EPA-600/3-79-009
January 1979
ACUTE AND CHRONIC TOXICITY OF HCN
TO FISH AND INVERTEBRATES
by
Lloyd L. Smith, Jr.
Steven J. Broderius
Donavon M. Oseid
Gary L. Kimball
Walter M. Koenst
David T. Lind
Department of Entomology, Fisheries, and Wildlife
University of Minnesota
St. Paul, Minnesota 55108
Grant No. R802914
Project Officer
Robert Drummond
Environmental Research Laboratory
Duluth, Minnesota 55804
This study was conducted
in cooperation with
University of Minnesota
Agricultural Experiment Station
St. Paul, Minnesota
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
DULUTH, MINNESOTA 55804
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory-
Duluth, 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.
-------
FOREWORD
This report represents nearly five years.of study on the acute and chronic
effects of cyanide on various aquatic animals, particularly fish. It should
serve as a reference for both administrators charged with setting permissible
levels of cyanide in United States surface waters and researchers planning to
conduct related studies.
Portions of the results herein have been published in several different
scientific journals. Many tables, figures and detailed comments had to be
editorially omitted from these journal articles to conserve space and reduce
publication costSo This report contains all pertinent data, including
previously published findings, as well as some new information never before
published.
Some preliminary studies on the photodecomposition of iron-cyanides,
which causes the release of HCN, also were started during this project. These
results will be published later in a separate EPA Ecological Report.
Donald I. Mount, Ph.D.
Director
Environmental Research Laboratory-Duluth
111
-------
DEDICATION
During the preparation of this report the senior investigator, Dr. Lloyd
L. Smith, Jr., died on June 17, 1978. His dedication and enthusiasm will be
missed.
iv
-------
ABSTRACT
The acute toxicity of HCN was determined at various temperatures and dis-
solved oxygen concentrations for different life history stages of seven
species of fish and two species of invertebrates. Eggs were the most tolerant
stage while fry and juvenile fish, the least tolerant stages, had similar sen-
sitivities. Temperature and dissolved oxygen concentration (DO) also have an
effect on acute toxicity; juvenile fish generally became more sensitive with
decreasing temperature and DO. In contrast to juvenile fish, invertebrates
were less sensitive to HCN with decreasing temperature. One species of inver-
tebrate had about the same sensitivity to HCN as juvenile fish, while the
other species was considerably more tolerant of HCN.
Long-term tests conducted with fathead minnows demonstrated that the con-
centrations of HCN having no adverse effect on egg production was between 12.9
and 19.6 yg/1. Chronic tests with brook trout demonstrated that on the basis
of spawning success the maximum acceptable toxicant concentration (MATC) was
between 5.7 and 11.2 yg/1 HCN. When compared with the mean number of eggs
spawned in the less productive of two controls, a 42.0 percent reduction was
observed at the latter concentration. Long-term tests with bluegills showed
that no spawning occurred at HCN concentrations as low as 5.2 ug/1. The sur-
vival of bluegill fry was drastically reduced at 19.4 ug/1 HCN in 57-day tests
which began with eggs.
Chronic experiments conducted with Asetlus and Gammafus demonstrated that
the highest concentration of HCN having no adverse effect was between 29 and
40 ug/1 for Asettus and between 16 and 21 ug/1 for Gammarus. When these two
invertebrate species were exposed to HCN simultaneously, the GammctFUS almost
eliminated the As&ttus from the controls and the lowest treatment level. How-
ever, there was also a downward shift in the no-effect/effect concentration
for GammoPus to between 4 and 9 yg/1. The values for Asettus remained approx-
imately the same as when tested alone.
The presence of sublethal concentrations of HCN caused increased predation
by green sunfish on fathead minnows, but it was not determined whether fat-
heads were easier prey or the green sunfish had greater appetites.
In acute tests where the HCN concentration is increasing with exposure
time, it was determined that the amount of mortality can be roughly predicted
from the relationship between exposure area (change in HCN concentration with
time) and median survival time for three types of time-concentration exposure
curves. Experiments conducted with intermittent and sublethal diurnal exposure
regimes demonstrated that adverse effects on early growth of the fathead minnow
are lessened as the exposure period is reduced.
v
-------
The lethal and sublethal effects of binary toxicant mixtures on fish were
evaluated according to existing models of multiple toxicity. Using the toxic
unit approach, it was determined that the Zn—HCN and ammonia-HCN mixtures were
more acutely toxic and the Cr-HCN mixture less toxic than what would he pre-
dicted from simply additive interaction. The sublethal joint action of
toxicant mixtures on early fish growth was not predictable from existing
multiple toxicity models, and in general, there was no joint interaction
between toxicants when compared with the observed responses elicited by indi-
vidual toxicants.
This report was submitted in fulfillment of Grant Number R802914 by the
Department of Entomology, Fisheries, and Wildlife, University of Minnesota,
under the sponsorship of the U.S. Environmental Protection Agency. This report
covers a period from November 1, 1973 to September 30, 1978.
vi
-------
CONTENTS
Foreword. i-ii
Dedication .;.............., iv
Abstract » V
List of Figures ix
List of Tables. x
Acknowledgments xiv
Sections
I. Conclusions ........*..... 1
II. Recommendations 2
III. Introduction. 4
IV. Acute Toxicity 6
Experimental Design. 6
Test Animals 6
Water Supply. 7
Laboratory Acclimation. 7
Test Apparatus. 9
Test Procedures 10
Statistical Analysis 12
Results o 12
Fathead Minnow 12
Bluegill. 14
Yellow Perch 14
Brook Trout 14
Other Fish Species 24
Effect of Temperature, 24
Effect of pH. 26
Invertebrates 31
Summary. 31
V. Chronic Toxicity. 33
Fathead Minnow 33
Experimental Design. 33
Results 35
First Generation 35
Survival 35
Growth. 35
Reproduction 35
vii
-------
Contents - Continued
FI Generation,.,,...,,.,.,,,,.,,, ,, , ,,,,,,.,,. 37
Survival , , 37
Growth. ...,..;, ,.,,..,,..„,, 37
Summary............................... 41
Brook Trout -. , 42
Experimental Design 42
Results 43
Adults 43
Growth and Survival. , 43
Reproduction 43
Embryos, Fry and Juveniles 44
Growth , 44
Survival. ,. 47
Summary. 47
Bluegill 49
Experimental Design 49
Results 51
Survival 51
Growth. 55
Reproduction.„, , — 55
Second Generation Hatching and Survival 55
Summary. 55
Invertebrate Populations 59
Variability Between Two Species 59
Experimental Design 61
Results 61
Summary ..,.., , 66
Response to Cyanide 66
Experimental Design 66
Results 68
Summary. 72
VI. Miscellaneous Tests ,. 77
Predator-Prey Relations 77
Experimental Design 77
Results 78
Exposure to Increasing Cyanide Concentrations..., 81
Experimental Design. 81
Results 84
Intermittent Sublethal Exposure Tests , 86
Experimental Design , 86
Results 87
Binary Mixtures with HCN. 92
Experimental Design. 92
Results 95
Summary 100
VII. References 110
VIII. Publications 114
viii
-------
FIGURES
Numb er
Relationship between test pH and 96-hr LC50 cyanide
concentrations for the fathead minnow at 20°C 30
Length-frequency diagrams of all Asellus in each test
chamber 62
Length-frequency diagrams of all Gcarmarus in each test
chamber 63
Length-frequency graphs of each test concentration for
chronic test—Garmavus alone 69
Length-frequency graphs of each test concentration for
chronic test—Asellus alone (2) 70
Length-frequency graphs of each test concentration for
chronic test—Asellus and Gammarus together 71
Representative curves for increasing HCN lethal exposure
tests normalized to time and concentration 83
Diurnal fluctuation of HCN levels in test chambers as a
function of their maximum concentrations 88
-------
TABLES
Number Page
1 Analysis of laboratory well water 8
2 Number of organisms per test chamber 10
3 Approximate initial size (mm total length) of fry, juveniles
and adults for toxicity tests 11
4 Acute toxicity of HCN to fathead minnow eggs expressed as
96-hr LC50 and median lethal concentrations at hatching 13
5 Acute toxicity of HCN to fathead minnow fry expressed as
96-hr LC50 and median lethal threshold concentrations 15
6 Acute toxicity of HCN to fathead minnow juveniles expressed
as 96-hr LC50 and median lethal threshold concentrations 16
7 Acute toxicity of HCN to bluegill eggs and fry expressed as
96-hr LC50 and median lethal threshold or hatching con-
centrations 17
8 Acute toxicity of HCN to bluegill juveniles expressed as
96-hr LC50 and median lethal threshold concentrations 18
9 Acute toxicity of HCN to yellow perch eggs and fry expressed
as 96-hr LC50 and median lethal threshold or hatching con-
centrations 19
10 Acute toxicity of HCN to yellow perch juveniles expressed
as 96-hr LC50 and median lethal threshold concentrations 20
11 Acute toxicity of HCN to brook trout eggs and sac fry
expressed as 96-hr LC50 and median lethal threshold or
hatching concentrations 21
12 Acute toxicity of HCN to brook trout swim-up fry expressed
as 96-hr LC50 and median lethal threshold concentrations 22
13 Acute toxicity of HCN to brook trout juveniles expressed
as 96-hr LC50 and median lethal threshold concentrations 23
x
-------
Tables - continued
Number Page
14 Acute toxicity of HCN to rainbow trout, largemouth bass
and black crappie juveniles expressed as 96-hr LC50
concentrations 24
15 Acute toxicity of HCN to wild stock fathead minnow juveniles
at 5°C intervals from 5° to 30°C expressed as 96-hr LC50
and median lethal threshold concentrations 25
16 Acute toxicity of HCN to brook trout juveniles at temperatures
from 4° to 18°C expressed as 96-hr LC50 and median lethal
threshold concentrations 27
17 Mean test conditions for composite free cyanide bioassays 28
18 Biological assay by the log-probit analysis method of fathead
minnow cyanide bioassays at 20°C grouped according to test
pH 29
19 Acute toxicity of HCN to Asellus aomnrunis and Gammapus
pseudolimnaeus adults expressed as 96-hr LC50 and median
lethal threshold concentrations 32
20 Survival and weight of first-generation fathead minnows after
28, 56, and 84 days of exposure to hydrogen cyanide 36
21 Egg production, hatchability, and terminal weights for first-
generation fathead minnows exposed to various concentrations
of hydrogen cyanide 38
22 Survival, length, and weight of FI fathead minnows after 28
and 56 days of exposure to hydrogen cyanide 40
23 Egg production of adult brook trout exposed to HCN for
144 days prior to start of spawning 44
24 Growth of embryo and juvenile brook trout exposed to various
levels of cyanide ' 45
25 Effect of HCN on the survival of brook trout eggs to hatching
and from hatching to 90-day-old juveniles 48
26 Percentage reduction in mean number of eggs spawned by
female brook trout exposed to HCN 49
27 Survival and fecundity of adult bluegills after 289 days,
Test 4 52
xi
-------
Tables - continued
Number Page
28 Percentage survival of bluegills from fertilized egg to the
57-day juvenile stage in various HCN concentrations for
Test 1 53
29 Percentage survival of bluegills from fertilized egg to the
57-day juvenile stage in various HCN concentrations for
Test 3 54
30 Mean lengths (cm) and mean weights (g) of adult bluegills
exposed to HCN for 289 days, Test 4 56
31 Terminal mean lengths and weights of bluegill juveniles at
various HCN concentrations in Tests 1 and 3 57
32 Percentage survival, length, and weight of bluegill juveniles
in Test 2 (mean initial weight = 0.66 g) 58
33 Test conditions in Gammavus and Asellus experiments 60
34 Population parameters in Aseltus and Gammams tests 64
35 Means for the population indices of five replications 65
36 Test conditions for simultaneous Gammar'us and Asellus
bioassays In HCN 67
37 Biological indices in the chronic test series (1) on
Asellus (both tests included) 68
38 Biological indices in chronic test series (2) on Asellus
(both tests included) 73
39 Biological indices In the chronic test series on Gammarus
(both tests included) 74
40 Biological indices in the chronic series with Asellus (A)
and Gammapus (G) together (both tests Included) 75
41 Comparisons of the no-effect/effect concentrations (pg/1)
of hydrogen cyanide for Asellus and GcormctTUS 76
42 Mean HCN concentrations in each experimental tank of each
test and the overall means for each tank at designated
nominal concentrations 78
43 Prey consumption (grams of prey consumed per gram of predator
present) for each test tank and the overall means for each
nominal HCN concentration 79
xii
-------
Tables - continued
Number
44 Prey consumed for each test tank as a percentage of the
control values and the overall means for each nominal
HCN concentration 80
45 Diluter and aquaria conditions as related to water turnover
times 82
46 Maximum HCN concentration, median survival time (MST) and
exposure area derived from the time-concentration curves
for the increasing HCN lethal exposure tests 85
47 Mean areas, standard deviations, and coefficients of
variation for the combined tests from Table 46 86
48 Growth of fathead minnows in 30 days when exposed to
continuous or intermittent HCN concentrations 89
49 Linear regression analysis of the toxicant dose response
curves expressing terminal mean dry weight in nig (Y) and
percentage terminal mean dry weight normalized to controls
(Y) as a function of log toxicant concentration in pg/1 (X) 91
50 Mean test conditions and description of test fish for acute
toxicity bioassays of individual toxicants and binary
mixtures 96
51 Dose-response parameters, lethal concentrations and additive
indices for acute toxicity bioassays of individual toxicants
and binary mixtures 97
52 Growth of fathead minnows in 30 days when exposed to
individual or mixture solutions of hexavalent chromium
and HCN 101
53 Growth of fathead minnows in 30 days when exposed to
individual or mixture solutions of zinc and HCN 103
54 Growth of rainbow trout in 30 days when exposed to
individual or mixture solutions of ammonia and HCN 105
55 Linear regression analysis of the toxicant dose response
curves expressing terminal mean dry weight in mg (Y) and
percentage terminal mean dry weight normalized to
controls (Y) as a function of log toxicant concentration
in mg/1 (X) 107
XLll
-------
ACKNOWLEDGMENTS
The authors wish to acknowledge the assistance of Ms. Mary Menke who main-
tained test stocks and conducted routine tests during the various bioassay
experiments. They also wish to thank the many student assistants who helped
collect field specimens, conduct bioassays, and perform chemical analyses and
calculations. We also wish to thank the Minnesota and Wisconsin Departments of
Natural Resources for their assistance in obtaining experimental fish.
xiv
-------
SECTION I
CONCLUSIONS
The results of experimental studies on the acute and chronic toxicity of
cyanide to aquatic organisms have demonstrated that HCN is toxic to fish and
invertebrates in low concentrations which are frequently found in polluted
waters. Lethal tests have shown that the susceptibility of fish to HCN is
dependent on their life history stage. Eggs are the most tolerant while fry
and juveniles, the least tolerant stages, have similar sensitivities. Acute
toxicity was demonstrated to vary among species, sources of fish, and with
environmental factors such as temperature, dissolved oxygen, and pH. The tox-
icity of HCN to all species of juvenile fish which were tested increased as
temperature decreased, yet the opposite was observed for an invertebrate
species. Decreased DO levels also increased the toxicity of HCN, but pH
levels within the range 6.8 to 8.3 had little effect.
In view of the uncertainty of predicting long-term adverse effects from
results of acute toxicity tests, the results of chronic exposures provide a
more reliable estimate of acceptable levels of HCN. On the basis of our tests
with representative freshwater organisms, it was demonstrated that egg produc-
tion (reproductive potential) and spawning success are reduced at low HCN
levels. From these experiments, it can be concluded that to insure cyanide
does not seriously impair the success of most freshwater organisms a contin-
uous and constant concentration of HCN should not exceed 5 yg/1. This level
should protect most aquatic populations during the most sensitive period of
population maintenance and can also be tolerated by all other life history
stages under reasonably adverse environmental conditions. If effects on repro-
duction and fry survival are not considered, our results suggest that concen-
trations up to about 25 yg/1 HCN can be tolerated without marked adverse
effects on natural populations of the species which were studied.
-------
SECTION II
KECOMMENDATI CMS
The acute toxicity to aquatic organisms of most simple and metallocyanide
complexes has been demonstrated in the scientific literature and from work con-
ducted during this investigation to be almost entirely due to the presence of
HCN as derived from ionization, dissociation, and photodecomposition of cyanide
containing compounds. Therefore, the analytically determined HCN concentration
in cyanide polluted waters is the most reliable chemical measurement of the
actual cyanide toxicity to aquatic organisms.
The relationship between total cyanide and HCN in natural waters has not
been well defined, but it is known to vary markedly with receiving water con-
ditions, the type of cyanide compounds present, the presence of other chemical
compounds, and the degree of exposure to natural light. The total cyanide con-
centration in some waters may consist almost entirely of free cyanide, or may
contain cyanides which may readily photodecompose or dissociate to yield HCN.
It must therefore be assumed that a standard total cyanide measurement is
indicative of the free cyanide concentration unless it can be demonstrated
otherwise. However, under most circumstances it would be expected that the
concentration of total cyanide will be greater than that of HCN. Since total
cyanide concentrations in most cases will not be toxicologically meaningful,
such a measurement would not be representative of the actual cyanide toxicity
but would most likely be a liberal estimate. Therefore, a simple but accurate
method which measures free cyanide or HCN directly in the micrograms per liter
range is needed and should be used by regulatory agencies in their implementa-
tion of receiving water cyanide standards that reflect sound water quality
criteria.
Determination of maximum acceptable toxicant concentrations of HCN should
be based on the most sensitive life history stage of an aquatic organism when
exposure is continuous throughout its lifetime. Any impairment of long-term
survival, growth rate, fecundity, or spawning behavior must be considered
detrimental to population maintenance unless ecological studies show that some
degree of impairment of these vital functions can be sustained on a continuous
basis without degradation of the aquatic population.
The effect of intermittent discharges of HCN at concentrations higher than
acceptable acute or chronic levels has not yet been fully determined. However,
the effect of low temperatures on acutely toxic levels of HCN to fish and the
extreme sensitivity of some life history stages demand caution in assuming that
temporary or seasonal deviations from demonstrated no-effect levels of HCN can
be safely tolerated by fish populations. Further testing is needed to deter-
-------
mine the long-term sensitivity of fish to HCN at low temperatures. Until this
is accomplished, it is recommended that a concentration no greater than 5 yg/1
HCN be considered a safe limit for the protection of freshwater fish and
invertebrates.
-------
SECTION III
INTRODUCTION
Compounds containing the cyanide group are present in many effluents,
including those from iron and steel processing plants, petroleum refineries,
and metal plating plants, and constitute a hazard to aquatic ecosystems in
certain waste receiving waters. These compounds usually exist in aqueous solu-
tions as free cyanide (i.e., CN ion and molecular HCN) or as metallocyanide
complexes. The toxicity of different forms of cyanide varies markedly. Cer-
tain cyanide complexes have relatively low toxicity, although they may photo-
decompose and dissociate into the highly toxic HCN. Wuhrmann and Woker (1948),
Bridges (1958), and Doudoroff et al. (1966) have concluded that HCN is the
principal toxic form of cyanide. However, Broderius et at. (1977) have demon-
strated that even though molecular HCN is more toxic than the CN ion, the
anion does contribute to the total toxicity in proportion to its concentration.
In aqueous solution the cyanide radical of simple alkali cyanides such as NaCN
hydrolyzes to form free cyanide. The molecular (un-ionized) component predom-
inates at pH values found in most natural waters with less than 6 percent of
free cyanide occurring in the ionic form below pH 8 at 25°C. As the pH of the
aqueous simple cyanide solution is increased, the percentage of free cyanide
present as the CN ion is increased to satisfy the equilibrium reaction
HCN J H+ + CN~.
The literature on the toxicity to fish of various cyanides has been
recently reviewed by Doudoroff (1976). The results vary widely among species
and among experimenters. Much of the data is difficult to assess because the
exact character of the test solutions, wastes, test conditions, test organisms,
etc. were often not clearly defined. Further, variations in the sensitivity
of organisms during different life history stages have not been well examined.
In view of the lack of comparative toxicity tests with fish in various life
history stages, and the absence of comprehensive chronic tests to determine
sublethal effects, the present study was undertaken using seven species of
fish and two species of invertebrates. The fish which were used in toxicity
tests were the fathead minnow, Pimephales promeles Rafinesque; bluegill,
Lepomis maovoGh-ivus Rafinesque; yellow perch, PeTQa flavesoens (Mitchill);
brook trout, Salvelinus font-inal-Ls (Mitchill); rainbow trout, Salmo gairdner-i
Richardson; black crappie, Pomoxi-s nigromaeulatus (LeSueur), and largemouth
bass, Mioropterus salmoides (Lacepede). The invertebrates Asellus comnunis
Say* and Gcanrnarus pseudolimnaeus Bousfield were also studied.
*Keyed out to Asellus militaris Hay in Pennak (1953) but to Asellus
oommunis Say in Williams (1970).
-------
Acute toxicity tests were conducted using all of the species listed above,
during various life history stages. The chronic toxicity of cyanide to fathead
minnows, bluegills, brook trout, Asellus, and Gammarus was also investigated.
In addition, a series of toxicity tests was conducted at different pH values,
using fathead minnows, to determine the relationship of hydrogen ion concentra-
tion and cyanide form on the toxicity of free cyanide. Tests were conducted to
determine the effect of temperature on the acute toxicity of HCN to fathead
minnows and brook trout. The lethal and sublethal effects of binary mixtures
of HCN and hexavalent chromium, zinc, or ammonia were evaluated. The response
of fathead minnows to continuously increasing and fluctuating sublethal levels
of HCN was also studied.
i
Iron-cyanides are relatively nontoxic, but when exposed to sunlight they
photodecompose to release HCN. Preliminary studies of this phenomenon were
begun during this project, but the results will be incorporated in a more
detailed study to appear in a future EPA report.
-------
SECTION IV
ACUTE TOXICITY
Acute toxicity tests were designed to determine the 96-hr median lethal
toxicant concentration (LC50) of HCN and the median lethal threshold concentra-
tion (LTC) (Sprague 1969) at various temperatures and dissolved oxygen concen-
trations. These tests were conducted with eggs, fry, and juveniles of four
species of fish, juveniles of three other fish species, and two species of
invertebrates.
EXPERIMENTAL DESIGN
Test Animals
For brevity in this paper, when referring to the various specific phases
in the life history of a fish as defined by Balon (1975), the less specific
common terms will be used. Toxicity tests using fish eggs encompassed both
the cleavage phase and the embryo phase in all cases. For brook trout the
term sac fry refers to the eleutheroemhryo phase, swim-up fry to the alevin
phase and juvenile to the smolt phase. For the other species, fry refers to
the eleutheroembryo phase and juvenile to the juvenile phase. Sexually mature
individuals were used for experiments with Asellus and Gammaz'us.
Brook trout and rainbow trout were procured as recently hardened eggs, or
as sac fry within 24 hr after hatch. They were provided by state hatcheries
(Lanesboro and French River, Minnesota, and Osceola, Wisconsin) where both the
parents and the offspring were,maintained at about 10°C. Trout eggs were
incubated in the laboratory in total darkness. Fathead minnows which were
used in toxicity tests were reared in the laboratory at 25°C from parents
obtained as eggs from the Environmental Research Laboratory at Duluth, Minne-
sota. Bluegills were collected from several local lakes. Eggs and fry were
obtained in the laboratory by artificially spawning adult bluegills which had
been collected the same day at temperatures between 24° and 28°C. The water
temperature measured during collection was maintained in the laboratory during
spawning. The male fish was killed with a blow to the head. The testes were
dissected out, rinsed with water, and placed in a cheesecloth bag. The female
was wiped dry, and the eggs were stripped into a wetted Petri dish. Sperm was
immediately squeezed from the bag and dripped onto the cluster of eggs. Water
was then added as the dish was moved to gently swirl the eggs and sperm. After
1 min a light suspension of bentonite clay was added to prevent clumping of the
eggs or their adherence to the dish. Juvenile bluegills (13-28 mm total
length) were collected about 2 weeks before being used in toxicity tests, when
-------
lake temperatures were between 13° and 28°C. Yellow perch eggs were obtained
from adults which had naturally spawned in large holding tanks in the labora-
tory. Eggs of yellow perch which had been spawned naturally and were deter-
mined to be less than 24 hr old were also collected from several northern Minne-
sota lakes having water temperatures between 9° and 16°C. Some of the eggs
were hatched in the laboratory, and the fry were used in toxicity tests. Juve-
nile perch were collected from Lower Red Lake, Minnesota about 2 weeks before
testing, when their total length was 48-62 mm and lake temperatures were 13° to
17°C. Largetnouth bass (48-56 mm total length) and black crappies (41-44 mm
total length) were collected from Round Lake, adjacent to Lake Phalen in Ramsey
County, at water temperatures between 21° and 25°C. Ase'llus were collected
from Rainy Lake, near Ranier, Minnesota, and Ganmavus from a smarll stream
running into the St. Croix River near Marine on St. Croix, Minnesota.
Water Supply
The laboratory well water came from the Jordan sandstone underlying the
Minneapolis-St. Paul area and was delivered through black iron pipes to an
epoxy-coated storage tank. From there it passed through a catalytic iron
removal filter and to the testing apparatus through PVC pipes. Table 1 lists
the chemical characteristics of the water at the point of delivery to the test
apparatus.
Laboratory Acclimation
The temperature of the water used for acclimation was controlled within a
range of -0.1°C by electric immersion heaters. Incoming well water was aerated
by its passing over a series of baffles. Tfeter having a specific oxygen con-
centration was obtained by siphoning it from a point along the series of
baffles. All acclimation chambers were constructed of glass and silicone seal-
ant.
Prophylactic treatments for disease control were administered as soon as
juvenile fish were brought into the laboratory. Neomycin and tetracycline
were used at a concentration of 20 mg/1. Fish were treated for 4 hr each day
for three consecutive days.
The type of food and frequency of feeding varied with the size and devel-
opment phase of the fish and species being acclimated. The various foods used
were (1) finely mashed, hard-boiled egg yolk, (2) "Glencoe" dry pellets,* (3)
live, recently hatched brine shrimp,t and (4) frozen, mature brine shrimp.f
All foods were supplied in quantities slightly in excess of what the fish would
eat in 5 min. AseHus and GcomaTus fed upon small, dead fish and dried leaves
of deciduous trees which had been soaked in flowing water for at least 30 days
just prior to use as food.
* Glencoe Mills, Glencoe, Minnesota
t San Francisco Bay Brand, Metaframe Inc.
1 Ibid.
-------
TABLE 1. ANALYSIS OF LABORATORY WELL WATER*
Item
Total hardness as CaC03
Alkalinity as CaC03
Calcium as CaC03
!
Magnesium as CaC03
Iron
Chloride
Sulfate
Sulfide
Fluoride
Total phosphorus
Sodium
Potassium
Copper
Manganese
Zinc
Cobalt, nickel
Cadmium, mercury
Ammonia nitrogen
Organic nitrogen
Specific conductivity
(umhos/cm at 25°C)
* Water taken from well head
aeration and heating; pH 7.
Concentration
(mg/1)
225
230
140
85
0.02
0.64
<5
0.0
0.16
0.021
4.8
1.6
0.0004
0.0287
0.0044
<0.0005
<0.0001
<0.20
<0.30
430
before
5
-------
The most rapid rate of change in temperature from that of collection to
the desired test temperature was 10°C/hr for eggs and fry, and 2°C/hr for juve-
nile fish. The maximum change in oxygen concentration after collection of fish
eggs and fry was 4 mg/1, and occurred in less than 1 min. Juvenile fish were
subjected to a maximum change in oxygen concentration of 1 mg/l/hr. Fish eggs
and fry were transferred to the test chamber immediately upon reaching the
desired test conditions of temperature and oxygen concentration. Juvenile fish
were held for 7 days under the desired conditions before being placed in the
test chambers. Holding temperatures for AsetTus and Gcamavus were adjusted to
test temperatures at a rate of l°C/hr and' were then maintained for a minimum of
7 days before testing.
Test Apparatus
Each test apparatus was supplied with water at the desired temperature and
oxygen concentration from a separate supply tank. Temperature was controlled
within a range of i0.1°C by a thermostat in conjunction with a solenoid valve,
which allowed hot water to pass through a stainless steel heat exchange coil.
Oxygen concentrations were adjusted and maintained by regulating the amount of
aeration in the head tank. The flow rate of air for aeration was adjusted
with a needle valve and monitored with a flow meter.
Toxicant water was delivered to each test chamber by a diluter modified
from that of Brungs and Mount (1970). The "C" cells were eliminated and "W"
cells dispensed a volume of 1.3 1. Water from the last "W" cell was used to
activate the water flow control bucket and was then wasted to the drain. Move-
ment of the water flow control bucket tripped a tnicroswitch which in turn acti-
vated a solenoid valve in the water supply line. The contents of each "W"
chamber flowed to a separate chemical metering apparatus (CMA) (Mount and
Warner, 1965). Each CMA dispensed a different volume (0.5 to 5,0 ml) of
sodium cyanide solution, so that each test chamber contained a different cya-
nide concentration. The sodium cyanide stock solution was maintained at a pH
of about 11.0 using sodium hydroxide. When the stock solution was mixed with
1.3 1 of test water at a pH of about 7.8, the resulting test solution was
within a few hundredths of the original test water pH. The dispensing volume
of each CMA was adjusted by moving a flexible plastic sleeve to increase or
decrease the length of the pipet bulb. The flow of water to the diluter was
adjusted so that 1.3 1 of water was introduced into each test chamber every
3 min. There were 12 diluters of the type described above, each supplying 1
control and 4 different treatment levels.
The test chambers used for experiments with invertebrates, juvenile fish,
and the swim-up fry of brook trout were constructed of glass the silicone seal-
ant. The tanks measured 50 x 25 x 20 cm high with an outlet 16 cm high, and
contained a volume of 20 1. For experiments with fish eggs and fry, except
for brook trout swim-up fry, a vertical, cylindrical, screen-bottomed, acrylic
plastic chamber was attached to the inside of the outlet tube of the 20-1 cham-
ber described above. The eggs or fry rested on the top side of the horizontal
screen. Water leaving the 20-1 chamber passed upward through the screen
(452-y opening, nylon filament cloth), past the eggs or fry and out the cham-
ber outlet. The cylinders measured 5.6 cm in diameter and 7.5 cm high. The
screened outlet was 2.3 cm from the bottom. To prevent surface stagnation in
-------
the cylinders, two 0.9-cm screened inlet ports were made at a height of 3 cm
and opposite the outlet hole. A series of twelve 0.3-cm holes were drilled
in the tube connecting the screened cylinder to the 20-1 chamber outlet tube.
By varying the number of holes that were open, the flow rate of water past the
eggs could be regulated. The flow rate was adjusted to a level which would
slowly move the eggs without lifting them off the screen.
The test chambers were lighted by two 4Q-W "Vita-lite"* fluorescent tubes
80 cm above the surface of the water. Light intensity at the water surface
was 590-840 lux. Timers were used to provide a constant 12 hr of light and
12 hr of darkness per day. In experiments with eggs, black bakelite covers on
top of the cylindrical chambers excluded all direct light from the eggs.
Test Procedures
The number of organisms placed in each test chamber depended on the spe-
cies and the developmental phase being tested (Table 2). Brook trout and blue-
gill eggs required no special treatment prior to exposure. Fathead minnow eggs,
obtained from natural spawnings of a laboratory-reared stock, were removed from
the underside of a cement-asbestos tile according to the method described by
Cast and Brungs (1973). To facilitate the handling of perch eggs, sections
which had the approximate number of eggs desired were cut from a strand. These
sections were then placed in flowing, aerated water for 1 hr. The viable eggs
in each section were then counted. The eggs were less than 24 hr old when
toxicity tests were started. The sizes of fry, juveniles and adults when tox-
icity tests began are listed in Table 3. Eggs and fry were exposed to cyanide
immediately when placed in the test chambers. Juvenile fish were held in the
test chambers for 3 days before the initial introduction of cyanide. Asellus
and Gammarus were placed in the test chambers 2 hr before cyanide was intro-
duced .
TABLE 2. NUMBER OF ORGANISMS PER TEST CHAMBER
Fry
Species Eggs Sac Swim-up Juveniles Adults
Brook trout 25 10 10 10
Fathead minnow 25-50 25 10 -
Bluegill 25 10-50 10-20
Yellow perch 20-70 25 10-20
Largemouth bass, - - 10 -
black crappie,
rainbow trout
Aseltus - - - - 10
Gammayus - - - - 20
*Luxor Lighting Products, New York
' 10
-------
TABLE 3. APPROXIMATE INITIAL SIZE (MM TOTAL LENGTH) OF FRY, JUVENILES
AND ADULTS FOR TOXICITY TESTS
Fry
Species Sac Swim-up Juveniles Adults
Brook trout 9-10 14-16 40-68
Fathead minnow 5-6 26-45
Bluegill 4 13-28
Yellow perch 4 48-62
Largemouth bass - 48-56
Black crappie — 41-44 -
Rainbow trout - 49-58
Asettus 9
Gammarus 11
Determinations of free cyanide, temperature, dissolved oxygen, pH, and
total alkalinity were made on samples taken from the center of the water mass
of each test chamber. All variables except total alkalinity (determined once
per test) were measured daily. Free cyanide concentrations were determined by
the Epstein colorimetric method (American Public Health Association et at.
1971). Calculated HCN concentrations were based on corresponding pH and tem-
perature measurements using dissociation constants of molecular HCN defined by
Izatt et at. (1962). Oxygen was determined by means of an oxygen specific
electrode, standardized by titration. The method of total alkalinity deter-
mination was that of Dobie and Moyle (1962).
During the cyanide exposure, brook trout swim-up fry and all juveniles
were fed twice per day in amounts slightly in excess of what they would eat.
Brook trout and fatheads were fed Glencoe dry pellets, and the other species
were fed mature, frozen brine shrimp. Asettus and GarmaTus could feed on pre-
soaked leaves placed in the tank for both food and cover.
Observations on the mortality of test organisms were made daily. Eggs
were determined to be dead when the egg contents began to turn an opaque white.
Fry mortality was based on the lack of a heart beat. Juvenile and adult mor-
tality was assumed when there were no opercular or swimming movements and no
response to a physical stimulus. Dead and surviving fish and invertebrates
from all tests, including fish fry which hatched during egg experiments, were
measured. The survivors of experiments with juvenile fish were also weighed.
11
-------
Statistical Analysis
Concentration-percentage mortality data were analyzed with a logarithmic-
probability (log-probit) program (Dixon 1973). Upper and lower 95 percent con-
fidence limits on the 96-hr median lethal toxicant concentration (LC50) and the
median lethal threshold concentration (LTC) were calculated from the equations
i —J?c ¥ i~»t~
log LC50 ± 1.96(l/b)(N'/2) 2
log LC50 - t 05.K_2)(l/b)(x2/D-2)^(N'/2)~^ for heterogeneous data.
The symbol, NT, refers to the number of organisms in treatments having ex-
pected mortality rates between 16 to 84 percent. The reciprocal of the log-
probit line's slope (1/b) is equivalent to the standard deviation of the loga-
rithm of the population's tolerance frequency distribution (a) or the logarithm
of Litchfield and Wilcoxon's (1949) slope function, S. The symbol, K, refers
to the number of treatment levels excluding controls, and x to the (Chi)2 value
calculated for goodness of fit of the regression line to the data points. If
this value is greater than the tabulated (Chi)2 with K-2 degrees of freedom,
the data are significantly heterogeneous. The formulas given above were
derived from Litchfield and Wilcoxon (1949) and Finney (1971). The median
lethal threshold concentrations were determined at the end of a time interval
when no mortality in any test treatment had occurred for at least 48 hr. When
the data were such that analysis by computer was not possible, the dose-
response line was graphically fitted to the data points on semilog paper (APHA,
1975).
RESULTS
The acute toxicity of HCN to fish depended on their stage of development.
Eggs were the most tolerant, and newly hatched fry and juveniles were the most
sensitive. The 96-hr LCSO's and LTC's, their 95 percent confidence limits, and
log-probit regression analysis of the concentration-percent mortality curves
for each exposure period are detailed in Tables 4 through 15.
Fathead Minnow
Tests using the fathead minnow during different life history stages were
conducted at about 15° to 25°C and in dissolved oxygen concentrations (DO) from
about 4 to 7 mg/1. Delayed embryonic development at low temperature and low
DO resulted in no apparent difference in tolerance with decreasing DO but an
increase in tolerance with decreasing temperature for 96-hr egg tests (Table
4). Median lethal HCN concentrations at hatching were markedly reduced for
tests at 25°C and when the DO was less than 5.52 mg/1. At similar DO levels
it appears that a reduction in temperature from 25.0° to 15.2DC resulted in
greater sensitivity at hatching. The time required for eggs to hatch at 25°,
20°, and 15°C was 6, 10, and 17 days, respectively.
In experiments with fry and juvenile fish, there was little difference
between the 96-hr LC50 and the median lethal threshold concentration. There
was no clear trend in sensitivity to cyanide associated with temperature in
fry experiments, but in tests at 25°C and 3.77 mg/1 DO, the fry were obviously
12
-------
TABLE 4. ACUTE TOXICITY OF HCN TO FATHEAD MINNOW EGGS EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL CONCENTRATIONS AT HATCHING
Mean test
conditions No.
°C
15.2
20.0
24.9
24.8
25.0
25.0
24.9
DO
mg/1
6.36
6.13
3.51
4.46
5.52
6.34
7.25
of
pH tests
7.86 2
7.88 11
7.72 2
7.95 2
7.90 3
8.00 5
7.99 1
Log-probit regression
analysis*
Treat-
ments t
8
5
40
36
5
8
6
6
9
10
12
10
-
a
- 5.033
-15.508
2.175
- 2.063
- 5.358
- 8.411
- 8.488
-13.291
- 6.262
- 5.288
- 4.612
-14.863
—
b
3.940
9.765
1.159
3.411
4.494
6.491
6.475
8.909
4.973
4.563
4.194
8.992
—
96-hr LC50
Ug/1
352
273
202
121
184
196
202
95 percent
confidence
limits
274-453
162-463
130-314
77.3-190
115-293
140-274
—
Hatch LC50
yg/i
126
118
116
113 '
180
162
187
95 percent
confidence
limits
90.9-174
97.3-142
86.6-157
83.0-154
122-266
135-193
—
A
*A probit of 4.0, 5.0, and 6.0 corresponds to 16, 50, and 84 percent mortality, respectively, when Y^
is the maximum likelihood probit value of percent mortality and X^ is log HCN concentration in yg/1
for the regression equation O , , ,, „ \
Y.._ = a + b(log Xi).
tNumber of HCN concentrations in the regression analysis.
-------
more sensitive (Table 5). The sensitivity of juvenile fathead minnows to
cyanide was unrelated to temperature or DO (Table 6).
Bluegill
The sensitivity to cyanide of bluegill eggs which had been incubated at
about 25°C and in DO's ranging from 6.90 to 3.39 mg/1 was not affected at the
time of hatching by differences in DO (Table 7). At cyanide concentrations
equivalent to the hatching LC50, essentially all the newly hatched sac fry
were premature and deformed in appearance. Fry were considerably more sensi-
tive to cyanide than eggs, and at 25°C, the 7-day LTC values for fry decreased
with a decrease in DO.
In tests using juveniles, conducted at 8.4° to 24.9°C and relatively con-
stant DO levels, the LTC values increased progressively from 61.7 to 120 yg/1
HCN with increase in temperature (Table 8). However, only a slight reduction
in sensitivity to cyanide with decreasing DO was observed for juveniles tested
at DO's ranging from 6.90 to 3.48 mg/1 and 25°C.
Yellow Perch
Results from yellow perch egg and fry experiments conducted at tempera-
tures from about 10° to 18°C and DO's from about 4 to 7 mg/1 were so variable
that specified median lethal concentrations cannot be considered definitive
(Table 9). For tests with juveniles at relatively constant DO levels, LTC
values increased progressively from 87.6 to 101 yg/1 HCN as test temperatures
increased from 15.0° to 21.0°C (Table 10). An increase in sensitivity with
decreasing DO was observed for juveniles tested at 7.10 to 3.56 mg/1 DO and
21°C; LTC values decreased from 107 to 75.0 yg/1 HCN over the range of DO's.
There was little difference between the 96-hr LC50 and the LTC values for
juvenile perch.
Brook Trout
In brook trout egg tests conducted at temperatures from about 7° to I3°C
and DO's from about 4 to 8 mg/1, the 96-hr LC50 values were all greater than
212 yg/1 HCN. The median lethal concentrations at the time of hatching for
tests conducted at about 6 mg/1 DO varied from 98.0 to 155 yg/1 HCN but did not
appear to be correlated with temperature (Table 11).
Tests using sac fry at 10.0°C and 3.50 to 7.96 mg/1 DO demonstrated that
dissolved oxygen concentrations markedly influenced HCN toxicity. The 96-hr
LC50 and LTC at 3.50 mg/1 DO were 108 and less than 71 yg/1 HCN while at 7.96
mg/1 DO they were 518 and 207 yg/1 HCN, respectively. At the intermediate
level of 6.03 mg/1 DO the 96-hr LC50 and LTC values were 350 and 169 yg/1 HCN,
respectively (Table 11).
The response to DO of swim-up brook trout fry was similar to that of sac
fry; their sensitivity to cyanide did not appear to be correlated to test tem-
perature. The LTC values at 3.90, 6.04, and 8.02 mg/1 DO were 53.9, 83.4, and
103 yg/1 HCN, respectively (Table 12). There was little difference between
the 96-hr and LTC values.
14
-------
TABLE 5. ACUTE TOXICITY OF HCN TO FATHEAD MINNOW FRY EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
15.0
20.0
24.6
24.7
24.9
DO
mg/1
6.38
6.14
3.77
5.14
6.17
PH
7.86
7.89
7.84
7.96
8.02
No.
of
tests
6
6
3
2
9
Log-probit regression
analysis
Treat-
ments
13
15
21
13
10
8
28
a
-12.743
-13.376
- 6.419
- 6.145
- 9.989
- 8.170
- 7.971
b
8.507
9.143
5.721
5.621
7.840
6.471
6.314
96-hr LC50
yg/i
122
99.1
81.6
108
113
95 percent
confidence
limits
104-143
88.9-111
71.2-93.6
90.3-130
96,5-133
yg/l
102
96.1
81.6
108
113
LTC
95 percent
confidence
limits
92.8-113
83.6-110
71.2-93.6
90.3-130
96.5-133
-------
TABLE 6. ACUTE TOXICITY OF HCN TO FATHEAD MINNOW JUVENILES EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
15.0
20.0
19.8
20.0
20.0
20.0
24.8
25.0
25.1
25.2
DO
mg/1
6.07
3.58
4.68
5.20
6.07
7.13
3.58
5.08
6.13
7.04
PH
7.86
7.70
7.80
7.78
7.91
7.90
7.75
7.83
7.98
7.96
No.
of
tests
4
4
2
4
4
4
2
2
8
2
Log-probit regression
analysis
Treat-
ments
15
15
11
11
4
8
9
13
9
7
7
28
6
a
-27.167
-28.565
- 8.512
- 9.924
-24.538
-21.086
-19.292
-20.485
-28.818
-18.494
-23.502
-20.032
-27.069
b
15.455
16.176
6.417
7.146
15.415
12.446
11.616
11.933
15.978
11.592
13.722
11.868
15.415
96-hr LC50
yg/1
121
-
128
-
82.4
125
-
137
131
106
119
129
120
95 percent
confidence
limits
116-125
—
109-149
—
76.4-88.9
117-133
—
122-153
124-138
87.9-129
111-129
124-133
113-128
yg/i
_
119
_
123
82.4
_
123
137
131
106
119
129
120
LTC
95 percent
confidence
limits
__
115-123
—
105-143
76.4-88.9
__
116-132
122-153
124-138
87.9-129
111-129
124-133
113-128
-------
TABLE 7. ACUTE TOXICITY OF HCN TO BLUEGILL EGGS AND FRY EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD OR HATCHING CONCENTRATIONS
Mean test
conditions
°C
Eggs
25.2
25.0
25.1
25.0
Fry
20.0
24.9
24.9
24.9
24.8
DO
mg/1
3.39
4.99
6.09
6.90
5.99
3.59
5.08
6.01
6.81
PH
7.70
7.79
7.92
7.90
7.89
7.72
7.80
7.93
7.90
No.
of
tests
2
2
2
2
6
3
4
8
4
Log-probit regression
analysis
Treat-
ments
6
6
8
6
11
11
4
10
7
22
28
9
5
a
- 1.349
- 4.902
-12.494
- 3.802
- 2.618
- 8.388
-40.381
- 2.756
-18.630
-13.580
-18.117
- 8.222
-11.820
b
2.236
3.630
6.157
3.185
2.973
5.791
22.256
3.279
10.872
7.611
9.883
5.433
7.355
96-hr LC50 LTC or hatch LC50
95 percent
confidence
yg/1 limits yg/1
690
535
693
580
365 188-709
205
109
232 147-366
149
276 241-316
218
271 200-368
194
95 percent
confidence
limits
461-1033
240-1192
572-841
343-980
—
156-270
99.9-120
__
117.189
__
193-247
__
110-340
-------
TABLE 8. ACUTE TOXICITY OF HCN TO BLUEGILL JUVENILES EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
oo
Mean test
conditions
°C
8.4
9.7
15.0
15.1
17.8
20.0
25.1
25.0
24.9
24.9
DO
mg/1
6.08
8.35
6.07
7.03
7.97
6.06
3.48
5.05
6.17
6.90
PH
7.80
7.94
7.83
7.92
8.12
7.86
7.71
7.78
7.92
7.86
No.
of
tests
2
1
3
2
1
6
6
6
5
6
Log-probit regression
analysis
Treat-
ments
5
-
6
3
-
10
9
12
15
16
12
a
-40.261
—
-44.847
-15.982
—
-31.160
-32.839
-16.173
-21.671
-19.485
-30.955
b
25.283
—
25.692
11.190
—
17.797
18.609
10.593
13.003
11.766
17.160
96-hr LC50
Ug/1
83.0
-
<92.0
87.1
75.0
99.0
108
-
99.7
113
120
125
95 percent
confidence
limits
__
—
—
81.0-93.8
66.0-85.2
—
103-112
—
86.0-116
55.8-227
109-133
115-135
yg/i
_
61.7
-
87.1
-
-
—
108
99.7
113
120
125
LTC
95 percent
confidence
limits
__
59.3-64.2
—
81.0-93.8
—
—
—
104-112
86.0-116
55.8-227
109-133
115-135
-------
TABLE 9. ACUTE TOXICITY OF-HCN TO YELLOW PERCH EGGS AND FRY EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD OR HATCHING CONCENTRATIONS
Mean test
conditions
°C
Eggs
14.0
14.1
14.0
14.1
18.0
Fry
10.0
14.0
14.0
14.0
18.1
DO
mg/1
3.36
4.96
6.04
7.2
6.17
7.04
3.84
5.32
7.44
7.00
PH
7.66
7.76
7.87
7.90
7.90
7.90
7.70
7.72
7.87
7.94
No.
of
tests
2
2
2
2
2
2
2
2
4
2
Log-probit regression
analysis
Treat-
ments a b
8 -11.968 6.900
3 -14.050 9.006
6 -14.491 8.849
6 - 5.099 4.813
8 - 0.072 2.340
_
_
8 -10.431 6.249
8 - 1.910 2.734
_
_
96-hr LC50
95 percent
confidence
Ug/1 limits
288 270-307
- —
>329
>317
>389
>276
>357
295 256-339
337 202-561
>395
>338
LTC or hatch LC50
yg/i
—
130
159
125
147
<109
<170
-
-
<179
<165
95 percent
confidence
limits
—
—
131-194
92.7-170
70.0-308
—
—
—
—
—
-------
TABLE 10. ACUTE TOXICITY OF HCN TO YELLOW PERCH JUVENILES EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions No.
DO of
°C mg/1 pH tests
15.0 6.10 7.82 6
18.0 6.05 7.83 6
21.4 3.56 7.69 5
21.0 5.09 7.74 5
21.0 6.13 7.82 6
21.1 7.10 7.82 6
Log-probit regression
analysis
Treat-
ments
20
20
22
22
13
11
13
13
20
21
17
18
-17
-21
-28
-30
-19
-21
-19
-19
-31
-30
-19
-19
a
.790
.367
.964
.573
.935
.323
.206
.711
.329
.843
.768
.512
b
11.
12.
17.
18.
13.
14.
12.
12.
18.
17.
12.
12.
650
575
215
062
266
037
214
538
082
868
172
069
96-hr LC50
Pg/1
90.4
94.0
75.8
95.9
102
108
95 percent
confidence
limits
85.8-95.2
91.8-96.2
69.8-82.2
88.8-104
98.7-106
102-115
Pg/1
87.6
93.2
75.0
93.5
101
107
LTC
95 percent
confidence
limits
83.4-92.
91.0-95.
68.7-82.
86.7-101
97.6-105
101-114
0
5
0
-------
TABLE 11. ACUTE TOXICITY OF HCN TO BROOK TROUT EGGS AND SAC FRY EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD OR HATCHING CONCENTRATIONS
Mean test
conditions
°C
Eggs
7.1
10.0
10.0
10.1
13.0
DO
mg/1
5.96
3.64
5.96
8.13
5.98
pH
7.80
7.71
7.79
7.88
7.77
No.
of
tests
2
2
3
3
1
Log-probit regression
analysis
Treat-
ments
6
-
10
-
4
a
-2.743
—
-1.225
—
-3.695
b
3.735
—
2.842
—
4.367
96-hr LC50
95 percent
confidence
jig/1 limits
>212
>232
>228
>242
>232
LTC or
yg/i
118
<98
155
-
98.0
hatch LC50
95 percent
confidence
limits
63.0-222
—
93.4-257
—
43.9-219
Sac fry
10.0
10.0
10.0
13.0
3.50
6.03
7.96
6.00
7.68
7.78
7.84
7.78
2
4
2
2
4
11
6
6
7
7
8
-2.114
-9.989
-7.149
-5.939
-3.095
-3.478
-8.993
3.502
5.891
5.453
4.030
3.496
3.518
6.327
108 77.1-150
350 290-423
518 403-665
257 193-342
<71
169
207
163
—
106-270
107-401
143-185
-------
N>
TABLE 12. ACUTE TOXICITY OF HCN TO BROOK TROUT SWIM-UP FRY EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
7.0
10.0
10.0
10.0
13.0
DO
mg/1
6.04
3.90
6.04
8.02
6.04
PH
7.83
7.73
7.80
7.85
7.79
No.
of
tests
6
6
14
6
8
Log-probit regression
Treat-
ments
14
20
10
10
37
39
14
15
19
21
analysis
a
- 7.954
- 9.524
-10.709
-12.962
- 8.898
- 7.793
-19.180
-11.872
-17.914
-12.874
b
6.695
7.582
8.996
10.372
7.137
6.659
11.944
8.377
11.663
9.195
96-hr LC50
yg/l
86.0
-
55.8
-
88.6
-
106
-
92.2
-
95 percent
confidence
limits
71.4-104
—
46.8-66.5
—
82.8-94.7
—
100-112
—
79.8-106
—
ug/l
_
82.3
_
53.9
_
83.4
_
103
_
- 87.9
LTC
95 percent
confidence
limits
_»
70.5-96.2
—
46.8-62.1
—
77.2-90.0
—
92.1-116
—
78.2-98.7
-------
TABLE 13. ACUTE TOXICITY OF HCN TO BROOK TROUT JUVENILES EXPRESSED AS 96-HR LC50
AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
6.9
6.8
10.0
10.0
10.0
10.0
13.0
13.1
DO
mg/1
6.01
9.26
4.02
6.00
8.06
8.82
6.04
8.40
PH
7.84
8.06
7.74
7.82
7.90
8.08
7.84
8.08
No.
of
tests
4
2
6
6
6
2
4
1
Log-probit regression
analysis
Treat-
ments
8
7
6
6
10
10
14
9
11
10
5
5
8
8
_
a
-41.017
-39.851
-50.061
-47.866
-31.995
-21.303
-59.000
-46.346
-19.910
-36.891
-35.896
-53.042
-42.817
-28.569
__
b
24.507
24.000
28.585
27.486
19.613
14.290
32.883
26.698
12.634
21.504
20.435
29.153
23.986
16.916
__
96-hr LC50
yg/i
75.4
-
84.4
-
76.9
-
88.4
-
93.7
-
100
-
98.5
-
112
95 percent
confidence
limits
71.
—
77.
—
71.
—
86.
—
85.
—
95.
—
94.
—
__
2-80.0
2-92.2
5-82.8
0-90.8
1-103
5-105
9-102
PgA
_
73.9
_
83.8
_
69.3
_
83.8
_
88.7
_
97.9
_
96.5
112
LTC
95 percent
confidence
limits
„
62.0-88.0
__
75.9-92.6
__
62.6-76.7
__
81.0-86.7
__
85.1-92.5
— —
93.2-103
__
91.9-101
__
-------
Toxieity tests with juvenile brook trout demonstrated little difference
between the 96-hr LC50 and LTC values. A positive correlation between resis-
tance to HCN and either temperature or dissolved oxygen was noted (Table 13).
When the test temperature was 6.9°, 10.0°, and 13 °C the LTC values, at DO
levels of about 6 mg/1, were 73.9, 83.8 and 96.5 yg/1 HCN, respectively. When
and 8.82 mg/1, the LTC values at 10.0°C were
respectively.
DO levels were 4.02, 6.00, 8.06,
69.3, 83.8, 88.7, and 97.9 pg/1.
Other Fish Species
Tests conducted with juvenile black crappies and largemouth bass indicate
that their sensitivity to HCN was similar to that of the other warm-water
species used in toxicity tests (Table 14). Juvenile rainbow trout were the
most sensitive of all species tested (Table 14). The difference in median
lethal concentration between wild stock fathead minnows (Table 15), the most
resistant species tested, and rainbow trout was approximately threefold. The
eggs of all species tested were the most tolerant life history stage. Pry and
juveniles, the least resistant stages, had similar sensitivities.
Slopes of the concentration-mortality curves were similar for all species
but were not the same for all life history stages. In general, the slope for
a specific species is smallest for egg bioassays and then increases with
successive developmental phases to the value determined for juveniles.
TABLE 14. ACUTE TOXICITY OF HCN TO RAINBOW TROUT, LARGEMOUTH BASS
AND BLACK CRAPPIE JUVENILES EXPRESSED AS 96-HR LC50 CONCENTRATIONS
Mean test conditions
DO
°C mg/1 PH
No.
of
tests
Log-probit regression
analysis
96-hr LC50
Rainbow trout
10.1 8.82
Largemouth
25.0 5.94
Bj.ack crappie
25.0 5.74
8.04
7.88
7.90
Treat-
ments
15 -35.492 23.039
yg/1
95 percent
confidence
limits
57.2 55.7-58.7
4 -46.487 25.676 101 95.7-107
3 -17.808 11.376 101 84.7-121
Effect of Temperature
A number of tests with different species indicated a marked positive cor-
relation between tolerance to HCN and temperature rather than a negative one
as might be anticipated. An attempt was made to evaluate the acute toxicity of
HCN to wild fathead minnows subjected naturally to different temperatures at
24
-------
TABLE 15. ACUTE TOXICITY OF HCN TO WILD STOCK FATHEAD MINNOW JUVENILES AT 5°C INTERVALS
FROM 5° TO 30°C EXPRESSED AS 96-HR LC50 AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
5.0
10.0
15.0
20.0
25.0
30.0
DO
mg/1
6.07
6.09
6.02
6.16
5.82
5.66
pH
7.87
7.82
7.88
8.02
7.99
8.15
No.
of
tests
2
2
2
2
2
2
Log-probit regression
analysis
Treat-
ments
6
7
5
8
8
8
6
6
a
-12.903
-15.532
-44.809
-17.799
-20.673
-31.483
-32.804
-21.242
b
9.128
9.259
23.374
9.995
11.331
16.282
17.160
11.957
96-hr LC50
yg/i
>167
165
-
191
-
174
160
157
95 percent
confidence
limits
—
140-194
—
178-205
—
158-192
140-182
126-194
Ug/1
91.5
_
135
_
184
174
160
157
LTC
95 percent
confidence
limits
52.0-161
__
129-141
— —
173-196
158-192
140-182
126-194
-------
different times of the year in a local lake. Fish were tested at 5°C intervals
from 30°C in August to 5°C in December 1975. For each test, a group of fish
was collected when the lake temperature was close to the desired test tempera-
ture. All tests were at a dissolved oxygen concentration of about 6.0 mg/1
(Table 15). A slight increase in tolerance to HCN was observed with a decrease
in test temperature from 30° to 20°C. The corresponding 96-hr LC50 values,
which were equal to the respective LTC's, increased from 157 to 174 yg/1 HCN.
The maximum tolerance to HCN was observed in October and at 15°C with the 96-
hr LC50 and LTC being 191 and 184 yg/1 HCN, respectively. As the test tempera-
ture decreased in the fall, threshold tolerance also markedly decreased so at
5°C the 96-hr LC50 was greater than 167 yg/1, but the LTC reached after 35 days
of exposure was 91.5 yg/1. The increase in sensitivity of field stock fathead
minnows at temperatures above 15°C, a relationship not observed for laboratory
reared individuals, was probably a result of environmental stresses imposed on
the wild fish during the hot summer months.
In a series of tests run during a 2-mo period with brook trout from a
single hatchery lot, the positive relationship between temperature from 4.0° to
18°C and tolerance to HCN was more apparent than it had been for fathead minnows.
The 96-hr LC50 varied from 53.0 yg/1 at 4.0°C to 143.0 yg/1 at 18°C and the LTC
from 51.8 to 138.0 yg/1, respectively (Table 16). The tolerance increased at a
relatively uniform rate with temperature increases to 12°C, but between 12° to
18°C the tolerance increased rapidly with successive 3°C intervals. In tests
run with bluegills at dissolved oxygen concentrations of about 6.0 mg/1 (Table
8), there was approximately a twofold increase in tolerance between 8.4° and
24.9°C.
Effect of pH
There are a number of fish surfaces where exchange of gases and ions
between blood and water can occur, but the gill epithelium is recognized as
the primary site. It is also generally accepted that ions have a lesser tox-
icity than more lipid-soluable, un-ionized molecules. This is because of
difficulty in penetrating strongly charged membranes by the relatively large
hydrated ions, and the resulting repulsion by or adsorption of ions to the
charged protein surface of the membrane. The acute toxicity to fish of solu-
tions containing free cyanide (i.e., HCN plus CN ) is mainly attributed to the
toxic action of molecular HCN, varying with the concentration of gas in solu-
tion and inversely with test pH, and not with the CN anions which are con-
siderably less toxic (Doudoroff and Katz 1950; Doudoroff 1976).
Because the toxicity of weak acids and bases is known to be pH dependent,
the change in tolerance of juvenile fathead minnows to free cyanide as a
function of pH was investigated. The fathead minnow was selected for these
tests as an experimental organism because it has a wide distribution in various
chemically diverse natural waters from acid bog lakes to lakes of high pH.
The minnows were reared in the laboratory under a constant photoperiod in
30-1 glass aquaria receiving a continuous supply of well water at 25°C and
with a pH of approximately 7.9. Six lots of fish were tested during a 15-week
period. All fish were approximately 13 weeks old at the time of testing, and
had a mean total length of 30.8 mm. The mean wet weight of survivors was
289 mg. The fish were fed Oregon Moist and Glencoe pellets twice daily until
26
-------
N>
TABLE 16. ACUTE TOXICITY OF HCN TO BROOK TROUT JUVENILES AT TEMPERATURES FROM 4° TO 18°C
EXPRESSED AS 96-HR LC50 AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
Mean test
conditions
°C
4.0
6.0
9.0
12.0
15.0
DO
mg/ 1
7.19
7.24
7.30
7.04
7.24
pH
7.92
7.83
7.88
7.94
8.02
No.
of
tests
2
2
2
2
2
Log-probit regression
Treat-
ments
5
4
_
4
_
4
-
5
5
analysis
a
-28.849
-32.301
-25.903
__
-31.135
—
-32.474
-26.234
b
19.
21.
_
17.
_
19.
-
18.
15.
626
757
_
403
_
827
-
840
737
96-hr LC50
95 percent
confidence
pg/1 limits
53.0 49.3-57.0
- —
61.8
_ —
68.3
-
72.5
97.5 69.3-137
_
LTC
95 percent
confidence
yg/i
_
51.8
_
59.7
„
66.3
72.0
_
96.6
limits
__
48.5-55.
—
55.0-64.
61.7-71.
—
__
79.3-118
3
8
2
18.0 7.06 8.00
143
138
-------
24 hr prior to exposure to the toxicant,
The 96-hr toxicity bioassays were conducted in three identical test
systems, each including one control and four treatment chambers with a similar
experimental design as previously described for acute toxicity tests. The flow
rate through each chamber provided 99 percent replacement in about 3 hr. The
pH of the test water was controlled by dispensing a sulfuric acid or sodium
hydroxide solution into the water supply reservoirs. The temperature of the
test water' was controlled at 20°C, and the water was aerated in the head reser-
voirs to maintain dissolved oxygen concentrations in the test chambers at about
7.5 mg/1. Each test chamber was illuminated for 12 hr each day by a 40-W incan-
descent bulb placed 25 cm above the chamber. Three days before beginning a set
of bioassays, 10 or 20 fish which had been acclimated to 20°C for 1 week were
randomly placed into each of the 12 treatment and 3 control chambers. Sulfuric
acid or sodium hydroxide was then slowly added to the head reservoirs to attain
the desired pH. The fish were acclimated to the specified pH for at least
2 days before the toxicant was introduced.
The relationship between test pH and the acute toxicity of free cyanide to
the fathead minnow was determined for pH values from about 6.8 to 9.3 at 20°C.
A summary of the test conditions and log-probit analysis of composite tests
grouped according to pH is presented in Tables 17 and 18 and Figure 1. It is
apparent that the 96-hr LC50 values for free cyanide and molecular HCN were
fairly constant within the pH range 6.8 to 8.3. Above this pH range to 9.3
the values diverged; free cyanide levels increased and HCN decreased.
TABLE 17. MEAN TEST CONDITIONS FOR COMPOSITE FREE CYANIDE BIOASSAYS;
STANDARD DEVIATIONS IN PARENTHESES
Test pH
6.830(0.028)
7.559(0.041)
8.286(0.034)
8.672(0.067)
8.974(0.068)
9.262(0.091)
Temperature
°C
20.1(0.12)
20.0(0.02)
20.0(0.01)
20.1(0.09)
20.0(0.12)
20.2(0.09)
DO, mg/1
7.65(0.06)
7.56(0.09)
7.66(0.06)
7.59(0.09)
7.71(0.10)
7.63(0,12)
Alkalinity
mg/1 CaCOa
Bicar-
bonate
Free CO-, in
test solution*
96
193
240
229
214
206
Total
96
193
240
244
243
250
mg/1
28.0
11.0
2.5
1.0
0.47
0.24
mm Hg
12.4
4.9
1.1
0.44
0.21
0.11
* Free C02 evaluated by nomographic method (APHA 1975) . Assuming
K = H2C03/P and log K = -1.41 at 20°C and one atmosphere, then
1 mg/1 C02 = 0.444 ram Hg C02 tension (Stumm and Morgan, p. 148, 1970).
28
-------
TABLE 18. BIOLOGICAL ASSAY BY THE LOG-PROBIT ANALYSIS METHOD OF FATHEAD
MINNOW CYANIDE BIOASSAYS AT 20°C GROUPED ACCORDING TO TEST pH
Test
PH
6.830
7.559
8.286
8.672
8.974
9.262
Log-prob:
Treat-
ments
6
7
6
9
17
10
It regression
a
-23.80
-23.47
-27.37
-33.46
-16.52
-28.73
analysis*
e
HCN
13.76
13.81
15.52
18.88
10.68
17.55
96-hr LC50
pg/1 as HCN
124
115
112
109
104
83
95 Percent
confidence
limits for
LC50, yg/1
106-144
102-130
115-128
105-113
96-112
80-87
Free Cyanide
6.830
7.559
8.286
8.672
8.974
9.262
6
7
6
9
17
10
-23.82
-23.60
-28.00
-34.59
-16.44
-28.57
13.76
13.82
15.54
18.64
9.83
15.29
124
117
133
133
152
157
107-144
102-135
126-140
128-138
140-165
142-173
*For equation Y. = a + g (log X.) when Y. is the maximum likelihood
probit value and X. is log cyanide concentration as pg/1 HCN (Dixon
1973).
29
-------
I60r
o:
ui
13
(O
<
UJ
o
I
o
FREE CYANIDE
CN"
HCN
PH
Figure 1.
Relationship between test pH and 96-hr LC50 cyanide concentrations
for the fathead minnow at 20° C
-------
At pH 8.7 and below the toxicity of free cyanide (HCN + CN ) is due
principally to HCN but CN apparently contributes to the observed toxicity at
higher pH values. Since pH is seldom greater than 8.7 in most natural waters,
it is proposed that the CN ion contributes very little to the toxicity of
solutions containing free cyanide (Broderius et a.1. 1977).
Invertebrates
In acute toxicity tests with Asellus conmunis Say at 18°C and DO of
7.52 mg/1 the 96-hr LC50 was 2328 yg/1 and LTC (at 11 days) was 1834 yg/1, both
with broad confidence limits (Table 19) . Tests with Gammopus ps&udot'irnnaeus
Bousfield at temperatures from 5.2°C to 20°C and DO of 6.70 to 7.88 mg/1,
indicated that the median lethal threshold concentrations of HCN varied from
58.1 jig/1 at 20°C to 183.8 yg/1 at 5.2°C. The linear regression of temperature
(X) versus LTC (Y) for Gammarus is expressed by the equation
Y = 225.7 - 8.117 (X)
with r2 = 0.96.
Summary
The 96-hr LC50 values for most species and their life history stages
occurred at a slightly higher HCN concentration than when acute toxicity is
measured at the median lethal threshold level. The life history stage most
sensitive to HCN varies among the fish species, but it was generally observed
that eggs are the most tolerant and juveniles the least tolerant. The response
of fish eggs is often so variable that an acute toxic concentration of HCN
cannot always be confidently assigned. The range of acutely toxic concentra-
tions of HCN among species is not great in the juvenile stage, except that
rainbow trout are appreciably more sensitive than brook trout or any of the
other species which were studied. At high dissolved oxygen levels and tem-
peratures above 9°C, the acute toxicity of HCN to juvenile fish varied from
57 yg/1 for rainbow trout to 191 yg/1 for wild fathead minnows. Temperature
has a marked effect on acute toxicity of HCN; juvenile fish in general become
more sensitive at lower temperatures. The sensitivity of wild stock fathead
minnows also increased at high temperatures. Dissolved oxygen levels below
5 mg/1 increased the sensitivity to cyanide of the juveniles of all species
tested. The implications of lowered oxygen and lower temperatures during cold
weather seasons in systems receiving cyanide wastes should be carefully eval-
uated in the development of standards set by regulatory agencies. No chronic
tests have been conducted incorporating low temperature cycles, but the
possibility of increased sensitivity and adverse effects from this factor
should not be ignored.
31
-------
TABLE 19. ACUTE TOXICITY OF HCN TO ASELLUS COMMUNIS AND GAMMARUS PSEUDOLIMNAEUS ADULTS EXPRESSED
AS 96-HR LC50 AND MEDIAN LETHAL THRESHOLD CONCENTRATIONS
NJ
Asellus
oommunis
Gammams
pseudo limnaeus
Mean test
conditions
DO
°C mg/1 PH
18.0 7.52 8.13
5.2 7.05 7.86
10.0 7.06 7.88
15.0 7.03 7.91
18.0 7.60 8.00
20.0 6.70 8.14
Log-probit
No. regression analysis
of Treat-
tests ments
2 6
6
2 8
2 7
2 8
8
3 11
11
2 8
8
a
- 4.756
-10.152
- 0.654
— *
- 5.760
- 3.422
- 2.342
- 1.113
1.149
- 0.158
b
2.898
4.643
2.497
— *
4.826
4.479
3.271
3.318
2.002
2.924
96-hr LC50
95 percent
confidence
yg/1 limits
1710-
2328 3168
—
— —
— —
142.8-
169.6 201.5
—
133.5-
175.6 231.0
—
63.7-
83.9 110.6
— —
LTC LC50
95 percent
confidence
pg/1 limits
— —
1834 1513-
2224
145.5-
183.8 232.1
140.0* — *
—
75.9 65.7-
87.7
—
69.6 53.0-
91.2
— —
58.1 38.8-
87.0
* LC50 calculated by graphical interpolation; computer program could not handle data
-------
SECTION .V
CHRONIC TOXICITY
Three species of fish, the fathead minnow, brook trout, and bluegill, and
two invertebrates, Asellus and Gamrnavus, were used in experiments to determine
the chronic toxicity of cyanide.
Little information is available on the chronic effects of hydrogen cya-
nide on fish. Neil (1957) and Broderius (1970) found that free cyanide con-
centrations of 10 yg/1 expressed as CN impaired the swimming performance of
salmonid fishes. Leduc (1966) measured the growth of juvenile cichlids
{Ci-oHlasoma bimaculatum) receiving unrestricted rations of tubificid worms and
subjected to various levels of free cyanide for 36 days at 25°C. Fish exposed
to free cyanide levels of 60 yg/1 and higher, expressed as HCN, grew more
slowly than control fish during the first 12 days of the experiments, but grew
faster than control fish during the final 12 days so that the effect of free
cyanide up to 100 yg/1, the highest treatment, on total weight gain was slight.
Leduc (1966) observed a similar pattern of growth with time in a single
24-day experiment at 16°C with juvenile coho salmon (Oneorhyneus kisutoh) fed
an unrestricted diet of earthworms. In this experiment, however, total weight
gain was markedly reduced compared to controls at both the lowest (10 yg/1)
and highest (80 yg/1) treatment levels but appeared to be unaffected at inter-
mediate levels of 20 ug/1 and 40 ug/1 free cyanide as HCN. The author did not
explain the poor growth in the lowest treatment level.
Since experimental designs were different for each species, descriptions
of apparatus and design are handled separately. The test water and analytical
procedures were the same as those described in the section on acute toxicity.
FATHEAD MINNOW
Experimental Design
The experiment was started in fifteen 20-1 glass chambers. A cyclic flow
of water was supplied to the test chambers at the rate of 1 1 every 3 tnin.
Twelve different treatment levels and three controls were randomly assigned to
test chambers. A toxicant metering system [Mount and Warner (1965)] intro-
duced sodium cyanide stock solutions to treatment chambers with each water-
delivery cycle. Nominal treatment levels were from 5 to 100 yg/1 HCN. Mean
temperatures in test chambers varied over the experimental period from 24.8°
to 25.1°C with standard deviations from 0.2° to 0.3°C. Mean pH readings
ranged from 8.06 to 8.09 with standard deviations from 0.04 to 0.12, and mean
33
-------
concentrations of dissolved oxygen were from 5.9 to 6.3 mg/1, with standard
deviations from 0.5 to 0.7 mg/1. Each test chamber was lighted with a 40-W
incandescent bulb 30 cm above the water surface. The photoperiod simulated
Evansville, Indiana day length starting at the December 15 day length of 10 hr,
30 min, and was adjusted biweekly.
Over a 4-day period, 80 newly hatched fathead minnow larvae of a strain
from the U.S. EPA Environmental Research Laboratory, Duluth, Minnesota, were
placed in each test chamber. When all chambers contained 80 larvae, sodium
cyanide was introduced into the treatment chambers. Free cyanide concentra-
tions in each chamber were determined three times weekly thereafter.
Through the first week of exposure to HCN, the fish were fed a fine,
hard-boiled egg yolk suspension three times daily. Beginning on the second
week, a supplement of finely ground Glencoe granules was fed twice daily. The
fish were fed newly hatched brine shrimp (Artemia) once daily beginning the
third week, and finely chopped lettuce was added to the egg yolk suspension in
the fifth week. During the first 4 weeks of the test, fish were fed an excess
ration of food. Fish were counted after 28 days of exposure to HCN, and sub-
sequent feedings were adjusted to the number of fish in each tank. After 56
days individual fish were weighed and randomly thinned to a maximum of 40 fish
per chamber. The individual fish were again weighed after 84 days. After 106
days the fish were transferred to a larger test unit consisting of 20 treat-
ment and 5 control chambers, each containing 35 1. The chambers were randomly
arranged in groups of five. Each group of chambers was illuminated by two
40-¥ Vita-Lite fluorescent tubes. One treatment (No. 7) was lost before trans-
fer to the larger apparatus. Of the remaining treatments, eight contained
enough fish to be duplicated in the new apparatus so that each chamber con-
tained approximately 20 fish. The three highest treatments each contained 20
fish or fewer and were not duplicated. Two of the three original control
chambers were duplicated in the new apparatus. A spawning substrate was pro-
vided by placing four 7.5-cm long halves of 7.5- and 10-cm diameter asbestos
drain tile in each tank.
The first spawning took place after 149 days. When the first spawning
occurred in a chamber, four mature males and three mature females were
selected to remain, and two spawning tiles were added. The 20 fish in one of
the previously unreplicated treatments were divided between 2 chambers after
the first spawning in the treatment.
Egg incubation experiments began after 192 days. When spawning occurred
in a chamber, some of the eggs were removed from the tile, and a sample of 50
undamaged eggs was placed in each of two 5.7-cm ID acrylic plastic cylinders
covered at one end with 351 p nylon monofilament screen. Cylinders were sus-
pended in the test water and oscillated at two cycles per minute through an
amplitude of 2.5 cm. Eggs not incubated were counted and discarded.
Simulated day length reached a maximum after 183 days and began to
decrease after 214 days. After 200 days of exposure, mortality, mostly
females, began to occur in the higher treatment levels. Spawning ceased after
256 days.
34
-------
After 227 days, a 56-day growth and survival experiment was started with
second generation larvae hatched in the parent experiment. The test conditions
in this experiment were identical to those imposed on the previous generation.
Photoperiod, however, was kept constant at the maximum day length of 15 hr,
45 tnin. Individual treatments were started on different dates, since all
treatments in the parent experiment did not produce eggs on the same day. Some
parent treatments did not produce fertile eggs in sufficient numbers to con-
tinue those treatments through the second generation. Fry which had been
spawned and hatched in control chambers were substituted in those treatments.
After 28 days of exposure to HCN, the second-generation fish were measured
photographically (Martin 1967) and were randomly thinned to a maximum of 40 per
chamber. After 56 days the fish were measured and weighed, and the experiment
wa s t ermina ted.
For statistical analysis, lengths and weights were transformed to loga-
rithms and percentages to angles (Steel and Torrie 1960). Lengths, weights,
ages at the onset of spawning, egg production, and hatchability data were sub-
jected to one-way analysis of variance. Treatment means were compared with
control means using Dunnett's procedure (two-tailed) with Kramer's modifica-
tion for unequal replication, at a significance level of a = 0.05. Means in
tables were calculated from transformed values.
Results
First Generation—
Survival—After 28 days, survival in the first-generation experiment
averaged 64 percent in the three controls, and .ranged from 80 percent to 11
percent in treatments (Table 20). A significant (a = 0.01) correlation
coefficient of -0.84 was found between survival and HCN concentration (controls
omitted). Lack of treatment replications in this phase of the parent experi-
ment precluded analysis of variance and treatment comparisons.
Over the period between 28 and 56 days, survival was at least 80 percent
in all chambers. Survival was at least 95 percent in all chambers between
56 and 84 days (Table 20). Survival in these two periods was not significantly
correlated with HCN concentration at the 0.05 level.
Growth—The mean weight of fish in treatments ranged from 130 percent to
65 percent of that in controls after 56 days, and from 128 percent to 86 per-
cent of mean weight in controls after 84 days. Although food was introduced
to each chamber according to the number of fish it contained after 28 days, the
tendency for variations in cumulative percentage survival to influence the
relationship between mean weight after 56 days and HCN concentration can be
seen in Table 20. For this reason no analysis of variance was performed.
Reproduction—The length of time from the start of HCN exposure to the
onset of spawning averaged 156 days in controls, and ranged from 148 to 206
days in treatment chambers. In no treatment did spawning begin significantly
earlier or later than in the controls.
35
-------
TABLE 20. SURVIVAL AND WEIGHT OF FIRST-GENERATION FATHEAD MINNOWS AFTER
28, 56, AND 84 DAYS OF EXPOSURE TO HYDROGEN CYANIDE*
Mean HCNt
concentration Percentage survivalf
Treatment
Control A
Control B
Control C
1
2
3
4
5
6
8
9
10
11
12
(yg/1) 0-28 d 28-56 d
5.9
11.4
17.9
24.7
32.8
40.5
57.5
66.8
75-3
88.9
98.1
64
71
58
80
59
60
51
46
59
49
49
29
19
11
94
100
100
100
100
98
100
100
100
97
87
91
80
100
56-84 d
100
100
100
98
100
95
100
97
100
97
100
100
100
100
Mean weight
(g)
56 'd
0.292
0.236
0.354
0.205
0.274
0.270
0.296
0.382
0.272
0.190
0.220
0.264
0.199
0.190
84 d
0.581
0.588
0.668
0.580
0.676
0.631
0.688
0.785
0.555
0.458
0.528
0.629
0.560
0.627
*Chambers originally contained 80 larvae. Numbers were reduced to a
maximum of 40 per chamber after 56-day measurements.
tEighty—four day period.
fSurvival of fish present at beginning of period.
36
-------
Total egg production and egg production per female in each chamber are
shown in Table 21. Due to the mortality of adult females in the higher treat-
ment levels, the following calculation of egg production per female was
required: the number of females surviving during each day of the 107-day
spawning period was totaled for each chamber and divided by 107; the quotient
was, in turn, divided into the total number of eggs produced in the chamber.
Egg production per female in treatments ranged from 72 percent of that in con-
trols to zero (Table 21). Egg production per female was significantly reduced
relative to controls in HCN treatments of 19.6 yg/1 (Treatment 3) and above.
The number of egg samples incubated in each chamber and the mean percent-
age hatch are shown in Table 21. Control B was omitted from statistical
analysis of hatchability because the mean percentage hatch (20,3 percent) of
egg samples spawned and incubated in that chamber was 17 standard deviations
lower than the mean percentage hatch in the other four controls (83.9 percent).
The mean percentage hatch of egg samples spawned and incubated in treatments
ranged from 84.7 percent to 10.0 percent.
Using one-way ANOVA with subsampling, differences in mean percentage hatch
between duplicate chambers within treatments were found to be significant
(Flljl62 = 2.27, a = 0.025). In Treatments 6 and 9, mean percentage hatch was
significantly lower than in the controls. The small number of samples incu-
bated in Treatments 8 and 10 decreased the sensitivity of statistical compar-
isons between those treatments and the controls so that the differences found
were not significant. When the parent experiment was terminated, the weights
of survivors of either sex in the treatment chambers did not differ signifi-
cantly from the weights of control fish (Table 21).
H?l Generation—
Survival—After 28 days survival was generally higher in the F]_ experiment
than in the parent experiment. Survival averaged 84 percent in the three con-
trols and ranged from 88 percent to 36 percent in treatments (Table 22).
Survival and HCN concentration were not significantly correlated at the 0.05
level. Over the period of 28 to 56 days, all chambers had at least 81 percent
survival. Survival from 28-56 days was not significantly correlated with HCN
concentration at the 0.05 level.
Growth—Mean length of fish in treatments after 28 days ranged from 116
percent to 64 percent of that in controls. The fish at 26.3 yg/1 were signifi-
cantly longer than control fish, and those exposed to 34.8 yg/1 (Treatment 5)
and above were significantly shorter than control fish.
Since the numbers of fish in experimental chambers were made approximately
equal after 28 days in the F} experiment, the effect of HCN on fish size after
56 days is more evident here than in the present experiment. Mean lengths of
fish in treatments after 56 days ranged from 105 percent to 81 percent of mean
length in controls, and mean weights ranged from 122 percent to 52 percent of
that in controls (Table 22). No treatments contained significantly longer or
heavier fish than did the controls. Mean lengths and weights of fish at 61.6,
70.5, 95.9 and 105.8 yg/1 HCN, respectively, were significantly reduced when
compared with controls. The differences between length and weight of the
37
-------
TABLE 21. EGG PRODUCTION, HATCHABILITY, AND TERMINAL WEIGHTS
FOR FIRST-GENERATION FATHEAD MINNOWS
EXPOSED TO VARIOUS CONCENTRATIONS OF HYDROGEN CYANIDE*
HCN concentra-
tion (UK/I)
Treatment
Control A
Control B
Control C
Control D
Control E
1 A
B
2 A
B
3 A
B
4 A
B
5 A
B
Mean
(256 d)
5,7
5.9
13.0
12.7
19.6
19.6
27.1
27.5
36.0
35.6
Std.
dev.
1.4
1.3
1.6
1.8
3.9
4,1
2.7
3.4
4.0
3.9
Hatchability
Eggs
per
female
2530
4483
3990
2718
3660
1886
3138
1701
1989
1694**
1241**
1093**
1640**
678**
1341**
Mean
percent-
age
86.2
20.3
87.9
79.4
82.1
38.8
84.3
81.2
81.4
84.7
28.0
34.8
43,8
38.3
62.9
Egg
samples
10
23
13
7
16
12
11
9
13
11
7
6
7
6
15
Mean weight of
surviving adults t
(g)
Males
4.78(4)
4.11(4)
3.93(4)
4.47(4)
4.85(3)
5.24(4)
3.76(4)
3.89(4)
5.45(4)
3.95(4)
4.67(4)
5.22(3)
4.55(4)
3.01(4)
5.06(4)
Females
2.36(3)
2.34(3)
2.59(3)
2.44(3)
2.52(3)
2.84(3)
2.37(3)
1.86(3)
2.37(3)
3.02(3)
1.64(2)
1.73(1)
2.39(1)
2.57(1)
2.42(3)
- Continued -
38
-------
TABLE 21. CONTINUED
HCN concentra-
tion (ug/1)
Mean
Treatment (256 d)
6 A 43.7
B 44.7
8 A 62.5
B 64.6
9 A 73.1
B 72.4
10 A 81.5
B? 79.8
11 § 96.1
12 # 105.4
Std.
dev.
7.0
5.1
8.0
9.2
8.1
10.5
12.1
14.1
9.3
10.1
Eggs
per
female
2054**
194**
74**
70**
573**
64**
266**
219**
0**
0**
Hatchability
Mean
percent- Egg
age samples
10.0** 18
15.6** 3
11.2 5
32.0 1
19.1** 8
20.0** 1
20.4 4
0
0
0
Mean weight of
surviving adultst
(g)
Males
4.48(3)
5.52(4)
5.00(3)
3.84(3)
4.60(3)
3.82(1)
3.85(3)
5.13(1)
4.58(4)
0
Females
2.39(1)
2.13(2)
2.26(3)
3.00(2)
2.41(3)
1.96(2)
2.54(2)
1.92(2)
1.89(5)
2.52(6)
* Except where noted, chambers contained four males and three females
when spawning began.
t Parenthesized values are numbers of survivors.
$ Started after 157 d with three extra males and three extra females from
10A.
§ Contained five males and six females after 106 d; too few for
replication.
# Contained two males and six females after 106 d; too few for replication.
** Values are significantly different from control values according to
Dunnett's procedure (two—tailed; a = 0.05).
39
-------
TABLE 22. SURVIVAL, LENGTH, AND WEIGHT OF Fx FATHEAD MINNOWS AFTER 28
AND 56 DAYS OF EXPOSURE TO HYDROGEN CYANIDE
Treatment
Control A §
Control B §
Control C §
1
2
3
4#
5
6§
7§
8§
9
10**
11§
12§
HCN
cone.'''
(Mg/D
5.7
12.2
20.5
26.3
34.8
43.0
52.2
61.6
70.5
81.0
95.9
. 105.8
Percentage
survival*
0-28 d
85
83
84
80
39
60
88
48
64
58
64
56
36
64
„ 40
28-56 d
100
98
95
100
94
100
95
100
100
98
100
95
95
100
81
Mean total Mean weight
length (mm) after 56 d
28 d
15.0
14.2
13.6
14.5
14.7
13.9
16.6$
12.5$
12. 0$
10. 8$
11.7$
10.5$
10. ot
10.4$
9.2$
56 d
29.2
28.7
26.7
28.7
28.0
28.6
29.6
27.6
27.1
26.8
26.7$
25.9$
24.8
24.8$
22.9$
(g)
0.238
0.238
0.205
0.234
0.258
0.221
0.277
0.184
0.212
0.199
0.172$
0.160$
0.163
0.150$
0.188$
*Except where noted, chambers originally contained 80 larvae. Numbers
reduced to a maximum of 40 per chamber after 28-d measurements.
«
t84—d period.
$ Survival of fish present at beginning of period.
§ Larvae spawned and hatched in control chambers of parent experiment.
//Began with 42 larvae.
**Began with 59 larvae.
$ Values are significantly different from control values according to
Dunnett's procedure (two-tailed; a = 0.05).
40
-------
survivors in Treatment 10 at 81.0 yg/1 HCN and that of the controls was not
significant.
Summary—
Most of the mortality in treatments and controls occurred during the first
28 days of both the parent and FI experiments (Tables 20 and 22). Survival
during this period was better in the Fj experiment and was significantly cor-
related with HCN concentrations only in the parent experiment. The disparity
between generations probably was not caused by development of resistance to
HCN by the second generation of exposed fish, since fish which had been
spawned in controls of the parent experiment and used in Treatments 6, 7, 8,
11, and 12 of the F^ expermient appear to have been as resistant to HCN as
fish which had been spawned in treatments. The staggered starting dates of
individual treatments or the fact that no two chambers contained fish from the
same spawning may have contributed to the greater random variation in survival
in the Fj experiment.
The combined effects of survival and HCN concentration on early growth of
the parental generation make it difficult to interpret treatment-to-treatment
differences in mean weight at 56 and 84 days. In the F} experiment, where
numbers of test fish were reduced after 28 days, 34.8 yg/1 HCN and higher
significantly reduced length after 28 days of exposure, and 61.6 yg/1 and
higher significantly decreased length and weight after 56 days. The tendency
of fish in treatments to approach the size of control fish between 28 and 56
days, a growth pattern also observed by Leduc (1966), may have been the result
of size-dependent physiological or behavioral phenomena or may have been an
experimental artifact caused by crowding of the larger fish in the controls and
lower treatments. Again, there is no evidence that the fish in Treatments 6,
7, 8, 11, and 12, which had been spawned in controls, were more or less resis-
tant to HCN than the fish that had been spawned in their respective treatment
levels.
The number of eggs produced per female was' significantly reduced compared
to controls at 19.6 yg/1 HCN and above. Egg hatchability was significantly
lower in Treatment 6 (44.2 yg/1) and Treatment 9 (72.8 yg/1) than in controls.
We estimate that the highest no-effect level of HCN for the fathead minnow
lies between 12.9 and 19.6 yg/1, based on statistical evaluation of egg pro-
duction. The lethal threshold concentration of HCN for juvenile fathead
minnows of the test strain at 25°C, pH 8, and 5-7 mg/1 DO, derived from acute
toxicity bioassays (Table 6) is approximately 120 yg/1, The application factor
therefore (no-effect level/LTC) lies between 0.11 and 0.16.
The statistical significance of an effect in the laboratory does not
necessarily imply ecological importance. Waller et al, (1971), in a computer
simulation study, found that a population of fathead minnows subjected to ran-
dom environmental variation would be driven to extinction by an 80 percent
reduction in the number of progeny recruited into the spawning stock. Assuming
that recruitment is proportional to fecundity, such a level of reduction could
be achieved between 44.2 and 63.6 pg/1 HCN on the basis of fecundity effects
alone. At these toxicant concentrations, reduction in hatchability and early
41
-------
survival could also be expected to influence the level of recruitment.
BROOK TROUT
Experimental Design
Adult brook trout were tested in 10 fiberglass tanks (210 x 54 x 54 cm)
with a 30-cm standpipe 12 cm from the downstream end. The tanks contained a
volume of 340 1. Each of two head tanks fed water to four cyanide treatments
and a control.
A diluter system similar to that described by Brungs and Mount (1970),
but utilizing a microswitch and two magnetic drive pumps, was used to control
water flow. The toxicant dispensing system was similar to the one described
by Mount and Warner (1965). The dilution system dispensed 4 1 of water every
2.5 rain, providing 99 percent replacement in 17 hr.
The adult fish in each tank were allowed to spawn in two boxes 50 x 3 8 x
15 cm deep described by Benoit (1974). Eggs were removed on the day of
spawning from each spawning box and placed in incubation cups made of acrylic
plastic cylinders 6.5 cm OD with Nitex screen bottoms. Eggs were incubated in
separate diluter systems employing 20-1 glass aquaria. The egg cups were
oscillated in the test water by means of a rocker-arm apparatus described by
Mount (1968). Brook trout fry and juveniles were subsequently kept in the
diluter system used for egg incubation. The diluter system was similar to that
described for adults, but it utilized two solenoid valves instead of pumps. A
water volume of 1.35 1 was dispensed every 2.5 min, and 99 percent replacement
occurred in 3 hr.
Temperature, pH and free cyanide concentrations in test waters were
measured three times per week, and dissolved oxygen was measured twice per
week. In all treatments the mean dissolved oxygen concentration ranged from
64-74 percent saturation, the pH from 7.94-8.01, and total alkalinity from
236-239 mg/1 as CaGOg. Water temperatures in the adult brook trout experiments
were varied on a seasonal basis with 14°C in May; 15° in June-August; 12°C in
September; and 9°C in October. ,A11 egg and juvenile tests were conducted at
9°C. Eggs were incubated and hatched in the dark. Fluorescent lights were
used in all experiments with other stages. Photoperiod was adjusted bimonthly
to simulate the day lengths in Evansville, Indiana.
Nineteen-month-old adult brook trout with mean weight of 103.9 g were
obtained from the Wisconsin State Fish Hatchery in Osceola, Wisconsin, on
April 29, 1975. Ten fish were randomly placed in each experimental tank on
the day of collection, and were kept at 14°C and treated 8 hr per day for 3
days with 20 mg/1 tetracycline. Water flow was stopped for 8 hr during treat-
ment, and the temperature rose 0.5°C during this period. After treatment the
fish were anesthetized with MS-222 for weighing and measuring. Fish were fed
dry pellets at a mean rate of 1.3 percent of their wet body weight per day.
Five days after fish were placed in the test tanks, the flow of toxicant was
started. The fish were weighed and measured at the beginning of each month
until they were thinned. On September 25 the adults were randomly thinned to
two males and four females per tank, except at 53.9 pg/1 HCN where three
42
-------
females were present. On the same day a spawning box was placed in each tank
at the downstream end, and a second box was placed in the middle of each tank
on September 30. Spawning boxes were inspected daily for eggs. The total
number of eggs, the number of viable eggs, and the number of spawnings were
recorded for each test tank. All spawners were sacrificed 3 weeks after the
last spawning or egg deposition. The gonads were inspected for development and
to determine which fish had spawned. From each spawning of 50 eggs or more, 50
randomly selected eggs were placed in incubation cups, and the remainder, up to
250, were placed in other incubation cups for determination of viability.
Viability was reported as the percentage of eggs which developed a neural keel
within 12 days at 9°C. Dead eggs were removed daily from the cups. At concen-
trations where no spawning occurred, eggs from controls were used to determine
hatchability, juvenile survival, and growth. when hatching was completed, 25
fry were randomly chosen for growth and survival tests. Weight at hatching
was determined from discarded sac fry. The sac fry were kept in the incubation
cups for 21 days and then released into the 20-1 tanks. Since the hatching
period was spread over 3-6 days, the median hatching date was used to establish
a growth period of 90 days. The temperature during incubation and growth of
juveniles was maintained at 9°C. The fish were measured at hatching and 30,
60, and 90 days after hatching, and were weighed at hatching and 90 days after-
ward. Length was measured photographically (Martin 1967) from the snout to the
end of the caudal peduncle. The juveniles in the growth tests were fed Glencoe
pellets two times per day and newly hatched brine shrimp once per day ad
libi-tim. The numbers of juveniles were reduced to 20 fish per tank at 60 days.
Growth data were transformed to logarithms and analyzed with one-way
analysis of variance followed by Duncan's new multiple range test (Steel and
Torrie 1960). One-way analysis of variance was used on the viability and hatch
results, with an arcsin square root transformation of the percentages.
Results
Adults—
Growth and Survival—Adult brook trout grew to a mean weight of 229.2 g
during 144 days of exposure to cyanide. No significant growth reduction was
noted at any concentration (p > 0.05). One fish died in each treatment at
53.9, 64.9, and 75.3 yg/1 HCN. The deaths occurred after the temperature was
lowered to 9°C.
Reproduction—Spawning began on September 27, 1975, in one control tank
and in the lowest HCN level, 5.7 yg/1 (Table 23). Spawning in other treat-
ments started a few days later, and the peak spawning rate occurred the second
week of October. Results from the treatment at 21.8 yg/1 HCN were omitted
from the spawning analysis because only one female developed eggs in this
treatment, and the deposited eggs were infertile.
The number of eggs spawned per female decreased with an increase in HCN
concentration above 5.7 yg/1 (Table 23). Regression analysis indicated that a
significant downward trend occurred with a resulting regression described as
Y = 552.8 — 6.49X, where: Y = mean number of eggs spawned per female and X =
HCN (yg/1). The correlation coefficient between HCN concentration and eggs
43
-------
TABLE 23. EGG PRODUCTION OF ADULT BROOK TROUT EXPOSED TO HCN
FOR 144 DAYS PRIOR TO START OF SPAWNING
HCN
(yg/i)
Control A
Control B
5.7
11.2
32.3
43.6
53.9
64.9
75.3
Standard
deviation
(yg/i)
-
0.9
1.3
3.8
3.9
6.8
7.3
8.8
* Formation of neural
spawned per
female was -0
Sex
ratio
M/F
2/4
2/4
2/4
2/4
2/4
2/3
2/3
2/3
2/2
keel
.84.
Mean eggs
Spawn- Percent spawned
ing viability* per female
5
11
6
3
3
3
3
1
0
93.6
93.4
89.9
78.1
72.9
86.6
64.1
0
0
502
744
513
291
246
442
262
124
0
Mean weight
of females
(g)
237.6
225.0
198.5
238.5
156,1
258.0
213.0
204.5
193.5
Egg viability was determined for all spawnings, and no fertile eggs were
found at 64.9 and 75.3 ug/1 HCN. Statistical testing of the viability of all
spawned eggs in the remaining treatments showed no significant effects due to
cyanide (p > 0.05). A composite sperm sample from several untested fish was
placed in each of the HCN concentrations used in the spawning tests with no
apparent reduction in motility (p > 0.05) at any level.
Embryos, Fry and Juveniles—
Growth—Length and weight of fry at hatching were not significantly
different (p > 0.05) in any treatment or control, but growth of juveniles in
HCN was reduced after hatch (Table 24). After 30 days of exposure no signif-
icant difference (p > 0.05) in length was found between fish exposed to 11 yg/1
HCN and fish in the control. Growth in all other treatments was significantly
different from control growth (p < 0.05). Growth was greater at 5.6 yg/1 than
in the controls and less at all treatments greater than 11 yg/1 HCN. Signifi-
cant differences in growth (p > 0.05) were not found between 22 and 44 yg/1,
33 and 55 yg/1, and 55 and 66 yg/1. At 77 pg/1 growth was less than that in
the controls and all other treatments. At 60 days the differences in linear
growth among treatments were greatest (F = 53.42). No significant difference
was found at this age between the controls, 11 yg/1, and 22 yg/1 (p > 0.05).
44
-------
TABLE 24. GROWTH OF EMBRYO AND JUVENILE BROOK TROUT EXPOSED
TO VARIOUS LEVELS OF CYANIDE
HCN
yg/l
Control
Control
Control
Control
Control
5.6
5.6
5.5
5.7
11.1
11.5
11.4
21.6
22.1
33.2
33.4
32.5
Stand.
dev.
yg/l
-
-
-
-
0.34
0.62
0.50
0.50
0.86
2.60
0.72
1.69
1.52
2.60
2.13
2.15
x Wt. at
hatch
(mg)
40
42
41
39
38
43
41
41
38
39
42
40
40
40
38
36
Hatch
13.6
13.7
13.8
11.7
13.7
14.4
12.9
12.7
12.6
13.1
13.6
12.7
13.0
12.0
12.3
12.8
Mean length (mm)
30 d 60 d
19.6 29.9
20.3 29.3
20.1 29.7
20.6 29.8
* *
20.4 31.8
19.3 29.3
20.6 29.2
19.7 29.5
*
19.3 29.5
18.7 26.0
* *
18.6 25.7
x Wt.
90 d
90 d (mg)
41.5 1031
39.8 910
41.9 1150
40.1 937
43.6 1238
42.4 1051
* *
44.1 1219
41.5 972
42.5 1067
37.1 681
* *
37.4 762
- Continued —
45
-------
TABLE 24. CONTINUED
HCN
Vg/1
43.6
43.5
45.0
55.9
54.7
55.9
66.8
67.5
77.0
77.4
77.5
Stand .
dev.
Pg/1
6.08
6.10
8.42
4.98
7.43
7.68
3.37
4.40
4.23
4.88
3.41
x Wt. at
hatch
(mg)
40
40
39
41
40
42
42
40
40
37
42
Hatch
12.9
13.3
13.0
12.1
12.5
12.2
12.6
11.0
12.2
11.7
12.5
Mean
30
19.
18.
18.
19.
18.
17.
16.
16.
length (ram)
d 60 d
7 26.4
* *
8 25.4
0 25.0
ft *
0 24.8
8 23.4
* *
8 23.2
1 21.8
* *
4 20.0
x Wt,
90 d
90" d (ing)
36.4 688
ft *
35.5 651
32.8 478
* ft
32.3 460
31.5 387
* *
28.8 284
25.7 196
* *
22.9 127
*Mean significantly different from the control (p - 0.05).
46
-------
Length of fish at 5.6 yg/1 was significantly greater than that in the controls.
Growth was legs than the controls (p < 0.05) at 33 pg/1 HCN and greater concen-
trations. No significant difference (p > 0.05) in length was found between 33
and 44 yg/1 and 55 and 66 yg/1, but at 77 yg/1 HCN it was less than all other
treatments. At 90 days the differences among treatments in length of fish were
marked. The lengths of fish in the controls, 5.6 Pg/1, and 22 yg/1 were not
significantly different. Growth at all other treatments was significantly
different from one another (p - 0.05, F = 41.38). Growth at 11 yg/1 was
greater than that in the controls, and growth at 33 yg/1 HCN and higher levels
was progressively less with increasing concentration. At 90 days, differences
in weight among treatments were less extreme, but showed the same pattern of
differences as lengths (F = 37.01). No difference (p > 0.05) was found among
the controls, 5.6 and 22 yg/1, between 5.6 and 11 yg/1, and between 33 and 44
yg/1. All treatments of 33 yg/1 and greater showed a weight reduction
(p - 0.05) when compared to the controls.
Survival—Survival of embryos and juveniles in each treatment is given in
Table 25. Eggs from controls were used in treatments in which no spawning
occurred (22, 66, and 77 yg/1 HCN). No significant differences (p > 0.05) in
hatchability of eggs exposed to cyanide was noted because of the high vari-
ability between replicates. Survival of 90-day-old juveniles decreased with
increasing cyanide concentrations. No significant differences were noted
between the control and 5.5, 11, 22, 33, and 44 yg/1 (p > 0.05). Survival at
55, 66, and 77 yg/1 was different from the survival in the control (p < 0.05).
No difference occurred between 55 and 66 yg/1, but survival at 77 yg/1 was
different from survival at all other levels (p - 0.05).
Summary—
Growth was significantly reduced by cyanide at the early life stages, but
no differences were noted in adults. During the early life stages from embryo
to juvenile, growth was progressively more influenced by the cyanide concentra-
tion as age increased. This increase in sensitivity was also shown in acute
toxic'ity tests (see Section IV). The level of cyanide having no adverse
effect on the growth of brook trout after hatching is between 22.1 and 33.2 yg/
1 HCN.
Based on a comparison of size of fish at each time period, growth of juve-
niles at 5.5, 11 and 22 yg/1 HCN was equal to or slightly better than that of
the control fish and less at higher HCN concentrations. Comparison of growth
increments during three time periods (0 to 30, 30 to 60 and 60 to 90 days)
showed that the percentage gain in length at concentrations of 33 to 66 yg/1
approached that of the control with an increase in time. This growth pattern
was similar to what Leduc (1966) found with coho salmon juveniles. The salmon
exposed to cyanide grew somewhat faster than the control fish during the
second half of the experiment. This increase in growth after initial impair-
ment was attributed by Leduc to an increase in food conversion efficiency.
Adult mortality occurred in the two highest treatments (64.9 and 75.3 yg/1)
when the temperature during the spawning tests was reduced from 12° to 9°C.
It was shown in Section IV that a decrease in temperature increases sensitivity
47
-------
TABLE 25. EFFECT OF HCN ON THE SURVIVAL OF BROOK TROUT
EGGS TO HATCHING AND FROM HATCHING TO 90-DAY-OLD
JUVENILES; RANGES ARE IN PARENTHESES
Nominal
HCN
(yg/D
Control
5.5
11
22"
33
44
55
66*
77*
Mean hatch
(percent)
72.5
(6.3-100.0)
46.5
(22.4-98.0)
70.8
(16.5-100.0)
72.0
(68.0-76.0)
70.0
(52.6-96.0)
72.9
(48.7-93.5)
81.3
(64.9-89.6)
86.9
(81.3-92.4)
76.7
(51.1-99.1)
Mean survival
from hatch
to juvenile
(percent)
98.6
(96.0-100.0)
100.0
100.0
94.0
(92-96)
100.0
100.0
84.0
73.5
(64.0-83.0)
30.0
(12.0-48.0)
Data for treatments at 22, 66 and 77 pg/1 were from
control eggs incubated and hatched at the indicated
test treatment concentrations.
48
-------
of fish to HCN, Mortality of juveniles occurred after the swim-up stage.
level of no adverse effect on survival occurred between 44 and 55 pg/1.
The
The number of eggs spawned per female was considerably lower at HCN con-
centrations of 11.2 yg/1 and higher. A study reported by Jensen (1971) showed
that a 5 percent increase in mortality of zero-age-group brook trout results
in a substantial decrease in yield. Jensen's theoretical analysis of the
effect of increased mortality rates on yield showed that with 50 percent addi-
tion mortality at the zero-age group the brook trout population would become
extinct. It is reasonable to assume that a decrease in spawning will also
bring about a decrease in the zero-age group unless the spawning population
deposits eggs in significantly greater numbers than are needed for maintenance.
Therefore, on the basis of the demonstrated effect of cyanide on spawning, a
decrease in the abundance of the zero-age class trout may be anticipated
(Table 26) in HCN concentrations of 11.2 ug/1 and higher.
TABLE 26. PERCENTAGE REDUCTION IN MEAN NUMBER OF EGGS
SPAWNED BY FEMALE BROOK TROUT EXPOSED TO HCN
HCN
(yg/D
Control A*
5.7
11.2
32.3
43.6
53.9
64.9
Mean eggs
spawned
per female
Percentage reduction in
spawned eggs per female
From lowest
From HCN cone.
control A* (5.7 ug/1)
502
513
291
246
442
262
124
42.0
5110
12.0
47.8
75.3
43.3
52.0
13.8
48.9
75.8
* Control A, used for comparison, was the least
productive control.
BLUEGILL
Experimental Design
Chronic toxicity tests were conducted with bluegill adults, juveniles,
and newly fertilized eggs. Two of the chronic tests (1 and 3) were started
from newly fertilized eggs and lasted for 57 days. Test 2 was conducted for
49
-------
29 days with young-of-the-year bluegills. Test 4 was conducted with first-
and second-year spawners. This test ran for 289 days and was carried through
90 days of the succeeding generation.
Adult bluegills were obtained from Marion Lake, Dakota County, Minnesota,
and young-of-the-year juveniles for Test 2 from Round Lake, Ramsey County,
Minnesota. Adult fish for Test 4 were given prophylactic treatment with 20 mg/
1 neomycin sulfate at 12°C before placement into test tanks. Two treatments
were given 1 week apart for 3 hr per day on each of three consecutive days.
One hundred fifty adults were equally and randomly distributed to 10 test
tanks. Water temperatures were raised approximately 2°C per day from 12° to
25°C and maintained at 25°C for 1 week prior to the beginning of the test.
Length and weight measurements were taken from fish anesthetized in MS-222 at
0, 70, 140, and 289 days. The number of fish in each tank was reduced to 10
at 140 days, with each tank having an approximately equal sex ratio. Two
controls and eight cyanide treatments were used in the adult test. The fiber-
glass test tanks were 210 x 54 x 54 cm, and a ,30—cm standpipe maintained a
volume of 340 1. Four liters were dispensed every 3 min to each tank. Ninety-
five percent of the test water was replaced every 12 hr. The chambers for
tests with eggs spawned in Test 4 were 20 1 glass aquaria into which oscil-
lating incubation baskets were placed for hatching tests. Every 3 min 1.35 1
was dispensed to each tank. The 95 percent water replacement time was 2.4 hr.
The test chambers and replacement times for chronic Test 2, using juveniles, and
Tests 1 and 3, using fertilized eggs, were the same as those used for the
second-generation experiment of Test 4. In Tests 1, 2, and 3, four controls
and eight replicated cyanide treatments were used. The photoperiod for Tests 1
and 2 was 12 hr of light and 12 hr of darkness. Tests 3 and 4 used the photo-
period of Evansville, Indiana. The temperature in all tests was held at 25°C,
and water analyses were made three times weekly from each test chamber for HCN,
pH, and dissolved oxygen. The water delivery system was similar to that
described by Brungs and Mount (1970), and the toxicant dispensing system was
similar to that of Mount and Warner (1965).
Fry from the swim-up stage through the first week of feeding were fre-
quently fed boiled, ground egg yolk suspension and brine shrimp nauplii
(ArtenrCa') . For the next 2 weeks they were fed brine shrimp nauplii only three
to five times daily, and after 4 weeks their diet was gradually changed to
frozen adult brine shrimp. The juveniles in Test 2 were also fed frozen brine
shrimp. Adults were initially fed minnows and Glencoe pellets, but later in
the test were fed cut beef liver and Purina floating pellets.
Spawning in the adult Test 4 took place on square concrete blocks (45 x
45 x 8.3 cm), each with a saucer-shaped depression 38 cm in diameter and 5 cm
deep. Spawning block areas were isolated by black plastic curtains to minimize
aggressive behavior. After each spawning the block was removed, keeping water
in the depression. The eggs were loosened from the edges with a paint brush
and placed in the block depression. Samples of eggs were removed for hatching
tests, and the remainder were preserved for counting. Eggs to be hatched were
placed into 6.5-cm OD acrylic cylinders with Nitex screen bottoms and oscil-
lated in the test tanks as described by Mount (1968) . As many as three
spawnings from each test tank were hatched, and fry were held for 90 days to
50
-------
determine survival and growth rates.
Tests 1 and 3 were each started with eggs from five artificially spawned
pairs. The eggs of a single female were dry fertilized by the sperm of one
male, and the fertilized eggs from the five pairs were combined. Two hundred
eggs were placed in each incubation chamber. The outlet of each chamber was
attached to the outlet on the test tanks. The number of holes in the outlet
stem of the chamber was adjusted so that the rate of water flow over the eggs
could be regulated as described in Section IV. Swim-up fry were released into
the 20-1 aquaria which held the incubation chambers. Subsamples of 20 sac fry
and 20 swim-up fry were taken from each tank for length measurements in Test 1.
The number of individuals in each chamber of Test 1 was reduced to 30 and 20
at 29 and 43 days, respectively. In Test 3 the number was reduced to 20 at
30 days. Survival and growth determinations were made at 29, 43, and 57 days
in Test 1 and at 30 days (survival only) and 57 days in Test 3. Growth and
survival of juveniles were determined at the 29-day termination of Test 2.
Results
Survival—
Survival of adult fish (Test 4) was not markedly affected until the start
of the spawning period. Deaths started to occur after a series of behavioral
changes which were first noted at the highest HCN concentrations. The first
symptoms were a lack of feeding activity and aggressiveness followed by an
impairment of ability to judge the distance to food and to capture it. These
symptoms were followed by gradual changes in swimming behavior starting with a
list to one side, proceeding to spiraling and occasional darting motions, and
terminating in a quiescent state, often upside down, on the tank bottom until
death. The time from the beginning of this sequence until death was about two
weeks. No deaths were observed at 9.8 yg/1 HCN and below. Survival of adults
when the test was terminated was greater than 84 percent up to 50.2 yg/1 HCN
and then decreased to 60 and 40 percent at 65.6 and 80.0 ug/1 HCN, respectively
(Table 27). Early stages of the previously described symptoms were noticable
in fish at 50.2 ug/1 HCN by the time the test was terminated.
In the tests which were started with fertilized eggs (1 and 3) , there was
considerable variability in survival (Tables 28 and 29). The mean survival in
the controls at termination of Test 1 was 18.7 percent (range 9.6 to 29.4).
Survival was significantly less than that in controls at 38.6 yg/1 HCN, using
one-way analysis of variance (p > 0.05) with arcsin transformation and Dunnett's
procedure (Steel and Torrie I960). The mean survival for the controls at the
termination of Test 3 was 23.3 percent (range 10.0 to 37.5), and survival at
19.4 pg/1 HCN and higher concentrations was significantly lower than that in
controls. The discrepancy in the no-effect concentration between the two tests
(1 and 3) appears to result from the high survival (39.8 percent) at 26.6 yg/1
HCN in Test 1, where one replicate had been lost. When the results of this
test concentration are omitted the survival trend of the combined tests (1 and
3) indicates that the level of significance determined in Test 3 (19.4 ug/1
HCN) fits this trend better than the result from Test 1 (Tables 28 and 29).
Most deaths of bluegill fry exposed to HCN occurred within the first 30 days
after hatching. No change in mortality rate was observed after 6 weeks. This
51
-------
TABLE 27. SURVIVAL AND FECUNDITY OF ADULT BLUEGILLS AFTER 289 DAYS, TEST 4*
Ln
NJ
HCN
cone.
yg/it
Con-
trol
Con-
trol
5.2
(1.1)
9.8
(1.2)
20.5
(1.8)
30.0
(2.4)
39.7
(4.5)
50.2
(4.5)
65.6
(4.0)
80.0
(4.7)
Survival
percent
100
100
100
100
90
90
84
90
60
40
No. females
after
thinning £
5
5
6
3
6
6
3
5
8
6
No. of
mature
females §
5
5
5
3
4
5
3
5
7
6
No. of
surviving
females
5
5
6
3
6
6
3
5
6
2
Number of
spawnings #
8
5
0
0
0
0
0
0
0
1
Total
eggs
spawned
59,394
26,791
0
0
0
0
0
0
0
6,896
Mean
eggs/
spawning
7,424
5,358
0
0
0
0
0
0
0
6,896
Number
eggs/g
female
110
54
0
0
0
0
0
0
0
12
*Mean of all treatments: temperature 24.7°-25.1°C; pH 8.07-8.11; dissolved 02 6.07-6.26 mg/1; total
alkalinity 236.0-236.3 mg/1.
tMean with standard deviation in parentheses.
tAfter thinning at 140 d from 15 to 10 fish per tank.
§ Declared mature if any sign of development.
i'Spawning defined as the eggs deposited on one spawning site within a 24-hr period.
-------
TABLE 28. PERCENTAGE SURVIVAL OF BLUEGILLS FROM FERTILIZED EGG TO THE 57-DAY
JUVENILE STAGE IN VARIOUS HCN CONCENTRATIONS FOR TEST 1*
Day 29 Days 43-57
HCN _ Percent _ Percent
yg/1 (SD) x survival x survival x
Control 30.4 29.4
Control 17.1 19.4 16.5 18.7
Control 19.8 19.1
Control 10.2 9.6
5.2(0.7) 5.3 40.0 23.3 37.3 21.9
5.3(1.5) 6.5 6.5
14.7(2.0) 15.6 lost 10.3 lost 9.0
15.6(2.2) 10.3 9.0
26.7(3.7) 26.6 22.7 33.5 lost 39.8
26.6(2.5) 44.2 39.8
38.5(4.1) 38.6 1.3 1.9t 0 0.7f
38.7(4.9) 2.5 1.3
49.6(5.3) 49.7 6.9 5.7 3.1 1.6
49.7(6.3) 4.4 0
60.3(7.2) 61.8 0 0.3t 0 Ot
63.2(7.0) 0.6 0
68.2(9.1) 70.3 1.3 2.3t 0.7 1.3t
72.3(7.8) 3.2 1.9
80.7(9.7) 81.2 0 Ot 0 Qt
81.7(8.4) 0 0
-Means of all treatments: temperature 24.8° to 25.0°C; pH 7.87 to 8.02;
dissolved 02 5.77 to 6.40 mg/1; total alkalinity 237.6 to 238.3 mg/1.
tMean is significantly different from control (p - 0.05).
53
-------
TABLE 29. PERCENTAGE SURVIVAL OF BLUEGILLS FROM FERTILIZED EGG TO THE 57-MY
JUVENILE STAGE IN VARIOUS HCN CONCENTRATIONS FOR TEST 3*
Day 30 Day 57
HCN _ Percent _ Percent
yg/1 (SD) x survival x survival x
Control 41.0 37.5
Control 20.0 25.0 20.0 23.3
Control 13.0 10.0
Control 26.5 25.5
4.8(0.9) 4.8 18.5 18.5 18.5 18.5
5.2(0.6) lost lost
8.9(2.1) 9.1 26.5 17.5 25.0 16.3
9.2(1.1) 8.5 7.5
19.2(1.7) 19.4 3.0 5.5t 3.0 2.8J
19.6(1.4) 8.0 2.5
28.5(4.5) 29.1 9.5 7.8 2.5 2.5^
29.7(4.7) 6.0 2.5
38.7(2.3) 39.5 8.5 15.3 3.0 3.8t
40.2(2.4) 22.0 4.5
49.3(3.3) 49.3 17.5 17.5 13.5 13.5
51.9(2.4) lost lost
61.8(4.1) 62.9 2.0 3.3t 0 0$
64.0(3.9) 4.5 0
80.4(2.3) 82.1 0 Ot 0 0$
83.8(6.2) 0 0
*Mean of all treatments: temperature 24.7° to 25.4°C; pH 8.02 to 8.07;
dissolved O2 5.40 to 6.22 mg/1; total alkalinity 236.6 to 237.4 mg/1.
is significantly different from control (p ^ 0.05).
fMean is significantly different from control (p - 0.01).
54
-------
observation is supported by results in Test 2, in which no deaths occurred to
wild young-of-the-year bluegills up to 53.1 yg/1 HCN (Table 29).
Growth—
Analyses of growth data are complicated by effects of low levels of HCN
on survival. If HCN does affect growth, the compensation created by the lower
densities due to greater mortality in the higher levels seemed to nullify this
effect. One-way analysis of variance on lengths and weights (logarithmic
transformation, p > 0.05) demonstrated no significant differences for any of
the chronic tests. The mean initial weight of all adult bluegills in Test 4
was 41.3 g, and mean terminal weight was 117.9 g (Table 30). The mean terminal
weight was 0.285 g in Test 1, and 0.127 g in Test 3 (Table 31). In Test 2
the mean initial weight was 0.66 g, and terminal weight was 1.11 g (Table 32).
Rep ro due t ion—
Spawning of adult bluegills in Test 4 occurred only in the controls except
for an early spawning which took place in the highest HCN level (80.0 yg/1).
After spawning started in the controls, the test was continued for 68 days. A
total of 13 spawnings occurred in the controls (Table 27); one control had 8
spawnings and the other had 5. The control with 8 spawnings produced 110 eggs/
g of female, and the control with 5 spawnings had 54 eggs/g of female. All
females in the controls were mature. The single spawning at 80.0 yg/1 HCN
yielded 12 eggs/g of female. No spawning occurred at 5 yg/1 HCN, and no lower
levels were tested. These data suggest that levels of HCN below 5 yg/1 would
probably inhibit the spawning of bluegills as well. Nearly 60 percent of the
males and 90 percent of the females in the test had partly or fully developed
gonads during the spawning period. Autopsies on the surviving adult bluegills
in Test 4 were done when the test was terminated. Weights and condition of
gonads and liver were noted. Organ weights and organ-to-body weight ratios
did not appear to be related to HCN concentration.
Second Generation Hatching and Survival—
The mean percentage hatch for controls in the second generation of Test 4
was 89 percent. Eggs from the single spawning at 80.0 yg/1 were incubated at
75.1 yg/1. Forty-three percent of the eggs hatched, but all of the fry were
dead within 2 weeks. The mean survival rates for second generation control
fish at 30, 60, and 90 days were 41, 36, and 34 percent, respectively. Means
of the total lengths of fish in controls at 30, 60, and 90 days were 14.9, 31.4,
and 46.9 mm, respectively.
Summary—
The long-term exposure of bluegills to HCN has demonstrated that, as is
the case with fathead minnows and brook trout, no-effect levels are much lower
than acutely lethal concentrations. Spawning is completely inhibited at
5.2 yg/1 HCN and, presumably, is inhibited to some extent at lower levels.
The effect on spawning appears to be both physiological and behavioral, since
fish exposed to low levels of cyanide exhibited gonadal development but did not
spawn when held in clean water for 2 weeks before final termination of the
55
-------
TABLE 30. MEAN LENGTHS (CM) AND MEAN WEIGHTS (G) OF ADULT BLUEGILLS
EXPOSED TO HCN FOR 289 DAYS, TEST 4
Length (cm)
Treat-
ment
1
2
3
4
5
6
7
8
9
10
yg/
HCN
1 (SD)
Control
Control
5.
9.
20.
30.
39.
50.
65.
80.
2(1.1)
8(1.2)
5(1.8)
0(2.4)
7(4.5)
2(4.5)
6(4.0)
0(4.7)
Days
0
13.6
13.5
13.8
13.3
13.0
13.1
13.6
13.3
13.2
13.6
70
14.4
14.3
14.5
14.2
13.8
13.9
14.2
14.0
13.9
14.2
140
15.1
15.0
15.1
15.0
14.7
14.6
14.9
14.7
14.6
14.9
289
17.7
17.6
17.5
18.0
17.3
17.0
18.0
17.5
16.2
17.6
0
43
42
44
42
35
40
41
40
38
44
.4
.8
.3
.5
.2
.3
.9
.1
.3
.3
Weight (g)
Days
70
54.6
54.6
57.1
53.6
48.4
50.0
50.1
49.8
51.6
53.7
140
71.4
70.4
72.4
70.8
65.1
66.1
67.2
66.8
69.3
67.7
289
118.5
117.3
117.6
130.8
118.7
122.0
122.5
119.7
85.0
110.5
*Mean temperature all treatments 24.7°-25.1°C (SD 0.3-0.4); pH 8.07-8.11
(SD 0.05); Dissolved 02 (rag/1) 6.07-6.26 (SD 0.19-0.3); total alkalinity
236 tng/1 (SD 16.1).
56
-------
TABLE 31. TERMINAL MEAN LENGTHS AND WEIGHTS OF BLTJEGILL JUVENILES AT VARIOUS
HCN CONCENTRATIONS IN TESTS 1 AND 3
Mean
HCN
cone.
(yg/D
Control
Control
Control
Control
5.2
5.3
15.6
26.6
38.7
49.6
Test 1
Mean
length
(mm)
28.8
29.3
29.2
31.3
28.4
28.8
30.9
27.7
30.5
25.6
Mean
weight
(g)
0.278
0.296
0.285
0.381
0.275
0.346
0.362
0.251
0.337
0.201
Mean.
HCN
cone.
(H8/1)
Control
Control
Control
Control
4.8
8.9
9.2
19.2
19.6
28.5
29.7
38.7
40.2
Test 3
Mean
length
(mm)
19.9
21.8
24.8
20.4
21.4
19.3
25.9
26.5
25.2
29.6
27.8
25.2
25.3
Mean
weight
(g)
0.094
0.115
0.184
0.099
0.102
0.077
0.210
0.302
0.259
0.388
0.326
0.213
0.223
49.3
21.2
0.124
57
-------
TABLE 32. PERCENTAGE SURVIVAL, LENGTH, AND WEIGHT
OF BLUEGILL JUVENILES IN TEST 2
(MEAN INITIAL WEIGHT = 0.66 G)
HCN
cone.
(Mg/1)
Control
Control
Control
Control
5.3
5.3
11.4
12.2
18.5
20.2
32.2
32.4
36.9
38.5
44.4
45.5
52.2
54.0
Terminal
percent
survival
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Mean
length
(mm)
42.5
42.4
41.1
41.3
41.3
41.8
40.9
41.4
42.5
41.6
41.6
41.5
41.0
41.2
41.6
42.1
42.3
41.7
Mean
weight
(g)
1.166
1.200
1.094
1.099
1.114
1.084
1.108
1.091
1.124
1.091
1.084
1.105
1.083
1.084
1.095
1.144
1.163
1.096
58
-------
experiment. During this period there were two spawnings in the controls.
Possibly the long delay past the regular spawning period, as defined by the
fish in controls, precluded any spawning. The survival of fry was adversely
affected at 15.6-19.4 pg/1 HCN, but juveniles were not affected below 53.1 ug/1.
The effects of low concentrations of HCN on the growth rates of juveniles could
not be measured, because of the overriding influence of crowding on growth
rates. Among adults, no significant differences in growth were observed.
In assigning a safe or no-adverse-effect concentration of HCN for blue-
gills, it is evident that this concentration lies below 5.2 yg/1 if a complete
life cycle is to occur in cyanide polluted waters. If reproduction and early
fry stages are not considered, concentrations of about 50 yg/1 may be tolerated
at higher temperatures. The seasonal implications of these differences must be
considered in pollution abatement. No long-term tests on young-of-the-year or
adults were run at low temperatures where acute toxicity tests (Section IV)
showed fish to be significantly more sensitive to HCN. The apparently safe
levels for juveniles (53.1 yg/1) at 25°C are lethal at lower temperatures.
Until further testing determines the long-term sensitivity of fish at low tem-
peratures, the apparently safe levels of cyanide for adults and young-of-the-
year should not be used as a basis for protecting these life stages at all
times of the year.
INVERTEBRATE POPULATIONS
Variability Between Two Species
An important consideration in selecting a test organism for bioassays is
its variability in response. An organism with low variability will give
greater precision and make comparisons of test results more reliable. Also,
researchers interested in doing tests on two species for studies of competition
and predator-prey interactions require test organisms which can both do well
under the same test conditions.
The experiments reported here compared the variability of Asetlus oormmnis
Say and Gcarmavus pseudol-imnaeus Bousfield and determined if both species pro-
duced viable populations under the same conditions when tested in separate
tanks.
Experimental Design—
The AselT-us used in the test were the second generation from adults hand
picked from bottom material scooped from Rainy Lake near Ranier, Koochiching
County, Minnesota. They were abundant along the shoreline in sheltered coves
where decomposing organic material had accumulated. The GcarmdTUS were col-
lected with a drift net from a small, spring—fed stream entering the Saint
Croix River at Marine on St. Croix, Washington County, Minnesota.
Before being used in experiments, Asellus and Gammarus were held sepa-
rately in 25-1 (50 x 24 x 25 cm high; water depth, 20 cm) tanks constructed
with double strength glass and silicone adhesive. Well water flowed through
the aerated tank at 700 ml/min. The dissolved oxygen concentration was main-
tained at saturation and the temperature at 18°C. A constant supply of fall-
59
-------
dropped leaves and dead minnows was provided as food. The leaves also provided
cover for the test animals. The leaves were soaked in flowing, aerated water
for about 30 days prior to use and were a mixture of white poplar, hackberry,
and cottonwood. The dead fish were freshly killed or frozen, and it appeared
that little feeding was done on the fish until some decomposition and fungus
growth had developed.
Water for the tests was adjusted to test temperature and saturated with
oxygen in a supply tank and then flowed by gravity to the five individual test
chambers. The 20-1 glass-silicone chambers were 50 x 25 x 21 cm, and a water
depth of 16 cm was maintained. The flow-through system provided 95 percent
water replacement in about 3 hr.
To begin the Asettus test, 40 individuals were randomly placed in each of
5 test chambers, and 40 were preserved to determine mean and range of total
length and the mean weight of the individuals. The same procedure was followed
for Gammarus in five separate tanks (Table 33). During the test, temperature
and pH were determined three times weekly, and total alkalinity and dissolved
TABLE 33. TEST CONDITIONS IN GAMMARUS AND ASELLUS EXPERIMENTS
Item
Date collected
Laboratory holding (days)
Test duration (days)
Size at start
Mean total length (ram)
Range
Mean weight (mg)
Temperature (°C)
pH
Dissolved oxygen (mg/1)
Total alkalinity (mg/1)
Light Intensity (lux)
Assitus
November 19, 1974
114
109
4.2
2.0-6.0
3.1
Gammavus
March 3, 1975
10
71
13.3
6.0-18.0
23.4
Means*
S.D.*
Means*
S.D.*
17.96-18.02 0.13-0.15
8.01-8.03 0.03-0.04
7.06-7.33
236
0.34-0.41
2
420-506
18.15-18.23 0.70-0.72
8.03-8.04 0.03
7.53-7.69 0.24-0.30
235 4
538-646
* Range of means and standard deviations are for all tanks in each
experiment.
60
-------
oxygen once per week. Each week all tanks were checked for the appearance of
young produced by the original stock. The test was terminated 45 days after
the first young were observed, and the contents of each tank were passed
successively through three screens (square mesh openings of 6.0, 2.0, and
0.4 mm) to separate debris and organisms into four size groups. The material
that passed through the smallest screen contained no organisms and was dis-
carded. The three larger groups were preserved and stained with 10 percent
formalin and rose bengal (0.6 g/1) . After a few weeks in the stain the organ-
isms were separated from debris and measured. The eggs and young were removed
from gravid females and counted. The entire population of adults, eggs, and
young was centrifuged at a low speed to remove excess water and weighed.
Results—
Length-frequency diagrams of all organisms in each test chamber were made
(Figures 2 and 3). Table 34 lists the population parameters for each test
chamber. The term "free individuals" refers to organisms not contained in the
brood pouch. Table 35 compares Asellus and Gcanmarus on the basis of the mean,
standard deviation and coefficient of variability for each of the population
indices for the five tanks combined.
The similarity of the general shape of the length-frequency diagrams
within each species indicates that the five tanks were fairly well synchro-
nized in terms of reproduction time and growth rates (Figures 2 and 3) . Com-
parison of the mean values for the five tanks of each species (Table 35) shows
that the number of free individuals, number of eggs and young in the brood
pouch, total number, and total weight were similar for the two species. The
values of Asetlus are all about 25 percent higher than for GammaPus, but the
coefficients of variability are about half as high as those for Gammarus. One
exception was the number of eggs and young in the brood pouch. In tank 3 of
the Asellus test reproduction began slightly earlier, there were many females
with empty brood pouches, and the number of free individuals in the 1.5 to
3.0 mm length group was high. In the Asellus culture tanks, where stocks have
been maintained since November 19, 1974, densities are much greater than in
the test tanks. Attempts to culture Ganvnapus at similar densities have failed.
Allee (1929) found that the Asellus can occur at high densities and remarks
that in general they are tolerant of many individuals in a limited space. By
contrast, papers by Anderson and Raasveld (1974) and Hynes (1954) report pre-
dation and cannibalism for species of Gammarus. Oseid (1977) studying the
effect of Asellus mi-litar-is* and Gatmtarus pseudolimnaeus on fish eggs found
that GammaPus consumed live fry and eggs of fish whereas Asellus did not. The
control tanks of toxicity bioassays of Nebeker and Puglisi (1974) and Oseid and
Smith (1974) contained means of 254 and 433 Gammarus, respectively, where test
conditions other than water supply and tank size were similar. The latter
study used water with a higher mineral content and larger tanks. In the
present study, the mean weight and mean number of eggs or young per gravid
female for Asellus and Gammarus were very similar, but again the coefficient
of variability was greater for Gammarus. The mean length of gravid females
* Keyed out to Asellus militafis Hay in Pennak (1953) but to Asellus
aommunis Say in Williams (1970).
61
-------
100
90
60
5O-
40
30-
20
10-
70
60
50
^^H> TANK I
JL
TANK Z
110
90
li_ 80
O
70
60
LJ
m
so
40
30
20
10
TANK 4
TANK 5
I.O 2.0 3.0 4.0 5X> 6O 7O SO 9.0 K>£>
LENGTH (MM)
Figure 2. Length-frequency diagrams of all Aseflus in each test chamber
62
-------
180
[70
ISO
[SO
(40
BO
120-
HO
too
90-
80
70
60
50
40
30
20
W
0
110
100
2 so
I 80
s£ 70
I 60
50
40
$ 30
20
IO
CC O
z>
Z 60
50
40
30
20
10
0
60
5O
4O
3O
2O
10
0
6O
50
4O
3O
20
IO
O
K5 2.03O4O5O6070BJ09.O [1.0 BO SO 17.0 BO
LENGTH (MMj
Figure 3. Length-frequency diagrams of all GammaPUS in each test chamber
63
-------
TABLE 34. POPULATION PARAMETERS IN ASELLUS AND GAMMARUS TESTS
Asellus Gconmarus
Tank Number Tank Number
Free individuals 510 526 702 496 334 612 513 289 340 237
Eggs or young in
brood pouch 162 120 0 163 103 65 60 117 77 31
Total free individ-
uals plus eggs and
young in brood pouch 672 646 702 659 437 677 573 406 417 268
Total weight of free
individuals plus
eggs and young in
brood pouch (g) 1.766 1.538 1.387 1.539 1.026 1.246 1.352 0.617 1.225 0.602
Mean weight of free
individuals (mg) 3.46 2.92 1.98 3.10 3.07 2.04 2.64 2.13 3.60 2.54
Mean number of
eggs or young per
gravid female 23.1 20.0 - 20.4 20.6 21.7 30.0 16.7 19.2 15.5
Mean length of
gravid females (mm) 5.50 5.58 - 5.50 5.50 11.8 13.0 11.6 11.8 12.5
-------
TABLE 35. MEANS FOR THE POPULATION INDICES OF FIVE REPLICATIONS
Asellus
x S.D. C.V.
Gammavus
x S.D. C.V.
Number of free individuals
Number of eggs and young
in brood pouch
Total number of free
individuals plus eggs
and young in pouch
Total weight of free
individuals plus eggs
and young in pouch (g)
Mean weight of free
individuals (mg)
Mean number of eggs or
young per gravid female
Mean length of gravid
females (mm)
514
110
624
1.451
2.91
21.0
5.52
131
67
106
25
61
17
298
70
468
158
31
159
40
44
34
0.274
0.55
1.4
0.04
19
19
7
1
1.008
2.59
20.6
12.14
0.367
0.62
5.8
0.59
36
24
28
5
-------
was much, smaller for Asellus. The growth rates for both species appeared to
be similar so that maturity at a small size allowed shorter generation time.
After 45 days from the first observation of young, the F^ generation of Asellus
had many gravid females, and tanks 2 and 3 had large numbers of ¥2 individuals.
After the same time Garnnarus Fj had not grown to the size of maturity, and only
the original stock of individuals, all 10.0 mm or larger, were gravid.
Summary—
These tests indicate that, as an individual test organism, Asellus has
several advantages. Asellus exhibited lower variability than GammaTUS in six
of seven population parameters; therefore fewer replications would be required
in experiments with Asellus to get the same precision. It can be easily
cultured in high numbers in the laboratory and possibly the time to complete a
generation is shorter, thereby reducing the time required to do full generation
tests. However, apart from variability, the tests also showed that in terms
of the population parameters (Tables 34 and 35) the two species produced
fairly similar populations. Therefore, when tested in the same test chamber
where the two species can interact, both species should have an equal chance
to develop a population apart from factors such as competition or predator-
prey relations.
Response to Cyanide
The tests described below were designed to determine the effect of cyanide
on two invertebrates when exposed separately or simultaneously to the toxicant.
GanmaTUS pseudol-inmaeus Bousfield was selected because it is aggressive and
competitive (Hynes 1954; Anderson and Raasveld 1974; Oseid 1977) and Asellus
eormunis Say because it is passive (Allee 1929; Oseid 1977). Short-term tests
were done with each species alone, and full-life-cycle tests were done with
each species alone and then with the two together.
Experimental Design—
Garmarus were collected 10 days before the beginning of each test from a
small, spring-fed stream tributary to the Saint Croix River at Marine on St.
Croix, Washington County, Minnesota. The Asellus were cultured in the labora-
tory from an initial group which was collected from Rainy Lake near Rainier,
Koochiching County, Minnesota and which provided subsequent generations for
all tests. Asellus and Gammavus were kept in the laboratory at 18°C with the
dissolved oxygen concentration near saturation. They were fed pre-soaked
deciduous tree leaves and dead fish. The days in the laboratory, the length
and weight of organisms at the beginning of each test, and the duration of
each test are noted in Table 36. Exposures were made with the apparatus and
chemical procedures described above in Section IV. Three acute toxicity tests
were performed with Ganrnarus and two with Asellus, and the 96-hr LC50 and
lethal threshold concentrations were calculated. For each series of chronic
exposures (.Asellus alone, No. 1, and No. 2; Gamtnapus alone; and Asellus and
Gammarus together in the same test tank), two simultaneous but separate tests
were conducted.
Nominal test conditions for both acute and chronic exposures were 18°C,
66
-------
TABLE 36. TEST CONDITIONS FOR SIMULTANEOUS GAW4ARUS AND ASELLUS BIOASSAYS IN HCN
Asellus*
(1)
Asellus*
(2)
Asellus and Gcomarus
Gconmarus* together
Asellus Gammavus
Days in laboratory
before testing
Size at start of test
300
429
12
360
17
Mean length
Length range
Mean weight
Duration (d)
(mm)
(mm)
(mg)
6
4
7
.3
.0-8.0
.7
115
Mean
Temperature (°
pH
Dissolved 0?
C) 18
7.
6.
.1-18
98-8.
03-6.
S.D.
.2 0.1-0.2
13 0.03-0.06
52 0.26-0.42
6
5
9
.7
.0-9.5
.2
112
Mean
18.0-18
7.92-7.
5.88-6.
S.D.
.1 0.1-0.5
96 0.05-0.10
14 0.26-0.40
8.7
3.0-14
10.7
83
Mean
18.0-18.1 0
7.99-8.02 0
6.77-7.00 0
.0
S.D.
.4-0.8
.04
.50-0.69
6.
3.
7.
98
1
5-8.0
1
9.
2.
14
98
Mean
18
7.
5.
.0-18.2
93-7.98
70-6.09
0.
0.
0.
7
5-17.0
.1
S.D.
1-0.2
04-0.05
30-0.41
(rng/1)
Total alkalinity 238
(mg/1)
Light intensity 481-660
(lux)
236 2 236
506-657 — 452-667
236
431-603
*Alone
-------
a pH of 8.0, and 6 mg/1 dissolved oxygen. Two 20-W Vita-Lite fluorescent tubes
above each tank provided approximately 500 lux at the water surface during the
16-hr daily photoperiod. Chronic tests were started by placing 40 individuals
at random into each of the test tanks. From the onset of reproduction, the
tests were continued 45 days. At that time all test specimens were preserved,
measured, checked for fecundity, and weighed as a group for each tank.
Results—
The 96-hr LC50 at 18°C for Asellus was 2,328 ug/1 HCN and for Garmarus was
175.6 ug/1. The respective LTC values (10-12 days) were 1,834 and 69.6 ug/1
(Table 19). The length-frequency graphs for each test tank of the three series
are depicted in Figures 4, 5, and 6. Tables 37, 38, 39, and 40 list the mean
HCN concentration and standard deviation, and the biological indices for each
test tank in the four test series.
TABLE 37. BIOLOGICAL INDICES IN THE CHRONIC TEST SERIES (1) ON ASELLUS
(BOTH TESTS INCLUDED*)
__ _ Tank number^ __
6 6 "" "~~3"' 3 5 "4 2
HCN (ug/1) mean 0 0 51 110 220 317 432
HCN (yg/1) S.D. 0 0 9 16 19 36 43
Number of free
individualst 1612 2467 262 109 _85_ __ 56 0
Number of eggs or
young in brood pouch 626 855 Tl 0 60 17 0
Total number of free
individuals plus eggs
and young in brood pouch 2238 3322 334 109 145_ 73 _ 0
Total weight (g) of free
individuals plus eggs
and young in brood pouch 3.703 4.612 0.725 0.292 0.106 0.142
Mean number of eggs or
young per gravid female 35 34 18 — 20 17
* Underlined values are lower than the mean of the controls by more than
two times the standard error of the mean
t Those individuals living separate from the brood pouch
68
-------
82 jJG/L HCN
80
70
60
50
40
30
20-
to-
0
pS/L HCN
CONTROL I
42 JJG/L HCN
zi JJG/L HCN
ii JJG/L HCN
CONTROL 2
1.0 2.0 3.0 4.0 5.O 6.0 7.O 8.0 9.0 10.0
12.0 >i3.5 1.0 20 SO 4j6 5.0 6.6 70 8,0 9.0 JO.O
LENGTH (MM)
>I3.5
Figure 4. Length-frequency graphs of each test concentration for chronic
test—Gaiwnapus alone 69
-------
TO JJ6/L MCN
77 ^jq/L MCN
LENGTH (MM)
JQO yS/L HCN
4T JJS/l. HCN
Figure 5. Length-frequency graphs of each test concentration for chronic
test—Asellus alone (2)
70
-------
ASELLUS
g no
4) /JG/L HCfJ
234567
GAMMARLIS
76 VG/L HCN
5S ffH/l. HCN
9 UI54S 1 2 3 4 5 6 7 9 9 fO 31 12
LENGTH (MM)
Figure 6, Length™frequency graphs of each test concentration for chronic
test—Aseltus and Garrmarus together
71
-------
For the first test series on Asellus there was a large reduction of
numbers and weight at the lowest tested concentration, 51 yg/1 (Table 37).
The highest concentration with survival and reproduction was 317 yg/1; there
was no survival or reproduction in the range 432 to 2,108 yg/1. For the
second series (Table 38) there was survival and reproduction at all concentra-
tions tested (5-100 yg/1). The numbers of free individuals, eggs, and young
were variable, making evaluation on the basis of these indices questionable.
The standard deviation for total weight was relatively low, and at concentra-
tions of 40 yg/1 and higher weights were lower than the mean of the two con-
trols by more than two times the standard error of the control mean. The mean
number of eggs per gravid female was reduced only at 100 yg/1. All other
indices were within two standard errors of the control mean. The highest no-
effect concentration lies between 29 and 40 yg/1, Table 41.
In the series on Gajnma.vus (Table 39) , numbers of free individuals were
variable, but weight was more consistent. At concentrations of 16 yg/1,
32 yg/1 and higher, weights were reduced by more than two standard errors of
the control mean. The low mean weight in the 16 yg/1 treatment was not thought
to be the result of cyanide toxicity. The no-effect/effect concentrations
were 16 and 21 yg/1 on the basis of total number of eggs or young in the brood
pouch and mean number of eggs or young per gravid female (Table 41) .
The effect of exposing Asellus and GammaTus in the same test chamber was
twofold (Table 40). The Garrmavus almost eliminated the Asetlus from the con-
trols and the lowest treatment (4 yg/1 HCN), and there was a shift downward of
the no-effect/effect concentrations for Gammarus. It is felt that predation by
Gammarus is the factor that almost depleted Asellus in the controls and lowest
treatment. On the basis of all the indices for Gamnar>us3 the highest no-effect
concentration lies between 4 and 9 yg/1, values which are lower than when
Gcrnnayus was exposed alone. Predation by large Asellus on small Garrmarus may
account for this reduced no-effect concentration. The highest no-effect con-
centration for Asellus in the same test was between 41 and 55 yg/1, approx-
imately the same as for the two series with Asellus alone.
Summary—
The attributes for survival of Asellus in a mixed community are their much
greater resistance to HCN for short periods (10-12 day LTC value of 1,834 yg/1)
and the ability of some fraction of the population to survive and reproduce at
high concentrations of HCN (up to 317 yg/1) after a continuous exposure of
115 days. By comparison, GarmaPUS have a 25-fold lower 10-day LTC, a 6-fold
lower concentration which permits any survival and a 19-fold lower level
permitting reproduction (Table 38). Therefore, the presence of low concentra-
tions (9 to 30 yg/1 HCN) probably shifts the' competitive advantage from the
aggressive and predaceous GammaPus to the more passive Asellus. Gomnafus are
probably excluded from cyanide polluted areas, which could be satisfactory to
them if they were alone, because they are unable to compete with more resistant
species whether typically predator or prey. Asetlus would be benefitted by
reduced competition and predation.
72
-------
TABLE 38. BIOLOGICAL INDICES IN CHRONIC TEST SERIES (2) ON ASELLUS (BOTH TESTS INCLUDED*)
Diluter number
34434434 343
HCN (yg/1) mean 005
HCN (yg/1) S.D. 001
10 19 29 40 47 58 70 77 100
3 2 5 4 10 8 7 12 12
Number of free
individualst
Eggs or young in
brood pouch
Free individuals
plus eggs and
young
Weight (g) free
individuals plus
eggs and young
758 2551 1622 2023 1091 1158 184 183 1335 376 453 655
154 732 67 578 198 394 0 72 621 124 123 51
912 3282 1689 2601 1289 1552 184 255 1956 500 576 706
2.495 3.221 6.918 5.466 3.646 3.146 0.488 0.520 1.454 1.418 0.867 0.973
Mean eggs or young
per gravid female 19.2 22.2 22.3 22.2 24.8 21.9
18.0 20.7 31.0 20.5 17.0
* Underlined values are lower than the mean of the controls by more than two times the standard
error of the mean
t Those individuals living separate from the brood pouch
-------
TABLE 39. BIOLOGICAL INDICES IN THE CHRONIC TEST SERIES ON GAMMARUS (BOTH TESTS INCLUDED*)
344
HCN (yg/1) mean 005
HCN (yg/1) S.D. 002
Number of free
individualst 137 554 764
Eggs or young in
brood pouch 6 135 113
Free individuals
plus eggs and
young 143 689 877
Weight (g) free
individuals plus
eggs and young 0.795 1.343 1.705
Mean eggs or young
per gravid female 3 27 19
Diluter number
344
11 16 21
236
331 82 335
13 16 0
344 98 335
0.726 0.255 0.629
13 16
3434
32 42 52 64
5 7 14 10
214 20 6 0
0000
214 20 6 0
0.273 0.110 0.033 0
- - - -
* Underlined values are lower than the mean of the controls by more than two times the standard
error of the mean
t Those individuals living separate from the brood pouch
-------
Ui
TABLE 40. BIOLOGICAL INDICES IN THE CHRONIC SERIES WITH ASELLUS (A) AND GAMMARUS (G) TOGETHER
(BOTH TESTS INCLUDED*)
Species
HCN (pg/1) mean
HCN (yg/1) S.D.
Number of free
individualst
Eggs or young in
brood pouch
Free individuals
plus eggs and
young
Weight (g) free
individuals plus
eggs and young
Mean eggs or young
per gravid female
A
G
A
G
A
G
A
G
A
G
1
0
0
27
354
0
334
27
688
0.216
1.965
—
17.6
2
0
0
52
591
0
386
52
977
0.419
3.142
—
16.8
1
4
1
34
745
0
359
34
1104
0.314
3.852
—
17.1
Diluter number
212
9
1
519
212
257
25
776
237
2.227
0.941
19.8
8^1
21
2
625
24
470
0
1095
24
1.734
0.086
23.5
31
5
903
3
363
0
1266
3
1.552
0.017
25.9
1
41
5
644
2
697
0
1341
2
1.522
0.011
25.8
2
55
5
132
0
127
0
259
0
0.568
—
15.9
1
62
11
114
0
34
0
148
0
0.369
—
17.0
2
76
6
437
0
224
0
661
0
1.176
—
18.7
* Underlined values are lower than the mean of the controls by more than two times the standard
error of the mean
t Those individuals living separate from the brood pouch
-------
TABLE 41. COMPARISONS OF THE NO-EFFECT/EFFECT CONCENTRATIONS (ug/1)
OF HYDROGEN CYANIDE FOR ASELLUS AND GAMMARUS
Exposed
alone ,
Series 1
Asellus
Exposed
alone ,
Series 2
Gammarus
Exposed
with
Gammapus
Exposed
alone
Exposed
with
Asellus
Number of free
individualst 0-51
Number of eggs or
young in brood pouch 0-51*
Total number of free
individuals plus eggs
and young 0-51*
Total weight (g) of
free individuals plus
eggs and young 0-51*
Mean number of eggs or
young per gravid female 0-51*
Highest concentration
with any survival 317
Highest concentration
with any reproduction 317
29-40*
77-100
41-55
41-55
41-55
41-55
41-55
34-42
16-21
32-42
21-32*
16-21
52
16
4-9*
4-9*
4-9*
4-9*
4-9*
41
' 9
* Those marked with an asterisk were lower than the mean of the two
controls for the respective test by at least two times the standard
error of the mean
f Those individuals living separate from the brood pouch
76
-------
SECTION VI
MISCELLANEOUS TESTS
PREDATOR-PREY RELATIONS
At 18°C the green sunfish (Lepomi-s oyaneZlus) has a greater tolerance of
HCN than does the laboratory reared fathead minnow (JPimephales promelas). The
96-hr LC50 values were 139 and 125 yg/1 HCN, respectively, and the LTC values
were 133 and 122 yg/1 HCN, respectively. It was considered possible that the
lesser tolerance of the fathead may make it more vulnerable to predation by the
more tolerant green sunfish. Previous toxicant tests using "Mirex" (Tagatz
1976) and mercury (Kania and O'Hara 1974) exposed the prey, but the predator
was not exposed, or was exposed for a much shorter period. In the tests
reported here, both the predator and prey were given equal exposure to cyanide.
Experimental Design
The system used to introduce water of controlled temperature, pH, and dis-
solved oxygen was the same as that described in Section IV. The test tanks
were the same as previously described, except that "Nitex" screen partitions
across the short side separated the tanks into two equal compartments. Three
devices to provide cover were placed in each half. After anesthetizing
(150 mg/1 MS 222) and weighing, one or two green sunfish were placed in one
compartment and 30-60 fathead minnows were placed in the other. The number of
fatheads varied depending on their size and the size and number of green sun-
fish. At the time of the randomization of prey for the test chambers, one
additional group was made up and preserved for determination of mean length and
weight. For the nine tests the mean size of green sunfish varied from 50 to
88 mm total length and fatheads from 15 to 21 mm total length. Both the pred-
ator and prey were fed during the 3 days of pre-exposure acclimation.
The tests were started by dispensing one-fourth of the amount of sodium
cyanide required to bring the tanks up to the desired concentration of HCN.
Next, the partition was removed that separated the two fish species; then
another addition of sodium cyanide was made to bring the HCN concentration up
to one-half of the aim. The method and frequency of tests for HCH, pH, tem-
perature, and dissolved oxygen was the same as that described in Section IV.
Fatheads which died during the pretest or test periods were removed. After
4 days of interaction the predators and remaining prey were counted, measured,
and weighed. The predation rate was calculated as grams of prey eaten for each
gram of predator, based on the beginning weights of both predator and prey.
77
-------
Results
The pH was maintained at 7.9 and the temperature at 18°C. The mean dis-
solved oxygen concentration varied from 5.5 and 5.9 mg/1. The mean concentra-
tion of HCN (yg/1) In each experimental tank for each test and the overall
mean concentration in each tank for the nine tests are presented in Table 42.
The consumption of minnows in each tank for the nine tests is presented in
Table 43. The overall means of these consumption rates ranged from 0.666 g of
minnow per gram of sunfish for controls to 0.942 g/g at a nominal HCN concen-
tration of 25 ug/1. The mean consumption rates of minnows in tests at the
nominal concentrations of 50 and 100 pg/1 were 0.859 and 0.846 g/g of green
sunfish. These prey consumption rates, expressed as a percentage of corre-
sponding control values, are given in Table 44.
The variation between prey consumption at similar HCN concentrations in
different tests was large. However, it was observed that for a single test
the comsumption rate was usually greater in the treatments than in the corre-
sponding control, with minor differences among the four treatments. These
tests show that the presence of sublethal concentrations of HCN caused
increased predation by green sunfish on laboratory reared fatheads, but do not
indicate whether fatheads were easier prey or the green sunfish had greater
appetites as suggested by Leduc (1966).
TABLE 42. MEAN HCN CONCENTRATIONS IN EACH EXPERIMENTAL TANK
OF EACH TEST AND THE OVERALL MEANS FOR EACH TANK
AT DESIGNATED NOMINAL CONCENTRATIONS
Test
1
2
3
4
5
6
7
8
9
Means
Nominal concentration (ug/1)
12.5
9.9
10.3
13.7
8.9
9.2
14.8
9.2
15.8
9.3
11.2
25
24.6
16.1
20.5
23.8
26.3
30.7
25.7
29.5
24.7
24.7
50
41.5
32.7
44.2
46.5
39.0
48.3
52.7
38.8
42.9
43.0
100
85.6
72.4
78.8
98.3
114.8
114.9
119.0
100.8
113.6
99.8
78
-------
TABLE 43. PREY CONSUMPTION (GRAMS OF PREY CONSUMED PER GRAM OF PREDATOR
PRESENT) FOR EACH TEST TANK AND THE OVERALL MEANS FOR EACH NOMINAL
HCN CONCENTRATION (WITHIN TEST RANK IN PARENTHESIS)*
Nominal concentration (yg/1)
Test
1
2
3
4
5
6
7
8
9
Means
(1)
(2)
(1)
(1.
(3)
(1)
(3)
(1)
(1)
0
0.
0.
0.
5)0.
1.
0.
0.
1.
1.
0.
296
236
261
803
045
206
284
320
452
666
12.5
(2) 0.
(3) 0.
(4) 0.
(1.5)0.
(1) 0.
(4) 0.
(4) 0.
(5) 2.
(2.5)1.
0.
298
323
370
893
855
279
296
904
980
911
25
(4)0.
(5)0.
(3)0.
(4)0.
(4)1.
(5)0.
(5)0.
(2)1.
(5)3.
0.
313
360
360
988
083
307
315
452
300
942
(3)
(4)
(5)
(3)
(2)
(3)
(2)
(3.
(4)
50
0.303
0.348
0.534
0.912
0.931
0.255
0.225
5)2.112
2.112
0.859
(5)
(1)
(2)
(5)
(5)
(2)
(1)
(3.
(2.
100
0.354
0.222
0.314
1.045
1.121
0.248
0.215
5)2.112
5)1.980
0.846
* Ranking of minnow consumptions among HCN concentrations within
' each test (highest rank - largest consumption).
79
-------
TABLE 44. PREY CONSUMED FOR EACH TEST TANK AS A PERCENTAGE OF THE CONTROL
VALUES AND THE OVERALL MEANS FOR EACH NOMINAL HCN CONCENTRATION
Test
1
2
3
4
5
6
7
8
9
0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
Nominal
12.5
100.7
136.9
141.8
100.0
81.8
135.4
104.2
220.0
136.4
concentration (i
25
105.7
152.5
137.9
110.6
103.6
149.0
110.9
110.0
227.3
ng/D
50
102.4
147.5
204.6
102.1
89.1
123.8
79.2
160.0
145.5
100
119.6
94.1
120.3
117.0
107.3
120.4
75.7
160.0
136.4
Means 100.0 128.6 134.2 128.2 116.8
80
-------
EXPOSURE TO INCREASING CYANIDE CONCENTRATIONS
In most toxicity tests, exposure to toxicants is continuous and constant.
The need to approximate the effects on fish of exposures more closely resem-
bling actual spills is apparent. It has been suggested by Larson and
Schlesinger (1977) that the median survival time of fish exposed to chlorine
may be expressed as a function of the area under a time-concentration curve.
This specific relationship was pursued in an investigation using hydrogen
cyanide and fathead minnows.
Experimental Design
The diluters (Brungs and Mount 1970) and aquaria used for testing are the
same as those described in Section IV. Fathead minnows of 9-10 weeks old and
reared in a laboratory hatchery were used in the experiments. The tests were
conducted at 25°C with oxygen concentrations above 60 percent saturation.
Different rates of HCN increase in test chambers were produced by altering
any of the following (Table 45): (1) the initial volume of fresh water in the
aquaria, (2) the volume of diluent water dispensed into the aquaria, (3) the
concentration of HCN dispensed, and (4) the time interval between introductions
of diluent water in successive cycles of the diluter (cycle time).
Each test (except Test 5) was conducted with one control and four treat-
ment levels, using ten test fish in each aquaria. Mortality times were
recorded to the nearest minute. HCN samples were taken at 15, 30, 60, 90, and
120 min and every 60 min thereafter until the cyanide concentration peaked. In
Tests 1, 2, and 4 the cycle time was the only factor manipulated in changing
the time-concentration curve. The initial and final volumes were 20 1 in each
aquarium. In Tests 7, 8, and 9 the initial aquaria water volumes were 5 1 each
and the final volumes were 20 1. The volume dispensed to each aquarium with
each diluter cycle was reduced from 1.2 1 to 0.5 1. The strength of the HCN
stock solution also was varied to obtain less convex curves. Essentially
straight-line increases in HCN concentration were obtained in Tests 10, 11, and
12 by starting with only 1.5 1 in each aquarium and employing the other modifi-
cations used in Tests 7, 8, and 9. In addition, the amount of toxicant dis-
pensed to each tank was increased at a geometric rate per unit time (e.g., if
1 ml of toxicant was dispensed for each of 3 diluter cycles, then each of the
next 3 cycles would receive 2 ml/cycle, the next 3 cycles 4 ml/cycle, and so
on). The three groups of diluter modifications outlined above produced curves
of HCN concentration like those in Figure 7 which have been standardized to
intersect at the same HCN concentration and point in time.
Test 5 was designed to estimate the latent period (i.e., the minimum amount
of time needed to produce mortality regardless of the toxicant concentration.
The test was static, conducted at 25°C and lasted 2 hr. Known amounts of
sodium cyanide stock solution were pipetted and mixed into 8 of 10 aquaria con-
taining 10 fish each. The other two aquaria were controls. Mortalities were
recorded as they occurred. HCN concentrations were recorded initially and at
the end of the test. The actual HCN concentrations diminished about 10 percent
in 2 hr.
81
-------
TABLE 45. DILUTEE AND AQUARIA CONDITIONS AS RELATED TO WATER TURNOVER TIMES
Test
no.
1
2
3
4
5
6
7
8
9
10
11
12
Initial
tank
volume
(1)
20
20
20
20
20
20
5
5
5
1.5
1.5
1.5
Final
tank
volume
(1)
20
20
20
20
20
20
20
20
20
20
20
20
Cycle
time
(min)
6
4
2
2
immediate slug
square slug -
4
4
4
4
4
4
Dispensed
volume
(ml)
1,200
1,200
1,200
1,200
- latent period
removal to fresh water
500
500
500
500
500
500
Test
dura-
tion
(min)
270
180
180
180
180
120
60
90
120
120
120
180
82
-------
TEST 6
O
TIME
Figure 7. Representative curves for increasing HCN lethal exposure tests
normalized to time and concentration
83
-------
Test 6 was designed to examine the responses of fathead minnows to a
given area of "square exposure" (Larson and Schlesinger 1977) at high cyanide
concentrations, returning them to fresh water after a specified amount of mor-
tality had occurred. Fish were acclimated for 2 days to test chambers (10 fish
per aquarium) before exposure tests. A sodium cyanide stock solution was
pipetted into the aquaria at concentrations of up to 1.176 mg/1 HCN. The di-
luter and toxicant dispensers were operated to maintain the HCN concentrations
for the test duration. Once the mortality reached 80 percent in any given
aquaria, the remaining fish were removed by dipnet to a nearby, identical
diluter operating with fresh water only. Any further mortalities were
recorded.
Results
It is generally assumed that median survival time exhibits less vari-
ability than any other point on the time-mortality curve. Exposure areas
bounded by the change in HCN concentration with time from initial exposure to
the median survival times are summarized in Table 46. It was anticipated that
the area under the time-concentration curve from time zero to the median sur-
vival time should be constant regardless of the type of time-concentration
exposure curve. In Table 47 the mean areas, standard deviations, and coeffi-
cients of variation (i.e., the ratio of the standard deviation to the mean,
expressed as a percentage) are given for the various tests, which are grouped
according to type of exposure (Test 3 had insufficient mortality). The
coefficients of variation ranged for the continuously increasing HCN tests
from 20.3 percent for all test concentrations combined (excluding Tests 3, 5,
and 6) to 16.6 percent for the combined concentrations of Tests 10—12.
Test 5 indicated a latent period of less than 22 min and somewhere near
15-18 min. Test 6 incorporated a square exposure as opposed to an increasing
concentration curve. The areas under the square exposure curves were three to
four times as great as the areas under the increasing concentration curves
(Table 46), probably because such large proportions of the areas under the
square exposure curves lay within the latent period. The coefficient of vari-
ability for the areas in the square exposure test was 9.5 percent, less than
those for the increasing concentration tests.
Despite a fair amount of variability, there seems to be some support for
the concept of toxicity based on area. Therefore, all percentage mortalities
were transformed to probits and plotted against the logarithms of their
respective expsoure areas, as was done by Larson and Schlesinger (1977).
These authors did not carry their data analysis beyond stating that toxicity
to fish was a function of area based on the plot of mortality probits versus
log area. Regression analysis was performed on the combined data of all tests
(excluding Tests 3, 5, and 6), Tests 1, 2, and 4, Tests 7-9, and Tests 10-12.
Although a linear relationship is indicated, the multiple correlation coeffi-
cients (R2) were only 0.41 for all combined tests; 0.51 for Tests 1, 2, and 4;
0.42 for Tests 7-9; and 0.49 for Tests 10-12. The variability is compounded
from several sources such as HCN determinations, accurate determination of the
actual time of death, and the fitting of the mortality data to the time-
concentration curves. The R2 values are not especially high, but the use of
areas as a basis for toxicity analysis cannot be disregarded until further
84
-------
TABLE 46. MAXIMUM HCN CONCENTRATION, MEDIAN SURVIVAL TIME (MST)
AND EXPOSURE AREA DERIVED FROM THE TIME-CONCENTRATION CURVES
FOR THE INCREASING HCN LETHAL EXPOSURE TESTS
Test
no.
1
2
3
4
5
6
HCN
(yg/D
390
596
674
820
243
348
401
466
227
293
362
430
256
382
464
536f
301
382
493
583
710
946
1387
1737
606
797
920
1176
MST
(min)
238
179
138
146
_
172
166
119
_
-
_
-
_
198
160
128
165.0
164.0
107.5
88.0
69.0
44.0
26.5
22.0
62.0
45.0
33.5
26.5
Area
(cm2)*
189.87
181.02
140.11
182.96
_
129.19
146.89
106.85
™_
-
_
-
_
184.35
179.26
164.08
_
-
-
—
_
-
_
-
587.06
560.39
481.56
496.13
Test HCN
no . (yg/D
7 1070
1423
1698
2090
8 593
-
836
1066
9 463
619
713
905f
10 554
703
794
1076
11 933
1290
1330t
1700f
12 512
664
736
1001
MST
(min)
57.0
49.0
40.5
38.5
79
_
59
53
_
119
102
68
_
124.0
114.0
96.5
92.5
84.5
74.5
69.0
184
166
154
117
Area
(cm2)*
117.08
129.57
111.69
134.09
97.69
_
102.78
122.92
_
167.09
163.33
118.90
_
140.16
134.05
128.56
104.06
120.81
95.75
107.81
152.17
161.47
154.00
126.59
* Area based on 1 cm2 = 40 yg/1 HCN x 8 min
t Estimates of the obtained maximum level
85
-------
TABLE 47. MEAN AREAS, STANDARD DEVIATIONS, AND COEFFICIENTS OF VARIATION
FOR THE COMBINED TESTS FROM TABLE 46
Test number
All tests combined
(excluding 3, 5, and 6)
Tests 1, 2, and 4
Tests 7-9
Tests 10-12
Test 6
Mean
area
(cm2)*
138.55
160.46
126.51
129.58
531.29
Standard
deviation
28.15
28.25
23.20
21.48
50.55
Coefficient
of variation
(percent)
20.3
17.6
18.3
16.6
9.5
* Area based on 1 cm2 = 40 vig/1 HCN x 8 min
investigations have been concluded which improve on the above conditions of
variability. The use of regression in this situation is important in its value
as a predictive tool. Should the exposure area to fish of a particular spill
be known, the amount of mortality could roughly be predicted from the
regression relationship.
INTERMITTENT SUBLETHAL EXPOSURE TESTS
The concentration of toxicants in natural waters frequently fluctuate in a
diurnal pattern. In certain situations there may be fluctuations from non-
measurable levels to nearly lethal levels. It was the objective of these
experiments to compare the sub^Lethal effect of HCN on early growth of fathead
minnows in a continuous and two different intermittent exposure situations.
Experimental Design
The sublethal tests were conducted in five identical diluters as previously
described for acute tests. The experimental glass chambers measured 25 x 50 x
30 cm high and contained 35 1 of test solution. The flow of water through each
chamber was at the rate of 250 ml/min, affording 95 percent replacement in about
7 hr. Water temperature was maintained at about 25°C. Aeration in the water
reservoir supplying the diluters maintained dissolved oxygen concentrations in
the test chambers at approximately 73 percent of saturation. The test chambers
were illuminated for 16 hr daily with two 40-W fluorescent tubes 45 cm above
the chamb er s.
Water temperature, dissolved oxygen (DO), pH, and free cyanide concentra-
tion in each test chamber were measured three times per week. The HCN concen-
trations were calculated as previously described. Determinations for each test
86
-------
tank were averaged and used as the reported values. The toxicant concentrations
in the various test tanks followed one of three regimes. In seven tanks the
concentrations were held constant. In the remaining 14 treatment chambers, the
concentrations fluctuated about two different patterns with seven test chambers
for each treatment regime. The overall mean of the maximum concentrations in
the pattern of diurnal fluctuation will be used when referring to specific test
treatments.
Growth tests were begun by randomly adding 50 newly hatched fathead minnow
larvae to each of 21 treatment-and 4 control test chambers. Through the first
week of exposure, the larvae were fed a fine, hard-boiled egg yolk suspension
three times daily. Beginning the second week a supplement of finely ground
Glencoe granules was added twice daily. The fry were fed newly hatched brine
shrimp (Artemia) twice daily beginning the third week, with finely chopped
lettuce also being added to the egg yolk suspension. After 30 days of exposure,
the test was terminated and the surviving fish were measured and weighed. Mean
dry weights were determined after the fish had been dried for 24 hr at 105°C.
Results
The diurnal fluctuations in HCN levels, expressed as percentages of the
maximum concentration for each test chamber, under the three treatment regimes
are presented in Figure 8. These relationships were determined from three time
exposure series which were run during the growth test, and represent the mean
patterns for the exposure regimes. The percentages were independent of the
maximum HCN concentration, and thus one response curve can be used to represent
the fluctuation for all concentrations of each regime. The vertical bars in
Figure 8 represent the range of percentages for all test tanks at a given time
of the diurnal exposure cycle. The relative exposure areas, compared to 100
percent for continuous arid constant concentrations, were 64.9 percent for the
pattern of 16 hr on and 8 hr off, and 34.5 percent for the pattern of 8 hr on
and 16 hr off. The on and off designations refer to when cyanide was being
metered into the test chamber and when it was not, respectively.
The results of 30-day growth tests with fathead minnows exposed to each of
three concentration regimes are presented in Table 48. The growth data were
analyzed by using linear regression (Table 49) to see if the relationships of
fish weight to log HCN concentration for the three exposure patterns were
similar and in some manner a function of the continuous exposure response.
Statistical analysis demonstrated that there is no significant difference
(P < 0.05) between the slopes of the concentration-response regression lines.
Therefore, the adverse effects of HGN on early growth of the fathead minnow for
the fluctuating diurnal regimes are for each exposure pattern some constant
function of the response observed with continuous exposure. This relationship
is not directly related to exposure area, mean, or maximum HCN concentration.
About all that can be concluded is that as the diurnal exposure period is
reduced, the adverse effect of HCN on early growth of the fathead minnow is
lessened. This is true even though the maximum concentrations attained in the
test chambers with fluctuating concentrations are the same as those sustained
in the continuous exposure aquaria.
87
-------
00
00
Z
UJ
o
z
o
o
u.
o
o
HI
o
I
LU
O
oc
UJ
0.
100
90
80
70
^ 60
50
40
30-
20-
CONTINUOUS
• I6JH ON/8H OFF
8H ON/I6H OFF
I2M 2 4 6 8 10 I2N 2 46 8 10 I2M
TIME
Figure 8. Diurnal fluctuation of HCN levels in test chambers as a function
of their maximum concentrations. Ranges indicated by vertical bars.
-------
TABLE 48. GROWTH OF FATHEAD MINNOWS IN 30 DAYS WHEN EXPOSED
TO CONTINUOUS OR INTERMITTENT HCN CONCENTRATIONS*
Test
solu-
tion
1
2
3
4
1
2
3
4
5
6
7
1
2
3
4
5
6
7
concen-
tration
Ug/1
-
-
-
20.4
26.8
41.8
43.2
60.7
71.4
82.4
22. 9#
28. 6#
34.2
49.8
57.5
66.1
73.8
Number
of
survivors Wet
Controls
24
26
25
30
32
30
31
25
26
16
22
16
24
24
29
26
28
24
29
42.5
40.9
32.5
24.9
Continuous
41.3
36.0
32.0
19.9
13.4
8.69
9.41
hr on - 8 hr off
29.9
35.6
38.9
34.0
30.6
30.1
22.3
Dry
6.29
6.81
4.84
4.40
5.53§
5.16
4.33
3.87
3.16
2.19
0.94
1.09
5.20
6.34
5.45
4.50
4.26
3.71
3.34
Percentt
of
controls
-
-
-
93.3
78.3
70.0
57.1
39.6
17.0
19.7
94.0
114.6
98.6
81.4
77.0
67.1
60.4
89
-------
Table 48 Continued
Test
solu-
tion
1
2
3
4
5
6
7
HCNt
concen-
tration
yg/i
23. 5#
33. 0#
45.8
49.6
61.6
63.4
85.1
Number
of
survivors
8 hr on -
25
24
26
26
25
26
27
Wet
16 hr
41.8
32.3
30.1
32.8
28.9
27.3
23.7
Mean weight (mg)
Dry
off
5.56
5.62
5.23
5.08
4.88
4.54
3.22
Percent^
of
controls
100.5
101.6
94.6
91.9
88.2
82.1
58.2
* Overall mean temperature was 23.5°C; DO, 6.24 mg/1; and pH, 7.99
t Concentrations corresponding to maximum values for intermittent
exposures
f Percentages based on dry weights
§ Weighted mean based on number of survivors in each replicate
# Values omitted from regression analysis (Table 49)
90
-------
TABLE 49. LINEAR REGRESSION ANALYSIS OF THE TOXICANT DOSE RESPONSE CURVES EXPRESSING TERMINAL
MEAN DRY WEIGHT IN mg (Y) AND PERCENTAGE TERMINAL MEAN DRY WEIGHT NORMALIZED TO CONTROLS (Y)
AS A FUNCTION OF LOG TOXICANT CONCENTRATION IN yg/1 (X)
Linear regression
Exposure regime
Continuous
16 hr on - 8 hr off
8 hr on - 16 hr off
Mean dry weight
a 6
14.63 -7.070
15.01 -6.195
17.47 -7.255
R2
0.9393
0.9862
0.8923
analysis*
Normalized
a
264.5
271.6
316.4
percent mean
3
-127.8
-112.2
-131.4
dry weight
R2
0.9393
0.9866
0.8934
* Values omitted from regression analysis are so indicated in Table 48.
Linear regression equation is Y = a + 3(log X).
-------
BINARY MIXTURES WITH HCN
Natural waters receiving wastes often contain several toxic or potentially
toxic substances. Relatively few investigations have defined the adverse
effect of mixtures of toxicants to fish, especially at the sublethal level.
Water quality standards governing the treatment and control of wastes dis-
charged into natural waters should reflect valid water quality criteria that
adequately protect valued aquatic populations. When more than one toxicant is
present in the same system, their combined, interactive effect can be exerted
in such a manner that the collective toxicity may be predictable from the
separate toxicities of the individual substances. This interactive effect may
occur as similar or independent joint action, and it may be greater or less
than predicted or show no interaction. If the response is less than predicted
on the basis of no interaction then antagonism occurs. Because of the impor-
tance of being able to predict toxicity in natural waters from a knowledge of
individual toxic constituents, the purpose of this investigation was to examine
the validity of various methods for predicting joint toxicity of binary mix-
tures of HCN with hexavalent chromium, zinc, or ammonia from lethal and sub-
lethal responses of fish to individual toxic components in the mixtures.
Experimental Design
Newly hatched larvae or juvenile fathead minnows (Pimephales promelas)
and fry or juvenile rainbow trout (Salmo ga-iz>dnei*i-i Richardson) were used as
test organisms. The minnows were cultured in the laboratory from a strain
originating from the U.S. Environmental Protection Agency's Environmental
Research Laboratory in Duluth, Minnesota. The minnows were approximately 8
weeks old when used, had mean total lengths of 19 mm, and mean wet weights
among survivors of 0.079 g. Numerous lots of juvenile minnows were used
during the acute toxicity tests. However, the sensitivity of all minnows to
any specific toxicant was demonstrated to be quite uniform throughout the
testing period.
Eyed rainbow trout embryos from eggs spawned in late April 1976 were
obtained in early July 1976 from the French River State Fish Hatchery, French
River, Minnesota. A second lot, spawned in late April 1977, was obtained from
the hatchery in early July 1977. The water temperature at collection and
during subsequent holding and testing of the trout was 10°C. The trout reared
from the 1976 stock were approximately 22 weeks old when used for lethal tests,
had mean total lengths of 53 mm, and mean wet weight among survivors of 1.48 g.
Fry reared from the 1977 stock were used in the sublethal tests.
The lethal and sublethal tests were conducted in identical diluters as
described for other acute tests in Section IV. The experimental glass chambers
measured 25 x 50 x 30 cm high and contained 35 1 of test solution. Water
flowed through each chamber at the rate of 250 ml/min, affording 95 percent
replacement in about 7 hr (Sprague 1969).
Water temperature in all tests with fathead minnows was maintained at 25°C
and with rainbow trout at 10°C. Aeration in the head water reservoir feeding
the diluters maintained dissolved oxygen concentrations in the test chambers
at approximately 80 percent saturation. A sulfuric acid solution was auto-
92
-------
matically dispensed into the reservoir to control the pH of the test water for
certain tests. The test chambers were illuminated for 16 hr daily with two
40-W fluorescent tubes 45 cm above the chambers. Stock toxicant solutions of
sodium cyanide, sodium dichromate, zinc sulfate and ammonium chloride were
prepared with reagent grade chemicals and deionized water. The ammonia tests
were conducted at 10°C to retard nitrification and prevent the buildup of
adverse nitrite levels in the test chambers.
Water temperature, dissolved oxygen (DO), pH, and toxicant concentrations
in each test chamber were measured daily during each lethal test and three
times per week during the sublethal tests. Determinations for each test were
averaged and used as the experimental toxicant concentration. Free cyanide con-
centrations were determined as in previous tests. Hexavalent chromium, zinc,
and free ammonia concentrations were determined by the diphenylcarbazide,
zincon, and direct nesslerization colorimetric procedures, respectively (APHA,
e~t ai. 1975). The concentration of molecular HCN or NH3 was calculated for
each free cyanide or ammonia determination using the daily pH and temperature
measurement and the pK equilibrium constants (i.e., -log Ka) calculated from
the equations:
= 3.658 + 1662/T (Broderius, unpublished data)
= 0.09018 + 2729.92/T (Emerson, et al. 1975)
where T is temperature in degrees Kelvin. The ratio [HCN]/[free cyanide] or
[NH3]/[free ammonia] is taken to be equal to the factor
1/(1 + 10pH " pKa) or 1/(1 + 10pKa " PH)
respectively. When both the molecular HCN or NH3 and free cyanide or ammonia
concentrations are expressed in the same units and as the molecular form, then
free cyanide or ammonia times the appropriate factor will equal molecular HCN
or NH3.
The 96-hr lethal tests were conducted with one control and four treat-
ments in a geometric series. Twenty fathead minnows or 10 rainbow trout
juveniles were randomly distributed into each treatment and control chamber.
The fish were acclimated to the test chambers for 2-3 days before introduction
of the toxicants. During the 24-hr period preceeding and throughout a lethal
test the fish were not fed. In each binary experiment the single toxicant
tests were completed before the joint toxicant tests. Since all the lethal
tests for each pair of toxicants were completed within a few weeks, changes in
fish susceptibility within a test period were assumed to be nonsignificant.
Percentage fish mortality after 96 hr of exposure to individual toxicants
and to binary mixtures containing fixed proportions of the constituents was
used as the response criterion.* Estimates of the concentration of toxicant
most likely to cause 50 percent mortality (LC50) and toxicity curve slopes
after 96 hr of exposure were made by a log-probit analysis computer program
(Dixon 1973). All the data from toxicity tests conducted for specific tox-
93
-------
icants and for their mixtures were composited for probit analysis. The 95 per-
cent confidence limits for LC50 values were computed according to formulas
proposed by Litchfield and Wileoxon (1949) and Finney (1971).
The manner in which the combined effect of mixtures of toxicants was cal-
culated by the toxic unit approach has been outlined by Sprague (1970). With
this procedure the proporitonal toxicity of each poison in a mixutre is
obtained by dividing the actual concentration in solution by its calculated
96-hr LC50. These values obtained for the two toxicants in a binary mixture
are then summed to give the proportion of the 96-hr LC50 in the mixture. The
concentrations selected for testing were usually such that each poison theoret-
ically contributed an equal proportion of toxic units to the mixture. The
toxic unit-percentage mortality relationships were analyzed by the computer
programs as outlined above.
The toxic unit (TU) or sum of biological activity index of joint toxicity
can be modified to represent an index of additive toxicity distributed about
zero, which represents simple additive toxicity. This additive index as
proposed by Marking (1977) is equal to [(1/TTJ) - 1.0] for TU i 1.0 and
[TU(-l) + 1.0] for TU i i.o. Thus toxicity greater than and less than simple
additive toxicity is represented by positive and negative index values, res-
pectively. The magnification factor is obtained by adding one to the numerical
Index value for greater than additive toxicity and taking the reciprocal of the
numerical index value for less than additive toxicity. The significance of
deviation from zero is determined by substituting selected values of the 95
percent confidence limits for the calculated toxic units into the appropriate
additive index formula to establish the deviation in the additive index range.
Simple additive toxicity is assumed whenever an index range overlaps zero.
If the linear response curves of individuals to two toxicants are not
parallel or if the modes of toxic action of the two constituents are known to
be different for toxicants which had similar slopes, then it is possible that
the separate responses of organisms to the individual concentrations of each
toxicant in a mixture can be summed. The mortality to be expected in such a
binary mixture exerting independent joint action (Bliss 1939) or response
addition (Anderson and Weber 1977) depends upon the correlation between the
susceptibilities of the individual organisms to each toxicant. According to
Plackett and Hewlett's (1948, 1952) mathematical models, if the correlation of
Individual tolerances is completely negative (r = -1) so that the proportion
of individuals most susceptible to one toxicant alone (Pj) are least suscep-
tible to the other (£2)> then the proportion of individuals expected to respond
to the mixture (Pm) can be represented by:
pm = P! + P2 if (Pi + P2 - 1)
With no correlation In susceptibility (r = 0) the relationship is expressed by:
Pm = P! + p2
If there was a complete and possible correlation (r = 1) so that individuals
most susceptible to the first toxicant would be correspondingly susceptible to
94
-------
the second, then
"D —, "TJ
m I
where PI is the response to the most toxic constituent in the mixture.
Growth tests were begun by randomly adding 50 newly hatched fathead
minnow larvae or 20, 4-week-old rainbow trout fry to each of the 21 treatment
and 4 control test chambers. From a subsample of 32 individuals, it was deter-
mined that the trout fry initially were 28.5 i 2.2 mm in mean total length, and
0.180 ± 0.059 g and 0.0278 g in mean wet and dry weight, respectively. The
treatments consisted of seven independent cyanide and seven second toxicant
levels, and seven binary mixtures.
Through the first week of exposure, the fathead minnow larvae were fed a
fine, hard-boiled egg yolk suspension three times daily. Beginning the second
week a supplement of finely ground Glencoe granules was added twice daily for
the remainder of the test. Beginning the third week, finely chopped lettuce
was added to the egg yolk suspension, and newly hatched brine shrimp {Aybem'La.)
were fed twice daily. During the first two weeks of the trout tests fry were
fed twice daily with ground Glencoe starter and three times daily with newly
hatched brine shrimp. Beginning the third week, Glencoe No. 1 and Oregon Moist
pellets were fed three times per day and brine shrimp twice daily. Food
slightly in excess of what the fish consumed in 5 min was added to each
chamber. After 30 days of exposure each test was terminated and the surviving
fish were measured and weighed individually. Mean weights of fish that had
been dried for 24 hr at 105°C were determined for each treatment.
For statistical analysis, individual wet weights were transformed to
natural logarithms and subjected to one-way analysis of variance. Treatment
means were compared with the overall mean from control replicates using
Dunnett's procedure (two-tailed) at a significance level of P < 0.05. Linear
regression analyses were applied to growth data to determine whether the rela-
tionships of mean dry weight or weight normalized to controls versus log
toxicant concentration were the same for a toxicant by itself and in a binary
mixture.
Results
The test conditions and results of lethal tests using individual toxi-
cants and binary mixtures are summarized in Tables 50 and 51. Probit analysis
of data from binary mixture tests, using the log summation of the individual
concentrations of toxicants in each treatment mixture, expressed as proportions
of their 96-hr LC50 values, as the dose metameter, generally gave 96-hr LC50
toxic unit values of slightly less than 1.0. The hexavalent chromium-HCN
mixture was the exception with a 96-hr LC50 toxic unit value of 1.312. Since
the 95 percent confidence limits for the toxic unit values do not bracket 1.0
in any case, nor does the additive index range overlap zero, the estimated
values for the 96-hr LC50 concentrations differ significantly from that
expected from the simple additive toxic unit hypothesis. Therefore, the acute
lethal toxicities of various binary mixtures of HCN and Cr, Zn, or ammonia
could not be predicted for the ratios tested by summing the fractional toxic-
95
-------
TABLE 50. MEAN TEST CONDITIONS AND DESCRIPTION OF TEST FISH FOR ACUTE TOXICITY BIOASSAYS
OF INDIVIDUAL TOXICANTS AND BINARY MIXTURES. (STANDARD DEVIATIONS IN PARENTHESES)
Toxicant (s)
HCN
HCN
Cr
Zn
Total
ammonia-N
NH3
Cr-HCN
Zn-HCN(l)
Zn-HCN(2)
(Total
ammonia-N) -HCN
NH3-HCN
Fish*
species
FM
RT
FM
FM
RT
RT
FM
FM
FM
RT
RT
No. of
tests
11
6
6
10
8
8
6
3
4
4
4
PH
Individual
7.76
8.04
7.80
7.69
7:96
7.96
Binary
7.90
7.74
7.72
7.98
7.98
Temp.
<°C)
Toxicants
25.0(0.12)
10.1(0.06)
25.0(0.18)
24.7(0.16)
10.0(0.07)
10.0(0.07)
Mixtures
25.1(0.13)
24.8(0.20)
24.6(0.15)
10.1(0.04)
10.1(0.04)
DO
(mg/1)
6.91(0.10)
8.82(0.09)
6.82(0.09)
7.12(0.11)
8.74(0.12)
8.74(0.12)
6.93(0.10)
6.95(0.18)
7.06(0.08)
8.93(0.05)
8.93(0.05)
Test
Length
(mm)
19.8
52.7
20.8
18.0
48.7
48.7
20.5
17.6
18.3
57.6
57.6
fish mean
Wet weight
survivors
(g)
0.0762
1.407
0.0950
0.0722
1.168
1.168
0.0988
0.0581
0.0733
1.852
1.852
* FM represents fathead minnows and RT, rainbow trout
-------
TABLE 51. DOSE-RESPONSE PARAMETERS, LETHAL CONCENTRATIONS AND ADDITIVE INDICES FOR ACUTE TOXICITY
BIOASSAYS OF INDIVIDUAL TOXICANTS AND BINARY MIXTURES (95 PERCENT CONFIDENCE LIMITS IN PARENTHESES)
Toxicant cone.* Dose-response regression
ratios, tox./HCN analysist
Fish Deter- Treat- 96-hr LC50
Toxicant(s) species Aim mined mentsT a 3 mg/1 or TU§
Additive index
and range^
Individual Toxicants
HCN
HCN
Cr
Zn
Total
ammonia-N
NHc
FM
RT
FM
FM
RT
RT
36 16.75 13.01
15 33.63 23.04
30 -2.787 5.119
32 2.590 5.776
0.125
(0.120-0.130)
0.0572
(0.0558-0.0586)
33.2
(31.1-35.4)
2.61
(2.49-2.75)
23 -9.069 9.063 35.7
(32.5-39.1)
23 6.325 8.460 0.697
(0.629-0.773)
- continued -
-------
TABLE 51. (continued)
Toxicant (s)
Toxicant cone.* Dose-response regression
ratios, tox./HCN analysist
Fish
species Aim
Deter- Treat-
mined ments'-^ a
96-hr LC50
mg/1 or TU§
Additive index
and range//
Cr-HCN
Zn-HCN (1)
FM 266
Binary Mixtures
235±20 20 3.906
9.277
FM 20.9 14.3+1.0 10 7.008 13.19
1.312
(1.222-1.408)
0.704
(0.650-0.763)
-0.312
-0.408 to -0.222
0.420
0.311 to 0.538
<£>
00
Zn-HCN (2)
(Total
ammonia-N) -HCN
NH3-HCN
FM 20 . 9
RT 624
RT 12 . 2
23.212.7
594+38
12.4±1.1
13
10
10
6.966
6.668
6.401
13.87
20.08
21.31
0.721
(0.692-0.
0.826
(0.790-0.
0.859
(0.824-0.
752)
864)
897)
0
0
0
0.387
.330 to
0.211
.157 to
0.164
.115 to
0.445
0.266
0.214
* Ratio of individual toxicant concentrations in different treatments of mixture tests. Aim
values are for equal proportions of toxic units.
/\ ^
t For equation Y^ = a + g(log X^) when Y^ is the predicted percentage mortality as probits and
Xi is log toxicant concentration in mg/1 or toxic units (Dixon 1973)
f Number of toxicant concentrations in the regression analysis
§ TU represents toxic units or the proportion of the expected 96-hr LC50 assuming simple additivity
# Calculations based on 96-hr LC50 in toxic units and 95 percent confidence limits
-------
ities of the particular poisons present in each mixture. In two of the three
cases, the mixtures exerted a supra-additive effect (i.e., toxicity was greater
than predicted) with the Zn-HCN and ammcmia-HCN mixtures being 1.4 and 1.2
times as toxic as predicted, respectively (Marking 1977). The exception was
hexavalent chromium whose interaction with HCN seems to be of the infra-
additive type since the mixture was 0.76 times as toxic as predicted. It
appears that for each fish species the dosage-mortality regression slope value
is nearly the same for HCN alone and mixture tests analyzed by the toxic unit
approach (Table 51), even when the individual toxicants differ both in potency
and in the slopes of their dosage-mortality curves. This result suggests that
in both systems HCN exerts its toxic action in a similar manner and is the
factor which defines the slopes of the mixture response curves.
For most of the individual toxicant and mixture tests a median lethal
threshold concentration could be determined in 96 hr of exposure. However,
for hexavalent chromium alone additional mortality of fathead minnows occurred
beyond this period. The 10-, 20-, and-30-day LC50 and 95 percent confidence
limts, determined under similar test conditions as reported for 96-hr tests
with Cr(VI) (Table 50), were 12.4 (11.0-14.0), 5.99 (5.14-6.96), and 4.36
(3.85-4.92) mg/1 as Cr, respectively.
In the present study, the slopes of the dose-response lines for components
of the mixtures were all dissimilar when tested separately. Parallelism
between response curves is apparently a prerequisite to concentration addition
as defined by Anderson and Weber (1977), but not for response addition. There-
fore, it is possible that this latter model, for toxicants which are thought to
primarily affect different biological systems, could predict responses compar-
able to those observed in the binary mixtures. When this approach was used to
analyze the data, it was determined that in every case the expected mortality
was significantly less using a Chi square test (P = 0.05) than was experi-
mentally obtained. Using the dose-response regression lines for the individual
toxicants and assuming the response addition model with no correlation in
susceptibility to the two toxicants, a mortality of about 20 percent was pre-
dicted .among test fish exposed to the Cr-HCN mixture which, in fact, killed
about 50 percent. For the Zn-HCN and ammonia-HCN mixtures in which the
observed mortality was about 50 percent, the predicted mortality was less than
1.0 percent.
Bliss (1939) has pointed out that for individual toxicants which act inde-
pendently and have dosage-mortality curves with different slopes, the response
to a binary mixture should have some compound discontinuity consisting essen-
tially of two straight line segments. The slopes of these rectilinear segments
should also approach the slopes of the regression lines for the individual
constituents. This result was not observed for the mixtures tested since in
all cases a single straight line adequately defined the dosage-mortality
relations. It was also observed that either toxicant could contribute to the
toxicity of the mixtures tested even when individually present at a concentra-
tion which would result in no mortality if the other were absent. However,
responses to toxicants acting upon different biological systems are predicted
to be additive only if each toxicant in a mixture is present at a. concentra-
tion above its respective threshold level. Therefore, it is proposed that the
response and concentration addition models (Anderson and Weber 1977) cannot be
99
-------
used to predict dosage-mortality curves for the mixtures studied.
The results of 30-day growth tests with fathead minnows and rainbow trout
exposed to individual toxicants or binary mixture solutions are presented in
Tables 52-55. The maximum percentage reduction in mean dry weight (Tables 52-
54) for concentrations approaching the incipient lethal level were all about
50 percent, with the greatest being 79 percent for 3.063 mg/1 of chromium
alone. The growth data were analyzed by linear regression to see if the slopes
and intercepts of the relationships defining mean dry weight on log toxicant
concentration are the same for a specific toxicant alone in solution and when
in a binary mixture. Since F tests show that there is no significant differ-
ence (P < 0.05) between appropriately paired regression lines defining the
relationship of weight to log concentration, no interactive sublethal effect on
growth could be demonstrated for the toxicants tested. This similarity of
regression lines can be seen in Table 55 by comparing the intercept and slope
parameters for individual toxicants with the parameters for those toxicants
when in binary mixtures. Hydrogen cyanide in combination with ammonia was the
only mixture that seemed to demonstrate some sublethal interactive effect on
growth of rainbow trout. When terminal mean dry weight was regressed on log
HCN, NH3, or total ammonia-N concentrations, the weight—concentration relation-
ship for a specific toxicant in the binary mixture indicated in all cases a
greater affect than that determined for the corresponding toxicant when tested
alone (Table 55). However, the only regression relationships that were statis-
tically different at an acceptable level of significance (F£ IQ = 3.67,
P < 0.063) were those calculated for the effect of HCN alone'compared to HCN
when in combination with ammonia. Therefore, the sublethal stress impose^ on
fish in most binary mixtures studied disrupted physiological functions so
that the relationships between concentration and reduction in growth were essen-
tially equal to those when the toxicants were present alone. The adverse
joint action of individual toxicants on growth were not predictable according
to existing models.
Summary
A thorough understanding of chemical reactions of constituents in a mix-
ture is necessary for prediction of joint effects. Where toxicants chemically
combine, prediction of joint toxicity may be most difficult. Chromium and zinc
can combine with cyanide to form coordination compounds. However, calculations
from stability constants showed that under the conditions prevailing in the
mixture solutions, less than 1.0 percent of the cyanide, and even smaller per-
centages of the metals, would be present as complex ions. Therefore, the
toxicants in each of the binary mixtures studied should be in essentially the
same form as they were in the solutions where their toxicities were individually
determined.
The physiological basis for interaction of toxicants in their effects on
fish has received some attention in recent years. As pointed out by Marking
(1977), projected theories for the mechanisms of toxicant interaction in mix-
tures include increase in rate of uptake, formation of toxic metabolites,
reduction of excretion, alteration of distribution, and inhibition of detoxifi-
cation. Lloyd and Swift (1976) studied whether the increased water uptake by
fish exposed to high ambient ammonia concentrations was accompanied by an
100
-------
TABLE 52. GROWTH OF FATHEAD MINNOWS IN 30 DAYS WHEN EXPOSED TO INDIVIDUAL
OR MIXTURE SOLUTIONS OF HEXAVALENT CHROMIUM AND HCN*
Test
solu-
tion
1
2
3
4
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Toxicant Number
concentration of
mg/1 survi—
HCN Cr vors
Controls
23
- 37
36
- 25
HCN
0.0176 - 26
0.0202 - 36
0.0300 - 27
0.0344 - 30
0.0517 - 27
0.0643 - 20
0.0842 - 21
Cr
0.050 27
0.105 35
0.186 40
0.432 38
0.803 36
1.448 33
3.063 22
Wet
48.3
37.7
49.9
54.9
46.9$
48.5
50.7
41.6
38.4
33. 9§
30. 2§
22. 3§
46.7
31.91
36.0
36.5
20. 4§
19.91
11. 8§
Mean weight
Dry
9.12
6.73
9.44
11.2
8.91*
9.08
10.0
7.86
7.43
6.11
5.10
3.76
9.37
5.66
6.82
7.11
3.61
3.61
1.91
(fflg)
Percent
of
controlst
-
_
-
101.9
112.2
88.2
83.4
68.6
57.2
42,2
105.2
63.5
76.5
79.8
40.5
40.5
21.4
101
-------
TABLE 52 Continued
Test
solu-
tion
1
2
3
4
5
6
7
Toxicant
concentration
inR/1
HCN
0.0149
0.0177
0.0252
0.0278
0.0398
0.0460
0.0497
Cr
0.030
0.051
0.100
0.213
0.436
0.563
0.796
Number
of
survi-
vors
Cr-HCN Mixture
28
31
25
28
33
34
32
Wet
50.4
42.9
46.9
35.0
30. 8§
26. 7§
26. 6§
Mean weight
Dry
9.96
8.19
9.24
6.89
6.24
5.35
4.81
(mg)
Percent
of
controls
111.8
91.9
103.7
77.3
70.0
60.0
54.0
* Overall mean temperature was 25.3°C; DO, 7.24 mg/1, and pH, 7.79
t Percentages based on dry weights
^ Weighted mean based on number of survivors in each replicate
i Values are significantly different from mean of controls according
to Dunnett's procedure ,(two-tailedj P < 0.05)
102
-------
TABLE 53, GROWTH OF FATHEAD MINNOWS IN 30 DAYS WHEN EXPOSED TO INDIVIDUAL
OR MIXTURE SOLUTIONS OF ZINC AND HCN*
Toxicant Number
Test concentration of
solu- mg/1 survi-
tion HCN
1
2
3
4
1 0.0232
2 0,0286
3 0.0344
4 0.0439
5 0.0535
6 0.0545
7 0.0865
1 ~
2
3
4
5
6
7 —
Zn vors
Controls
25
30
31
28
HCN
30
27
23
27
29
25
26
Zn
0.299 33
0.462 32
0.465 34
0.632 26
0.689 21
0.759 19
0.840 13
Mean weight (mg)
Perce'
o
Wet
61
57
51
45
53
45
42
40
36
32
27
26
56
46
40
38
37
35
42
.2
.5
.3
.8
.8t
.1
.2
.1
.9
.0§
.15
.85
.4
.1
.9
.8
.5
.9§
.4
Dry
13.
11.
10.
9.
10.
9.
8.
7.
7.
6.
5.
5.
11.
9.
8.
7.
6.
7.
7.
2
5
2
11
9t
27
22
93
44
07
32
12
9
28
09
54
86
11
92
nt
f
concrolst
_
-
_
—
85
75
72
68
55
48
47
109
85
74
69
62
65
72
.0
.4
.8
.3
.7
.8
.0
.2
.1
.2
.2
.9
.2
.7
103
-------
TABLE 53. Continued
Test
solu-
tion
Toxicant
concentration
mg/1
HCN
Zn
Number
of
survi-
vors
Mean weight (mg)
Wet
Dry
Percent
of
controlst
Zn-HCN mixture
1
2
3
4
5
6
7
0.167
0.0242
0.0283
0.0352
0.0390
0.0448
0.0571
0.226
0.288
0.419
0.566
0.655
0.620
0.863
26
35
29
23
28
18
11
56.6
51.9
42.4
46.6
35. 8§
44.4
36.4
12.2
11.0
8.14
9.39
6.79
8,89
6.55
111.9
100.9
74.7
86.1
62.3
81.6
60.1
* Overall mean temperature was 25.2°C; DO, 7.01 mg/1; and pH, 7.78
t Percentages based on dry weights
f Weighted mean based on number of survivors in each replicate
§ Values are significantly different from mean of controls according
to Dunnett's procedure (two-tailed; P < 0.05)
104
-------
TABLE 54. GROWTH OF RAINBOW TROUT IN 30 DAYS WHEN EXPOSED TO INDIVIDUAL
OR MIXTURE SOLUTIONS OF AMMONIA AND HCN*
Test
solu-
tion
1
2
3
4
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Toxicant concentration Number
mg/1 of
Total survi-
HCN NH3 ammonia-N vors
Controls
- - 20
20
- - 19
20
HCN
0.0086 20
0.0148 19
0.0176 20
0.0203 - - 17
0.0262 17
0.0347 16
0.0440 6
NH3 or total N
0.053 2.49 20
0.102 5.45 19
0.121 6.53 20
0.171 9.25 20
0.221 11.8 19
0.253 14.5 13
0.318 18.0 14
Mean weight (mg) __
Percent
of
Wet Dry controls
815
824
842
863
836±
880
833
870
805
757
615#
441#
907
747
846
726
672
584#
459#
160
154
156
157
157*
1661 105.75
154 98.1
163 103.8
149 94.9
136 86.6
112 71.3
79 50.3
1655 105.11
139 88.5
156 99.4
128 81.5
119 75.8
100 63.7
76 48.4
105
-------
TABLE 54. Continued
Test
solu-
tion
1
2
3
4
5
6
7
0
0
0
0
0
0
0
Toxica
HCN
.0072
.0134
.0161
.0159
.0221
.0248
.0279
nt c<
m,
]
0
0
0
0
0
0
0
oncentr.
s/i .
NH3
NHg O
.057
.098
.152
.132
.161
.166
.187
ation
Number
of
Mean weight
Total survi-
aromonia-N vors Wet
r total N-HCN mixture
2,78 20 888
4.
6,
7.
9.
9.
10.
80
89
57
09
27
6
19
20
18
19
18
14
830
755
700
630
625
546#
Dry
I67i
153
138
125
111
111
93
(mg)
Percent
of
controls
106. 4§
97
87
79
70
70
59
.5
.9
.6
.7
.7
.2
* Overall mean temperature was 10.1°C; DO, 8.0 rag/1, and pH, 7.93
t Percentages based on dry weights
i Weighted mean based on number of survivors in each replicate
§ Values were omitted from regression analysis summarized in Table 55
# Values are significantly different from mean of controls according to
Dunnett's procedure (two-tailed; P < 0.05)
106
-------
TABLE 55. LINEAR REGRESSION ANALYSIS OF THE TOXICANT DOSE RESPONSE CURVES EXPRESSING TERMINAL
MEAN DRY WEIGHT IN MG (Y) AND PERCENTAGE TERMINAL MEAN DRY WEIGHT NORMALIZED TO CONTROLS (Y)
AS A FUNCTION OF LOG TOXICANT CONCENTRATION IN MG/L (X)
Linear regression analysis*
Toxicant(s)
g HCN
Cr-HCN
HCN
Zn-HCN
HCN
NH3-HCN
or total N
Mean dry weight
a 3
Weight
-5.053 -8.481
-6.387 -8.853
-3.590 -7.762
-5.876 -10.01
-134.4 -165.3
-157.0 -162.8
R2
on log HCN
0.9594
0.8469
0.9067
0.7332
0.9014
0.9100
Normalized
a
-56.70
-71.75
-32.81
-53.79
-85.71
-100.3
percent mean
B
-95.17
-99.40
-71.13
-91.78
-105.4
-103.8
dry weight
R2
0.9595
0.8468
0.9060
0.7331
0.9018
0.9096
-------
TABLE 55. Continued
o
00
Linear regression analysis*
Mean dry weight
Toxicant (s)
Cr
Cr-HCN
Zn
Zn-HCN
NH3
NH3-HCN
Total N
Total N-HCN
a
3.996
4.720
5.938
6.073
18.24
-39.67
251.3
274.0
3
Weight on
-3.583
-3.395
-9.798
-9.021
-137.5
-193.5
-131.5
-170.7
R2
log X - second
0.7992
0.8987
0.7789
0.8133
0.8466
0.7891
0.8582
0.9506
Normalized percent mean dry weight
a
toxicant
44.83
52.95
54.47
55.73
11.63
-25.46
160.1
174.7
3
-40.23
-38.12
-89.90
-82.70
-87.57
-123.5
-83.77
-108.9
R2
0.7991
0.8986
0.7779
0.8135
0.8459
0.7894
0.8575
0.9508
Linear regression equation is Y = a + 3(log X)
-------
Increased rate of phenol uptake in ammonia—phenol mixtures. They concluded
that the rate of uptake and concentration of phenol in fish muscle is not
influenced by elevated ambient ammonia levels nor is the urine flow rate
increased for rainbow trout exposed to ammonia-phenol solutions. Therefore,
no physiological bases could be found to account for the empirical, additive
toxic action of ammonia and phenol. The effects measured in the binary mixture
were those corresponding to either ammonia or phenol alone and the presence of
one did not appear to enhance the effect of the other.
In the present study, it was thought that the toxic action of the toxi-
cants investigated was dissimilar. However, adverse lethal effects are appar-
ently not independent since each toxicant in our binary mixtures did not act
to produce its own separate stress which, when predicting joint toxicity, can
be summed according to existing multiple toxicity models. The interactive
nature of toxicants studied was also a function of the particular response
measured. Therefore, since it has not been established whether the adverse
effects of toxicant mixtures on fish can be predicted from individual toxicant
effects, a need still exists for development of a valid multiple toxicity
approach to evaluate the toxicity of chemical combinations.
109
-------
SECTION VII
REFERENCES
Allee, W.C. 1929. Studies in Animal Aggregations of the Isopod, Asellus oam-
munis. Ecology 10:14-36.
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1975. Standard Methods for the
Examination of Water and Wastewater. 14th ed. Am. Public Health Assoc.,
Washington. 1193 pp.
American Public Health Association, American Water Works Association, and
Water Pollution Control Federation. 1971. Standard Methods for the
Examination of Water and Wastewater. 13th ed. Am. Public Health Assoc.,
Washington. 874 pp.
Anderson, R.S., and L.G. Raasveldt. 1974. Gammapus and ChaoboTUS Predation.
Can. Wildl. Serv. Occas. Pap. No. 18. 24 pp.
Anderson, P.D., and L.J. Weber. 1977. The Toxicity to Aquatic Populations of
Mixtures Containing Certain Heavy Metals. Proc. Int. Conf. on Heavy
Metals in the Environ. Toronto, Ontario, Canada 2:933-953.
Balon, E.K. 1975. Terminology of Intervals in Fish Development. J. Fish.
Res. Board Can. 32:1663-1670.
Benoit, D.A. 1974, Artificial Laboratory Spawning Substrate for Brook Trout
(Salvelinus fonti,nalis Mitchill). Trans. Am. Fish Soc. 103:144-145.
Bliss, C.I. 1939. The Toxicity of Poisons Applied Jointly. Ann. Appl. Biol.
26:585-615.
Bridges, W.R. 1958. Sodium Cyanide as a Fish Poison, U.S. Fish Wildl. Serv.,
Spec. Sci. Rep. Fish. No. 253. 11 pp.
Broderius, S.J. 1970. Determination of Molecular Hydrocyanic Acid in Water
and Studies of the Chemistry and Toxicity to Fish of the Nickelocyanide
Complex. M.S. Thesis, Oregon State University, Corvallis, Oregon. 93 pp.
Broderius,, S.J., L.L. Smith, Jr., and D.T. Lind. 1977. Relative Toxicity of
Free Cyanide and Dissolved Sulfide Forms to the Fathead Minnow (Pi-me-
phales promelas Rafinesque). J. Fish. Res. Board Can. 34:2323-2332.
110
-------
Broderius, S.J. 1978. (Unpublished data). University of Minnesota, St. Paul,
Minnesota.
Brungs, W.A., and D.I. Mount. 1970. A Water Delivery System for Small Fish
Holding Tanks. Trans. Am. Fish. Soc. 99:799-802.
Dixon, W.J. 1973. BMD Biomedical Computer Programs. 3d ed. University of
California Press, Berkeley, Calif. 733 pp.
Dobie, J., and Moyle, J.B. 1962. Methods Used for Investigating Productivity
of Fish-rearing Ponds in Minnesota. Minn. Dep. Conserv. Fish. Res. Unit,
Spec. Publ. No. 2. 62 pp.
Doudoroff, P. 1976. Toxicity to Fish of Cyanides and Related Compounds. A
Review. Ecol. Res. Ser. EPA-600/3-76-038. U.S. Environmental Protection
Agency, Cincinnati, Ohio. 155 pp.
Doudoroff, P., G. Leduc, and C.R. Schneider. 1966. Acute Toxicity to Fish of
Solutions Containing Complex Metal Cyanides in Relation to Concentrations
of Molecular Hydrocyanic Acid. Trans. Am. Fish. Soc. 95:6-22.
Doudoroff, P., and M. Katz. 1950. Critical Review of Literature on the Toxi-
city of Industrial Wastes and Their Components to Fish. - I. Alkalies,
Acids and Inorganic Gases. Sewage Ind. Wastes 22:1432-1458.
Emerson, K., R.C. Russo, R.E. Lund, and R.V. Thurston. 1975. Aqueous Ammonia
Equilibrium Calculations: Effect of pH and Temperature. J. Fish. Res.
Board Can. 32:2379-2382.
Finney, D.J. 1971. Probit Analysis. 3d ed. Cambridge University Press,
Cambridge. 333 pp.
Cast, M.H., and W.A. Brungs. 1973. A Procedure for Separating Eggs of the
Fathead Minnow. Prog. Fish-Cult. 35:54.
Hynes, H.B.N. 1954. The Ecology of Gammarus duebeni, Lilljeborg and its
occurrence in freshwater in western Britain. J. Anim. Ecol. 23:38-84.
Izatt, R.M., J.J. Christenson, R.T. Pack, and R. Bench. 1962. Thermodynamics
of Metal-cyanide Coordination. I. pK, AH° and AS0 Values as a Function
of Temperature for Hydrocyanic Acid Dissociation in Aqueous Solution.
Inor. Chem. 1:828-831.
Jensen, A.L. 1971. The Effect of Increased Mortality on the Young in a Popu-
lation of Brook Trout, a Theoretical Analysis. Trans. Am. Fish. Soc.
100:456-459.
Kania, H.J., and J.O. O'Hara. 1974. Behavioral Alterations in a Simple
Predator-prey System due to Sublethal Exposure to Mercury. Trans. Am.
Fish. Soc. 103:134-136.
Ill
-------
Larson, G.L, and D.A. Schlesinger„ 1977. Effects of Short-term Exposures to
Total Residual Chlorine on the Survival and Behavior of Largemouth Bass
(Miavopteipus salmoides). Pages 55-70 in R.A. Tubb, ed. Recent Advances
in Fish Toxicology, a Symposium. Ecol. Res. Ser. EPA-600/3-77-085. U.S.
Environmental Protection Agency, Cincinnati, Ohio.
Leduc, G. 1966. Some Physiological and Biochemical Responses of Fish to
Chronic Poisoning by Cyanide. Ph.D. Thesis, Oregon State University,
Corvallis, Oregon. 146 pp.
Litchfield, J.T., Jr., and F. Wilcoxon. 1949. A Simplified Method of Evalu-
ating Dose-effect Experiments. J. Pharmacol. Exp. Ther. 96:99-113.
Lloyd, R., and D.J. Swift. 1976. Some Physiological Responses by Freshwater
Fish to Low Dissolved Oxygen, High Carbon Dioxide, Ammonia and Phenol with
Particular Reference to Water Balance. Pages 47-69 in A.P.M. Lockwood,
ed. Effects of Pollutants on Aquatic Organisms. Society for Experimental
Biology Seminar Series, Vol. 2. Cambridge University Press, Great Britain.
Marking, L.L. 1977. Method for Assessing Additive Toxicity of Chemical
Mixtures. Pages 99-108 in F.L. Mayer, and J.L. Hamelink, eds. Aquatic
Toxicology and Hazard Evaluation. ASTM STP 634. American Society for
Testing and Materials, Philadelphia.
Martin, J.W. 1967. A Method of Measuring Lengths of Juvenile Salmon From
Photographs. Prog. Fish-Cult. 29:238.
Mount, D.I. 1968. Chronic Toxicity of Copper to Fathead Minnows (Pimephales
promelas Rafinesque). Water Res. 2:215-223.
Mount, D.I., and R.E. Warner. 1965. A Serial-dilution Apparatus for Contin-
uous Delivery of Various Concentrations of Materials in Water. U.S.
Public Health Serv. Publ. 999-WP-23. 16 pp.
Nebeker, A.V., and F.A. Puglisi. 1974. Effects of Polychlorinated Biphenyls
(PCB's) on Survival and Reproduction of Daphnia^ Ganmarus and Tanytco?sus.
Trans. Am. Fish. Soc. 103:722-728.
Neil, J.H. 1957. Some Effects of Potassium Cyanide on Speckled Trout
(.Salvelinus fontinalis") . Paper presented at Fourth Ontario Industrial
Waste Conference. Water and Pollution Advisory Committee, Ontario Water
Resources Commission, Toronto, Canada.
Oseid, D.M. 1977. Control of Fungus Growth on Fish Eggs by Asellus militaris
and Gammarus pseudoZimnaeus. Trans. Am. Fish. Soc. 106:192-195.
Oseid, D.M., and L.L. Smith, Jr. 1974. Factors Influencing Acute Toxicity
Estimates of Hydrogen Sulfide to Freshwater Invertebrates. Water Res.
8:739-746.
Pennak, R.W. 1953. Freshwater Invertebrates of the United States. The Ronald
Press, New York. 769 pp.
112
-------
Plackett, R.L., and P.S. Hewlett. 1948. Statistical Aspects of the Indepen-
dent Joint Action of Poisons, Particularly Insecticides. I. The Toxicity
of a Mixture of Poisons. Ann. Apppl. Biol. 35:347-358.
Plackettj R.L., and P.S. Hewlett, 1952. Quantal Responses to Mixtures of
Poisons. J. Roy. Stat. Soc., Series B (London), 14:141-154.
Sprague, J.B. 1969. Measurement of Pollutant Toxicity to Fish. I. Bioassay
Methods for Acute Toxicity. Water Res. 3:793-821.
Sprague, J.B. 1970. Measurement of Pollutant Toxicity to Fish. II. Utilizing
and Applying Bioassay Results. Water Res. 4:3-32.
Steel, R.G.D., and J.H. Torrie. 1960. Principles and Procedures of Statis-
tics. McGraw-Hill, New York. 481 pp.
Stumm, W., and J.J. Morgan. 1970. Aquatic Chemistry. An Introduction
Emphasizing Chemical Equilibria in Natural Waters. John Wiley and Sons,
Inc., New York. 583 pp.
Tagatz, M.E. 1976. Effect of "Mirex" on Predator-prey Interaction in an
Experimental Ecosystem. Trans. Am. Fish. Soc. 105:546-549.
Waller, W.T., M.L. Dahlberg, R.E. Sparks, and J. Cairns, Jr. 1971. A Com-
puter Simulation of the Effects of Superimposed Mortality Due to Pollutants
on Populations of Fathead Minnows '(Pimephales promelas') . J. Fish, Res.
Board Can. 28:1107-1112.
Williams, W.D. 1970. A Revision of North American Epigean Species of AselZus
(Crustacea: Isopoda). Smithson. Contrib. Zool. 49:1-80.
Wuhrmann, K, and H. Woker. 1948. Beitrage zur Toxicologie der Fishche. II.
Experimentelle Untersuchungen liber die Ammoniak- und Blausaurevergiftung.
[Contributions to Fish Toxicology. II. Experimental Investigations of
Ammonia and Hydrocyanic Acid Poisoning]. Schwiz. Z. Hydrol. 11:210-244.
113
-------
SECTION VIII
PUBLICATIONS
Lind, D.T., L.L. Smith, Jr., and S. J. Broderius. 1977. Chronic Effects of
Hydrogen Cyanide on the Fathead Minnow (P-imephales ppomelas). J. Water
Pollut. Control Fed. 49(2):262-268.
Koenst, W.M., L.L. Smith, Jr., and S.J. Broderius. 1977. Effect of Chronic
Exposure of Brook Trout to Sublethal Concentrations of Hydrogen Cyanide.
Environ. Sci. Technol. 11(9):883-887.
Broderius, S.J., and L.L. Smith, Jr. 1977. Relationship Between pH and Acute
Toxicity of Free Cyanide and Dissolved Sulfide Forms to the Fathead
Minnow. Pages 88-117 -in Richard A. Tubb, ed. Recent Advances in Fish
Toxicology, A Sympo'sium. Ecol. Res. Ser. EPA-600/3-77-085. Corvallis
Environ. Res. Lab., U.S. Environ. Prot. Agency, Corvallis, Oregon.
Broderius, S.J., L.L. Smith, Jr., and D.T. Land. 1977. Relative Toxicity of
Free Cyanide and Dissolved Sulfide Forms to the Fathead Minnow (Pimephales
promelas). J. Fish. Res. Board Can. 34(12):2323-2332.
Kimball, G.L., L.L. Smith, Jr., and S.J. Broderius. 1978. Chronic Toxicity of
Hydrogen Cyanide to the Bluegill. Trans. Am. Fish. Soc. 107(2):341-345.
Smith, L.L., Jr., S.J. Broderius, D.M, Oseid, G.L. Kimball, and W.M, Koenst.
1978. Acute toxicity of hydrogen cyanide to freshwater fishes. Arch.
Environ. Contain. Toxicol. 7(3) : 325-337 .
Oseid, D.M. 1978. A Comparison of the Variability of Asellus aomrnunis
(Crustacea: Isopoda) and Gammarus pseudol-irmaeus (Crustacea: Amphipoda)
and Suitability for Joint Bioassays. Bull. Environ. Contain. Toxicol.
20(5): (In press).
Oseid, D.M., and L.L. Smith, Jr. 1979. The Effects of Hydrogen Cyanide on
Asetlus communis and GarnnaPus paeudol-imnaeus and Changes in Their Compet-
itive Response When Exposed Simultaneously. Bull. Environ. Contam.
Toxicol. (Scheduled for Volume 21, Issue 4, March 1979).
Broderius, S.J., and L.L. Smith, Jr. Lethal and Sublethal Effects of Binary
Mixtures of Cyanide and Hexavalent Chromium, Zinc, or Ammonia to the
Fathead Minnow (P-imephales pvome1as~) and Rainbow Trout (Salmo
J. Fish. Res. Board Can. (Submitted for Review 5/24/78).
114
-------
Plackett, R.L., and P.S. Hewlett. 1948. Statistical Aspects of the Indepen-
dent Joint Action of Poisons, Particularly Insecticides. I. The Toxicity
of a Mixture of Poisons. Ann. Apppl. Biol. 35:347-358.
Plackett, R.L., and P.S. Hewlett, 1952. Quantal Responses to Mixtures of
Poisons. J. Roy. Stat. Soc., Series B (London), 14:141-154.
Sprague, J.B. 1969. Measurement of Pollutant Toxicity to Fish. I. Bioassay
Methods for Acute Toxicity. Water Res. 3:793-821.
Sprague, J.B. 1970. Measurement of Pollutant Toxicity to Fish. II. Utilizing
and Applying Bioassay Results. Water Res. 4:3-32.
Steel, R.G.D., and J.H. Torrie. 1960. Principles and Procedures of Statis-
tics. McGraw-Hill, New York. 481 pp.
Stumm, W., and J.J. Morgan. 1970. Aquatic Chemistry. An Introduction
Emphasizing Chemical Equilibria in Natural Waters. John Wiley and Sons,
Inc., New York. 583 pp.
Tagatz, M.E. 1976. Effect of "Mirex" on Predator-prey Interaction in an
Experimental Ecosystem. Trans. Am. Fish. Soc. 105:546-549.
Waller, W.T., M.L. Dahlberg, R.E. Sparks, and J. Cairns, Jr. 1971. A Com-
puter Simulation of the Effects of Superimposed Mortality Due to Pollutants
on Populations of Fathead Minnows *(Pimephales promelas') . J. Fish. Res.
Board Can. 28:1107-1112.
Williams, W.D. 1970. A Revision of North American Epigean Species of Asellus
(Crustacea: Isopoda). Smithson. Contrib. Zool. 49:1-80.
Wuhrmann, K, and H. Woker. 1948. Beitr^ge zur Toxicologie der Fishche. II.
Experimentelle Untersuchungen iiber die Ammoniak- und Blausaurevergiftung.
[Contributions to Fish Toxicology. II. Experimental Investigations of
Ammonia and Hydrocyanic Acid Poisoning]. Schwiz. Z. Hydrol. 11:210-244.
113
-------
SECTION VIII
PUBLICATIONS
Lind, D.T., L.L. Smith, Jr., and S. J. Broderius. 1977. Chronic Effects of
Hydrogen Cyanide on the Fathead Minnow (P-imephalss ppomelas). J. Water
Pollut. Control Fed. 49(2):262-268.
Koenst, W.M., L.L. Smith, Jr., and S.J. Broderius. 1977. Effect of Chronic
Exposure of Brook Trout to Sublethal Concentrations of Hydrogen Cyanide.
Environ. Sci. Technol. 11(9):883-887.
Broderius, S.J., and L.L. Smith, Jr. 1977. Relationship Between pH and Acute
Toxicity of Free Cyanide and Dissolved Sulfide Forms to the Fathead
Minnow. Pages 88-117 -in Richard A. Tubb, ed. Recent Advances in Fish
Toxicology, A Sympo'sium. Ecol. Res. Ser. EPA-600/3-77-085. Corvallis
Environ. Res. Lab., U.S. Environ. Prot. Agency, Corvallis, Oregon.
Broderius, S.J., L.L. Smith, Jr., and D.T. Lind. 1977. Relative Toxicity of
Free Cyanide and Dissolved Sulfide Forms to the Fathead Minnow (P-imep'ha.les
promelas). J. Fish. Res. Board Can. 34(12):2323-2332.
Kimball, G.L., L.L. Smith, Jr., and S.J. Broderius. 1978. Chronic Toxicity of
Hydrogen Cyanide to the Bluegill. Trans. Am. Fish. Soc. 107(2):341-345.
Smith, L.L., Jr., S.J. Broderius, D.M. Oseid, G.L. Kimball, and W.M. Koenst.
1978. Acute toxicity of hydrogen cyanide to freshwater fishes. Arch.
Environ. Contam. Toxicol'. 7(3) : 325-337 .
Oseid, D.M. 1978. A Comparison of the Variability of Asetlus Gomrnunis
(Crustacea: Isopoda) and GarmaPus pseudol-inmaeus (Crustacea: Amphipoda)
and Suitability for Joint Bioassays. Bull. Environ. Contam. Toxicol.
20(5): (In press).
Oseid, D.M., and L.L. Smith, Jr. 1979. The Effects of Hydrogen Cyanide on
Aseltus cormunis and Ganmavus pseudol-Lmnaeus and Changes in Their Compet-
itive Response When Exposed Simultaneously. Bull. Environ. Contam.
Toxicol. (Scheduled for Volume 21, Issue 4, March 1979).
Broderius, S.J., and L.L. Smith, Jr. Lethal and Sublethal Effects of Binary
Mixtures of Cyanide and Hexavalent Chromium, Zinc, or Ammonia to the
Fathead Minnow (Pimephales promelas') and Rainbow Trout (Salmo
J. Fish. Res. Board Can. (Submitted for Review 5/24/78).
114
-------
TECHNICAL REPORT bATA
{Please read Instructions on the reverse before completing)
2.
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
January 1979 issuing date
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1BA608
11. CONTRACT /GRANT NO.
Grant No. R802914
13. TYPE OF REPORT AND PERIOD COVERED
Final, 11/1973 - 9/1978
14. SPONSORING AGENCY CODE
EPA/600/03
1. RhPORT NO.
EPA-600/3-79-009
4. TITLE AND SUBTITLE
Acute and Chronic Toxicity of HCN to Fish and
Invertebrates
7. AUTHOR(S)
Smith, L. L., Jr., S. J. Broderius, D. M. Oseid,
G. L. Kimball, W. M. Koenst, and D. T. Lind
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Dept. of Entomology, Fisheries, and Wildlife
University of Minnesota
219 Hodson Hall
St. Paul, MN 55108
12. SPONSORING AGENCY NAME AMD ADDRESS
Environmental Research Laboratory — Duluth, MN
Office of Research and Development
U.S. Environmental Protection Agency
Duluth, Minnesota 55804
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Acute and chronic toxicity of hydrogen cyanide to seven fish species and two invertebrates was
determined in dynamic flaw-through bioassays. The 96-hr LC50 concentrations of HCN for juvenile fish
ranged from 57 ^g/1 for rainbow trout at 10°C to 191 pg/1 for field stock Fathead minnows at 15°C.
The fry and juvenile fish were similar in their sensitivity to HCN with eggs the most tolerant stage.
The 96-hr 1X50 concentrations for invertebrates ranged from 176 ug/1 for (kuarurus to 2328 jjg/1 for
Asellus at 18°C. Long-term tests conducted with fathead minnows demonstrated that the concentrations
of HCN having no adverse effect on egg production was between 12,9 and 19.6 ug/1. Chronic tests with
brook trout demonstrated that on the basis of spawning success the maximum acceptable toxicant concen-
tration (HATC) was between 5.7 and 11.2 pg/1 HCN. Long-terra tests with bluegills showed that no
spawning occurred at HCN concentrations as low as 5.2 vig/1. Chronic experiments with invertebrates
demonstrated that the highest concentration of HCN having no adverse effect was between 29 and 40 yg/1
for Asellus and between 16 and 21 yg/1 for Gaimarue. Experiments conducted with intermittent and
sublethal diurnal exposure regimes demonstrated that adverse effects of early growth of the fathead
mlnnuK are lessened as the exposure period is reduced. Using the toxic unit approach, it also was
demonstrated that the Zn-HCN and nmmonia-HCN mixtures were more acutely toxic and the Cr-HClI mixture
less toxic than what would be predicted from simply additive interaction. This report covers a period
from 'Jovember 1, 1973 to September 30, 1978.
I17-
fa.
DESCRIPTORS
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN ENOED TERMS C. COSATI Field/Group
Freshwater fishes
Aquatic animals
Toxicity
Cyanide
Hydrogen cyanide
Behavior
Temperature
18- OiSTRIBUTION STATEMENT
RELEASE TO PUBLIC
Dissolved oxygen
pH
Growth
Reproduction
Bioassay
Predation
Exposure time
Binary mixtures
Intermittent exposures
Diurnal fluctuations
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
20 SECURITY CLASS (This pagef
UNCLASSIFIED
06C
06F
06T
21,
------- |