O
u">
O
100
60
I
BROOK TROUT
FATHEAD MINNOW
CHANNEL CATFISH
OGOLDFISH
QFLAGFISH
OBLUEGILL
20
10
10
20 60
EXPOSURE TIME, hr
100
200
600
Fig. 4. Relationship between median lethal concentration (LC50) of sodium arsenite
and exposure time for six species of freshwater fish
-------
Median lethal concentrations, their 95% confidence limits, and
statistical information for the concentration-percent response curves
for given exposure times are detailed in Appendix Tables 34 to 39 for
all six test species. The 96-hr LC50 estimates were 72. 0 mg/£
for bluegill, an average of 48. 5 mg/jfcfor flagfish, 44. 9 mg/4 for
goldfish, 31. 2 mg/jfcfor channel catfish, 27. 0 mg/A for fathead minnow,
and 25. 8 mg/j&NaAsO2 for brook trout. A median lethal threshold of
32. 0 mg/X was manifested in 336 hr with goldfish and one of 18. 0 mg/£
developed after 262 hr with brook trout. In both cases, median lethal
thresholds were approximately 30% lower than the 96-hr LC50. The
sensitivities of flagfish fry, juvenile goldfish and channel catfish
were very similar, particularly in the concentration raig e of 45 to 70
mg/JL
TOXICITY OF BERYLLIUM SULFATE TETRAHYDRATE
Water Quality
Beryllium sulfate precipitated when introduced into the test
containers, and large amounts accumulated at the bottom during the
course of the test. Beryllium sulfate was probably reacting in the
ambient watet to form beryllium hydroxides of low solubility. A
likely hydrolysis product is Be(OH)2i which has a solubility of 2 ppm
(Lange ). The probable reaction is nBe^+nH^O^1^ (BeOH)n+ 4- nH+
37
(Everest ), which proceeds strongly to the right at the average experi-
mental pH of 7. 7 to 7. 8. Since all tests were conducted at concentra-
tions exceeding the theoretical solubility of beryllium sulfate or beryll-
ium" hydroxide, mortality of the test specimens could be expected to be
essentially equivalent between concentrations. Because graded, con-
centration-dependent mortality was observed, the lethal effects of
beryllium may have been caused either before or during its precipita-
tion or by the precipitated material or both. In order to determine the
rate of what appeared to be a slow precipitation reaction, a chamber
37
-------
similar to those used for testing was filled with 10 SL of filtered water
(Whatman No. 1 filter paper), spiked with sufficient BeSO4 • 4H2O
to provide an initial concentration of 150 mg/j6 (89 mg/i, BeSO4) and
sampled serially thereafter for five days. Upon collection, samples
were filtered through a fine (4. 0 - 5. 0 (a ) fritted disc into a test tube
containing one milliliter of concentrated hydrochloric acid. In trial
runs, precipitated beryllium was not observed to be passed through
this filter. After mixing, the sample was measured for elemental
beryllium on an atomic absorption spectrophotometer. As shown in
Figure 5, there was a rapid decline in dissolved beryllium (as BeSO.)
to 26. 7% of nominal within one minute. Thereafter, concentrations of
dissolved beryllium remained relatively stable for 30 to 60 min, then
gradually declined to 2% of nominal one day after introduction. This
experiment indicated that precipitation of beryllium sulfate proceeded
slowly, confirming laboratory observations and the findings of Everest
The presence of beryllium sulfate or its probable hydrolysis
product, beryllium hydroxide, in the laboratory water reduced pH,
total alkalinity, and acidity in a manner similar to that described
for selenium dioxide; however, the dimunitions in pH and total alkalinity
were less. The minimum pH (6. 60) was observed at a measured con-
centration of 59. 0 mg/j£BeSO4, while the lowest total alkalinity, 107
mg/4, was observed at a measured concentration of 64. 0 mg/£ in
the test using brook trout (Table 9). Hardness, specific conductance
and dissolved oxygen concentrations were not observed to be affected
by the addition of beryllium sulfate.
Average values for eight water quality variables monitored during
the toxicity tests of beryllium sulfate are given in Table 10. Because
of the effect of the chemical on pH, total alkalinity and acidity, values
cited represent measurements in uncontaminated (control) water only.
Water temperatures varied within + 0. 5° C of the specified temperatures
of 15 C for brook trout and 25 C for the remaining species. Dissolved
38
-------
CO
CO
JU
•=>?
"M 25
; <
2 20
2
H
oi 15
JCENT:
h—
O
f-\
O
0 5
0
0)
]
1 1 1 1 1 1 1 III 1
0
00 0
/
.0 ° - 00
- ® 0
00
1 III III 1 1 ° 1 A
1 2 6 10 20 60 100 200 600 1000 2000 6000 1
1), 000
Fig.
TIME, min
5. Changes in the concentration of dissolved beryllium (as BeSO4) spiked into
10 SL of diluent water at a level of 89 mg/j£ BeSO4
-------
Table 9. MINIMUM LEVELS OF pH AND
TOTAL ALKALINITY OBSERVED IN TOXICITY TESTS
OF BERYLLIUM SULFATE (BeSO4)
Species
Measured
concentration
of BeSO4,
pH Total
alkalinity,
mg/>e CaCO.
Fathead
minnows
Goldfish
Channel
catfish
Brook
trout
47.8
59.0
47.8
64.0
6.95
6.60
6.95
7.46
114
124
114
107
40
-------
Table 10. WATER QUALITY DURING TOXICITY TESTS
OF BERYLLIUM SULFATE
Species
Fathead*
minnow
Goldfish
Channel
catfish
Brook
trout
Water
temper -
atyre,
C
24. 8b
+ 0.4
25.2
±0.4
24.9
+ 0.4
15.2
+ 0.3
Dissolved
mg/Ji
7.4
+0.2
6.5
+0.3
7.3
+0.3
7. 1
±°-6
oxygen,
%
satura-
tion
88.5
+ 2.5
76.9
+ 4.2
87.5
+ 3.0
70.0
+ 6.0
PH
Total
alka-
linity,
Acidity,
mg/J&CaCO,
mg/j6CaCO3
8.04
±0. 17
7.57
+0.09
8.04
+0. 17
7. 90
+ 0.30
146
+ 6
161
+ 6
146
+ 6
144
+ 5
4.6
+2. 5
c
4.6
+ 2.5
6.6
^2.9
Total
hard-
ness,
mg/J&CaCQ-
140
± 4
147
+ 8
140
± 3
137
± 4
Specific
conduct-
ance,
\JL mhos/cm
372
+ 28
392
+ 13
372
+ 28
364
Tests using one day-old flagfish fry were conducted concurrently in the same
containers.
Means + 1 standard deviation are given.
No observation
-------
oxygen concentrations were 70% of air saturation in all tests. The
range of mean pH was 7. 57 jf 0. 09 to 8. 04 _+ 0. 17; that of total alkalinity,
144 _f 5 to 161 J- 6 mg/4 CaCO3; and that of acidity, 4. 6 _+ 2. 5 to 6. 6
_+ 2. 9 mg/£ CaCO,. Meantotal.hardness varied from 137 + 4 to 147
•f 8 mg/Jl CaCOo and specific conductance from 364 to 392 (J mhos/cm.
ToxicjLty_
Beryllium sulfate had a relatively low toxicity to the freshwater
fish tested, being more toxic than lead chloride, but less toxic than
the pentachlorophenol, cyanide, selenium and arsenic compounds.
The descending order of toxicity was fathead minnow, flagfish fry,
goldfish, brook trout, and channel catfish.
Toxicity curves encompassing exposures for up to 336 hr were
obtained for juvenile fathead minnows, flagfish fry and juvenile goldfish.
Juvenile channel catfish and adult brook trout were exposed for 96 hr
to 59. 3 mg/^BeSO. without mortality. Ten percent mortality in brook
trout after 120 hr exposure accounted for their placement ahead of
channel catfish in terms of sensitivity. Two types of curves may have
characterized the toxicity of beryllium sulfate to the freshwater fish
studied (Figure 6). A median lethal threshold of 25. 4 rag/Ji BeSO>
was reached in 336 hr in the toxicity test using fathead minnows. Linear
toxici ty curves portrayed the relationship between LC50 estimates and
toxicant exposure times for juvenile goldfish and flagfish fry, although
averaging of the concentration-percent response data and an insufficient
number of observations precluded anything but a tentative conclusion for
flagfish. Although the toxicity curve for goldfish suggested linearity,
the 216-hr and 240-hr LC50 estimates for goldfish may have indicated
increased mortality rates after exposure for these periods, similar to
those observed for selenium dioxide.
42
-------
100
w
H
w
s
D
w
pq
UH
o
o
m
U
60
20
FATHEAD MINNOW
Q FLAGFISH
O GOLDFISH
10
20
60
100
200
600
EXPOSURE TIME, hr
Fig. 6. Relationship between median lethal concentration
(LC50) of beryllium sulfate and exposure time for three
species of freshwater fish.
43
-------
For the species tested, beryllium sulfate was acutely lethal
down to 25. 4 mg/4 which was the 336-hr LC50 for fathead minnows
and about twelve times the theoretical solubility limit of beryllium
hydroxide in distilled water. The 336-hr LC50, also designated as the
median lethal threshold, was only 33% lower than the 96-hr LC50 of
37.9 mg/j? (Appendix Table 40). Ninety-six hour LC50 estimates of
46. 3, 41. 1 and 41. 1 mg/j6BeSO4 were determined for three groups
of flagfish initially exposed within 24 hr of their hatching (Appendix
Table 41). These estimates, which were pooled and averaged (42. 8
mg/H BeSO,) for graphical presentation, were only slightly higher
than the 96 hr estimate for minnows. The 96 hr JLC50 for juvenile
goldfish was 55. 9 mg//, 1.5 times that for minnows. The lowest
LC50 estimate calculated for goldfish was 38. 4 mg/Ji BeSO4 for
240 hr exposure (Appendix Table 42 and Figure 6).
TOXICITY OF LEAD CHLORIDE
Lead chloride was not found to be acutely toxic to juvenile fat-
head minnows or adult brook trout at concentrations up to 100 mg/X
PbCl- and exposures up to 14 days. Preliminary static tests utilizing
up to 500 mg/4 PbCl- indicated no acute lethality to fathead minnows
in four day periods. Upon mixing with the diluent water, the toxicant
precipitated immediately and rapidly settled out in the diluter apparatus
and in the test containers. Under the test conditions, which were
essentially the same as described for the othe r chemicals, there
was little evidence that coagulation of the fish's mucus occurred.
44
-------
SECTION VI
DISCUSSION
TOXICITY OF SODIUM PENTACHLOROPHENATE
The acute toxicity of the sodium salt of pentachlorophenol
(NaPCP) was characterized by two types of toxicity curves, one
resembling a rectangular hyperbola and the other a sigmoid shape.
Only in the tests of brook trout was a rectangular hyperbola designa-
ting an acute median lethal threshold observed. Sigmoid-shaped
curves characterized the responses of fathead minnows, goldfish
and bluegill to the chemical. A change in slope of the toxicity curve
commonly is interpreted to indicate a change in the mechanism or
transmission of toxicity or a change in the resistance of the test
specimens (Sprague ). In light of the known functional effects of
NaPCP on fish, detailed below, it is likely that the fish's resistance
was diminished through a progressive deterioration in physiological
condition which ultimately resulted in their mortality.
One of the salient physiological consequences of sodium penta-
chlorophenate intoxication is an increase in oxygen consumption (Crandall
38
and Goodnight ), which is due to the uncoupling of oxidative phos-
phorylation. In studies of the effects of sublethal concentrations of
sodium pentachlorophenate (0. 1 mg/£) on selected enzymes of glycolysis,
the pentose-phosphate shunt, the Embden-Meyerhof pathway, the tri-
carboxylic acid cycle, and the cytochromes in Atlantic eel (Anguilla
_anguilla), Bostrom and Johansson found that the activity of enzymes
associated with anaerobic catabolism of carbohydrates (glycolysis) was
reduced, but that enzymes associated with the aerobic pathways (i. e.
pentose phosphate shunt, TCA cycle, and the cytochromes) were enhanced.
The augmented, NaPCP-induced energy demands of fasting fish are
supplied initially by fatty acids (e.g. palmitic, stearic and oleic acid)
40 41
from triglycerides (Hanes, et al ' ). In many cases, extensive
diminution or exhaustion of saturated and monounsaturated fatty acids
45
-------
is followed by utilization of proteins and long-chain polyunsaturated fatty
42
acids to fulfill the energy demands (Saddler, Koski and Cardwell ).
Utilization of proteins and phospholipids may be regarded as a lethal
physiological manifestation since the structural integrity of the animal
is being severely compromised.
Thus, in the present studies, the heightened energy demands of
pentachlorophenate-exposed fish, coupled with those associated with
starvation over a cumulative 17 day period, could have resulted in the
exhaustion of the energy stores of the fish and led to the fatal utilization
of other substrates essential for the animal's biochemical integrity. If
the minnows, bluegill, and goldfish had proportionally less depot fat than
the adult brook trout, they may have exhausted their supplies earlier and
subsequently perished. Growth of guppies (Poecilia reticulata Peters)
38
(Crandall and Goodnight ) and of underyearling sockeye salmon
43
[Oncorhynchus nerka (Walbaum)](Webb and Brett ) exposed for 45
to 90 days to sublethal concentrations of sodium pentachlorophenate has
been found to have been depressed even though the fish were fed during
43
the period. Webb and Brett also found conversion efficiency to be reduced.
Curves resembling that obtained for brook trout have been obtained
by others. A rectangular hyperbola described the acute toxicity of sodium
pentachlorophenate to yearling sockeye salmon which were exposed for 504 hr
43
in a flow-through system (Webb and Brett ). However, close inspection
of the curve revealed what may have been a slight shift in the curve's
slope between 100 and 200 hr, similar to what -we observed for bluegill,
minnow, and goldfish. No such change was observed for underyearling
sockeye, which had been found to have a sensitivity equal to that of the
44
older fish. Norup also described a rectangular hyperbola toxicity curve
for guppies exposed for 168 hr to NaPCP.
46
-------
The acute toxicity tests conducted by our laboratory with an inter-
mittent-flow bioassay system indicated that the median lethal threshold
for the four species tested was at least 0. 11 mg/£NaPCP. This figure
is one-half or less of those stated in much of the early literature, but
is somewhat higher than the median lethal threshold of 0. 057 mg/£
NaPCP and the levels of 0. 00174 to 0. 0018 mg/i, affecting growth and
food conversion efficiency in 50% of the underyearling sockeye salmon
43
exposed in a continuous-flow system (Webb and Brett ). As noted
earlier, Bostrom and Johansson observed significant alterations in
the activities of certain enzymes in Atlantic eels exposed for 96 hr to
0. 1 mg/£ NaPCP in a static system. One of the highest lethal thresholds
45
is that of 0. 75 mg/£PCP, given by Bandt and Nehring for rainbow trout.
The lethal threshold for 72 hr exposure for five species of minnows
(cyprinidae) is between 0.2 and 0. 4 mg/£ NaPCP (Goodnight ). More
44
recently, Norup determined that the median lethal threshold for guppies
was of the order of 2 nag/A He also presented a toxicity curve repre-
senting 25 fish species which indicated a composite lethal threshold of
less than 0. 1 mg/Ji NaPCP for fish in general.
For shorter periods of exposure, comparison of LC50 estimates
derived from the present work with those derived by others indicates
closer agreement regardless of whether the tests were conducted in
static or continuous-flow systems. For the bluntnose minnow [Pimephales
notatus (Rafinesque)], a close relative of the fathead minnow, Goodnight
calculated that 0. 4 mg/4 NaPCP would produce 100% mortality in 7
to 45 hr. This was surprisingly close to the LC50 value of 0.47 mg/-6
NaPCP we calculated for the same period for fathead minnows. Crandall
47
and Goodnight calculated a 96-hr LC50 of 0. 35 mg/£ NaPCP for 50 mm
fathead minnows. This figure is only slightly higher than the interpolated
estimate of 0. 29 mg/£ determined in this laboratory for the same species.
However, the 24-hr L.C50 for rainbow trout and brown trout has been
estimated to be around 0. 01 mg/4 (Alabaster ), a much lower figure
than that found for brook trout.
47
-------
The disparity in tolerance between rainbow and brown trout and that
of brook trout is so great that some of the difference is probably
ascribable to variations in the characteristics of the experimental fish,
teat conditions and water quality.
The toxicity of sodium pentachlorophenate is known to increase
with a decline in experimental pH. Goodnight noted that fish sur-
vived longer at pH 7. 6 than at 6. 6. Later studies by Crandall and
47
Goodnight confirmed the earlier observation. They found that the
/
median lethal time for fathead minnows exposed to 1. 0 mg/A NaPCP
at a pH of 5. 9 to 6. 0 averaged 21 to 38 min, but increased to 72 to 93
min at a pH of 7. 5 and 7. 6, and to at least 1440 min at a pH of 8. 9 to
9. 0. According to these investigators, earlier studies by Blackman
49
et al., which indicated that the physiological effects of substituted
phenols was greatest when the pH of the water was closest to the pK
of pentachlorophenol (4. 8), confirmed their observations. At pH 4. 8
pentachlorophenol is present in equal quantities of free acid and conjugate
base. It thus appears that the free acid is more toxic than the phenoxide
47
ion (Crandall and Goodnight ). However, the change in NaPCP toxicity
may not be as great as indicated by their work since the minnows were
not acclimated to the test temperatures but rather introduced abruptly.
Nevertheless, the median lethal times for exposure to 1. 0 mg/X
NaPCP of 225 to 281 min indicated much longer survival at 10°C than at
18° C (LT50 * 72 to 93 min) or 26° C (LT50 = 35 to 57 min).
TOXICITY OF SODIUM CYANIDE
Toxicity curves for the five species of freshwater fish indicated
that acute mortality essentially ceased after approximately 100 hr.
That • odium cyanide had a median lethal threshold is consistent with
its well known physiologic.action of causing tissue hypoxia and ultimately
anoxia through competitive inhibition of enzymes (e.g. cytochrome
oxi4a*e) participating in oxidative phosphorylation.
48
-------
Since the effect is based on competition for active sites on
\
enzymes, the magnitude of hypoxia depends on the relative mole
fractions of cyanide ion and the competing molecules. Several other
investigators have also found that sodium cyanide has a median lethal
threshold for fish. Wuhrmann calculated that thresholds of 0. 06,
0.08, 0. 10,and 0. 30 mg/Xprussic acid (hydrocyanic acid, HCN) existed
for minnows (Phoxinus laevis), perch (Perca sp. ), tench (Tinea tinea),
and chub (Squalius c_ephalusji respectively, with the threshold for P^
laevis being reached in 22 hr. The data of Doudoroff indicate that a
median lethal threshold of 0. 23 mg/£ cyanide ion (CN~) was reached in
a maximum of 96 hr in static toxicity tests using 50 mm juvenile fat-
head minnows. In general, although the cyanide concentrations at which
the median lethal thresholds were observed were of the same order as
those found in our studies, they manifested themselves earlier. The
difference can be probably traced to the fact that cyanide concentrations
decline temporarily as a result of the hydrolysis of cyanide to hydro-
cyanic acid (i. e., NaCN + HpO^HCN + OH" + Na*) and concomitant
loss of hydrocyanic acid from the system because of volatilization.
Doudoroff was aware of the loss of HCN, the most toxic form of
cyanide, noting that minnows perished within 14 hr when introduced
into fresh cyanide solutions (e.g., 0. 32 mg/j£CN ), but incurred only
20% mortality within an equivalent interval when introduced into a solu-
tion which had been allowed to stand for 24 hr after preparation. The
net effect of hydrocyanic acid loss is to produce an apparent median
lethal threshold in a toxicity test, unless additional cyanide is added
to stabilize toxicant levels. In our intermittent-flow toxicity tests,
temporal monitoring of the total cyanide concentrations facilitated
maintenance of toxicant concentrations at prescribed levels through
manipulation of toxicant renewal rates and daily preparation of fresh
stock toxicant solutions. The range of acute lethality of sodium cyanide
was also shown to be slightly broader than found by others using static
test conditions, again probably ascribable to the stable toxicant
concentrations.
49
-------
Much of the data describing the toxicity of cyanides to freshwater
fish have been derived from tests conducted under static conditions,
with or without renewal or measurement of the toxicant. Although the
more sophisticated methods and toxicity testing equipment developed
within the last decade and utilized in the present studies were designed
to provide more accurate results than the earlier methods, comparison
of LC50 values determined in this laboratory with those obtained by
earlier workers indicates good agreement. For example, the 96-hr
L.C50 for fathead minnows exposed to sodium cyanide has been variously
determined to be 0.23 mg/j2CN" (Doudoroff ), 0. 19 mg/^CN" in hard-
52
water (Henderson, Pickering, and Lemke ) and 0. 12 mg/j6CN in
52
soft water (Henderson, et ol. ). In one of the few tests where sodium
cyanide concentrations were continually renewed, the 96-hr L.C50 for
50 mm bluegill was approximately 0. 15 mg/JL HCN, equivalent to the
median lethal threshold (Doudoroff .et^ajU ) . We determined that the
96-hr LC50 for fathead minnows was less than half this value (0. 065
mg/.e CN~).
Further search of the literature also indicated fairly good agree-
ment of toxicity data for bluegill exposed to sodium cyanide under
52
static and intermittent-flow conditions. Henderson, et al. calculated
the 96 hr LC50 to be 0.08 mg/lfor bluegill exposed in hardwater. This
value is only slightly higher than the; interpolated L.C50 value of 0. 068
mg/Jt, CN determined here. Ninety-six hour median lethal concentra-
tions for bluegill exposed to potassium cyanide by Cairns and Cairns
and Scheier ' ' were essentially the same as estimates based
on sodium cyanide, regardless of whether the diluent water was hard
(range 0. 14 to 0. 17 mg/^CN") or soft (0. 13
50
-------
The small variation in the median lethal concentrations of the
simple cyanide salts (NaCN and KCN) for fresh water fish indicates a
uniformity in toxicity which is not significantly affected by water quality,
the type of bioassay system employed, or by the size, age or condition of
the test fish, even though the bioassays were conducted by investigators
2
using different water qualities, stocks of test fish, and methods. Sprague
has commented that because of this variation, lathal levels for toxicants
in general may vary a thousand-fold. Variation in toxicity of simple
cyanides is obviously much lower.
Although pH is known to dramatically affect the toxicity of
metal cyanide complexes, it does not have much of an affect on the toxicity
of sodium or potassium cyanide in waters of approximately neutral pH,
since cyanide is present predominately as the free acid (HCN). The
influence of pH may be notable when the cyanide is present in a complex
ion. An increase in the hydrogen ion concentration will cause liberation
of HCN from the complex. These relationships have been examined
51 35
extensively by Doudoroff and Doudoroff, et al.
51
-------
TOXICITY OF SELENIUM
Selenium is a relatively common element which comprises
approximately 6x10 percent of the earth's crust. In some areas
of the United States, particularly in the midwest, seleniferous deposits
are extensive and are major sources of selenium pollution of some
, 57
fresh waters. In Nebraska, for example, Engberg found that over
33% of the wells and 25% of the surface waters contained elemental
selenium in excess of the 10 Mg/£ upper limit established by the U.S.
Public Health Service for drinking water. Selenium is commonly pro-
duced from the anode muds of electrolytic copper refineries. Indus-
trially, the various allotropic forms of selenium are used in a variety
of processes and components, including photocells, solar cells, and
semiconductors. It is also employed in the dye and pigment industry
and, because of its biological potency, as an insecticide in certain
applications.
Although selenium is similar chemically to sulfur and tellurium,
its toxicity is most similar to that of arsenic. The toxicology of selenium
to mammals is much better known than that to fish because it is respon-
sible for the "Alkali Disease'1 and "Blind Staggers" syndromes of grazing
animals. These maladies result when animals graze on certain types
of vegetation in soils containing high concentrations of selenium. When
the element is abundant, certain plants accumulate large concentrations
and store it in various organic forms. The pathologies associated with
the two syndromes differ from that observed for inorganic selenium
poisoning (selenosis) because the organic selenides are more toxic to
mammals than the inorganic forms. The effects of the various types
of selenium poisoning on animals and the nutritive importance of selenium
• ^ ft
are reviewed oy Rosenfeld and Beath
52
-------
One of the first recorded studies of the effects of selenium on
59
aquatic animals was completed by Ellis, et al. using 80 mm goldfish
and 160 mm channel catfish. In one experiment, goldfish were exposed
to 2 mg/-6 Se as sodium selenite for 46 days. Every 48 hr the fish were
transferred to a fresh toxicant solution and fed. Under these conditions,
fish ceased feeding within eight days, and the first animals perished
in 18 days. Most mortalities were recorded between 25 and 37 days.
The authors noted that goldfish died in four to ten days when exposed
to 5 mg/£ Se as sodium selenite. Intraperitoneal injections of 3. 0 mg
Se/kg into 54 g catfish were found to be fatal to the fish in about 7 hr
at 10° C. Injections of lesser amounts (0.2 mg/kg/day/5 days) pro-
duced exophthalmia, anemia, leukopenia and degeneration of liver,
kidney and spleen. More recently, Weir and Hine compared median
lethal concentrations of selenium dioxide with those significantly affecting
a conditioned behavior in 40 to 80 mm goldfish using static test conditions.
The 9-day LC50 of 15. 9 mg/,6 SeO2, which was derived from 48 hr
exposure followed by 7 days holding in uncontaminated freshwater at
23° C, was 48 times greater than the concentration (0. 035 mg/S, SeO^)
disrupting the ability of goldfish to respond to a light stimulus and seek
a dark area.
i
The 9-day L.C50 estimate obtained by Weir and Hine is com-
parable to the LC50 of 17.2 mg/£SeCX> which we obtained for 168 hr
Li
continuous exposure, but about one-third that of our 48-hr LC50 of
46.5 mg/£ SeO9 for the same species. Although the water used by
13
Weir and Hine was much softer (deionized water was reconstituted
with 50 mg/j£ CaCO,) and of lower pH (range 6. 0 to 6. 9), it is probable,
in light of the external pathology they noted, that substantial latent
mortality occurred.
53
-------
Our results indicate that for the given water quality and experi-
mental conditions, selenium dioxide was most toxic to fathead minnows
and least toxic to bluegills. Use of test speciments of equivalent age
and size would affect the order of species sensitivity to some extent, as
i - _
evidenced by the fact that the 96-hr LC50 for the much smaller fathead
minnow fry was 40% that for juvenile minnows exposed for a comparable
period. Nevertheless, direct comparison was possible for channel
catfish, goldfish, and bluegill, which were of similar size and age.
The nature of the toxicity curves provides further insight into
the biological action of selenium dioxide in addition to describing its
differential effect on the various species. Because no lethal thresholds
were observed for exposure periods up to 336 hr, it is evident that both
toxicant concentration and exposure time influence toxicity. This con-
clusion is strengthened further by the observations that the fathead
minnows, goldfish and bluegill, which were exposed for more than
168 hr, died at faster rates than would have been anticipated from
extrapolation of the slopes of toxicity curves derived from exposures
less than 168 hr. Selenium poisoning in fish evidently produces physio-
logical dysfunction which manifests a lethal character upon prolonged
exposure of the fish to the toxicant. Because of the external hemorrhag-
ing observed by us for fathead minnow, channel catfish, brook trout
and goldfish, and by Weir and Hine for goldfish, it is likely that the
peripheral blood, including either the blood or capillaries or both,
was involved. This hypothesis is strengthened by the observation
of erythrocyte destruction in catfish dosed with sodium selenite (Ellis,
59
et al. ), and the common findings of damage to the hematopoietic
system in mammals given inorganic selenium.
TOXICITY OF SODIUM ARSENITE
There are three important factors to consider in assessing the
results of the toxicity tests of sodium arsenite. First, the presence
of the toxicant altered the quality of the water, and in so doing, prob-
ably altered to some extent the magnitude of the responses of the test
54
-------
animals. Secondly, there was no precipitation of the toxicant upon
combination with the diluent; hence, the question of what form of the
chemical was causing death was not as complicated. Finally, at
least two modes of toxicity were evident from the curves relating
median lethal concentration to exposure time.
In contrast to the acid conditions resulting from hydrolysis of
the selenium, beryllium and lead compounds, hydrolysis of sodium
arsenite rendered the water alkaline. At the highest toxicant con-
centrations employed, pH levels reached 8. 9. pH values between
5. 0 and 9. 0 are generally not acutely lethal for most fish species
(Doudoroff and Katz ; European Inland Fisheries Advisory Commission ),
although chronic exposure to intermediate levels (pH 6. 6) has been
found to depress egg production and spawning of at least one freshwater
ft 2 ft 3
fish species (Mount ). Jordan and Lloyd determined that 15 days
were required to kill 50% of a population of rainbow trout at a pH of
9. 5, while four day exposures to pH 10. 5 were necessary to produce
64
50% mortality in 39 mm bluegill (Cairns and Scheier ). In light of
these findings, it appears that the highest levels of pH reached in the
toxicity tests were near the lethal limits, and that sublethal pH stress
may have interacted synergistically or additively with the toxicant
to augment the responses of the test animals. Moreover, the sudden
increase in ambient pH encountered by the fish, which had been accli-
mated to a level of approximately 7. 5 to 8. 0, might also act as a
stresser. Jordan and Lloyd found that the 24-hr median lethal pH
levels were 9. 86, 9. 91 and 10. 13 for rainbow trout previously accli-
mated to pH values of 6. 55, 7. 50 and 8. 4. It is apparent that the
possibility.of a pH-toxicant interaction cannot be ignored for the high
sodium arsenite concentrations. But since the log-probit, concentration-
percent-mortality regression lines were calculated for the range of
concentrations tested, including the lower concentrations where the pH
interaction would likely be negligible, the bias would be effectively
mitigated.
55
-------
4
According to McKee and Wolf , the toxicity of arsenic is similar
to that of selenium, although arsenic antagonizes the toxicity of selenium.
In these studies, bimodal toxicity curves characterized the relationship
between LC50 and exposure time for three of the species tested, fathead
minnow, channel catfish and bluegill, whereas unimodal toxicity curves
(resulting in median lethal thresholds) were observed for brooktrout and
goldfish. Differences in sensitivity, mediated perhaps by innate tolerance
or a capability for metabolizing or excreting the chemical, may have
distinguished the trout and goldfish from the other species. For the
other species, the changes in slopes of the toxicity curves, each
signifying increased mortality upon prolonged exposure to the lower
concentrations, may represent the manifestation of a different toxicity
mode, perhaps one denoting severe functional deterioration. Arsenic
has been stated to be a cumulative poison (Jones ). Previous workers
have either not conducted toxicant exposures sufficiently long to demon-
strate whether a change in slope occurs or have conducted the tests in
static systems wherein toxicant concentrations probably declined rather
than remained stable. Grindley , in a study elaborated by Wuhrmann ,
exposed minnows (Phoxinus laevis) to sodium arsenite for 36 hr and
observed a straight line relationship between the median time required for
manifestation of ataxia and toxicant concentration. Holland, et al.
exposed 0. 30 g pink salmon fry [Oneorhynchus gorbuscha (Walbaum)]
to sodium arsenite in a static system for ten days. When their percent
mortality data were plotted against initial toxicant concentrations,
mortality appeared to drop off and approach a median lethal threshold
between 150 and 200 hr. Unfortunately, measured concentrations declined
temporally, e.g., from 16.25 to 9.25 mg/^at the highest level tested
and from 1. 50 to 0 mg/jfcat the lowest.
Median lethal concentrations for sodium arsenite obtained by
others in laboratory and field experiments were generally lower than
VL Q
those determined in this program. Clemens and Sneed determined
that the 24-and 48-hr LC50 values for channel catfish were 47. 9 and
56
-------
25. 9 mg/& NaAsCX. respectively. Mortalities ceased after 48 hr.
The 48-^and 96-hr LC50 estimates for catfish exposed in this study were
56. 9 and 31. 2 mg/4 NaAsCX, respectively. Gilderhus estimated
96-hr LC50 values of 34 mg/^NaAsO2 for goldfish and 35 mg/£ NaAsO,
for blue gill in laboratory tests, then determined that as little as 4. 0
mg/4 sodium arsenite inhibited growth of both 0. 5 g immature and 15. 8
adult bluegill and survival of the immature fish when sodium arsenite
was applied at various intervals to outdoor pools. The 96-hr median
lethal concentration determined for bluegill in our laboratory was twice
as high (72 mg/4) as the LC50 determined by Gilderhus , while the
336-hr LC50 (31. 6 mg/It) was more than 1, 000 times higher than the
concentration of 0. 03 mg/£NaAsO2 that Gilderhus determined would
not affect survival or growth of bluegill when the toxicant was applied
at a weekly rate for the control of aquatic vegetation.
TOXICITY OF BERYLLIUM SULFATE
Beryllium as beryllium sulfate was one of the two metals tested
at concentrations exceeding its solubility in the moderately hard,
alkaline diluent water. The most probable reaction is between
beryllium and hydroxyl ions, with consequent rise in hydrogen ion
concentration (i. e. nBe+* + nH^Ozzfi (BeOH)11"1" -h nHf )(Everest37).
L* > ri
Since the solubility of beryllium hydroxide in distilled water (2 mg/Z,
Lange ) is of the same order as the lowest concentration found to be
lethal in the toxicity tests (336-hr LC50 of 2. 2 mg/4 Be in bioassay
using fathead minnows), it is evident that the toxicity tests had to be
conducted above the solubility limit, i. e., in the presence of beryllium
hydroxide.
57
-------
In these toxicity tests, beryllium concentrations were measured
as total metal from water samples collected in the middle of the test
chambers. Because of precipitation, however, probably only a fraction
of the nominal concentration was causing the toxicity. Since beryllium
uptake in fasted freshwater fish would be primarily if not entirely via
the gills, it would be expected that only dissolved beryllium could be
causing the toxicity. Since all toxicity tests were conducted above the
diluent water solubility of this chemical and since graded, concentra-
tion dependent kills were observed, it is possible that some of the
precipitated material may have contributed significantly to mortality.
Lloyd and Herbert have advanced an hypothesis that may well account
for the toxicity of mixtures of precipitated metals in general, which
may include beryllium. Dissolution of the precipitated metal adhering
to the branchial epithelium may be effected by the transiently high
hydrogen ion concentrations (stemming from carbonic acid) existing
at the gill-water interface because of excretion of carbon dioxide.
Mount recently invoked this hypothesis as a partial explanation of
the toxicity of mixtures of precipitated zinc sulfate to 1 to 2 g fathead
72
minnows. Tabata , however, believed that precipitated heavy metals
would not harm most aquatic animals.
The change in pH of the diluent water was found to vary with the
total concentration of beryllium sulfate. Hydrogen ion concentrations
increased with increased beryllium sulfate concentrations, although
the magnitude of the increase measured in the intermittent-flow system
was not as great as that observed in static systems. The lowest pH
levels recorded by Slonim were approximately 3. 7 and 4. 1 at 200
and 100 mg/Ji BeSO>, respectively, in toxicity tests of guppies in hard
water (hardness = 400 to 500 mg/
-------
the chambers being sufficient to keep the precipitation reaction in an
intermediate stage by displacing old toxicant and adding fresh. As
demonstrated earlier, the precipitation reaction was much slower than
the 90% molecular displacement time of nine hours.
There is little doubt that the increased acidity of the diluent water
caused by the presence of beryllium sulfate also contributed to the
toxicity of the metal, even though the maximum displacement in pH
was only about one unit. Although the pH range of 5 to 9 is not directly
lethal to fish (European Inland Fisheries Advisory Commission ),
abrupt variation from the optimum, which occurred in the present
toxicity tests as a function of toxicant concentration, is not anticipated
to be well tolerated by most fish species and would constitute a stresser.
Coupled with the increased solubility of beryllium under more acid
conditions, the net effect probably would be to increase somewhat the
toxicity of the beryllium mixtures at the lower levels of pH. Many other
investigators have documented the positive effect of decreased pH on
metal toxicity to fish (Wuhrmann ; Mount ; Freeman and Everhart ).
f>?
In Mount's recent investigation of the chronic effects of low pH on.
fathead minnows, he observed decreased egg production by adult fathead
minnows chronically exposed to a pH of 6. 6 relative to controls which
had been reared at pH 7. 5. Decreased fecundity is probably one of
the functional manifestiations of sublethal stress.
The few investigations of the toxicity of beryllium to fish have
o
been conducted under static conditions. Pomelee added daily a complex
of beryllium sulfate and tartaric acid to aquaria containing goldfish,
minnows and snails for 12 days and did not observe any toxic effects on
the animals at a total beryllium concentration of 28. 5 mg/£. Tarzwell
74
and Henderson studied the acute toxicity of beryllium sulfate to fat-
head minnows and blue gill. Ninety-six hour median lethal concentrations
were estimated to be 0. 2 mg/4 Be in soft water (20 mg/£ CaCGs hard-
ness) and 11 mg/j&Be in hard water (400 mg/£ CaCO., hardness) for
fathead minnows, and 1. 3 mg/ i Be in soft water and 12 mg/4 Be in
59
-------
hard water for blue gill. Slonim examined the relationship between
hardness and beryllium sulfate toxicity using the adult and fry stages
of guppies. Adult guppies were 100 times more tolerant in water hav-
ing 400 to 500 rng/Atotal hardness (96-hr LC50 = 27 mg/Ji Be) than in
water having only 20 to 25 mg/4 hardness (96-hr LC50 = 0. 23 mg/£Be).
Neither age nor size appeared to affect sensitivity since survival times
of fry and adults in beryllium solutions were similar (Slonim ). In
a later study using adult guppies, Slonim and Slonim examined the
relationship between beryllium toxicity and four levels of hardness.
The 96-hr LC50 estimates were 0. 16, 6.1, 13. 7 and 20. 0 mg/X Be
for 22, 150, 275 and 400 mg/£CaCO, hardness, respectively. For
96 hr toxicant exposure, the relationship between LC50 and water
hardness could be expressed by the linear regression equation: log
LC50 = - 3. 044 + 1. 706 (log hardness).
The median lethal concentrations detailed above are from 1. 5
to 7 times higher on a beryllium ion basis than those found in our tests.
For example, the 96-hr LC50 of 11 mg/Ji Be obtained for fathead
74
minnows exposed in hard water by Tarzwell and Henderson was 3. 4
times greater than the comparable LC50 of 3. 25 mg/4 Be determined
in our laboratory. Hardness appears to exert a substantial influence
on beryllium toxicity as the data of Slonim and Slonim suggests.
72 75
Calcium is known to reduce the toxicity of many metals. (Tabata ' ).
Under the intermittent-flow test conditions employed in the present
study at a hardness averaging approximately 141 mg/jfcCaCO, and pH
averaging 7. 84, two types of curves may have characterized the toxicity
of beryllium sulfate. In the toxicity test using juvenile fathead minnows,
a rectangular hyperbola, designating a median lethal threshold in the
region of 240 hr, was in evidence. In the test using goldfish, the curve
appeared to be sigmoid-shaped. Because of insufficient observations
in the tests of flagfish, a straight line was used to tentatively establish
the relationship between LC50 and beryllium toxicity in the 48 to 96-hr
exposure interval. The data indicated a steadily declining L.C50 between
60
-------
72 and 240 hr and are in contrast with the observations of Slonim
and Slonim and Slonim , who found little change in percent mor-
tality of guppies after 24 hr exposure. The difference may be due
to the use of an intermittent-flow system in our studies and a static
system without toxicant renewal in theirs. Although Slonim indi-
cated that there was little toxicant loss from the system in 96 hr,
it is possible that some beryllium precipitated and settled out, as was
demonstrated in the experiment described in the Materials and Methods
Section and in Figure 5, leaving only a small amount of dissolved ion
in the supernatant water at its solubility limit. As noted earlier,
precipitated metal enmeshed in the gill filaments probably contributes
to toxicity upon its dissolution by carbonic acid and uptake by the fish.
TOXICITY OF LEAD CHLORIDE
Lead chloride was not found to be acutely lethal in a four day period
to the two most sensitive freshwater fish tested, brook trout and fathead
minnows. Precipitation of lead occurred rapidly, probably in the form of
lead carbonate. According to Davies and Everhart , the solubility
of lead in hard water (353 mg/4 CaCOj), is of the order of 30
The solubility of lead as PbCl2 in our laboratory water was found to be
approximately 1. 5 mg/A. Because the toxicity of lead is limited by its
low solubility in natural waters, it is not considered to be a very important
pollutant in terms of acute toxicity (Aronson ) . However, the 96-hr
LC50 for lead chloride to fathead minnows has been estimated to be
2.4 mg/A Pb in soft water (20 mg/A CaCO,) and greater than 75 mg/A Pb
74
in hard water (400 mg/4 CaCO-) (Tarzwell and Henderson ), suggesting
a substantial acute toxicity in soft waters. In studies of the short-and long-
78
term effects of lead nitrate on juvenile brook trout, Dorfman and Whitworth
calculated a 96-hr LC50 of 3. 12 mg/4 Pb for much smaller specimens (8 to
10 g) than employed for our studies. They used a reconstituted, deionized
water having a pH of 6. 8 to 7. 2 and alkalinity of 27 to 34 mg/X CaCO,.
61
-------
When single doses of lead nitrate were administered daily (5 times/week)
and allowed to flush out, the level not affecting growth and survival of the
fish was between 10 and 15 mg/4 Pb. Even though fish may not experience
mortality upon acute exposure to lead in hard waters, functional distur-
bances may result. In sea water spiked with 10 mg/jC lead as lead nitrate,
79
Jackim observed a 22% decline in the activity of the enzyme, -y-aminole-
vulinate, after 96 hr in mummichogs [Fundulus heteroclitus (Linnaeus)] and a
66% decline in the enzyme's activity in winter flounder [ps eudopleur onectes
-, 875
americanus (Walbaum)j exposed for one week. Dawson employed massive
concentrations of lead acetate (100, 000 ppm) to invoke severe disturbances in
the peripheral blood and hematopoietic system of brown bullheads [ Ictalurus
nebulosus (Lesueur)] after 16 to 183 days.
62
-------
SECTION VII
REFERENCES
1. Mount, D. I. and C. E. Stephan. A Method for Establishing
Acceptable Toxicant Limits for Fish-Malathion and the
Butoxyethanol Ester of 2, 4-D. Trans. Amer. Fish. Soc.
_96^ 185-193, 1967.
2. Sprague, J. B. Measurement of Pollutant Toxicity to
Fish. I. Bioassay Methods for Acute Toxicity. Water
Res. (Oxford). 3: 793-821, 1969.
3. Sprague, J. B. Measurement of Pollutant Toxicity to
Fish. II. Utilizing and Applying Bioassay Results. Water
Res, (Oxford). 4: 3-32, 1970.
4. McKee, J. E. and H. W. Wolf. Water Quality Criteria.
Publication No. 3-A, Second Editioti. Sacramento, California
State Water Quality Control Board, 1963. 548 p.
5. Committee on Methods for Toxicity Tests with Aquatic
Organisms. Method for Acute Toxicity Tests with Fish
and Macroinvertebrates using the Static Technique.
Unpublished Manuscript, 1973. 25 p.
6. Committee on Methods for Toxicity Tests with Aquatic
Organisms. Method for Acute Toxicity Tests with Fish and
Macroinvertebrates using the Flow-Through Technique.
Unpublished Manuscript, 1973. 23 p.
7. Doudoroff, P., B. G. Anderson, G. E. Burdick, P. S.
Galtsoff, W. B. Hart, R. Patrick, E. R. Strong, E. W. Surber
and W. M. Van Horn. Bioassay Methods for the Evaluation
of Acute Toxicity of Industrial Wastes to Fish. Sewage Ind.
Wastes. 23: 1380-1397, 1951.
8. American Public Health Association. Standard Methods
for the Examination of Water and Wastewater. 13th Edition.
New York, American Public Health Association, 1971.
874 p.
9. Pomelee, C. S. Toxicity of Beryllium. Sewage Ind. Waste
25: 1424-1428, 1953.
63
-------
10. Slonim, A. R. Acute Toxicity of Beryllium Sulfate to the
Common Guppy. J. Water Pollut. Control Fed. 45:
2110-2122, 1973.
11. Slonim, C. B. and A. R. Slonim. Effect of Water Hardness
on the Tolerance of the Guppy to Beryllium Sulfate. Bull.
Environ. Contam. Toxicol. 10; 295-301, 1973.
12. Ohio River Valley Water Sanitation Commission,Committee
on Toxicities, Metal Finishing Industries Action Committee.
Report No. 3, 1950.
13. Weir, P. and C. H. Hine. Effects of Various Metals on
Behavior of Conditioned Goldfish. Archiv. Environ. Health.
2£: 45-51, 1970.
14. Becker, C. D. and T. O. Thatcher. Toxicity of Power Plant
Chemicals to Aquatic Life. Battelle, Pacific Northwest
Laboratories. Richland, Washington. Report No. WASH-1249.
United States Atomic Energy Commission. 1973.
15. Brown, V. M. Concepts and Outlook in Testing the Toxicity
of Substances to Fish. In: Bioassay Techniques and Environ-
mental Chemistry, Glass, G. E. (ed. ). Ann Arbor, Ann Arbor
Science Publishers, Inc. 1973. p. 73-95.
16. Snieszko, S. F. and G. L. Bullock. Freshwater Fish Diseases
Caused by Bacteria Belonging to the Genera Aeromonas and
Pseudomonas. U.S. Bur. Sport Fish. Wildl. Fish." Leaflet
No:"4~5~9~,~~1962, 7 p.
17. Patterson, E. B. Terramycin for the Control of Fish Diseases,
Chas. Pfizer & Co. Tech. Inform. 118, 1970, 19 p.
18. Herman, R. L. Oxytetracycline Toxicity to Trout. U, S, Bur. Sport
Fish. Wildl. Tech. Papers, No. 33, 1969, 4 p.
19. Herman, R. L. Oxytetracycline in Fish Culture - A Review.
U.S. Bur. Sport Fish. Wildl. Tech. Papers, No. 31, 1969, 9 p.
20. Wedemeyer, G. The Stress of Formalin Treatments in Rainbow
Trout (Salmo gairdneri) and Coho Salmon (Oncorhynchus
kisutchj^ JTFish. Res. Board Can. (Ottawa). 28: 1899-
T9~tF47T971. ~~
64
-------
21. Mount, D. I. and W. A. Brungs. A Simplified Dosing Apparatus
for Fish Toxicology Studies. Water Res. (Oxford). 1: 21-29,
1967. ~
22. Houston, A. H. , J. A. Madden, R. J. Woods, and H. M. Miles.
Some Physiological Effects of Handling and Tricaine
Methane sulphonate Anesthetization upon the Brook Trout,
Salvelinus fontinalis. J. Fish. Res. Board Can. (Ottawa).
• ~
23. Houston, A. H. , J. A. Madden, R. J. Woods and H. M. M:Ues.
Variations in the Blood and Tissue Chemistry of Brook
Trout, Salvelinus fontinalis , Subsequent to Handling, Anesthesia
and Surgery. J. ^Flsh. Rei". Board Can. (Ottawa). 23:
635-642, 1971.
24. Wedemeyer, G. Some Physiological Consequences of Handling
Stress in the Juvenile Coho Salmon (One orhync hus kisutch)
and Steelhead Trout (Salmo gairdner7J~."j. Fis~hT~R~eY.
Board Can. (Ottawa). ~2^T~ iTW-Tf^, 1972.
25. Environmental Protection Agency. Methods for Chemical
Analysis of Water and Wastes. Cincinnati, U. S. Environmental
Protection Agency, Analytical Quality Control Laboratory, 1971.
312 p.
26. Perkin- Elmer Corporation. Analytical Methods for Atomic
Absorption Spectrophotometry. Norwalk, Conn., Perkin- Elmer
Corporation, 1971.
27. Rudling, L. Determination of Pentachlorophenol in Organic
Tissues and Water. Water Res. (Oxford). _4: 533-537, 1970.
28. Weast, R. C. and S. M. Selby. Handbook of Chemistry and
Physics. 48th Edition. Cleveland, The Chemical Rubber
Company, 1967.
29. Litchfield, J. T. , Jr. and F. Wilcoxon. A Simplified Method
of Evaluating Dose- Effect Experiments. J. Pharmacol. Exp.
Ther. ^6: 99-113, 1949.
30. Dixon, W. J. (ed. ). BMD Biomedical Computer Programs.
Berkeley, University of California Press. 1973. 773 p.
65
-------
31. Bliss, C. I. Confidence Limits for Biological Assays.
Biometrics Bull. _!:. 57-65, 1945.
32. Steel, R. G. D. and J. H. Torrie. Principles and Procedures
of Statistics. New York, McGraw-Hill Book Company, Inc.,
1960. 481 p.
33. Lee, G. F. Chemical Aspects of Bioassay Techniques for
Establishing Water Quality Criteria. Water Res. (Oxford).
T_: 1525-1546, 1973.
34. Black, J. A., R. F. Roberts, D. M. Johnson, D. D. Minicucci,
K. H. Mancy, and H. E. Allen. The Significance of Physico-
Chemical Variables in Aquatic Bioassays of Heavy Metals.
In: Bioassay Techniques and Environmental Chemistry,
Glass, G. E. (ed. ). Ann Arbor, Ann Arbor Science
Publishers, Inc., 1973. p. 259-275.
35. Doudoroff, P., G. Leduc, and C. R. Schneider. Acute
Toxicity to Fish of Solutions Containing Complex Metal
Cyanides in Relation to Concentrations of Molecular Hydrocyanic
Acid. Trans. Amer. Fish. Soc. 95: 6-22, 1966.
36. Lange, N. A. Handbook of Chemistry. New York, McGraw-Hill
Book Company, 1967. 2001 p.
37. Everest, D. A. The Chemistry of Beryllium. Amsterdam,
The Elsevier Publishing Company, 1964. 151 p.
38. Crandall, C. A. and C. J. Goodnight. Effects of Sublethal
Concentrations of Several Toxicants on Growth of the Common
Guppy, Lebistes reticulatus. Limnol. Oceanogr. 7: 233-
239, 196T:~
39. Bostrom, S. L. and R. G. Johansson. Effects of Pentachloro-
phenol on Enzymes Involved in Energy Metabolism of the Eel.
Comp. Biochem. Physiol. (Oxford). jll_J3 : 359-369, 1972.
40. Hanes, D. , H. Krueger, I. Tinsley and R. Lowry. Penta-
chlorophenol Effects on Growth and Fatty Acids of Coho
Salmon. Pharmachologist. 9: 207, 1967.
66
-------
41. Hanes, D. , H. Krueger, I. Tinsley and R. Lowry. Alterations
of Fatty Acid Metabolism in Coho Salmon by Diet and
Pentachlorophenol. Pharmacologist. 10: 337, 1968.
42. Saddler, J. B. , K. V. Koski and R. D. Cardwell. Fatty Acid
Alternations during Migration and Early Sea Water Growth
of Chum Salmon (Oncorhynchus keta). L/ipids. 7: 90-95,
1972< ~
43. Webb, P., W. and J. R. Brett. Effects of Sublethal Concentrations
of Sodium Pentachlorophenate on Growth Rate, Food Conversion
Efficiency, and Swimming Performance in Underyearling
Sockeye Salmon (Oncorhynchus nerka). J. Fish. Res. Board
Can. (Ottawa). 3p:~~399-507, 1"9?3T "
44. Norup, B. Toxicity of Chemical In Paper Factory Effluents.
Water Res. (Oxford), b: 1585-1588, 1972.
45. Bandt, H. J. and D. Nehring. Die Toxicologische Wirkung von
Pentachlorophenol-Natrium and Hexachlorophen auf Fische
und Niedere Wassertiere. Z. Fisch. Deren Hilfswiss.
(Berlin). JU): 5435-5438, 1962.
46. Goodnight, C. J. , Toxicity of Sodium Pentachlorophenate
and Pentachlorophenol to Fish. Ind. Eng. Chem. 34: 868-872,
1942.
47. Crandall, C. A. and C. J, Goodnight. The Effect of Various
Factors on the Toxicity of Sodium Pentachlorophenate. Limnol.
Oceanogr. _4: 53-56, 1959.
48. Alabaster, J. S. Toxicity of Weedkillers, Algicides and Fungicides
to Trout. Proc. 4th Brit. Weed Control Goaf. , Society of Chemical
Industry (London), I960, p. 1-20.
49. Blackman, G. E. The Physiological Activity of Substituted
Phenols II. Relationship Between Physical Properties and
Physiological Activity. Arch. Biochem. Biophys. 54:
55-71, 1955.
50. Wuhrmann, K. Sur Quelques Principes de la Toxicologie du
Poisson. Bull. Cent, beige Docum. Eaux. (Liege) No. 15:
49, 1952. Fish. Res. Board of Can. Transl. Ser. No. 243,
1959.
67
-------
51. Doudoroff, P. Some Experiments on the Toxicity of
Complex Cyanides to Fish. Sewage Ind. Wastes. 28:
1020-1040, 1946. '
52. Henderson, C. , Q. H. Pickering and A. E. Lemke. The
Effect of Some Organic Cyanides (Nitriles) on Fish. Proc.
15thlnd. Waste Conf., Purdue Univ. 45: 120-130, 1961.
53. Cairns, J., Jr. Environment and Time in Fish Toxicity.
Ind. Wastes. 2: 1-4, 1957.
54. Cairns, J. , Jr. and A. Scheier. The Effects of Periodic Low
Oxygen Upon the Toxicity of Various Chemicals to Aquatic
Organisms. Proc. 12th Ind. Waste Conf., Purdue Univ. 42:
165-167, 1958.
55. Cairns, J., Jr., and A. Scheier. The Relationship of Bluegill
Sunfish Body Size to Tolerance for Some Common Chemicals.
Proc. 13th Ind. Waste Conf., Purdue Univ. 43: 243-252,
1959.
56. Cairns, J. , Jr. and A. Scheier. Environmental Effects upon
Cyanide Toxicity to Fish. Notulae Natur. No. 361, 1963,
Up.
57. Engberg, R. A. Selenium in the Water Resources of Nebraska
in Comparison to Public Health Standards. Amer. Water Res.
Assoc. Publ., Proc. Ser. No. 16: 1-2, 1972.
58. Rosenfeld, I. and O. A. Beath. Selenium Geobotany, Bio-
chemistry, Toxicity and Nutrition. New York, Academic
Press, 1964. 411 p.
59. Ellis, M. M. , H. L. Motley, M. D. Ellis and R. O. Jones.
Selenium Poisoning in Fishes. Proc. Soc, Exp. Biol. Med. 36:
519-522, 1937.
60. Doudoroff, P. and M. Katz. Critical Review of Literature on
the Toxicity of Industrial Wastes and their Components to
Fish. I. Alkalies, Acids and Inorganic Gases. Sewage Ind.
Wastes. 28: 1432-1458, 1950.
68
-------
61. European Inland Fisheries Advisory Commission. Water
Quality Criteria for European Freshwater Fish - Extreme pH
Values and Inland Fisheries. Water Res. (Oxford). 3:
593-611, 1969.
62. Mount, D. I. Chronic Effect of Low pH on Fathead Minnow
Survival, Growth and Reproduction. Water Res. (Oxford).
7_: 987-998, 1973.
63. Jordon, D. H. M. and R. Lloyd. The Resistance of Rainbow
Trout (Salmo gairdneri Richardson) and Roach (Rutilus rutilus
L.) to Alkaline Solutions. Int. J. Air Water Pollut. (London)
8; 405-409, 1964.
64. Cairns, J., Jr., and A. Scheier. The Relation of Bluegill
Sunfish Body Size to Tolerance for Some Common Chemicals.
Ind. Wastes. J3: 126, 1958.
65. Jones, J. R. E. Fish and River Pollution. In: River
Pollution II. Causes and Effects, Kline, L. (ed. ). London,
Butterworths, 1962, p. 254-310.
66. Grindley, J. Toxicity to Rainbow Trout and Minnows of Some
Substances Known to be Present in Waste Water Discharged
to Rivers. Ann. Appl. Biol. (London) 33: 102-112, 1946.
67. Holland, G. A., J. E. Lasater, E. D. Neumann, and W. E.
Eldridge. Toxic Effects of Organic and Inorganic Pollutants
on Young Salmon and Trout. Wash. State Dept. Fish. Res.
Bull. No. 5, I960, 264 p.
68. Clemens, H. P. and K. E. Sneed. Lethal Doses of Several
Commercial Chemicals for Fingerling Channel Catfish.
U.S. Fish Wildl. Serv. Spec. Sci. Rep. Fish. No. 316, 1959,
10 p.
69. Gilderhus, P. A. Some Effects of Sublethal Concentrations
of Sodium Arsenite on Bluegills and the Aquatic Environment.
Trans. Amer. Fish. Soc. 95: 289-296, 1966.
70. Lloyd, R. and D. W. M. Herbert. The Influence of Carbon
Dioxide on the Toxicity of Unionized Ammonia to Rainbow
Trout (Salmo gairdneri Richardson). Ann. Appl. Biol.
(London) 48: 399-404, I960.
69
-------
71. Mount D. I. The Effect of Total Hardness and pH on Acute
Toxicity of Zinc to Fish. Int. J. Air Water Pollut. (London)
_10: 49-56, 1966.
72. Tabata, K. Studies on the Toxicity of Heavy Metals to Aquatic
Animals and the Factors to Decrease the Toxicity - I. On the
Formation and the Toxicity of Precipitate of Heavy Metals.
Bull. Tokai Reg. Fish. Res. Lab. (Tokyo). No. 58:
203 - 214, 1969. -
73. Freeman, R. A. and W. H. Everhart. Toxicity of Aluminum
Hydroxide Complexes in Neutral and Basic Media to Rainbow
Trout. Trans. Amer. Fish. Soc. 100: 644-658, 1971.
74. Tarzwell, C. M. and C. Henderson. Toxicity of Less Common
Metals to Fishes. Ind. Wastes, _5: 12, I960.
75. Tabata, K. Studies on the Toxicity of Heavy Metals to Aquatic
Animals and the Factors to Decrease the Toxicity - II. The
Antagonistic Action of Hardness Components in Water on the
Toxicity of Heavy Metal Ions. Bull. Tokai Reg. Fish. Res. Lab.
(Tokyo). No. 5J3: 215-232, 1969.
76. Davies, P. H. and W. H. Everhart. Effects of Chemical
Variations in Aquatic Environments: Volume III. Lead
Toxicity to Rainbow Trout and Testing Application Factor
Concept. U.S. Environmental Protection Agency Ecol. Res.
Ser. No. EPA-R3-73-011C, 1973, 80 p.
77. Aronson, A. L. Biologic Effects of Lead in Fish. J. Wash.
Acad. Sci. 6^: 124-128, 1971.
78. Dorfman, D. and W. R. Whitworth. Effects of Fluctuations
of Lead, Temperature and Dissolved Oxygen on the Growth
of Brook Trout. J. Fish. Res. Board Can. (Ottawa).
_26: 2493-2501, 1969.
79. Jackim, E. Influence of Lead and Other Metals on Fish
Y - Aminolevulinate Dehydrase Activity. J. Fish. Res.
Board Can. (Ottawa). _3J); 560-562, 1973.
80. Dawson, A. The Hemopoietic Response in the Catfish,
Ameiurus nebulosus, to Chronic Lead Poisoning.
Biol. Bull. 68: 335-346, 1935.
70
-------
SECTION VIII
APPENDIX
Appendix Table No. Page
1. Characteristics of Test Species 75
Exposed to Sodium Pentachlorophenate
2. Characteristics of Fish Exposed to 76
Sodium Cyanide
3. Characteristics of Fish Exposed to 77
Selenium Dioxide
4. Characteristics of Fish Exposed to 78
Sodium Arseiiite
5. Characteristics of Fish Exposed to 79
Beryllium Sulfate
6. Reproducibility and Accuracy of 80
Chemical Analyses of Selenium
Dioxide.
7. Reproducibility and Accuracy of 81
Chemical Analyses of Sodium
Arsenite
8. Reproducibility and Accuracy of 82
Chemical Analyses of Beryllium
Sulfate
9. Reproducibility and Accuracy of 83
Chemical Analyses of Lead Chloride
10. Reproducibility and Accuracy of 84
Chemical Analyses of Sodium Cyanide
71
-------
Appendix Table Page
11. Reproducibility and Accuracy of 85
Chemical Analyses of Sodium
Pentachlorophenate
12. Measured Concentrations of Sodium 86
Pentachlorophenate (NaPCP) in
Toxicity Tests
13. Measured Concentrations of Sodium 87
Cyanide (NaCN) in Toxicity Tests
14. Measured Concentrations of Selenium 88
Dioxide (SeGO in Toxicity Tests
15. Measured Concentrations of Sodium 89
Arsenite (NaAsO?) in Toxicity Tests
16. Measured Concentrations of Beryllium 90
Sulfate (BeSOJ in Toxicity Tests
17. Standard Concentration-Per cent 91
Mortality Data Supplied by Committee
on Methods for Toxicity Tests with
Aquatic Organisms
18. Comparison of Median Lethal 92
Concentrations and 95% Confidence
Limits with Other Aquatic Toxicology
Laboratories
19. Median Lethal Concentrations (LC50) 93
for Adult Brook Trout Exposed to
Sodium Pentachlorophenate
20. Median Lethal Concentrations (LC50) 94
for Juvenile Fathead Minnows Exposed
to Sodium Pentachlorophenate
21. Median Lethal Concentrations (LC50) 95
for Goldfish Exposed to Sodium
Pentachlorophenate
72
-------
Appendix Table
22. Median Lethal Concentrations (LC50) 96
for Bluegill Exposed to Sodium
Pentachlorophenate
23. Median Lethal Concentrations (LC50) 97
as CN~ for Fathead Minnows Exposed
to Sodium Cyanide
24. Median Lethal Concentrations (LC50) 98
as CN~ for Bluegill Exposed to
Sodium Cyanide
25. Median Lethal Concentrations (LC50) 99
as CN~ of Adult Brook Trout Exposed
to Sodium Cyanide
26. Median Lethal Concentrations (LC50) 100
as CN~ for Channel Catfish Exposed
to Sodium Cyanide
27. Median Lethal Concentrations (LC50) 101
as CN~ for Juvenile Goldfish Exposed
to Sodium Cyanide
28. Median Lethal Concentrations (LC50) 102
for Juvenile Fathead Minnows Exposed
to Selenium Dioxide
29. Median Lethal Concentrations (LC50) 103
for Adult Brook Trout Exposed to
Selenium Dioxide
30. Median Lethal Concentrations (LC50) 1°4
for Juvenile Channel Catfish Exposed
to Selenium Dioxide
31. Median Lethal Times (LT50) for 105
Juvenile Flagfish Exposed to Selenium
Dioxide (SeO2)
73
-------
Appendix Table_ Page
32. Median Lethal Concentrations (LC50) 106
for Juvenile Goldfish Exposed to
Selenium Dioxide
33. Median Lethal Concentrations (LC50) 107
for Juvenile Bluegill Exposed to
Selenium Dioxide
34. Median Lethal Concentrations (LC50) 108
for Adult Brook Trout Exposed to
Sodium Arsenite
35. Median Lethal Concentrations (LC50) 109
for Juvenile Fathead Minnows Exposed
to Sodium Arsenite
36. Median Lethal Concentrations (LC50) HO
for Juvenile Channel Catfish Exposed
to Sodium Arsenite
37. Median Lethal Concentrations (LC50) H1
for Juvenile Goldfish Exposed to Sodium
Arsenite
38. Median Lethal Concentrations (LC50) 112
for Flagfish Fry Exposed to Sodium
Arsenite
39. Median Lethal Concentrations (LC50) l1^
for Juvenile Bluegill Exposed to Sodium
Arsenite
40. Median Lethal Concentrations (LC50) 114
for Juvenile Fathead Minnows Exposed
to Beryllium Sulfate
41. Median Lethal Concentrations (LC50) H5
and Median Lethal Times (LT50) for
Flagfish Fry Exposed to Beryllium Sulfate
42. Median Lethal Concentrations (LC50) for H&
Juvenile Goldfish Exposed to Beryllium
Sulfate
74
-------
Appendix Table 1 . CHARACTERISTICS OF TEST SPECIES
EXPOSED TO SODIUM PENTACHLOROPHENATE
en
Approximate Develop- Test specimen Wet body
Species age at testing, mental density, Total length, weight,
months stage g fish/ SL mm g
a d
Fathead minnow 3 J 0. 06 22. 7
+3.2
Bluegill 12 J 3.15 70.4
+7.2
Bluegill 12 J 0. 50 43. 6
+6. 1
Brook trout 18 Ab 25.25 218.0
+16.0
Goldfish 12 A° 1.70 65.9
+12.4
Goldfish 12 A° 4.24 80.5
+14.6
Juvenile
b. , ,
d
0. 098
+0. 047
5.2
+1.6
0.84
+0. 49
101. 0
+21.0
2.8
+1.3
7.0
+3.0
cAdult
Means jf standard deviation are given
-------
Appendix Table 2. CHARACTERISTICS OF FISH
EXPOSED TO SODIUM CYANIDE
-j
a>
Approximate
Species age at testing, Developmental
months stage
Fathead 3 Ja
minnow
Goldfish 6 J
Brook trout 18 A
Channel 6 J
catfish
Bluegill 6 J
Juvenile.
bAdult.
1
Test specimen Total.
density, length,
g fish/4 mm
0. 12 28. 1 °
+ 4.4
1.76 54.3
+ 7.9
20.33 211.0
+11.0
0.69 51.8
+ 7.1
0.12 23.1
+ 4. 4
Wet body
weight,
g
0. 205°
+0. 105
2.9
+1.3
81.3
+16.2
1.14
+0.57
0.20
+0. 10
'Means + standard deviation are given.
-------
Appendix Table 3 • CHARACTERISTICS OF FISH
EXPOSED TO SELENIUM DIOXIDE
Species
Fathead
minnow
Fathead
minnow
Flagfish
Brook
trout
Channel
catfish
Goldfish
Bluegill
Approximate
age at testing,
months
(1 day)
3
2
18
6
6
6
Develop-
mental
stage
yolk- sac
fry
Ja
J
Ab
J
J
J
Test specimen
density,
g fish/ji
0. 05
0. 04
24.90
1.45
1.45
2.42
Total length,
mm
5.0C
+ 1.0
20.7
+ 3.0
+ 14.7
- 2.9
211. 0
+13.0
64.5
+ 6.9
62.0
+ 4.9
65.3
+ 6.4
Wet body
weight,
g
0.085
+0. 045
,0.059
-0. 039
99.6
+18. 5
2.4
+0.8
2.4
+0.5
4. 0
+ 1.4
Juvenile.
Adult, but not sexually mature. Gonads of male trout comprised 0. 0096 mg/g body
weight, while those of female trout comprised 0. 0288 mg/g.
c
Means + 1 standard deviation are given.
-------
Appendix Table 4 . CHARACTERISTICS OF FISH
EXPOSED TO SODIUM ARSENITE
CO
Species
Goldfish
Fathead
minnow
Brook
trout
Bluegill
Channel
catfish
Approximate
age at testing,
months
6
3
18
6
6
Develop-
mental
stage
Ja
J
Ab
J
J
Test specimen
density,
g fish/4
0. 15
0. 02
21. 18
1. 27
1.46
Total length,
mm
62. Oc
+ 4.9
21.0
+ 2.7
200.0
+14.0
51.8
+ 7.7
66.4
+ 8.7
Wet body
weight ,
g
2.4C
+0.5
0.085
+0. 035
84.7
+H. 0
2. 1
+ 1. 1
2.4
+0.8
Juvenile.
bAdult.
CMeans + 1 standard deviation are given.
-------
Appendix Table 5 . CHARACTERISTICS OF FISH EXPOSED
TO BERYLLIUM SULFATE
Species
Flagfish
Fathead
minnow
Goldfish
Channel
catfish
Brook
trout
Approximate
age at testing,
months
(1-4 days)
3
6
6
18
Develop-
mental
stage
yolk- sac
fry
Ja
J
J
Ab
Test specimen
density, Total length,
g fish/A mm
c
3.0
+ 0.5
0.08 25.4
+ 2.9
1. 52 64. 3
+ 7.0
1.27 63.9
+ 6.4
31.00 229.0
^10.0
Wet body
weight,
g
0. 139°
+0. 050
2.5
+1.1
2. 1
+0.6
124. 0
+22. 0
Juvenile.
bAdult.
'Means J- 1 standard deviation are given.
-------
Appendix Table 6 . REPRODUCIBILITY AND ACCURACY
OF CHEMICAL ANALYSES OF SELENIUM DIOXIDE
Absorbance
of spiked
concentrations of
as Se (mg/A )a
selenium
dioxide
„ b . Standards
Replicate ****£
1 0.0000
2
3
4
5
Mean 0. 0000
Standard
deviation
Standard
error
Coefficient of
variation, %
5C
0.0168
0.0168
0.0168
0.0173
0.0170
0.0169
0.0002
0.0001
1.1
5 10
0.0168 0.0313
0.0170 0.0301
-
-
-
0.0169 0.0307
-
-
15
0. 0434
0.0432
-
-
-
0.0433
.
-
20
0.0550
0.0558
-
-
-
0. 0554
-
-
aRead at a wavelength of 196 nm.
Distilled water.
GSpiked into filtered (Whatman No. 1 filter paper) water collected
from tank containing control fish.
80
-------
Appendix Table 7 . REPRODUCIBILITY AND ACCURACY OF
CHEMICAL ANALYSES OF SODIUM ARSENITE
Absorbance of
Reagent
Replicate blank
1 0.0000
2
3
4
5
Mean 0. 0000
Standard
deviation
Standard
error
Coefficient of
variation, %
spiked concentration of sodium arsenite
As (rag/* )a
Standards
2C 25 10
0.0057 0.0054 0.0139 0.0278
0.0061 - -
0.0057
0.0054 - - -
0.0054 - -
0.0057 0.0054 0.0139 0.0278
0.0003 -
0.0001 -
5.3
as
20
0.0562
-
-
-
-
0.0562
-
-
-
Read at a wavelength of 193. 7 nm.
Distilled water.
CSpiked into filtered (Whatman No. 1 filter paper) water collected from
tank containing control fish.
81
-------
Appendix Table 8 . REPRODUCIBILITY AND ACCURACY OF
CHEMICAL ANALYSES OF BERYLLIUM SULFATE
Absorbance of spiked concentrations of beryllium sulfate
as Be (mg/£ )a
_ ,. . Reagent
RepUcate ^^
1 0.0000
2 -
3
4
5
Mean 0. 0000
Standard
deviation
Standard
error
Coefficient of
variation, %
1.0C
0. 1249
0. 1249
0. 1249
0. 1238
0. 1232
0. 1242
0.00074
0.00033
0.59
Standards
0.5 1.0
0.0610 0.1249
. -
-
-
-
0.0610 0.1249
_ _
-
- _
Read at a wavelength of 234. 9 nm.
Distilled water.
CSpiked into filtered (Whatman No. 1 filter paper) water collected
from tank containing control fish.
82
-------
Appendix Table 9 . REPRODUCIBILITY AND ACCURACY OF
CHEMICAL ANALYSES OF LEAD CHLORIDE
oo
00
Replicate
1
2
3
4
5
Mean
Standard
deviation
Standard
error
Coefficient of
variation, %
Absorbance of spiked concentrations
Jxcagcnc „
blank 5 5
0.0004 0.0379 0.0362
0.0379
0.0379
0.0372
0.0372
0.0004 0.0376 0.0362
0.0004
0.0001
1.1
of lead chloride as Pb (mg/4 )
Standards
10 15 20
0.0716 0.1073 0.1427
_
-
_
- - -
0.0716 0.1073 0.1427
-
-
.
*Read at a wavelength of 283. 3 nm.
Distilled water.
cSpiked into filtered (Whatman No. 1 filter paper) water collected from tank
containing control fish.
-------
Appendix Table 10. REPRODUCIBILITY AND ACCURACY OF
CHEMICAL ANALYSES OF SODIUM CYANIDE
00
Replicate
1
2
3
4
5
Mean
Standard
deviation
Standard
error
Coefficient of
variation, %
Absorbance of spiked concentrations of sodium cyanide as CN" (
Standard
Blank" Control" 5" 5 10 15
0.000 0.003 0.123 0.118 0.264 0.399
0.130 -
0.133 -
0.131 -
0. 131
0.000 0.003 0.130 0.118 0.264 0.399
0.004 -
0.002 - - -
3.1
Ug/A )a
20
0.575
-
-
-
-
0.575
-
-
aRead at a wavelength of 620 nm.
b
Distilled water.
cWater taken from tank containing control fish.
dSpiked into filtered (Whatman No. 1 filter paper) water collected from tank
containing control fish.
-------
Appendix Table 11. REPRODUdBILITY AND ACCURACY OF
CHEMICAL ANALYSES OF SODIUM PENTACHLOROPHENATE
Peak height, mm
Reagenta
Replicate blank
1 0
2
3 -
4
5
Mean 0
Standard
deviation
Standard
error
Coefficient of
variation, %
0. 2 mg£ji
Control Spike
0 110.6
101.2
101.3
100.1
94.4
0 101.52
5.82
2.60
5.73
0.2 mg/JL
Standard
101.1
-
-
-
-
101.1
-
-
-
Distilled water.
Spiked into filtered (Whatman No. 1 filter paper) water collected
from tank containing control fish.
85
-------
00
05
Appendix Table 12. MEASURED CONCENTRATIONS OF
SODIUM PENTACHLOROPHENATE (NaPCP) IN TOXICITY
TESTS
Measured NaPCP concentration, mg/£
Fish No. measure- Tank Tank
Species ments/tank 1 2
Brook trout 3 Oa 0.091
+0.014
Fathead minnow 3 0 0. 125
+ 0. 050
Goldfish
Test 1 3 0 0. 082
+0. Oil
Test 2 1 0 ---
Bluegill
Test 1 2 0 0. 313
+ 0. 007
Test 2 000. 075
Tank
3
' 0
±°
0
+ 0
0
±°
-
0
±°
0
. 110
. 023
.212
. 040
. 107
. 006
-.-
.375
.023
. 100
Tank
4
0. 162
+0. 057
0.260
-1-0. 102
0. 156
-1-0, 002
0.432
0. 502
+0. 099
0. 133
Tank
5
0.
1°.
0.
+ 0.
0.
+0.
0.
0.
0.
230
042
408
094
188
004
650
719
030
178
Tank
6
0.
+0.
0.
+ 0.
0.
+ 0.
0.
0.
0.
347
046
581
141
232
019
921
810
018
237
No pentachlorophenate detected.
Means + 1 standard deviation are given.
-------
Appendix Table 13.
MEASURED CONCENTRATIONS OF SODIUM CYANIDE (NaCN)
IN TOXICITY TESTS
CO
Measured concentration of NaCN, mg/£
No. measure
Fish species ments/tank
Fathead minnow
Test 1 3
Test 2 1
Bluegill 1
Brook trout
Test 1 3
Test 2 4
Channel catfish 2
Goldfish
Test 1 4
Test 2 2
Tank
1
Oa
-
0
0
0
.
0
-
0
'
0
-
0
-
Tank
2
0. 077b
+0.015
0.268
0. 160
0. 16
+_0. 02
0.093
+0. 054
0.146
+_0. 083
0.293
+0. 1 14
0.616
+0. 020
Tank
3
0. 100
+_0. 023
0.342
0.184
0.22
+_0. 04
0. 178
+0. 015
0.240
+0. 047
0.383
+0. 165
0.817
+0. 090
Tank
4
0. 146
+0. 044
0.424
0.252
0.28
+ 0.03
0.224
+0. 009
0.254
+0. 066
0.480
+0. 196
1.12
+.0. 04
Tank
5
0. 173
+0.061
0.572
0.364
0.39
+.0.05
0.306
+0. 02 1
0.458
+_0. 018
0.634
+0. 217
1.34
+0. 02
Tank
6
0.245
+0. 056
0.783
0.440
0.49
+_0. 03
0.397
+0. 042
0.577
+0. 064
0.840
+ 0. 352
2. 13
+_0. 01
No cyanide detected.
Me ana +_ 1 standard deviation are given.
-------
Appendix Table 14 . MEASURED CONCENTRATIONS OF SELENIUM DIOXIDE (SeO2)
IN TOXICITY TESTS
00
00
Measured SeO2 concentration, mg/
Fish Species
Fathead minnow
Test 1
Test 2
Brook trout
Channel catfish
Flagfish
Goldfish
Test 1
Test 2
Bluegill
Test 1
Test 2
No. measure-
ments/tank
3
2
2
2
2
5
2
2
4
Tank
1
oa
0
0
0
0
0
0
0
0
Tank
2
+ l.S
17.6
25.6
+_ 0. 4
17.9
11.2
+_ 1.2
7.0
+ 1.4
38.8
+_ 1.6
- - - - .
8.8
+ 1.3
Tank
3
4.2
+ 1.5
22.4
+ 5.4
33.7
+_ 0. 8
25.4
+_ 4. 8
16.9
+ 1.6
10.4
+ 1.3
51.7
+_ 1.6
_ _ .. _
11.9
+_ 1.2
Tank
4
5.6
+ 1.8
29. 7
+ 7.4
51. 1
+_ 4. 0
36. 5
+_ 0
21.8
+ 2.2
14.0
+ 0.4
72. 0
+_ 0. 8
70.9
+ 9.4
18.2
+_ 3.2
Tank
5
8.2
+ 1.7
40. 1
+_ 8.2
67.6
+_8. 3
45.4
+_ 0. 6
27.9
+ 2. 3
20.0
+ 3.2
86. 3
+ 1.2
95.4
+ 2.6
23.9
+_ 2.2
SL
Tank
6
11.2
+ 2. 1
54. 1
+_16. 8
102.2
+_ 7. 1
63.2
+ 0
37. 6
+ 0.2
33.0
+ 4.4
114. 1
+ 0
133.0
+ 0.3
33. 1
+ 2.9
No selenium detected.
Means +_ 1 standard deviation are given.
-------
Appendix Table 15. MEASURED CONCENTRATIONS OF SODIUM
ARSENITE (NaAsO2) IN TOXICITY TESTS
Measured concentration of NaAsO_,
Fish Species No. measure-
ments/tank
Brook trout
Test 1 4
Test 2 2
Fathead minnow
Test 1 4
g Test 2 1
Channel catfish 2
Goldfish
Test 1 3
Test 2 1
Flagfish 2
Bluegill 3
Tank
1
oa
-
0
-
0
-
0
0
"
0
•
0
0
-
0
~
Tank
2
8.
±3'
27.
+ 1.
6.
±2-
25.
19.
+ 1.
21.
+2.
66.
19.
±L
16.
+ 2.
6b
5
7
1
7
4
5
8
4
4
3
6
8
4
3
6
Tank
3
10
±3
38
±1
8
4-1
33
25
±°
24
±3
90
25
•1-0
23
+ 2
.6
. 5
.4
.8
.8
.2
. 7
. 1
.4
.6
.3
.8
. 1
.4
.9
.2
Tank
4
15
±3
42
±2
13
-1-2
44
41
±°
32
4-2
129
41
±°
31
+ 0
. 5
.3
.4
.3
. 3
.0
. 1
. 5
.4
. 1
.6
.7
.5
.4
.6
. 7
mg/je.
Tank
5
19.
+2.
71.
jJL
17.
±3-
59.
51.
I1-
40.
I1'
182.
51.
I1'
41.
+ 2.
9
8
5
7
3
1
9
7
6
1
2
6
7
6
5
4
Tank
6
25.
±3'
84.
±2<
25.
±4-
82:
72.
±8'
57.
+2.
272.
72.
+8.
67.
+ 2.
9
1
7
9
6
3
1
4
9
4
9
0
4
9
9
5
arsenic detected.
3Means + 1 standard deviation are given.
-------
Appendix Table 16. MEASURED CONCENTRATIONS OF BERYLLIUM
SULFATE (BeSO4) IN TOXICITY TESTS
Measured BeSO. concentration, mg/£
CD
0
Fish
species
Fathead
minnow
No. measure- Tank Tank Tank
ments/tank 123
3 Oa 18. 8b 25.1
+ 2. 5 + 2. 7
Tank
4
28.6
+ 2. 0
Tank
5
37. 6
+ 2. 3
Tank
6
47.8
+ 2.2
Flagfish 3 0 18.8 25.1 28.6 37.6 47.8
±2'5 ±.2.7 +_2.0 £2.3 £2.2
Goldfish 3 0 18.8 24.6 36.8 46.9 59.0
+ 3.0 +0.7 + 4i 6 +3.9 +13.0
aNo beryllium detected.
Means + 1 standard deviation are given.
-------
Appendix Table 17. STANDARD CONCENTRATION -
PERCENT MORTALITY DATA SUPPLIED BY COMMITTEE ON
METHODS FOR TOXICITY TESTS WITH AQUATIC ORGANISMS
Percent Mortality
Toxicant Concentrations, ng/J&
Data
Set
A
B
C
D
E
Control
0
0
0
0
0
7.8
0
0
0 .
0
0
13
0
0
0
0
0
22
10
70
10
20
20
36
100
100
40
70
30
60
100
100
100
100
100
100
100
100
100
100
100
91
-------
Appendix Table 18. COMPARISON OF MEDIAN LETHAL CONCENTRATIONS AND 95%
CONFIDENCE LIMITS WITH OTHER AQUATIC TOXICOLOGY LABORATORIES
co
CO
Our laboratory
Data Litchfield and Wilcoxon (1949}
set LC50,
Mg/*
A 25.4
B 21.2
C 35.5
D 29.8
E 35.5
95% Confidence
limits
22.
18.
28.
23.
27.
1
8
2
9
1
for LC50
- 29.
- 23.
- 44.
- 37.
- 46.
2
9
7
2
9
Computer program <
Average of eight
5ther laboratoriesa
LC50, 95% Confidence LC50. 95% Confidence
tag/A limits for LC50 ng.
25.
+1.
20.
+0.
35.6 30. 3-41.7 35.
+2.
29.5 25.0 - 34.7 29.
+0.
35.5 29. 1 - 43.3 36.
+3.
It limits
3
9
4
9
3
4
4
8
5
4
20.
+5.
14.
+4.
27.
+3.
23.
+1.
28.
±3-
4
1
8
0
5
3
5
9
2
5
for LC50
- 32.
+4.
- 49.
+ 57.
- 45.
+3.
- 36.
+1.
- 46.
+4.
0
8
3
0
1
6
9
2
8
3
Results obtained through use of both manual and computer methods.
Means + 1 standard deviation are given for concentrations in pg/4.
^
'Calculations not made since computer was not programmed to handle data having only
one partial kill.
-------
Appendix Table 19. MEDIAN LETHAL CONCENTRATIONS
(LC50) FOR ADULT BROOK TROUT
EXPOSED TO SODIUM PENTACHLOROPHENATE
Exposure
time,
hr
24
32
48
72
96
152
219
336
LC50,
mg/je
0.315
0.230
0. 180
0. 153
0. 138
0.128
0. 118
0.118
95% Confidence
limits for LC50,
mg/je
0.294
0.215
0. 168
0.143
0.129
0. 119
0. 110
0. 110
- 0.337
- 0. 246
- 0. 193
- 0. 164
- 0. 148
- 0. 138
- 0. 126
- 0. 126
Slope
function
1.07
1.07
1.07
1.07
1.07
1.08
1.07
1.07
All median lethal concentrations calculated according to
29
Litchfield and Wilcoxon . The slope function, S, is the
antilogarithm of the standard deviation of the population
tolerance frequency distribution.
93
-------
Appendix Table 20. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR JUVENILE FATHEAD MINNOWS
EXPOSED TO SODIUM PENTACHLOROPHENATE
CD
Exposure
time,
hr
8
12
21
72
96
120
173
216
240
288
336
LC50,
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
469
358
343
309
285
276
255
235
217
185
153
95% Confidence
limits for LC50,
mg/A
0.437
0. 334
0.315
0.285
0.267
0.258
0.237
0.210
0.194
0. 155
0.123
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
- 0.
504
383
374
336
305
295
274
263
243
221
188
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
o.
a
a
0721
0797
0872
0953
0769
0758
0844
1315
1307
1772
1680
Log-probit ,
regression equation
9.
10.
10.
10.
12.
12.
12.
9.
10.
9.
9.
56 +
61 +
33 +
34 +
09 +
38 +
04 +
78 +
08 +
13 +
86 +
13.87
12. 55
11.47
10.49
13.01
13.20
11.85
7.60
7.65
5.64
5.95
(logx^)
(logx.)
(logx.)
(logx^
(logXj)
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
Standard deviation of the logarithm of the population tolerance
frequency distribution.
Probit yj = a + b (logx.).
-------
Appendix Table 21. MEDIAN LETHAL CONCENTRATIONS
(LC50) FOR GOLDFISH EXPOSED TO SODIUM PENTACHLOROPHENATE
CO
Ul
Exposure
time, LC50,
hr mg/jfc
21
46
120
168
268
336
0.369
0.270
0.253
0.241
0.202
0. 189
95% Confidence
limits for LC50,
mg/Jfc
0.336
0.246
0.231
0.219
0.171
0. 162
- 0.406
- 0.297
- 0.277
- 0.265
- 0.239
- 0.219
*a
a
0. 1100
0.1099
0.0910
0.0961
0.2180
0. 1955
Log -pr obit
regression
equation
8.94 + 9.09 (logx.)
10.18 + 9. 10 (logx.)
11.56 +11.00 (logx.)
11.43 +10.41 (logx.)
8.19 +4.59 (logx.)
8.71 + 5. 12 (logx.)
' Standard deviation of the logarithm of the population tolerance
frequency distribution •
Probit y^ - a + b (logx.).
-------
Appendix Table 22. MEDIAN LETHAL CONCENTRATIONS (LC50) FOR
BLUEGILL EXPOSED TO SODIUM PENTACHLOROPHENATE
Exposure LC 50, 95% Confidence
time, mg/jfc limits for LC50,
hr mg /£
Log - probit regression
equation"
0.828 0.798-0.860
6.5 0.719 0.691-0.749
0.0372
0. 0403
7.21 + 26.90 (log
8. 56 + 24.80 (log
9.5 0.534 0.497-0.574
0.0716
8. 80 + 14. 00 (log x.)
CO
O»
30
0.303 0.283-0.324
0.0564
14.20 + 17.73 (log x^
243
0.251 0.222-0.284
0.1236
9.86 + 8.09 (log x
313 0.226 0.196-0.261
0.1443
9.48 + 6.93 (log x^
390
406
0.207 0.170-0.252
0.188 0.162-0.218
0.1971
0.1470
8.47 + 5.07 (log X.)
9.95 + 6.80 (log x.)
Standard deviation of the logarithm of the population tolerance frequency distribution.
Probit yi = a+b (log x i).
-------
CO
Appendix Table 23. MEDIAN LETHAL CONCENTRATIONS (LC50)
AS CN" FOR FATHEAD MINNOWS EXPOSED TO SODIUM CYANIDE
Exposure
time,
hr
1.5
2.5
3.0
6.0
33.0°
120.0
144.0
192.0
240.0
LC50,
nag /A
0.326
0.216
0. 196
0. 181
0.142
0.120
0. 117
0..114
0.114
95% confidence
limits for LC50,
mg/jfc
0.299
0.208
0. 189
0. 174
0.106
0. 104
0. 102
0.102
- 0. 356
- 0.225
- 0.205
- 0. 187
- 0.136
- 0.131
- 0. 127
- _0. 127
^a
rr
0. 0870
0. 0449
1.0710
1.0640
0. 1255
0.1156
0. 1066
0. 1066
Log-probit ,
regression equation
10.59
19.81
12.33
13.07
13.84
13.84
+ 11.49 (logx.)
+ 22. 25 (logx^
+ 7.97 (logx.)
+ 8. 65 (logx.)
+ 9. 38 (log x.)
+ 9. 38 (logx.)
aStandard deviation of the logarithm of the population tolerance frequency distribution.
bProbit yi = a + b
cMedian lethal time.
-------
co
oo
Appendix Table 24. MEDIAN LETHAL, CONCENTRATIONS (LC50)
AS CN~ FOR BLUEGILL EXPOSED TO SODIUM CYANIDE
Exposure
time,
hr
1.5
2.0
6.0
24.0
48. 0C
120.0
168. 0
LC50,
95% confidence
limits for LC50,
a
mg/4 mg/jp.
0.326
0.216
0. 183
0. 149
0. 134
0. 124
0.116
0.299 -
0.208 -
0. 176 -
0. 143 -
0. 126 -
0. 108 -
0. 103 -
0.356
0.225
0. 191
0. 154
0. 142
0. 142
0. 130
0.0870
0. 0449
1.0720
1.0587
1.0887
0. 1735
0. 1515
Log-probit ,
regression equation
10.59 + 11.49 (logx.)
19.81 + 22.25 (logx.)
10.23 + 5.77 (logx.)
11. 19 + 6.60 (logx.)
Standard deviation of the logarithm of the population tolerance frequency distribution.
Probit y. = a + b (logx.), where x. = concentration of CN .
LC50 based on truncated data.
-------
Appendix Table 25. MEDIAN LETHAL CONCENTRATIONS (LC50) AS
CN~ OF ADULT BROOK TROUT EXPOSED TO SODIUM CYANIDE
CO
CO
Exposure
time,
hr
Test No. 1
6
12
24
48
96
168
240
240b
264
264 b
Test No. 2
7
10
21
264
288
LC50,
mg/jfc
0.260
0.210
0. 196
0.163
0.156
0.152
0. 141
0.131
0. 127
0. 124
0.223
0.170
0. 158
0.133
0. 126
95% confidence
limits for LC50,
mg/A
0.245 -
0.196 -
0. 185 -
0.152 -
0.145 -
0.142 -
0.103 -
0.117 -
0.100 -
0.111 -
0.184 -
0.159 -
0.148 -
0.110 -
0. 105 -
0.277
0. 225
0.209
0. 175
0. 167
0. 164
0.193
0.147
0.162
0.139
0.269
0.182
0.168
0.159
0. 151
Slope
function,
sa
1.063
1.083
1.074
1. 083
1.083
1. 084
1.512
0. 1320
1. 377
0. 1284
1. 248
1. 192
1.075
1.277
1. 277
aAntilogarithm of £ •
Calculations by computer program. Log-probit regression equations were
Probit y. = 11. 68 + 7.58 (logx.) for 240-hr LC50 and Probit y. = 12. 06 + 7.79 (logx.)
for 264-nr LC50. * V
-------
o
o
Appendix Table 26. MEDIAN LETHAL CONCENTRATIONS (LC50) AS
CN~ FOR CHANNEL CATFISH EXPOSED TO SODIUM CYANIDE
Exposure 95% confidence
time, LC50, limits for LC50,
. hr mg/A mg/4
6 0.249 0.212 - 0.293
10 0.187 0.169 - 0.207
20 0. 166 0. 153 - 0. 180
26 0. 161 0. 149 - 0. 174
a
/\
a
0. 1865
0. 1181
0. 0930
0. 0997
Log -pr obit ,
regression equation
8. 24 + 5. 36 (logx.)
11. 17 + 8.47 (logx.)
13.40 + 10.75 (logx^
12.96 + 10.03 (logx.)
Standard deviation of the logarithm of the population tolerance frequency distribution.
Probit y . = a + b (logx..).
-------
Appendix Table 27. MEDIAN LETHAL CONCENTRATIONS (LC50) AS
CN~ FOR JUVENILE GOLDFISH EXPOSED TO SODIUM CYANIDE
Exposure
time,
hr
4
6
8
12
24
24C
36
48
96
120
168
240
336
LC50,
mg/;
1.134
0.856
0.595
0.403
0.330
0. 446
0.350
0.345
0.318
0.309
0. 298
0. 278
0.261
95% confidence
limits for LC50,
mg/Jt
0.962 -
0.730 -
0.531 -
0. 355 -
0.287 -
0.415 -
0.335 -
0.329 -
0.300 _
0.290 -
0.281 -
0.260 -
0.238 -
1.337
1.004
0.665
0.457
0.378
0.480
0.366
0.361
0.337
0.329
0.317
0.298
0. 287
A a
a
0. 1646
0. 1600
0. 1447
0.1636
0. 1581
0.0732
1.0865
1.0739
0.0670
0.0711
0. 0696
0.0786
0. 0947
Log-probit ,
. . . b
regression equation
...
9. 79 + 13. 66 (logx.)
---
12. 43 + 14. 92 (logx.)
12.16 + 14. 06 (logx.)
12.56 + 14.36 (logx.)
12.07 + 12.73 (logx^
11. 16 + 10.56 (logx.)
aStandard deviation of the logarithm of the population tolerance frequency distribution.
Probit yj = a + b
°Statistics calculated from a second toxicity test conducted for 336 hr.
-------
Appendix Table 28. MEDIAN LETHAL CONCENTRATIONS (LC50) FOR JUVENILE
FATHEAD MINNOWS EXPOSED TO SELENIUM DIOXIDE
o
to
Exposure
time,
hr
18.5
21.5
24.5.
30
42
72
96
120
144
168
220C
LC50,
mg/JL
31.
27.
24.
19.
15.
10.
7.
4.
3.
2.
2.
2
9
3
5
6
9
3
5
2
9
9
95% Confidence
limits for LC50,
26.
24.
16.
16.
13.
9.
5.
3.
2.
2.
9
1
4
8
4
4
7
4
6
5
- 36.3
- 32. 4
- 36. 0
- 22.7
- 18.3
- 12.7
- 9.2
- 6.0
- 3.9
-3.3
•• •"
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
a
1945
1917
1846
1931
1806
2181
3062
3777
2477
1657
~ ~
Log-probit regression
equation
-2.
-2.
-2.
-1.
-1.
o.
2.
3.
2.
2.
68
55
50
68
61
24
19
27
97
22
+ 5. 14
+ 5.22
-1- 5.42
+ 5.18
+ 5.54
+ 4.59
+ 3.27
-1- 2.65
+ 4.04
+ 6.03
"• —
(logx.)
(logxj)
(logx^)
(logx-)
(logx.)
(logxj)
(logXj)
(logx.)
(logx.)
(logx.)
la is the standard deviation of the logarithm of the population tolerance frequency distribution.
3probit yi = a + b (log34).
:Median lethal time for a concentration of 2. 9 + 1. 5 mg SeO2/4 calculated to have 95% confidence
limits between 129 and 374 hr and a A value of 0. 308.
o
-------
Appendix Table 29. MEDIAN LETHAL CONCENTRATIdNS
(LC50) FOR ADULT BROOK TROUT EXPOSED TO
SELENIUM DIOXIDE
o
CO
Exposure
time,
hr
6
7
24
48
64
96
LC50,
mg/Ji
87.3
70.3
36. 3C
23. 8°
23. Od
14.3
95% Confidence
limits for LC 50, a
mg/4 a
74.2 - 102.6 0. 1622
63.7 - 77.6 0.0986
_ _ _ — -
___
13. 1-15.5 0.0688
Log-probit ,
regression equation
- 6. 97 + 6. 17 (logx.)
-13.74 + 10.14 (logXi)
___
_ _ _
-11.78 + 14. 54 (logx.)
Standard deviation of the logarithm of the population tolerance frequency distribution.
bProbit y = a + b (logx^.
Derived from regression equation log y. (LC50) = 3. 95 - 1.65 logXi (exposure time).
based on LC50 values for 6, 7, 64, Ind 96 -hr. X
Median lethal time of 64. 0 hr for a concentration of 23. 0 nag/A.
-------
Appendix Table 30. MEDIAN LETHAL, CONCENTRATIONS
(LC50) FOR JUVENILE CHANNEL CATFISH EXPOSED TO
SELENIUM DIOXIDE
Exposure
23
28
52
94 C
95% Confidence Loe-orobit
. . . . A ci rcfirc s sion GovLSLtion
ins / * ^^S ' &
46.7 42.1 - 51.8 0.1028 -11.24 + 9.73 (logx^
40.2 37.8 - 42.8 0.0617 -21.00 H- 16. 20 (log x.)
24.9 20.0 - 31.1 0.0592 -8.27 + 9. 50 (logx^
19. 1 17. 1 - 21.4 1. 1504
aStandard deviation of the logarithm of the population tolerance frequency distribution.
bProbit Yi = a + b (logXj).
CLC50 calculated by method of Litchfield and Wilcoxon
-------
Appendix Table 31. MEDIAN LETHAL TIMES (LT50) FOR
JUVENILE FLAGFISH EXPOSED TO SELENIUM
DIOXIDE (SeO2)
o
01
Measured
SeO,
concentration, LT50,
mg/J& hr
11.2
16.9
21.8
27.9
37.6
83.1
67.2
68.3
55.4
44.3
95% Confidence
limits for LT50,
hr
55.1 -
48.7 -
43.9 -
42.3 -
35.7 -
125.2
92.8
106.3
72. C
55.0
a
0.2371
0. 1863
0.2557
0. 1565
0. 1250
Log-probit ,
regression equation
-3.10
-4.81
-2.17
-6. 14
-8.17
+ 4. 22 (logxi)
+ 5. 37 (logxi)
+ 3. 91 (logxi)
+ 6. 39 (logxi)
•f 8. 00 (logxi)
Standard deviation of the logarithm of the population tolerance frequency
distribution.
3Probit yi = a + b (logxj).
-------
Appendix Table 22. MEDIAN LETHAL, CONCENTRATIONS
(LC50) FOR JUVENILE GOLDFISH EXPOSED TO
SELENIUM DIOXIDE
Exposure
time,
hr
12C
18C
24
36
48
60
96
120
144
168
216
264
336
LC50,
mg/jfc
110.
76.
71.
60.
46.
41.
36.
32.
22.
17.
13.
11.
8.
0
5
?
2
5
2
6
7
3
2
0
5
8
95% Confidence
limits for LC50,
100.
70.
65.
54.
42.
36.
26.
22.
16.
13.
10.
9.
7.
7 -
6 -
8 -
6 -
6 -
1 -
7 -
7 -
4 -
1 -
1 -
9 -
8 -
120. 1
82.9
77.2
66.3
50.7
46. 9
50.2
47. 0
30. 3
22.5
16.6
13.3
9.9
1.
1.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
a
125
184
1022
1122
1120
1515
3159
3626
3523
3123
2873
1707
1194
Log -pr obit b
regression equation
-13.
-10.
- 9.
- 5.
+ 0.
+ 0.
+ 1.
+ 1.
+ 1.
- 1.
- 2.
14
86
89
66
05
82
17
05
13
21
92
f 9. 79
f 8.92
f 8.93
f 6.60
f 3.17
f 2.76
f 2.84
f 3.20
t- 3.48
f 5. 86
<- 8.38
(logxi)
(logxi)
(logxi)
(logxi)
(logxi)
(logxi)
(logxi)
(logxi)
(logxi)
(logxj)
(logxi)
Standard deviation of the logarithm of the population tolerance frequency
distribution.
bProbit yi = a. + b (logXi).
c 29
Calculated according to method of Litchfield and Wilcoxon
-------
Appendix Table 33. MEDIAN LETHAL CONCENTRATIONS
(LC50) FOR JUVENILE BLUEGILL EXPOSED TO
SELENIUM DIOXIDE
Exposure
time,
hr
8
18
24
168
192
240
288
336
LC50,
mg/£
126.6
82. 1
77.3
30.7
27.7
23.6
20.5
17.6
95% Confidence
limits for LC50,
mg/A
104.5 -
76.3 -
72.1 -
27.2-
24.3 -
20.8 -
18.6 -
16.4 -
153.4
88.4
82.8
34.5
31.6
26.9
22.7
18.9
^a
a
0.1921
0. 0743
0.0709
0. 1360
0.1519
0. 1620
0. 1249
0. 0826
Log-probit ,
regression equation
-5.95-1- 5.21 (logx^
-20.75 + 13.45 (logxi)
-21. 64 + 14. 11 (logx^
- 5. 93 + 7. 35 (logXi)
- 4. 50 .+ 6.58 (logxi)
- 3.48 -1- 6. 17 (logxi)
- 5. 50 -1- 8, 01 (logx^
-10.10 +12.11 (logxi)
a is the standard deviation of the logarithm of the population tolerance frequency
distribution.
b Probit y. = a + b (log^) •
-------
Appendix Table 34. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR ADULT BROOK TROUT EXPOSED TO SODIUM ARSENITEa
o
00
Exposure
time, hr
22
24
30
31
48
93
144
164
262
LC50,
mg/£
54.
53.
40.
.42.
27.
25.
20.
19.
18.
1
9
8
2
8
8
0
4
0
95% Confidence
limits for LC50,
mg/4
28.5 -
49.2 -
39.6 -
36.4 -
23.9 -
23.7 -
18.3 -
17.8 -
17.1 -
102.
59.
42.
49.
32.
28.
21.
21.
19.
9
1
1
0
4
1
8
2
0
0.
0.
0.
0.
0.
0.
0.
0.
0.
b
A
a
1840
0913
0315
1723
1522
0836
0870
0888
0506
Log -pr obit
regression equation0
- 4.
-13.
-46.
- 4.
- 4.
-11.
- 9.
- 9.
-19.
42 +
97 +
13 +
44 +
49 +
89 +
95 +
50 +
83 +
5.
10,
31.
5.
6.
11.
11.
11.
19.
• 43 (logx.)
• 96 (logx.)
. 74 (logx.)
81 (logx.)
57 (logx!)
96 (logx.)
49 (logx.)
26 (logx.)
77 (logx.)
One test conducted for 48 hr and one for 262 hr.
Standard deviation of the logarithm- of the population
tolerance frequency distribution*
1 Probit yi = a + b (logx.).
-------
o
CD,
Appendix Table 35. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR JUVENILE FATHEAD MINNOWS EXPOSED
TO SODIUM ARSENITE
Exposure
time,
hr
14
16
24
48
72
96
187
283
336
LC50,
mg/jj
45.3
41.7
36.2
31.3
29.2
27.0
24.9
21.7
18.2
95% confidence
limits for LC50,
i
39.6
37.9
33.2
29.1
27.0
24.7
22. 1
16.5
13.3
mg/£
- 51.9
- 45.9
- 39.5
- 33.8
- 31.6
- 29.4
-28.0
-28.6
- 25.0
* a
0. 1749
0. 1228
0. 1123
0.0875
0.0795
0.0861
0.0961
0. 3555
0. 4082
Log -pr obit ^
regression equation
- 4. 47 +
- 8.20 +
- 8.88 +
-12.09 +
-13.42 +
-11.62 +
- 9.52 +
1.24 +
1.91 +
5.72 (logx.)
8. 14 (logx.)
8.91 (logx.)
11.43 (logx.)
12.57 (logx.)
11.61 (logx.)
10. 40 (logx.)
2.81 (logx.)
2. 45 (logx.)
Standard deviation of the logarithm of the population tolerance frequency distribution.
b.
Probit y. = a + b (logx.,).
-------
Appendix Table 36. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR JUVENILE CHANNEL CATFISH EXPOSED TO SODIUM ARSENITE
Exposure
time, LC50,
hr
. 43
48
67
77
92
96
mg/4
60.8
56.9
47.5
44.5
33.3
31.2
95% confidence
limits for LC50,
* a
a
mg/jfc
55.4 -
52.5 -
42. 0 -
39.8 -
29.0 -
27.6 -
66.8
61.8
53.6
49.8
38.3
35.4
0.0937
0.0822
0. 1405
0.1291
0. 1775
0.1612
Log-probit ,
regression- equation
-14.03 H
-16.36 H
- 6.93 H
- 7.77 H
- 3.58 H
-4.27 i
h 10.67 (logx.)
h 12. 17 (logx.)
h 7. 12 (logx.)
h 7. 75 (logx.)
h 5.63 (logx.)
h 6.21 (logx.)
aStandard deviation of the logarithm of the population tolerance frequency distribution.
b.
Probit y. = a + b (logx^.
-------
Appendix Table 37. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR JUVENILE GOLDFISH EXPOSED TO SODIUM ARSENITE
Exposure
time, hr
6
18
24
36
48
72
96
168
240
336
211.
103.
89.
60.
54.
50.
44.
36.
33.
32.
95% Confidence
LC50, limits for LC50
mg/* mg/*
2
1
6
8
6
0
9
2
1
1
179.
92.
78.
54.
48.
44.
40.
33.
18.
29.
3 -
6 -
8 -
8 -
9 -
5 -
3 -
4 -
4 -
9 -
248.8
114.8
101.9
67.4
60.9
56.0
50. 1
39.4
59.6
34.4
0.
0.
0.
0.
0.
0.
0.
0.
o.
0.
/•a
a
1637
1508
1666
1344
1107
1159
1095
0953
0711
0716
Log-probit ,
regression equation
- 9.
- 8.
- 6.
- 8.
-.10.
- 9.
-10.
-11.
-16.
-16.
20 +
36 +
72 +
28.+
69 +
66 +
09 +
37 +
38 +
03 +
6.11
6.63
6.00
7.44
9.04
8.63
9.13
10.50
14.07
13.96
(logx^
(logx.)
i
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
(logx.)
Standard deviation of the logarithm of the population
tolerance frequency distribution-
b
Probit y.= a + b (logx.)-
-------
Appendix Table 38. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR FLAGFISH FRY EXPOSED TO SODIUM ARSENITE
Exposure
timei hr
29
- 67
93C
93C
93C
LC50,
mg/A
69.8
60.8
55.5
40.4
49.6
95% Confidence
limits for LC50,
mg/X
57. 1 - 85.4
53.0 - 69.8
42. 1 - 73.2
23.7 - 68.8
38.5 - 63.9
a
e
0. 1427
0. 1597
0.2341
0.1561
0.2320
Log-probit
regression equation
-7.92 + 7.01 (logx.)
-6. 17 + 6.26 (logx.)
-2.45 + 4.27 (logx.)
-5.29 + 6.40 (logx.)
-2.31 -1- 4.31 (logx.)
Standard deviation of the logarithm of the population tolerance frequency
distribution.
b Probit y. = a + b (logx.).
c The average LC50 of the 93-hr LC50 estimates, 48.5 mg/J&, was used for
plotting the toxicity curve for this species. The 93-hr LC50 values are
equivalent to 96-hr LCSO's and were derived from three replicates.
-------
Appendix Table 39. MEDIAN LETHAL CONCENTRATIONS (LC50)
FOR JUVENILE BLUEGILL EXPOSED TO SODIUM ARSENITE
Exposure LC50,
time.hr mg/4
96
120
168
240
264
288
336
72.
67.
61.
47.
42.
37.
31.
0
3
7
8
2
0
6
95% Confidence
limits for LC50,
mg/ t>
55.5
58.1
51.9
41.2
35.6
32.0
27.9
- 93.
- 77.
- 73.
- 55.
- 50.
- 42.
- 35.
5
8
2
6
0
9
8
1.
0.
0.
0.
0.
0.
0.
a
A
a
5299C
1460
1717
1738
2193
1896
1617
Log-probit .
regression equation
-7. -5 2
-5.42
-4.66
-2.41
-3.27
-4.27
•»
+ 6.85
+ 5.82
+ 5. 75
+ 4.56
+ 5.27
+ 6.18
(logx.
(logx.
(logx.
(logxi
(logxi
(logx.
)
)
)
)
)
)
a o is the standard deviation of the logarithm of the population
tolerance frequency distribution.
b
Probit yi = a + b
c 29
Computations performed by the method of Litchfield and Wilcoxon
a is expressed as the slope function, S, the antilogarithm of
-------
Appendix Table 40. MEDIAN LETHAL CONCENTRATIONS
(LC50) FOR JUVENILE FATHEAD MINNOWS EXPOSED TO
BERYLLIUM SULFATE
Exposure
time,
hr
75d
92e
96e
121
164
192
283
336
LC50,a
mg/i
47.8
40.2
37.9
30.8
27.7
27.4
26.1
25.4
95% confidence ^
limits for LC50, a
mg/jj
27.6
27.5
29.4
26. 1
25.9
24.4
23.9
_ ^
- 58.5
- 52.3
- 32.3
-29.3
- 29.0
- 27.9
- 27.0
.
0.0910
0. 0764
0. 0623
0.0733
0.0723
0.0777
0.0691
Log-probit
regression equation
-12.62 H
-15.66 H
-18.91 H
-14.66 H
-14.88 H
-13.24 H
-15.32 H
—
h 10.98 (logx.)
h 13. 09 (logx.)
h 16.06 (logx.)
h 13.64 (logx.)
h 13.83 (logXi)
h 12. 87 (logXi)
h 14.46 (logx.)
As
Standard deviation of the logarithm of the population tolerance frequency
distribution.
:Probit yt = a + b (logx.).
Median lethal time-
3Heterogenous concentration - percent mortality data.
-------
Appendix Table 41. MEDIAN LETHAL CONCENTRATIONS (LC50) AND
MEDIAN LETHAL TIMES (LT50) FOR FLAGFISH FRY
EXPOSED TO BERYLLIUM SULFATE
Group
Median A a
response 95% confidence a
estimate limits
Log-probit ,
regression equation
96-hr LC50 (mg/J& BeSO4)
IC
II
III
LT50 for
I
II
III
46. 3 mg/£ 43. 9-48.8
41. 1 mg/Jfc 37.2 - 45. 3
41. 1 mg/A 38.4 - 44.0
47. 8^2.2 mg/JB BeSO4
41.3hr 10.5-162.1
74.8hr 61. 1-91.4
55.4 hr 48. 3-63.5
mg/A 1.053
mg/jfc 0.0801
mg/i 0. 0488
hr 0.4996
hr 0.0821
hr 0. 0560
1.77 + 2.00 (logx.)
-17.81 + 12. 18 (logx.)
-26. 13 + 17.86 (logx )
i
Standard deviation of the logarithm of the population tolerance frequency distribution.
bProbit y. = a + b
c 29
Calculation by method of Litchfield and Wilcoxon
-------
Appendix Table 42. MEDIAN LETHAL, CONCENTRATIONS
(LC50) FOR JUVENILE GOLDFISH EXPOSED TO
BERYLLIUM SULFATE
Exposure
time,
hr
96
120
168
186
216
240
LC50,a
mg/ji
55.9
49. 3
48. 3
46.5
41.6
38.4
95% confidence
limits for LC50,
a
mg/A
49.0 -
44.0.-
42.7 -
40.8 -
37.2 -
34.4 -
63.7
55.3
54. 6
53. 1
46.6
43.0
0.1510
0. 1325
0. 1595
0. 1526
0, 1122
0.1115
Log-probit
regression equation
-6.57
-7.78
-5.56
-5.93
-9.44
-9.21
-h 6. 63 (logx.)
+ 7.55 (logx.)
+ 6. 27 (logx. )
+ 6. 55 (logx.)
+ 8. 92 (logx.)
+ 8. 97 (logx.)
dAs Be SO4 .
Standard deviation of the logarithm of the population tolerance frequency distribution.
CProbit Yi = a + b (logx.,).
------- |