EPA-R3-73-OHb               ECOLOGICAL RESEARCH SERIES
FEBRUARY 1973
   Effects  of Chemical  Variations
   in  Aquatic Environments
   Vol. II
   Toxic  Effects of
   Aqueous Aluminum to Rainbow Trout
                        Office of Research and Monitoring
                      U.S. Environmental Protection Agency
                               Washington, D.C. 20460

-------
            RESEARCH REPORTING SERIES
Research reports of the  office  of  Research  and
Monitoring,  Environmental Protection Agency, have
been grouped into five series.  These  five  broad
categories  were established to facilitate further
development  and  application   of   environmental
technology.   Elimination  of traditional grouping
was  consciously  planned  to  foster   technology
transfer   and  a  maximum  interface  in  related
fields.  The five series are:

   1.  Environmental Health Effects Research
   2.  Environmental Protection Technology
   3.  Ecological Research
   4.  Environmental Monitoring
   5.  Socioeconomic Environmental Studies

This report has been assigned  to  the  ECOLOGICAL
RESEARCH  series.   This series describes research
on the effects of pollution on humans,  plant  and
animal   species,  and  materials.   Problems  are
assessed   for   their   long-   and    short-term
influences.    Investigations  include  format!en,
transport, and pathway studies  to  determine  the
fate  of  pollutants and their effects.  This work
provides the technical basis for setting standards
to  minimize   undesirable   changes   in   living
organisms   in   the   aquatic,   terrestrial  and
atmospheric environments.

-------
                                                 EPA-R3-73-011b
                                                 February  1973
EFFECTS  OF CHEMICAL VARIATIONS IN AQUATIC ENVIRONMENTS:

                        Volume II

  Toxic  Effects of Aqueous Aluminum  to  Rainbow Trout
                            By

                     W.  Harry Everhart
                     Robert A. Freeman
      Colorado State University, Fort Collins, CO

                     Project 18050 DYC
                      Project Officer

                    J. Howard McCormick
            National Water Quality  Laboratory
                    6201 Congdon Blvd.
                 Duluth, Minnesota  55804
                       Prepared  for

            OFFICE  OF RESEARCH  AND  MONITORING
          U.S. ENVIRONMENTAL PROTECTION AGENCY
                 WASHINGTON, D.C. 20460
 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402
             Price 75 cents domestic postpaid or SO cents GPO Bookstore

-------
                          EPA Review Notice
This report has been reviewed by the Environmental Protection
Agency and approved for publication.  Approval does not signify
that the contents necessarily reflect the views and policies of
the Environmental Protection Agency, nor does mention of trade
names or commercial products constitute endorsement or recommenda-
tion for use.
                                 ii

-------
                             ABSTRACT
Fertilized eggs, fry, and fingerlings were exposed to aqueous aluminum
complexes in neutral and basic media under constantly flowing, controlled
conditions of aluminum concentration, pH, and temperature.  Toxicities of
various concentrations were highly pH dependent.  Dissolved concentrations
over 1.5 ppm aluminum caused physiological and behavioral aberrations as
well as acute mortality.  Toxic effects of suspended aluminum, though
greater at lower concentrations, do not increase as much as the effects of
dissolved aluminum with higher concentrations.  Growth of trout exposed
to high dosages of aluminum was reduced only as long as or slightly longer
than the exposure continued.

Egg and fry bioassays were conducted with exposures in trays and simulated
natural redds.  Fertilization was not affected by any concentrations
tested, and most mortalities occurred during hatching and in the post-
swim-up stage.  Trends in toxicity were similar to those found with finger-
lings indicating dissolved aluminum to be more toxic than equivalent
suspended amounts.

This report was submitted in fulfillment of Grant Number WP-01398-01,
under  (partial) sponsorship of the Water Quality Office, Environmental
Protection Agency.
                                iii

-------
                              CONTENTS
Section




I




II




III




IV




V




VI




VII




VIII
Conclusions




Recommendations




Introduction




Methods




Results




Acknowledgment s




References




Publications
 1




 3




 5




 9



21




37




39



hi

-------
                               FIGURES
1.  Solubility of aluminum hydroxide complexes (Modified from
    Hem, 1967).
                                                                Page
2.  Schematic diagram of experimental apparatus.                 10

3.  Arrangement of artificial redds (A,A) and tray divided       18
    into two sections (B,B) in troughs.

k.  Growth of fingerlings associated with aluminum exposures     23
    at pH 8.0.

5.  Growth of fingerlings associated with aluminum exposures     25
    at pH 8.5 and 9-0.

6.  Growth of fingerlings associated with aluminum exposures     28
   . at pH 7.0.

7.  Growth rates of fingerlings recovering from exposure to      31
    5.2 ppm Al at pH 8.5 and 9-0.

8.  Growth rates of fingerlings recovering from exposure to      32
    9-52 and 5.2 ppm Al at pH 7-0.
                                  vi

-------
                               TABLES
Mo.


1.  Summary of water supply analyses.                            15

2.  Summary of recovery growth rates                             30

3.  Comparison of growth rates, H: Slope     = Slope             33
                                        X SS U
k.  Fertilization and early life history with controls and       35
    treatments at pH 9-0.

5.  Fertilization and early life history with controls at        36
    pH 8.0 (troughs 1 and 2), 0-52 ppm Al at pH 7.0 (troughs
    3 and U), 5.2 ppm Al at 7.0 (troughs 5 and 6), and 5.2
    ppm Al at pH 8.0 (troughs 7 and 8).
                                  VII

-------
                             SECTION I

                            CONCLUSIONS

1.  Dissolved and suspended forms of aqueous aluminum are toxic to
rainbow trout,

2.  Debilitated condition of fish after only a week's exposure to 5.2
ppm aluminum at any pH tested would have made it difficult for them to
withstand a flow rate characteristic of a trout stream or river during
such activities as migration, feeding, or escaping from predators.

3.  Extremely acute mortalities occurred at 5.2 ppm dissolved aluminum
while mortality rates were more chronic with equivalent suspended amounts.
Effects on activity and coloration were more pronounced in dissolved con-
centrations.  Both forms advanced gill hyperplasia to the point of near
total fusion of filaments.

4.  At 0.52 ppm the occurrence of first symptoms of distress did not ap-
pear until after three to six weeks of exposure.  Eventual mortalities
were much higher in number at pH 7.0  (44%)  than at 8.0  (8%) where all
aluminum was soluble.

5.  At 0.05 ppm of dissolved aluminum, growth was excellent and behavior
normal.  It is evident that  this concentration can be tolerated without
any obvious effects on growth or behavior.  This amount represents sat-
uration at pH  7.0 and is  totally soluble at all higher pH's.

6.  Mortality rates at 5.2 ppm at pH  7, 0.52 ppm at pH 7, and 5.2 ppm at
pH 8 were all similar indicating toxic effects begin at lower concentra-
tions of suspended aluminum  than dissolved  aluminum.  The concentration
dependency of suspended aluminum toxicity is of very low order while sol-
uble toxicities are strongly concentration  dependent.

7.  Loss of appetite, greatly decreased activity and gill hyperplasia
initiate sooner at 5.2 ppm than 0.52  ppm (pH 7 and 8), but after six weeks
there was little difference  between the fish of the two groups.

8.  Neither brief nor extended contact with lethal amounts of both soluble
and/or insoluble aluminum produces irreversible physiological damage as
far as growth is concerned to young trout surviving the exposure.

Fertilization and Early Life History

9.  Comparison of mortalities from egg lots fertilized in the aluminum con-
centrations and in the control water  did not show any significant advantages
or disadvantages to either.

-------
10.  At pH 9.0 only the 5.2 ppm aluminum concentration demonstrated a
higher mortality than the control.

11,  Aluminum concentration of 5.2 ppm at pH 8.0 demonstrated the highest
overall mortality.  Aluminum at 5.2 ppm and pH 7.0 was next toxic.
Twenty-nine percent of the pH 8.0 mortality occurred at hatching when
young rainbows died just as the embryo was breaking through the shell.
At pH 7.0 eight percent of the mortality occurred just as the head of the
embryo emerged from the egg shell.

-------
                            SECTION II

                         RECOMMENDATIONS

General

Basic chemical research is needed to produce more accurate methods of
chemical analysis in natural waters.  Much is lost with instruments
capable of precise determinations if we don't know the form of the chem-
ical or if natural complexes interfere with analysis.

The problem of recommending safe concentrations for aluminum is much more
difficult than for most other metals.  Methods of determining aluminum
in solution are commonly insensitive to colloidal precipitates.  The
aluminum methods proposed by Standard Methods are additionally somewhat
insensitive to polymeric aluminum hydroxide ions.  Atomic absorption
methods are less sensitive to aluminum than other common metals.

In the practical sense, the enforcement of regulations regarding permis-
sible aluminum concentrations will be very different since toxic effects
occur at concentrations that will be difficult to analyze in unknown
waters where interference also adds to detection problems.  It is certain,
however, that the old rule of permitting 1/10 of the 96 hour TLcQ concen-
tration will not hold.  Under some conditions 1.5 ppm aluminum would be
more toxic than 5 ppm under other conditions.  The effect of complexation
by ligands other than hydroxide  is, as yet, another total unknown.

If hydroxide complexed aluminum  is present only as anionic and neutral or
near neutral precipitated forms, a condition that should hold for most
natural waters with pH greater than 5.5, tolerable concentrations of
either form probably should not  exceed 0.10 ppm if rainbow trout are to
survive and grow normally.

-------
                            SECTION III

                           INTRODUCTION

A real need in the management of our water resources is basic informa-
tion on which to base recommendations for establishing water standards
or criteria for pollutants.  The importance of our sport and commercial
fisheries emphasizes the need for detailed studies of the effects of
toxicants on the fish populations.  Among the objectives of this grant
were to investigate the toxicity to rainbow trout (Salmo gairdneri) of
soluble anionic species and neutral precipitates of aluminum hydroxide
and lead nitrate complexes under known conditions of pH and concentra-
tion.  The aluminum studies were divided into effects of aluminum on
fingerling rainbows and effects of aluminum on fertilization and early
life history.

Aluminum is abundant in the earth's crust, but its natural occurrence
is limited to highly insoluble complex minerals.  As a result, the con-
centration of aluminum in natural waters is minute.  The importance of
aluminum as a pollutant has been reemphasized by the impending develop-
ment of oil shale mining in several Rocky Mountain States.  The
liberation of aluminum from Dawsonite  (NaAl(OH),. C0_, which is present in
some oil shale ore, is a biproduct of  some oil shale mining processes.

Behavior of aluminum in aqueous solution is extremely complex; it forms
a variety of complexes with water, hydroxide, fluoride, silicate, and
sulfate.  The free aluminum ion is rare occurring only in very minute
amounts.  Possible forms exhibit vastly different structures and ac-
tivities and will certainly be found to produce different effects on
living organisms.

The chemical nature of aluminum in dilute solution is controversial.  Slow
reaction rates, the formation of metastable solids and gels, colloidal and
subcolloidal suspensions, and marginal analytical techniques add to the
problems of determining water associated structures.  The reports of Hem
and Roberson (1967), Hem  (1968) and Roberson and Hem  (1969) provide lucid
explanations of the behavior of aluminum in natural waters.

Solubility of aluminum under conditions where hydroxide complexes dominate
(the vast majority of natural waters)  is, briefly, a direct function of pH,
decreasing toward 5.5 and increasing toward both extremes of the pH scale
(Figure 1).  Soluble forms in acidic environments are polymeric and cation-
ic while monomeric anions are present  in basic media.  The critical
importance of pH is exemplified by its control of complexation, polymeriza-
tion, hydrolysis, and solubility.

In acidic conditions, monomeric hydrolysis of the hexaquo complex is fol-
lowed by the joining of two such monomers to form a dimer, sharing the
two OH ions on a common edge and eliminating two water molecules (Hem, 1968).

-------
2.0
3.0
4.0
5.0
6.0
7.0
8.0
 9.0
      CATIONIC
      POLMERS
                              i	r
 NEUTRAL
 POLYMERIC
 PRECIPITATES
                    ANIONIC
                    MONOMERS
      • Test Conditions Investigated
                     J__	L
   3.0       4.0
5.0       6.0       7.0      8.0       9-0
       FIG. I   SOLUBILITY OF ALUMINUM HYDROXIDE
               COMPLEXES (Modified from Hem, 1967)

-------
This shared edge is the basic building linkage in the continuing pro-
cess of polymerization with no apparent limit on polymer size.  As
the polymer enlarges, chains and then six membered planar rings of
complexes are formed.  The ring systems become less soluble as the poly-
mers enlarge.  The basic complexes retain a monomeric configuration but
hydrolize at four sites to produce the anion, (Al (OH),(H_0)- .

Previous investigations into aluminum toxicity have shown little regard
for aluminum's complex chemistry, resulting in widely divergent results.
For example, Jones (1964) reported that 0.07 ppm was the limiting safe
concentration of aluminum to sticklebacks whereas Schaut (1939) stated
that as long as aluminum did not result in critically low pH's (via
hydrolysis) it should have no toxic effects on river fish.  Many studies
have utilized conditions which exceeded the solubility of aluminum
with no account of the precipitates.  Others have investigated conditions
where pH could have been a lethal factor.

Previous studies with fish include Thomas  (1915), Ebeling (1928), Oshima
(1931), Ellis  (1937), Sanborn (1945), Minkina (1946), Pulley  (1950),
Wallen et al  (1957), and Jones  (1964).
                                  T

-------
                             IV METHODS

The "bioassay apparatus was of constant flowing design.  Principal com-
ponents were eight troughs , three aluminum administering devices , a
hydroxide diluter, heating and cooling units, and a pressure regulating
reservoir (Figure 2).

Water for the system was tapped directly from a 10" municipal main.
All piping for the system was either rigid or flexible polyvinyl chloride
except for one section of steel pipe carrying high temperature water from
the heaters to a temperature control valve.  It was also necessary to use
brass gate valves in this section (See Table l) .  All other valves and
fittings were plastic or polyethylene.

Two heaters connected in series with 110,000 BTU/hr were used to heat
a portion of the water to about 60  C for eventual mixing to attain a
temperature of 13 C  (55 F) for the  experiments.  Mixing was achieved with
 a Powers UliO photopanel temperature  control valve.  To prevent supersat-
 uration of the heated water, the  input to the pressure regulating reservoir
 was  splashed vigorously and directly so as to keep the water swirling and
 liberate excess  gases.  When ambient water temperature exceeded 55 F, the
 entire flow was  passed through two 1 H.P. Min-o-cool refrigeration units.

 All  water was passed through two  activated carbon filters connected in
 series for dechlorination.  The piping of the dechlorinators was arranged
 to permit the removal and refilling  of one unit without interrupting flow
 to the system.

 The  pressure regulating reservoir was equipped with a 1 1/2" diameter
 standpipe which  was  constantly just  overflowing.  This assured a constant
 head to the system and constant flow through the hydroxide diluter and to
 the  troughs .

 At full scale operation,  the system  required about 22 liters per minute:
 Two  liters per minute per trough, a  small portion of which came from the
 hydroxide diluter  which required  3.5 liters per minute, and one liter
 per  minute diverted  from  between  the dechlorinators to a 3-gallon tank con-
 taining several  f ingerlings .  These  fish were used to monitor the perform-
 ance of the first  dechlorinator .  As long as there was no mortality in this
 tank, the first  dechlorinator was effectively removing chlorine and the
 second dechlorinator was  not being consumed.  The remainder of the 22 liters
 overflowed the standpipe  of the pressure regulating reservoir.

 The  aluminum stock solution was dosed to the system by three constant flow
 devices, one per set of duplicate conditions, described by Freeman (1971).

 A constant flow  was  metered into  the hydroxide diluter where it was mixed
 with filtered concentrated WaOH solution metered from a. 2-liter Mariotte

-------
                                                  TEMPERATURE
          A
        MAIN

CO 	 »-
HEATING

	 \
-1
&
— • —



^l

».

\\ 1 —
diib
\
CO

— w

vt





                                                                DECHLORINATORS
ALTERNATE
INPUTS
      PRESSURE
      REGULATING
      RESERVOIR
               EXPERIMENTAL
               TROUGHS
                                                                                                  ©
                                   REFRIGERATION UNITS
                                           \
HYDROXIDE DILUTER

 /                Al.
                                                                                      RECORDING
                                                                                      THERMOMETER
                       FIG. 2.  SCHEMATIC  DIAGRAM OF EXPERIMENTAL APPARATUS

-------
bottle.  The diluter was constructed of 6.35 mm (1/4") plexiglass,
forming a rectangular trough about 35 cm x 6.2 cm x 7.5 cm.  Standpipes
of soft glass tubing inserted in rubber stoppers were placed in appro-
priate holes in the bottom of the trough of the diluter.  The effluent
from each standpipe was collected in a glass funnel suspended below the
diluter and carried via rubber tubing to the appropriate trough.  Two
baffles at the head of the diluter induced turbulence to assure uniform
mixing of the concentrated NaOH.  The flow from the standpipes could be
regulated by diameter of the tube and depth of placement.  Enough water
was passed through the diluter to maintain a constant level in its trough
at all settings of the standpipes.  At times, small plexiglass filter
boxes packed with glass wool were placed over the smaller standpipes to
filter particles of iron hydroxide precipitated from the water supply
(Table 1) by the high pH of the diluter.  This prevented adhesion of the
iron hydroxide to the inside surfaces of the standpipes and the resulting
disruption of desired constant flow rate.

The reagents were mixed and diluted in large orifice boxes at the head
of each set of duplicate troughs.  Each box was fitted with a perforated
vertical and sloped horizontal baffle to induce turbulence and insure
uniform dispersion of the reagents.  Holes on opposite sides of the
boxes delivered equal amounts of test solution to each of the duplicate
troughs.  Each box was calibrated to insure maintenance of correct dilu-
tion and flow rate.  The remainder of the 2 liters per minute per trough
was drawn directly from a manifold connected to the pressure regulating
reservoir.  Turnover time for the troughs was 69 minutes.

The fiberglass troughs measured 5' x 1' x 1'  (152.4 cm x 30.5 cm x 30.5
cm).  A mean depth of 29.7 cm was maintained by a standpipe at the end
of each  trough.  A perforated weir blocked the standpipe from the rest
of the trough, and 1/4" mesh screens were placed over the troughs to pre-
vent fish from jumping out.

A sensor of a Foxboro recording thermometer was placed midway in one
trough to monitor the temperature of the system.

A Corning Model 12 pH meter, graduated to 0.005 units, was used to
monitor pH and a Gallet 7 jewel stopwatch graduated to 1/10 second was
used to monitor the various flows of the system.

The NaOH was a purified flake from J. T. Baker and Co.  The aluminum for
the system was supplied by anhydrous aluminum chloride from Allied Chem-
ical Co.

This study was designed to examine the toxic effects of aluminum at dif-
ferent pH's and concentrations.  Concentrations chosen were significant
in terms of possible solubility within the range of pH to be investigated,
The range of conditions investigated and corresponding behavior of the
aluminum were:
                                 11

-------
Total Aluminum
Added

5.2 ppm

0 . 52 ppm

.052 ppm
PH
7.0
1% dis

99% sus
10% dis
90% sus
100% dis
PH
8.0
10% dis

90% sus
100% dis

100% dis
PH
8.5
32.% dis

68% sus
100% dis

100% dis
PH
9.0

100% dis

100% dis

100% dis
dis - dissolved
sus - suspended
The mean solubility product prescribed by Hem (1967), K^ =  1.95+ 1 x lO-1?,
was used to calculate the amounts of aluminum necessary to be added to
achieve desired conditions, as well as those forms present under observed
conditions.  The rapid turnover time of the experimental tanks plus the
movement of the fingerlings were sufficient to prevent significant settling
of the precipitates at supersaturated conditions.  When settling was ob-
served it occurred behind the perforated weirs at the lower end of the
troughs.

The procedure supplied by R. W. Smith and J. D. Hem  (M.S.) involving the
kinetic complexation of aluminum with ferron was employed to determine the
various forms in which the aluminum species exists in aqueous media.  The
three forms described by Smith and Hem are the monomeric form, Al , des-
cribed as Al  , (A1(OH)_  or Al(OH),-, the polynuclear form, Al , described
as a six-membered aluminum hydroxide ring, and the colloidal form Al ,
described as Al(OH) , aluminum hydroxide.
The only significant deviation from the procedure outlined by Smith and Hem
was that anhydrous aluminum chloride was used to prepare the aluminum
standard stock solution because of the difficulty encountered in preparing
the solution from aluminum metal.  This should not in any way affect the
results and may in this case actually be desirable since the source of
aluminum in the toxicity experiment was also anhydrous aluminum chloride.

Using this procedure on fresh solutions containing 5.2 ppm aluminum, the
curves similar in shape to those described by Smith and Hem were obtained.
The following data were then derived from the curves and the calibration
curve using the methods described by Smith and Hem.

In general, this method entails extrapolation of the flat portion of the
curve to zero time and estimating the amount of Al  + Al  from the absorb-
ance.  The amount of Al  is determined by the extrapolation of the curved
portion of the curve to zero time.  Al  is determined by subtraction of Al
+ Al  from the total aluminum in solution, in this case 5.2 ppm.
                                 12

-------
Initial pH             A1H              Alb           Al°         %A1 diss*

    6.0                1.1              2.2           1.9            21
    7.0                1.3              2.4           1.5            25
    8.0                1.8              2.4           1.0            35
    9.0                4.0              0.8           0.4            77

*% Al dissolved - Ala /Al  total X 100%
Although they do not agree absolutely, the values for the percent dissolved
aluminum as a function of pH follow the order of the calculated values.

Fluoride ion was periodically measured using a fluoride specific ion elec-
trode and was consistently found to be very near 1.0 ppm, the value the
City of Fort Collins claims to maintain.  The method did not work at 0.52
ppm.  Smith and Hem note that fluoride ion interferes with the analysis
at low aluminum concentrations.

An interesting experimental result was observed by allowing the ferron to
react completely with the aluminum species.  In every case, regardless of
pH, if the ferron-aluminum complex absorbance was measured 12-24 hours after
the beginning of the experiment, all  solutions were observed to have a final
absorbance reading equivalent to a concentration of 5.1;K1 ppm.  This
measurement is representative of the  total concentration of aluminum species
in solution and agrees with the atomic absorption spectroscopy measurements
in which the total aluminum concentration of solutions was found to be
5.0ppm+%.

Considering the difficulty in establishing steady state conditions in the
troughs and the slow approach to equilibrium of the aluminum distribution,
the agreement between calculated and  experimentally determined aluminum dis-
tributions is better than might be expected.  Either set of values illus-
trates that experimental conditions were chosen to span the range of
distribution from dissolved to suspended aluminum.

Aluminum chloride was dissolved in tap water in the reservoirs of the
constant flow devices.  These devices were stopped for about 10 minutes
for the refilling process.  The hydrolysis of A1C1., results in highly acidic
stock solutions.  About 20 ml of concentrated hydrochloric acid was added
per 200 liters of stock solution to insure rapid and total solution of the
aluminum chloride when the reservoirs were refilled.  Concentrations were
adjusted so that either 18, 19, or 20 ml per minute, depending on which
device was used, would supply the correct concentrations when diluted to
two liters in the headboxes.  Flows of these devices were monitored at
three or four times daily during the  experiment.  The mean and variance of
the concentrations of the experiments were calculated from these data.
This was possible since all factors influencing test concentrations, namely
concentrations of the stock solutions, delivery rates, and dilution rates
were known.
                                13

-------
Ambient pH of the input water plus the addition of various concentrations
of the acidic aluminum solutions required that a basic additive be intro-
duced to bring the test pH to the range desired and convert the cationic
aluminum to its anionic form.

The pH of the system was controlled by dilute NaOH solution from the
hydroxide diluter.  A concentrated solution of 9.8 N NaOH was delivered
at 1.4 ml per minute into 3.5 liters of water at the head of the diluter.
Various amounts of sodium hydroxide were diluted via the diluter standpipes
to bring pH's to test levels.  The pH of the system was recorded three or
four times daily and used to calculate the pH mean and variance of the
system.

Data were taken from the continuous trace of the recording thermometer
at three-four intervals to calculate mean and variance of temperature.

During the course of the tests, water samples were collected at about two
week intervals for analysis of dissolved solids, dissolved oxygen, alkalin-
ity, chloride, nitrate, silicate and sulfate (Table 1).  Determinations
were made by the procedures described in Standard Methods for the Examina-
tion of Water and Wastewater (1965).  Metallic cations were determined by
atomic absorption and emission spectroscopy.

Fingerlings

Limitations in laboratory size and  available water made  it necessary  to
raise the fingerlings in a water supply other  than that  used for  the
toxicity tests.  Trout were hatched and reared to test size in well water
and exposed to aluminum, after acclimation, in tap water  (Table 1).  All
trout used in these experiments were hatched in  the laboratory.   Eyed eggs,
spawned from infectious-pancreatic-necrosis-free brood  stock, were super-
ficially disinfected with Merthiolate and hatched in  tray incubators.  The
eggs were hatched and fry raised to test size  in well water.  All fish used
in the experiments had a recorded disease free history.  Minor conditions
of gill hyperplasia were prevalent  when fish were held in well water.   This
condition rapidly disappeared when  the trout were transferred to  tap  water
for acclimation.  Troughs holding the experimental subjects were  cleaned
daily.  The well water supplied 10' x 1' x  I1  (305 cm x  30.4cm) troughs
at mean depth of  27.2 cm at a rate  of 8 liters per minute.  Fry and
fingerlings were  fed at all  times according to a feeding chart modified
from Deuel et al  (1942).  Clark's pelletized commercial  trout chow was  the
food source.

Fry up to about a month old were fed  six times per day.   Older fry and
fingerlings both  before, during, and  after  testing were  fed four  times  per
day.  Growth of all  test animals was  recorded  at approximately two-week
intervals and feeding rates  adjusted  according to  growth.

Pretest weights were determined by  weighing a  sample  of  the pretest  finger-
lings.  Weights of  test animals during  the  test  and during  recovery  were
determined by weighing  the entire group.

-------
                           Table 1.   Summary of water supply analyses


 This water supply was used for the  aluminum toxicity determinations.  Concentrations are in mg/1,

Date
1/21/70


2/2/70


2/21/70
3/25/70

4/13/70



4/27/70
5/7/70

5/23/70

6/16/70


Dissolved
Oxygen
10.6


9.6


10.0
10.0

9.0



8.6
8.8

8.4

9.2


Total
Alkalinity3
40.3


40.9


40.9
46.2

50.6



51.0
39.3

17.1

13.6



£*£
7.65


7.50


7.60
7.72

7.75



7.62
7.44

6.85

6.90



Temp. (°C)a
5.0


4.4


3.3
3.4

7.2



8.3
11.7

12.5

12.9



Cl
2.65


7.75


3.50
0.75

2.50



—
1.50

3.00

—



NO., SO, S:00
•"•"•" j "^4 ^
0.25 2.88 12.75 ^ H


0.30 *** 11.40


0.13 *** 10.80
0.16 *** 10.53

0.20 *** 12,30



— — — — — ~
0.22 *** 12.20

0.23 *** 10.20

—
cu ro
en
00 rt
• CO
O
• pi
rt
oo H
LD CO
rt
&> CO
"•O P)
• rt
O
SB
H
en
rt
•^J CO
O P
rt
T)
CS
 — Not determined.
  a Before adjustment.
  * Less than 0.05 ppm.
 ** Less than 0.01 ppm.
*** Not detectable.

-------
                                            Continuation of Table 1
H
ON
Date
1/21/70
2/2/70
2/21/70
3/25/70
5/7/70
5/25/70
Ca (2)
16
20
10
11
22
22
Mg (10)
5.0
1.0
0.2
0.2
0.4
0.4
Na (10)a
2.9
1.0
2.0
1.1
4.5
4.5
K (1)
4.4
*
*
*
*
A
Al (l)a
A
*
*
*
A
A
Fe (10)  Zn (1)





 0.20     **


 0.20     **



 0.15     **



 0.45     **


 0.45     **
                                                                                                     CU (1)
                                                                                                       A



                                                                                                       A
                                                                                                       A



                                                                                                       A
        Parentheses  indicate confidence factors for data,  i.e.  the true value lies between that reported

        multiplied by or  divided  by the factor in parenthesis.

-------
Test animals were transferred without tempering from the well water to
the tap water for pretest acclimation.  Fish for all tests, except at
pH 7.0, were acclimated to tap water and pH for 8 days prior to intro-
duction to test conditions.  At pH 7.0 acclimation was limited to 3 days.
In all cases, fish adapted to the change in water without any noticeable
behavioral or physiological problems.  Mortalities during the acclimation
never exceeded 1%.  During this time the pH was raised at the rate of 0.2
units per day to the test pH.  At the end of the acclimation, healthy
trout were selected from this group and placed directly into the various
test conditions.

Behavior, growth, and mortality were recorded.  Controls were weighed
once or twice during the test to adjust feeding rates.  The rates for all
trout were then adjusted according to the growth of the controls rather
than weighing the test fish and possibly altering mortality rates.  Test
fish were fed the prescribed food amount or as much as they would consume
daily, whichever amount was smaller.

At the end of the exposures all groups were weighed.  Some recovery sub-
jects were transferred directly back to well water while others remained
for a time in dechlorinated tap water.  Growth of these fish was deter-
mined at about 10-day intervals by weighing the entire group.  From these
data the long term effects of aluminum on growth could be determined.

Time of death, weight of fish at death, and trough number for each mor-
tality were recorded.  From these data it was possible to determine the
time of mortality of 50% of the group.  Averages from the duplicates pro-
vided  the reported TL50 time.  Smirnov's statistic  (Birnbaum, 1962), a
non-parametric statistic used to compare two unknown distribution functions,
was used  to compare the empirical distribution functions of mortality
rates  between sets of duplicate conditions.  The suspected dependence of
survival  time on size was also investigated by a simple regression; however,
no significant relationship was discovered.

Fish were held prior to testing and for recovery in well water of the fol-
lowing analysis: D. 0., 5.2-5.8 ppm; total alkalinity, 301-335 ppm; pH,
7.2-7.5; temperature, 16.5-16.7 C; chloride, 11-20 ppm; nitrate 2.9-3.2
ppm; sulfate, 62-86 ppm; silicate 14-17 ppm; calcium, 200-220 ppm; sodium,
125 ppm; potassium, 0.4-1.0 ppm; aluminum, iron, zinc and copper all less
than 0.05 ppm.

Fertilization and very early life history

Eggs and sperm were collected from IPN certified free brood stock and fer-
tilized in control water and in the aluminum concentrations.  Fertilizing
in control and treatment waters gave a chance to observe the effects of
the aluminum on eggs, sperm, and the fertilized egg.  In a further effort
to approach natural conditions half of the eggs were reared in artificial
redds of gravel and the other half in trays   (Figure 3).  An effort was
                                 17

-------
FIG.  3.   ARRANGEMENT OF  ARTIFICIAL  REDDS  (A,A) AND TRAY
         DIVIDED INTO  TWO  SECTIONS (B, B)  IN  TROUGHS.
                           18

-------
made to deposit 200 eggs in each redd and in each tray.  However, there
was some variation in the actual number which was determined volumetrieally
to reduce any handling stress.  Concern over the hatching success in the
artificial redds dictated the use of trays.  Dead eggs could not be removed
from the eggs in the gravel and fungus could well have been the cause of
loss rather than the aluminum.  Eggs in the trays were examined daily and
dead eggs removed.

In the first run, controls and experimentals were all at pH 9-0 with
treatments of 0.052, 0.52, and 5-2 ppm aluminum.  Replicates of each were
provided.  Half of the eggs were fertilized directly in the treatment water
and the other half of the eggs were fertilized in the control water and
then transferred to the treatment tanks, but separated in redds and trays.

In the second run controls were carried at pH 8.0, and the treatment tanks
at pH T-0 with aluminum  concentration of 0.52 ppm, pH 7-0 and aluminum con-
centration of 5-2 ppm, and pH 8.0 and aluminum concentration of 5.2 ppm.
                                  19

-------
                             V RESULTS
Fingerlings

Aluminum Toxicity at pH 8.0
Duplicate concentrations of 0.05 ppm, 0.52 ppm, and 5.2 ppm aluminum
were tested.  Age of test fish was approximately 6 months at the start
of the test.  All exposures in this experiment were conducted for 45
days (1080 hours).

The control troughs received approximately.^. 39 mg NaOH per liter to
maintain a mean pH of 8.02  (S_ = 8.4 x 10  ).  No mortality occurred
among the control fish for  the duration of the test.  These fish ap-
peared healthy and active throughout the test.

At pH 8.0, 90% of the 5.2 ppm aluminum was insoluble.  The insoluble
aluminum hydroxide remained suspended in the troughs giving a slight
whitish irridescent cast to the water.  The coloration of slight accu-
mulations of settled precipitate behind the weir at the downstream end
of the troughs indicated that some iron was being coprecipitated by the
aluminum hydroxide.  The mean concentration of total aluminum added was
5.23 ppm (s  = 0.051).  The mean pH for the test was 8.02 (s  = 8.9 x
10  ), thus a mean concentration of 0.55 ppm dissolved aluminum was
maintained.

Within 24 hours an overall  slowdown in activity was observed in the 5.2
ppm troughs, and almost all interest in feeding had disappeared.  On no
day of the test was more than half of the prescribed amount of food con-
sumed.  After five days gill hyperplasia was evident in many of the trout
in both troughs.  By the sixth day, many fish were experiencing momentary
equilibrium problems or swimming on their backs breaking the surface,
increasing in severity until the fish completely lost equilibrium and
remained upside down for a  period of from 3 to 24 hours preceding death.
The coloration of these fish was very dark, many nearly black.  During
the second week of exposure, periods of increased activity and feeding
occurred for durations of up to one day.  After the fourteenth day, how-
ever, no increases in activity were ever observed.  Overall physical
condition steadily deteriorated with deaths occurring regularly.  Within
30 days these fish lost all negative photoaxis to sunlight and all fright
reaction to approach to the troughs by humans.  Late in the test, trout
could be prodded without attempting to flee.  Respiratory rates were con-
siderably higher than any other of the tests at pH 8.0.

After 45 days, weights of the 12 survivors averaged 30.2% of the control
weight as opposed to averaging 106.5% of the control weight at the start
of the test  (Figure 4).  Prior to release in well water, for recovery,
experimental fish were tempered to it for a half hour.  From this
                                  21

-------
additional  stress, however, 11 of the 12 died within 24 hours.  The one
survivor had returned to normal activity and feeding after six days.
                                             2           _3
A mean concentration of 0 .514 ppm aluminum  (s  = 5.8 x 10  ) was main-
tained at a pH of 8.03  (s2 = 7.2 x 10~ Representing total saturation
with soluble aluminum.  Under these conditions, fish were slightly less
active than the controls after a single day.  After three days, a definite
intermediate condition between the controls and the 0.52 ppm test existed.
By the sixth day, consumption of food was considerably reduced.  These fish
consumed their total prescribed diet on about one half of the test days,
feeding little if at all after 30 days.  Darker coloration was evident in
this test also, although not to the degree of the 5.2 ppm test.  After 10
days exposure, there was a noticeable decrease in the fright reaction of
the fish in these troughs.  Periodic fluctuations in overall activity
were observed throughout the test.  Gill hyperplasia was evident in about
half of the fish at 21 days.  At the conclusion of the test, low power
microscopic examination revealed that nearly all fish were showing hyper-
plasia to some degree.

Although only two deaths occurred very late in these tests, symptoms
preceding mortality were identical to those exhibited by fish in the higher
concentrations.  Although these fish showed a net weight increase the aver-
age increase was 38% of the control increase.  Their weights averaged 58.1%
of the control weight following the exposures as opposed to 95.9% of the
control weight at the start of the test (Figure 4).  Continued exposure
under these conditions would undoubtedly produce high mortalities.

At the conclusion of the test, five fish from each trough were placed
without tempering into well water for recovery.  Within 48 hours all symp-
toms of the exposure except the gill hyperplasia had disappeared.  In the
following 16 days, these fish showed extremely rapid weight increase
comparable  to the controls.  No additional mortalities occurred and at the
conclusion  of the recovery period the fish were normal in all appearances.

A set of duplicate concentrations were also run at a mean concentration of
0.0516 ppm  aluminum (s  « 6.0 x 10  ) at pH 8.04 (s  = 3.3 x 10  ).  These
conditions  required the addition of 2.55 mg NaOH per liter to maintain pH.
This amount of aluminum is the saturation point at pH 7.0 and is 100%
soluble at  all higher pH's.

These fish were healthy and active throughout the test, showing fast growth
and normal behavior at all times.  They averaged 101% of the control weight
after the test as opposed to 97.2% of the control weight at the start of
the test.   Further testing of this concentration was not conducted since
in the range of our investigations the form of this amount of aluminum
would be unchanged.

Aluminum Toxicity at pH 8.5

At pH 8.5,  about 3.19 mg NaOH per liter was necessary to maintain the control
pH, while the test troughs required about 27.0 mg per liter.  Fifty finger-
lings were  placed in each trough.  Of the.mean 5.14 ppm aluminum (s  = 8.8 x
  ) added at mean pH 8.48 (s  = 8.0 x 10  ) a mean of 1.57 ppm aluminum
                                 22

-------
UJ
      UJ
                       8/27 9/10 9/24 10/8 10/22  11/22
I/I
1/8
1/50
2/21
3/8
                                                        DAYS
                   FIG.  4.   GROWTH OF  FINGERLIIMGS  ASSOCIATED WITH ALUMINUM EXPOSURES
                            AT  pH 8.0.

-------
was in solution.  Trout used in these tests were about 6 weeks»old at
the beginning.  Mean temperature for these tests was 12.6 C (s  = 0.14)

On introduction to the test conditions, fish showed immediate stress,
darkening in coloration, and a total loss of interest in food.  Mortal-
ities began to occur almost immediately with all the symptoms previously
described.  In addition, fecal casts were obvious from all fish.  This
test was terminated after 222 hours (9 days 6 hours).  Aluminum addi-
tions were stopped and pH was allowed to return to ambient, 7.7.  Within
12 hours, about half of the fish had returned to normal coloration and
began taking small amounts of food.  After 72 hours, however, there was
still no fright reaction in these fish.  Overall normality seemed to have
returned after five days, except for persisting gill hyperplasia.  On
the 8th day these fish were transferred without tempering into well
water.  For the next 10 days they showed no weight increase.  After 18
days recovery, weighings indicated a near normal growth rate had resumed.
At this time the weights of these fish averaged 55.2% of the control
weight as opposed to averaging 106.9% of the control weight at the start
of the test (Figure 5).  After 49 days recovery they averaged 63.9% of
the control weight.  Mortalities during the first 60 days of recovery
were 3% as opposed to 5% among controls.  In the next 80 days there were
no additional mortalities.  After 140 days recovery, test fish averaged
82.3% of the control weight.

Aluminum Toxicity at pH 9.0

At pH= 9.0 approximately 5.07 mg NaOH per liter was added to maintain con-
trol pH while 35 mg per liter was required to maintain the test pH.  The
mean control pH,was 8.99 (s  = 5.5 x 10~ ).  The mean test pH was 8.99
(s? = 8.1 x 10 „) with a mean aluminum concentration of 5.2 mg per liter
(s  = 8.0 x 10  ).  Under these conditions the mean concentration of
soluble aluminum was 5.5 ppm (97.15%) and 0.15 ppm suspended (2.85%).
Six week old fingerlings were used for this test.  Mean temperature for
the test was 12.6°C (s  = 0.18).

These fish showed immediate shock and heavy mortality within 48 hours.
Fecal casts were wide spread and gill hyperplasia was universal within
three days.  No fish ever accepted food.  Coloration was very dark and ac-
tivity was minimal.  Test conditions were terminated after 113 hours (4
days, 17 hours) by stopping the aluminum addition and allowing pH to
return to ambient.  Within 36 hours some fish began feeding and about one-
half showed lightening of coloration.  No fright reaction to approach ap-
peared until after 5 days.  By the sixth day, however, activity and
feeding were near normal.  On the eighth day some fish showed an apparent
relapse as some test symptoms reappeared.  On the tenth day the entire
group was transferred to well water.

These fish also showed a loss in weight for the first two weeks of recovery
resuming normal growth after 30 days recovery.  Before testing, their
weights averaged 116.75% that of the controls, but this dropped to 50.0%
                                  2k

-------
ro
                                                 9.0 CONTROL
                                                 8.5 CONTROL
               FIG. 5.  GROWTH OF FINGERLINGS  ASSOCIATED WITH ALUMINUM EXPOSURES
                      AT  pH  8.5  AND 9.0.

-------
before climbing to 76.8% after 60 days recovery (Figure 5).   During the
first 140 days of recovery there were no mortalities among the recovering
fish.  After 140 days recovery, test fish averaged 85% of the control
weight.

Aluminum Toxicity at pH 7.0

A final series of tests was carried out at a nominal pH of 7.0  At the
beginning of these tests, the ambient input pH to the system was about
7.4.  Since the pH controlling mechanism is suitable only for controlling
pH's by either raising or lowering at a given time, not both simultaneously,
it was decided to run the control and 0.52 ppm test without hydroxide
addition and control only the 5.2 ppm test.  (The addition of 0.52 ppm alum-
inum did not reduce the pH of the system enough to permit control by
hydroxide addition.  At 5.2 ppm approximately 13.4 mg NaOH was added per
liter for pH control.  Solubility of aluminum in the pH range from 6.5 -
7.0 is practically constant so the absolute control of pH in this range
is not nearly as critical as at a higher pH.

Complications were encountered when  the water flow to  the entire system
became reduced for some unknown reason.  As a result,  the flow to the
control troughs was reduced to 1.4 liters per minute for  the test.  All
flow rates were in excess of that needed to support the  quantities of fish
used.  No stress reactions were observed among  the controls as a result of
this adjustment.

On  the seventh day of these tests, hydrated aluminum sulfate was introduced
to  the municipal water supply  to precipitate  the  increased turbidity due
to  heavy snow melt run off in  the Cache la Poudre River.  These additions
were continued for the duration of  the  test.  According  to treatment plant
personnel,  the alum was  added  empirically, but  averaged  250 Ibs. per
million gallons of water.  Such additions resulted in  a  total concentration
of  about 2.4 ppm  aluminum with about 0.04 ppm soluble.   Since the plant
utilized large settling  basins and  is situated  about 25  miles from the
University,  it is assumed that only  the soluble portion  was present  in  the
input  to the  system.   Such an  amount is negligible  in  comparison  to  the
additions of  the  tests  themselves.   Some curious  behavior on  the  part of
the control trout, namely refusal  to feed, was  noted on  about  5 days of
the test.   The alum  additions  may  be accountable  for  this.  As  a  result  of
these  additions,  the input pH to  the system  fluctuated far more than
normally.   The  pH of the 5.2  ppm test was  difficult  to regulate on this
account.

The pH of  the input  water eventually became  reduced  to levels where  the
hydroxide  diluter could have  been used  to  raise the pH of the controls  and
 0.52 ppm tests.   The solubility of aluminum  is  nearly  constant  in the
 region of  the test pH.   Sudden increases  in  the pH of  the tests would ex-
 pose the fish to conditions  unlike those being tested if the input pH
 should arise sharply and hydroxide additions were continued,  so no hydroxide
was added.   For the same reason, the pH of the 5.2 ppm test  was held below
                                  26

-------
7.0.  This precaution proved to be well founded as on several occasions
the input pH increased to well above 7.0, and fish would have been ex-
posed to acutely toxic conditions invalidating the test, had hydroxide
additions been in progress.

The trout were eleven weeks old at the start of the test.  All exposures
were conducted for 45 days  (1080 hours).  The mean temperature for the test
was 12.8°C  (s  = 0.28). Twenty-five fish were used in each trough.  The
mean pH for the controls was_2.44 (s  = 2.9 x 10  ) for the first seven
days and 6.85  (s  = 4.4 x 10  ) thereafter.
                                                      2           -3
At a mean concentration of  0.513 ppm aluminum added (s  = 6.3 x 10  ),
the pH was  1.27 (s  = 2.8 x 10~ ) for the first seven days of exposure
and 6.52  (s  = 4.0 x 10  )  thereafter.

Under these conditions, fish showed a darkening of coloration and reduced
feeding activity beginning  on the seventh day.  Gill hyperplasia was ob-
served on some individuals.  These symptoms gradually increased in incidence
and severity for the first  five weeks of exposure.  Mortalities began to
occur on  the twenty-third day and occurred regularly thereafter.  At the
conclusion  of  the  exposure  (45 days)  these tests  had not achieved 50% mor-
tality.   The average TL n value reported for  these conditions is  the
average of  times projected  by linear  regression of the mortality  rates.  At
the conclusion of  the  test  these  trout  were feeding very erratically.   Their
average weight was 32.5% of the control average as opposed  to 94.1% of  the
control average at the start of the  tests  (Figure 4).  After  74 days recov-
ery the average weight of  test  fish  remains only  43.1% of  the control
average.

Another pair  of-duplicate  exposures  received  a mean  5^14 ppm aluminum
 (s  = 8.1 x 10  )  at  a mean pH  of  6.80  (s  =  6.8  x 10   ).   Under  these  con-
ditions feeding  stopped after four  hours exposure.  After  four days, ac-
tivity was  drastically reduced  and  coloration was very  dark.  Gill hyper-
plasia was  observed on all  individuals  within seven  days.   These  trout
exhibited a small  degree  of fright  reaction  throughout  the test and began
feeding very  lightly  for  the last  three weeks of  exposure.   Mortalities
occurred  at a  regular  rate throughout the test again preceded by  a  total
loss  of equilibrium.   No  fecal  casts were observed.

Survivors of  the  test  were extremely emaciated,  averaging  only  24.1%  of the
control weight as  opposed  to 91.8%  of the control average  at the  beginning
of the  test (Figure 6).   A single mortality  occurred among the  controls
during  this test.   Resumption of  growth among these  fish proceeded very
slowly  as they averaged 29.5% of  the control weight  after  74 days recovery.

Smirnov's statistic was used to verify the agreement of mortality rates in
sets  of duplicate conditions.   These tests indicated no significant  differ-
ence among  mortality  rates  of any set of duplicates  (a  = .10) except  for
 the test  at pH 9.0.   Although by  the conclusion of that study mortality

-------
ro
CO
               FIG. 6. GROWTH OF  FINGERLINGS ASSOCIATED WITH  ALUMINUM  EXPOSURES

                      AT  PH 7.0.

-------
rates were identical, early differences caused the tests to be sig-
nificantly different  (a =.01).  The severity and short duration of this
exposure are undoubtedly causing these variations.

As a result of noting the significant reduction in growth rates of the
fingerling rainbows during the actual bioassay determinations experimental
fish were maintained for several months following exposures of aluminum
(Table 2).

Statistical analysis comparing the slopes of the recovery growth plots
(Figures 7 and 8), as suggested by Dixon and Massey  (1969) reveal no sig-
nificant differences  (a = .10) between growth rates  of any group of test
fish when compared to controls at the same weight (Table 3).  The immediate
implications of these findings suggest that neither  brief or extended con-
tact with lethal amounts of both soluble and/or insoluble aluminum will
produce irreversible physiological damage as far as  growth is concerned,
to young trout surviving the  exposure.  Mortality rates returned to the
control rate in all cases except for the test at pH  8, 5.2 ppm aluminum.
In these conditions, 11 of 12 survivors succumbed during the first 24 hours
of recovery, presumably due to their extremely debilitated condition, stress
of handling, weighing and changing the water supply.  Effects of maturation
and fertility were not studied and may be significant.

Thus, the growth of trout exposed to high dosages of aluminum under these
conditions will be reduced only as long as or slightly longer than the
exposure is continued.  In the case of a 45 day exposure,  this means that
the survivors are essentially 45-60 days behind in their development.  In
young fish this may easily account for a 100% weight difference even after
an extensive recovery period. Applying this to natural situations, the
size differential and subsequent delay in development could be important
factors in susceptibility to  predation, procuring available food, negotiat-
ing currents and  spawning.


Fertilization and early life  history

The research to study the effects of aluminum on  the fertilization and early
life history was designed to  determine the toxicity  by exposing eggs in
simulated natural redds and in submerged screen trays  (Figure 3) .

Culturing the eggs in the simulated natural redds was unsuccessful as very
heavy mortalities  (80% to complete) indicated eggs and sac fry were addi-
tionally stressed by  conditions  (likely poor water flow through the gravel)
in the redd along with the aluminum exposure.  The few eggs that  did develop
to hatching lagged behind the eggs in the trays.  Simulated natural conditions
precluded disturbing  the gravel for accurate counts  at the eyed and hatching
stages.  Dead eggs could not  be removed for the same reason, and  fungus
developed in all redds to provide an additional source of  mortality.  Likely
a more sophisticated  system of forcing water  through the  gravel would have
improved  the conditions.
                                  29

-------
                             Table 2.   Summary of Recovery Growth Rates




                                                     Tests
pH 8.0
Concentration control
Exposure time
(days)
*Average weight
start bioassay (gins)
*Average weight
end bioassay (gms)
*Mortality during
bioassay (%)
Average weight
end recovery (gms)
Recovery time (days)
^-Mortality during
recovery (%)
45.0
7.4
21.6
.0
30.0
16.0
—
8.0 8.0 8.5
.52 5.2 ppm control
45
7.
12.
8.
17.
16.

45.0
1 7.9
5 6.5
0 77.0
5
0 1.0
0 83
9.25
2.0
3.0
.0
73.5
161.0
5
8.5 9.0
5.2 ppm control
9.25
2.2
1.5
51.0
58.3
161.0
3
4.
1.
3.
t
73.
165.
2
7
9
5
0
5
0

9.0 7.0
5.2 ppm control
4.7
2.2
2.2
67.0
65.3
165.0
2
45.0
3.4
9.6
2.0
205.9
290.0
3
7.0
0.52 ppm
45.0
3.2
3.1
44.0
109.0
290.0
5
7.2
5 . 2 ppm
45.0
3.1
2.3
58.0
96.4
290.0
16
* Average of 2 replicates




+ Adjusted to allow for periodic random thinning of groups to prevent overcrowding

-------
                pH 8.5 CONTROL
                pH 8.5 TEST
                pH 9.0 CONTROL
                pH 9.0 TEST
   0
40    80    120    160
  DAYS  RECOVERY
200
FIG.  7.   Growth Rates of  Fingerlings
 Recovering from Exposure to 5.2ppm
 Al at pH  8.5 and 9.0.

-------
6.0
                     CONTROL 	
                     0.52 ppm	
                     5.2 ppm  	•	
    0    60    120    180    240   300
            DAYS RECOVERY
  FIG.  8.  Growth Rates of  Fingerlings
   Recovering from  Exposure to 0.52 and
   5.2 ppm  Al at pH 7.0.
                   32

-------
                   Table 3

Comparison of Growth Rates During Recovery Periods

    Hypothesis:  Slope        =  Slope
                       Test            Control
             Al ppm
Slope
Control
Test
Control
Test
Control
Test
Test
8.5
8.5 5.2
9.0
9.0 5.2
7.0
7.0 0.52
7.0 5.2
.0167
.0212
.0163
.0158
.0145
.0137
.0160
.932
.973 .890 n.s.
.930
.894 .097 n.s.
.955
.973 .176 n.s.
.971 .330 n.s.
                       33

-------
Eggs and fry exposed in the submerged trays were studied in two experimental
series: the first was made with controls and treatments at pH 9.0, and the
second was made with controls at pH 8.0, with the 0.52 ppm Al at pH 7.0,
and with the 5.2 ppm Al replicated at pH 7.0 and pH 8.0.  Of first concern
was the possible effects of aluminum on the fertilization. In both series
all the control eggs were fertilized in the control water.  One tray in
each treatment trough was supplied eggs fertilized in control water and
one tray was supplied with eggs fertilized in the appropriate aluminum con-
centration.  Tables 4 and 5 present the data on egg mortality to the eyed
stage for both series.  None of the data would support any claim that alum-
inum toxicity reduced fertilization.  In fact, in the first series the
mortality to the eyed stage is lower for the eggs fertilized in the aluminum
concentrations.  The data in Tables 4 and 5 also confirm the high fertility
of the eggs.

Although the figures for the egg and fry survival would seem to encourage
the application of statistical comparisons, problems with the rearing and
with maintaining constant pH's reduce the power of the early life history
data.  Consequently only general observations seem warranted.  For example,
total mortality from fertilization to termination for series one at standard
pH 9.0 indicate a mortality of 74% for the controls, 60% for the 0.052 ppm
Al, 55% for the 0.52 ppm Al, and 78% for the 5.20 ppm Al.

In the second series of the early life history exposures, total mortality
from fertilization to termination was 50% for controls, 58% for the 0.520
ppm Al at pH 7.0, 68% for the 5.2 ppm Al at pH 7.0, and 92% for 5.2 ppm
Al at pH 8.0.

A general conclusion is that dissolved aluminum is more toxic as evidenced
by higher mortality of 5.2 ppm at pH 8 and pH 9, but that high concentra-
tions of suspended aluminum (5.2 ppm, pH 7.0) are also nearly as toxic.
The trend is the same as for the fingerling rainbows.

-------
             Table 4.  Fertilization and  early  life history with controls and treatments at pH 9.0.
u>
VJI
Trough
1
1
2
2
3
3
4
4
5
5
6
6
7
7
8
8
A =
C =
A =
Treatment
Control
Control
Control
Control
0.052 *
0.052
0.052
0.052
0.520
0.520
0.520
0.520
5.200
5.200
5.200
5.200
pptn. aluminum concentration
fertilized in control water pH 9.0
fertilized in appropriate aluminum
Total Eggs at
Start each
Tray
C 188
C 201
C 165
C 190
A 276
C 196
A 211
C 183
A 337
C 171
A 301
C 174
A 226
C 220
A 335
C 198
concentration pH 9.0
Number Egg
Mortality to
Eyed Stage
36
36
26
42
8
21
2
35
3
32
0
22
12
35
1
29

Percent
Mortality
to Eyed Stage
19
17
16
22
3
11
1
19
1
19
0
13
5
16
1
15


-------
                 Table 5.  Fertilization and early life history with controls at pH 8.0 (troughs 1 and 2), 0.52
                           ppm Al at pH 7.0 (troughs 3 and 4), 5.2 ppm Al at pH 7.0 (troughs 5 and 6), and 5.2
                           ppm Al at pH 8.0 (troughs 7 and 8).
         Trough
Treatment
LO
cr\
Total Eggs at
Start each
Tray
             * = ppm. aluminum concentration
             C « fertilized  in control water  pH 8.0
             A - fertilized  in appropriate aluminum concentration and pH
Number Egg
Mortality to
Eyed Stage
Percent
Mortality
to Eyed Stage
1 Control
1 Control
2 Control
2 Control
3 0.520*
3 0.520
4 0.520
4 0.520
5 5.200
5 5.200
6 5.200
6 5 . 200
7 5.200
7 5.200
8 5.200
8 5.200
C 299
C 244
C 274
C 245
A 289
C 225
A 230
C 262
A 276
C 214
A 216
C 245
A 257
C 212
A 152
C 232
37
31
34
29
24
22
14
20
20
20
12
36
29
25
15
31
12
13
12
12
8
10
6
8
7
9
6
15
11
11
10
13

-------
                            SECTION VI

                          ACKNOWLEDGMENTS
The Department of Fishery and Wildlife Biology, Colorado State University,
initiated the grant proposal, provided the facilities for the aluminum
studies, and handled the administration.  Dr. W. Harry Everhart, Professor
and Chairman Fishery Major, was Project Director.

Mr. Robert Freeman, Graduate Student, designed the test facility, con-
ducted the fingerling studies, and worked toward solving the aluminum
chemistry problems.

Dr. Janet Osteryoung, Quantitative Chemist, was technical consultant for
chemical problems.  Mr. Kenneth Olsen, Graduate Student, Department of
Chemistry, monitored the pH and prepared aluminum concentrations for the
fry portion of the aluminum studies.

The support of the project by  the Water Quality Office, Environmental
Protection Agency, and assistance from Mr. J. Howard McCormick, Grant
Project Officer, are acknowledged.
                                 37

-------
                             SECTION VII

                              REFERENCES
 1.  American Public Health Association et al.   1965.   Standard Methods
    for the Examination of Water and Wastewater.   12th ed.  American
    Public Health Association, N. Y., N. Y.   769  pp.

 2.  Birnbaum, Z. W.  1962.  "Introduction to Probability and Mathematical
    Statistics", Harper & Bros., N. Y., N. Y.,  325 pp.

 3.  Deuel, C. R., D. C. Haskell and A. U. Tunnison.  1942.   The  New York
    State Fish Hatchery Feeding Chart.  Fisheries Research Bulletin No.
    3,  New York State Conservation Department.

 4.  Dixon, W. J. and F. J. Massey, Jr.  1969.   Introduction to  Statistical
    Analysis.  McGraw-Hill Book Co., N. Y., N.  Y.  638 pp.

 5.  Ebeling, G.  1928.  Uber die gifligkeit einiger schwermetallsalze
    an Hand eines Falles aus der Prasix.  Zeits,  Fisherei 26:49-61.
    Abstract in J.A.W.W.A.  23:1626  (1931).

 6.  Ellis, M. M.  1937.  Detection and measurement of stream pollution.
    Bur. Fish. Bull. 22:365-437.

 7.  Freeman, R. A.  1971.  A constant flow delivery device for  chronic
    bioassay.  Trans. Am. Fish. Soc.  100(1):135-136.

 8.   Hem, J. D.  1968.  Graphical methods for the study of aqueous aluminum
     hydroxide, fluoride and sulfate complexes.   U.S.G.S. Water  Supply
     Paper 1827-B.

 9.   Hem, J. D. and C. E. Roberson.  1967.  Form and stability of aluminum
     hydroxide complexes in dilute solution.  U.S.G.S. Water Supply Paper
     1827-A.

10.   Jones, J.R.E.  1964.  "Fish and River Pollution".  Butterworth Scientific
    Publications, Washington, D. C.   203 pp.

11.  Minkina, A. L.  1946.  On the action of iron and  aluminum on fish.
     Trudy Moskov.  Zooparka 3:23-26.   (In Russian).   Quoted in: Doudoroff,
     D. and M. Katz, 1953.  Critical  review of literature on to toxicity
     of industrial wastes and their components to fish.  II.  The metals, as
     salts.  Sewage and Ind. Wastes,  25, 7: 802-839.

12.   Oshima, S.  1931.  On the toxic  action of dissolved salts and  their
     ions upon young eels  (Angilla japonica).  J. Imp. Fish. Exp. Sta.
     2:1939.
                                   39

-------
13.  Pulley, T. E. 1950.  The effects of aluminum chloride in small con-
     centration on various marine organisms.  Texas J. of Science 2:405-411.

14.  Roberson, C. E. and J. D. Hem.  1969.  Solubility of aluminum in the
     presence of hydroxide, fluoride and sulfate.  U.S.G.S. Water Supply
     Paper, 1827-C.

15.  Sanborn, N. H.  1945.  The lethal effect of certain chemicals on
     fresh water fish.  Canning Trade 67(49):10-12.

16.  Schaut, G. G.  1939.  Fish catastrophes during droughts.  J.A.W.W.A.
     31:771-822.

17.  Snedecor, G. W.  1966.  Statistical Methods.  The Iowa State Univ.
     Press, Ames, Iowa.  534 pp.

18.  Thomas, A.  1915.  Effects of certain metallic salts upon fishes.
     Trans. Amer. Fish. Soc. 44(1):120-124.

19.  Wallen, I. E., W. C. Greer and R. Lasater.  1957.  Toxicity to Gambusia
     affinis of certain pure chemicals in turbid waters.  Sewage and Ind.
     Wastes, 29:695-711.
                                 1*0

-------
                           SECTION VIII

                           PUBLICATIONS
Freeman, Robert A. and W. Harry Everhart
     1971  Toxicity of aluminum hydroxide complexes in neutral and basic
           media to rainbow trout.  Trans. Amer. Fish. Soc., 100(4):644-658,

Freeman, Robert A.
     1971.  A constant flow delivery device for chronic bioassay.  Trans.
           Amer. Fish. Soc., 100(1) -.135-136.

Freeman, Robert A.
     1972.  Recovery of rainbow trout from aluminum poisoning.   (Submitted
           to Trans. Amer. Fish.  Soc.).

-------
1
Accession Number
w
5
Organization
2

Department of
Subject Field & Group
05C
Fishery
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
and Wildlife Biology

             Fort Collins, Colorado 80521
   Title
             EFFECTS OF CHEMICAL VARIATIONS IN AQUATIC ENVIRONMENTS:
             Toxic effects  of aqueous aluminum to rainbow trout.
10

A u thorns)
Everhart ,
Robert A.
W. Harry and
Freeman
16

21
Project Designation
18050-2DYC
TVoJe
22
   Citation
              Environmental Protection Agency report
              number, EPA-R3-73-011b,  February 1973.
   Descriptors (Starred First)
         *Rainbow trout,   ^aluminum,   *toxicity , *pH
25
Identifiers (Starred First)


      *Rainbow trout,   ^aluminum,    *toxicity
27
Abstract
    Fertilized eggs, fry, and fingerlings were exposed to aqueous  aluminum complexes
  in neutral and basic  media under constantly flowing,  controlled conditions of aluminum
  concentration, pH,  and temperature.  Toxicities  of various concentrations were highly
  pH dependent.  Dissolved concentrations over  1.5 ppm aluminum caused physiological and
  behavioral aberrations as well as acute mortality.   Toxic effects of suspended aluminum,
  though greater at lower concentrations, do not increase as much as the effects of
  dissolved aluminum  with higher concentrations.   Growth of trout exposed to high dosages
  of aluminum was reduced only as long as or slightly longer than the exposure continued.
       Egg and fry bioassays were conducted with exposures in trays and simulated natural
  redds.  Fertilization was not affected by any concentrations tested, and most mortalities
  occurred during hatching and in the post-swim-up stage.  Trends in toxicity were similar
  to those found with fingerlings indicating dissolved aluminum to be more toxic than
  equivalent suspended amounts.
Abstractor
          Harry Everhart
                           institution Cornen University
 WRSIC
      (REV. JULY 1969)
                            SEND. WITH COPY OF DOCUMENT. TO: *ATER RESOURCES SC JENT |F 1C
                                                   U.S. DEPARTMENT OF THE INTERIOR
                                                   WASHINGTON, D. C. 20240
   «U.S. GOVERNMENT PRINTING OFFICE: 197.3-514-154. 264-1-3

-------