|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
EFFECTS OF POTASSIUM PERMANGANATE ON THE SAILFIN MOLLY, Poecilia
latippinna, AT VARYING SALINITY LEVELS
EMILY N. MARECAUX
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Emily N. Marecaux
To my family and friends who have supported me throughout my college career.
My committee members, Ruth Francis-Floyd, B. Denise Petty, Scott P. Terrell,
Kathleen H. Hartman, and Roy P. E. Yanong, have provided significant contributions to
this project. Jeff Hill provided extremely important help with the statistical analysis of
this project. Sherry Giardina provided helpful guidance through the graduate student
process. Scott Graves of the University of Florida Tropical Aquaculture Laboratory
helped me set up the aquaculture tank system used for this project. Jamie Holloway from
the Department of Fisheries and Aquatic Sciences helped with the pilot study and
gathering of supplies for this project. Patricia Lewis and Don Samuelson from the
University of Florida College of Veterinary Medicine allowed for the use of the histology
laboratory which was an integral part of my project. Robert Leonard, Tina Crosby, Jen
Matysczak helped gather behavioral observation data. Chris Langeneck took the pictures
of the sailfin mollies. Finally, the generous support from Segrest Farms was very much
appreciated and the project could not have been completed without their donation. Thank
you to everyone that helped this project come to completion.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S ................................................................................................. iv
LIST OF TABLES .......... .......................................... ................ ........ vii
LIST OF FIGURE S ........ ........ .......................................... .............. viii
ABSTRACT .............. .................. .......... .............. ix
1 L ITER A TU R E R E V IEW ......... ................... ......... ........................................1
Introduction ................. .................. ...........1....
The Sailfin Molly Poecilia latipinna........................................... .................. 3
Potassium Permanganate KMnO4...................................... .. ............... 5
T ables and F figures ................................................................... .......................... . 9
2 PILO T STU D Y ........... ............................... ...... ...... .............. .. 11
E xperim mental D design ........................................................ ........ .. .... ........ 11
R e su lts .................................................................................................................... 1 3
D isc u ssio n .............................................................................................................. 1 4
3 M ATERIALS AND M ETHOD S ........................................... .......................... 15
Experim ental Procedures ......... .......................................... ....... ...............15
Experimental Fish and Quarantine Procedure.................................. ...............15
Acclimation Procedure ............................ ......... .. ..... ............... 16
R espon se V ariables ............................ ........................ .. ....... .... ............17
Experim mental D design ................................. ......... .......... ....... 19
KM nO 4 Treatm ent ........... .. ................................. .... .......... ...... .... 19
E xperim ental P procedure ........................................ .........................................20
Statistics ..................... ............ .............................. 21
4 R E S U L T S .............................................................................2 2
W after Q uality................................................. 22
Behavior ......................... ..............................22
H isto lo g y ........................................................................................2 4
M o rta lity ..............................................................................2 7
5 DISCUSSION AND CONCLUSIONS ............... ......... ................................31
A PROCESSING SCHEDULE FOR THE SHANDON EXCELSIOR
AUTOMATIC TISSUE PROCESSOR.................................. ........................ 39
B TYPICAL COMPOSITION OF INSTANT OCEAN SALT....................................40
L IST O F R E F E R E N C E S ........................................................................ .....................4 1
B IO G R A PH IC A L SK E TCH ..................................................................... ..................45
LIST OF TABLES
1-1 The 96h LC5o levels for potassium permanganate (KMnO4) determined for
juvenile and larval striped bass, Morone saxatilis. ...............................................10
1-2 Dosages and target pathogens of KMnO4 based on literature review........ ........ 10
3-1 Experimental treatment combinations of KMnO4 concentrations and salinity
le v e ls ............................................................................ ................ 1 9
3-2 Calculated KMnO4 demand for each salinity level ...........................................20
4-1 Select water chemistry values for experimental tanks (n = 12) prior to KMnO4
treatm ents. ...........................................................................22
4-2 Kruskal-Wallis analysis of six-hour behavior scores by treatment combinations
(salinity KMnO4 concentration) (N = 3 tanks per treatment).............................23
4-3 Dunn's multiple comparison test for behavior scores shows significant treatment
group p s. .............................................................................. 24
4-4 Kruskal-Wallis analysis of the slide score by the salinity/KMnO4 treatment
com bination s. ...................................................... ................. 2 5
4-5 Dunn's multiple comparison test for histology slide scores shows significantly
different treatm ents. ...................... .................. ... .... ................. 25
4-6 Two-way ANOVA of the total percentage mortality by the salinity and the
K M nO 4 concentration. ..................................................................... .................. 28
4-7 One-way ANOVA analysis of the total mortality percentage by the treatment
com bination s. ...................................................... ................. 2 8
A-1 Processing schedule for the shandon excelsior automatic tissue processor ............39
B-l Typical com position of instant ocean salt ............................................................. 40
LIST OF FIGURES
1-1 The sailfin m olly .................. .. .. ............................................ .... .9
4-1 Gills from sailfin mollies in 2 g/L salinity water treated with KMnO4 ................26
4-2 Gills from sailfin mollies in 15 g/L salinity water treated with KMnO4................26
4-3 Gills from sailfin mollies in 30 g/L salinity water treated with KMnO4................27
4-4 Cumulative Total Percentage Mortality Over 7 days by Salinity (fresh 2 g/L,
brackish 15 g/L, and salt 30 g/L) and KMnO4 Concentration (0.0, 0.5, 1.0, and
3.0 m g/L) ................................ ................................. ........... 29
4-5 The cumulative total percentage mortality shown over time for the 2 g/L
treatm ents. ...........................................................................29
4-6 Cumulative total percentage mortality shown over time for the 15 g/L
treatm ents. .......................................................................... 30
4-7 Cumulative total percentage mortality shown over time for the 30 g/L
treatm ents. .......................................................................... 30
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
EFFECTS OF POTASSIUM PERMANGANATE ON THE SAILFIN MOLLY, Poecilia
latipinna, AT VARYING SALINITY LEVELS
Emily N. Marecaux
Chair: Ruth Francis-Floyd
Major Department: Fisheries and Aquatic Sciences
Potassium permanganate (KMnO4) is used in fish culture for disease treatment,
water clarification, rotenone detoxification, and historically for management of oxygen
depletion. Most commonly, KMnO4 is used in freshwater systems at 2 mg/L to control
ectoparasites, bacteria, and fungi. Effective concentrations are determined by the KMnO4
demand of the water being treated. Although safe use of KMnO4 in freshwater systems is
well documented, its toxicity to fish in saltwater systems is less well known.
The sailfin molly, Poecilia latipinna, a euryhaline species, was used as a model to
test the toxicity of KMnO4 at varying concentrations and at different salinity levels.
Target KMnO4 concentrations of 0.0, 0.5, 1.0, and 3.0 mg/L plus the KMnO4 demand
were tested. Toxicity was tested at salinity levels of 2, 15, and 30 g/L. Mortality rates
and fish behavior were monitored throughout the experiment and tissue samples for
histological analysis were taken at time zero, immediately post-treatment (12 hours), and
at the end of the monitoring period (7 days).
The mortality rate was significantly higher in the 30 g/L salinity, 3.0 mg/L KMnO4
treatment group than in any other treatment group (p < 0.001). The 15 g/L salinity, 3.0
mg/L KMnO4 treatment group was also found to be significantly different from the 15
g/L salinity 0.0 and 0.5 mg/L KMnO4 treatment groups (p < 0.001). The 2 g/L salinity,
3.0 mg/L KMnO4 treatment was not found to be significantly different. Dunn's multiple
comparison test indicated that treatments of 2 g/L, 15 g/L, and 30 g/L salinities treated
with 3.0 mg/L showed significant changes in behavior resulting in the loss of
equilibrium. Dunn's multiple comparison test also indicated that treatments 15 and 30
g/L salinity, 3.0 mg/L KMnO4 concentration showed significantly different gill damage
as indicated by secondary lamellar fusion, lifting of epithelial cell lining by expansion of
the lamellar interstitium by inflammation and edema, and necrosis.
Results from this study suggest that KMnO4 at concentrations of 0.5 and 1.0 mg/L
may be safe for use in water containing sailfin mollies in water of salinities of 2, 15, and
30 g/L. However, KMnO4 should not be used at concentrations of 3.0 mg/L in 2, 15 or
30 g/L salinity water on the sailfin molly until further research is conducted. Toxicity of
potassium permanganate increased in the higher salinity groups (15 and 30 g/L)
compared to the low salinity group (2 g/L).
Potassium permanganate, KMnO4, is an oxidizing agent that is used in fish culture
for disease treatment, water clarification, rotenone detoxification, and historically for
management of oxygen depletion in fish ponds. Potassium permanganate has been used
to treat external pathogens including fungus, bacteria, and some parasites (Lay 1971;
Masser and Jenson 1991; Noga 1996; Francis-Floyd and Klinger 1997; Plumb 1999;
Carpenter et al. 2001; Stoskopf 1993; Bishop 2001; Straus and Griffin 2002; Thomas-
Jinu and Goodwin 2004). KMnO4 can also be used for water clarification purposes (Lay
1971) by oxidizing organic material in the water, forming precipitates that can be
removed from the water by a filter. When rotenone is used in water treatments, KMnO4
can be used as a counter agent to detoxify the water (Lawrence 1956). KMnO4 was also
used to add oxygen to aquaculture ponds in situations of low dissolved oxygen levels in
water (Lay 1971). This method is no longer practiced as KMnO4 did not raise oxygen
levels significantly and may cause oxygen levels to decrease by killing the oxygen
producing algae in the pond (Tucker and Boyd 1977).
Many different treatment regimes have been developed over the years for the
different uses of KMnO4 in aquaculture. However, use of KMnO4 in salt water is less
common than in freshwater systems so less is known about what happens when it is used
as a treatment in the marine environment. There is debate over the use of KMnO4 for
treatment of marine fish and concern regarding the toxicity threshold to KMnO4 in
marine species held in saltwater.
Marking and Bills (1975) found that KMnO4 was more toxic to fish in waters with
pH between 8.5 and 9.5 and total hardness of 300 mg/L as CaCO3 (compared to water
with a total hardness of 12 mg/L as CaCO3). Stuart (1983) and Noga (1996) have
suggested that KMnO4 is toxic to fish in saltwater because of the higher pH typically
associated with saltwater, which may cause manganese dioxide to precipitate onto the
gills. Natural seawater typically has a pH between 7.8 and 8.2 (Spotte 1992), which may
be higher than the pH of some freshwater systems, which are commonly between 6.8 and
7.2. In a freshwater efficacy study using KMnO4 at a dosage of up to 1.5 mg/L to treat
ichthyophthiriasis in channel catfish, Ictaluruspunctatus, the pH was maintained at 8.5
+/- 0.1 (Straus and Griffin 2002), and the authors reported that no fish died in the
effective treatment group (1.25 mg/L). This suggests that pH levels above those typically
found in freshwater may not be the sole cause of KMnO4 toxicity to fish.
Reardon and Harrell (1994) determined the KMnO4 concentration that caused
50% mortality of exposed population over 96 hours, 96h LC5o, at different salinity levels
up to 15 g/L for juvenile and larval striped bass, Morone saxatilis. In this study juveniles
were most tolerant of KMnO4 at a salinity level of 5 g/L, while larvae were most tolerant
of KMnO4 at a salinity level of 3 g/L. Both juvenile and larval striped bass were least
tolerant of KMnO4 at salinity levels of zero. The 96h LC5o levels that Reardon and
Harrell (1994) reported are depicted in Table 1-1 (see page 10).
There was a significant decrease in the 96h LC5o level at salinity levels of 0, 10,
and 15 g/L compared to the 5 g/L treatment for juveniles with KMnO4 being most toxic
to juveniles at the 0 g/L salinity followed by the 15 g/L salinity. It was suggested that the
greatest toxicity occurred in 0 g/L water because this was the salinity where the greatest
osmotic imbalance occurred. This experiment showed that margin of animal safety may
narrow when KMnO4 is applied at higher salinities, but it did not demonstrate if the trend
continued as salinity increased above 15 g/L.
The Sailfin Molly Poecilia latipinna
The sailfin molly, Poecilia latipinna, formerly described and named Mollienesia
latipinna by Charles Alexandre Lesueur in 1821 (Robins 2003), is from the family
Poeciliidae, comprising over 190 species (Parenti and Rauchenberger 1989). The natural
distribution of the sailfin molly is fresh, brackish, and salt waters of Florida, Mexico,
Texas, South and North Carolina, and Virginia (Petrovicky 1988; Robins 2003;
Courtenay and Meffe 1989). Non-indigenous populations are established in the western
U.S. (Arizona, California, and Nevada), Hawaii, Canada, Central America, Singapore,
Australia, New Zealand, Guam, and the Philippines (Courtenay and Meffe 1989). The
sailfin molly prefers lowland areas such as marshes, lowland streams, swamps, and
estuaries (Robins 2003).
The sailfin molly is a fusiform shaped small fish (15-53 mm total length) with a
small head and upturned mouth (Robins 2003)(see Figure 1-1, page 9). The dorsal fin is
greatly enlarged in mature males compared to those of mature females. The dorsal fin is
used as a display to attract females for reproduction. Only dominant males display the
dorsal fin. Subordinate males use the "ambush" technique for breeding (Balsano et al.
1989). The "ambush" technique refers to the chance that the dominant male becomes
distracted, allowing the subordinate male to breed with the female before being chased
away. Males have a modified anal fin called the gonopodium, which is used for internal
fertilization. At rest the gonopodium points caudally, but during reproduction, the
gonopodium is pointed forward and is inserted into the female in a quick motion, which
results in sperm being deposited into the female. The female molly can store the sperm
deposited by the male. The gestation period is 3-4 weeks. Females are viviparous and
give birth multiple times during the year (Robins 2003).
The sailfin molly, Poecilia latipinna, is a euryhaline species that can tolerate
salinities as high as 70 to 80 g/L (Nordlie et al. 1992). Gustafson (1981) used short-term
salinity acclimation as a method to evaluate the influence of salinity on plasma
osmoregulation and routine oxygen consumption. Gustafson (1981) altered the salinity at
a rate of 4.14 g/L per day for sailfin mollies of brackish water origin (mostly 10-20 g/L,
but ranged from 4-35 g/L) and for sailfin mollies of freshwater origin (no information on
the salinity level given). Frank Nordlie (The University of Florida, personal
communication) recommended salinity level adjustments at a rate of 5 g/L every five
days for proper osmoregulation balance. Sailfin mollies are highly adaptable to changing
salinity ranges that are found in their natural habitat.
For aquaculture purposes, salinity up to 3 g/L is considered freshwater (Chapman,
The University of Florida, personal communication), while brackish water ranges from
3.0 g/L to 29 g/L and saltwater is 30 g/L and above. Some species are more sensitive to
salinity than others, therefore, it is necessary to know the limits of the species being used
The sailfin molly is primarily an herbivore, eating plants and algae, but is also
opportunistic and will eat other food items including detritus or insect larvae and
cannibalism has been reported (Meffe and Snelson 1989). The sailfin molly is a prey
item for many predators. Predators that eat this fish include reptiles, birds, other fishes,
amphibians, and insects. The sailfin molly is also popular in the aquarium trade and is
available in a wide variety of colors through domestication. It has also been used for
research and for biological control of mosquitoes (Courtenay and Meffe 1989).
Potassium Permanganate KMnO4
Historically, the Environmental Protection Agency (EPA) (1985), registered
KMnO4 for use in cooling towers, evaporative condensers, air wash systems, ornamental
ponds, cooling fountains, aquaria, human drinking water, poultry drinking water, for
surface disinfection, sanitization, and as a deodorizer. However, KMnO4 is now exempt
from registration by the EPA because it does not pose unreasonable risks to public health
or the environment.
Currently KMnO4 is a U.S. Food and Drug Administration (FDA), investigational
new animal drug (INAD) under investigation by Carus Chemical Company (Peru, IL)
and Stuttgart National Aquaculture Research Center (SNARC) (Stuttgart, AR). FDA
approval will allow for the legal use of KMnO4 in water containing food fish. To gain
approval from the FDA, Carus Chemical Company is responsible for research on product
chemistry and mammalian toxicology, while SNARC is responsible for research on
efficacy, human food safety, target animal safety, and environmental safety (Straus,
SNARC, personal communication). Regulatory action on KMnO4 has been deferred
pending the outcome of current research.
As a therapeutant for fish, KMnO4 has been used as an external bactericidal and
fungicidal agent. Formulations of KMnO4 are available as ready-to-use liquids, pellets or
tablets, powder, or crystals. The active ingredient is the permanganate ion. It functions
as a strong oxidizing agent that is corrosive and burns any organic material it comes into
contact with. For this reason it is effective if used in fish culture to control external
disease-causing agents including bacteria, fungi, and some parasites. The oxidizing
activity is also the primary problem for treated fish.
As an oxidizing agent, KMnO4 is able to add oxygen, remove hydrogen, or
remove electrons from an element or compound (Cams Chemical Company 2004). For
example in drinking water treatments, KMnO4 is able to oxidize soluble manganese and
iron into manganese dioxide and iron oxide, which are insoluble and can be removed by
The effectiveness of KMnO4 is related to the amount of oxidizable material in the
water, i.e. organic material and other elements (inorganic) that may be oxidized. This is
referred to as the KMnO4 demand of the water (Tucker 1984). Engstrom-Heg (1971)
developed a test, using a spectrophotometer, to determine the KMnO4 demand of the
water to be treated. Later, Boyd (1979) developed a quick visual test to determine the
KMnO4 demand of the water to be treated. He treated 1,000 mL samples of water with 0,
1, 2, 3, 4, 6, 8, 10, and 12 mg/L of KMnO4 in separate containers; after fifteen minutes,
the container with the lowest concentration that was still pink was considered the KMnO4
demand of the water.
Potassium permanganate is rendered inactive by organic material (Tucker and
Boyd 1977), therefore, in a recirculating system the biological filter may be affected
because organic material, including the nitrifying bacteria, may be oxidized. However, it
is unclear if KMnO4 has a significant effect on the efficiency of the biofilter (Spotte
1992). One study (Levine and Meade 1976) demonstrated that a treatment of KMnO4
inhibited nitrification 86%, while another study (Collins et al. 1975) demonstrated that
KMnO4 had no effect on nitrification. If the filter has a build-up of organic material, it
will increase the KMnO4 demand of the water if the filter is left online during treatment
and the biological portion of the filter may be damaged.
Tucker (1989) developed a method to estimate the required treatment of KMnO4
based on the 15-minute KMnO4 demand of the water. The calculation from his work is
2.5 multiplied by the value obtained in the fifteen-minute test. However, the level of
KMnO4 needed to control an ichthyophthiriasis outbreak in an efficacy study was 1.25
mg/L (Straus and Griffin 2002). In that study it was determined that using Tucker's
method, the treatment rate indicated would be 1.0 mg/L. This implies that Tucker's
method is not a "fail-safe" method for determining treatment dosages. Another
recommendation for compensation of the KMnO4 demand is to add 2 mg/L to the
KMnO4 demand of the water (Plumb 1999). Other treatment recommendations found in
the literature are summarized in Table 1-2 (see page 10).
The toxicity margin of KMnO4 is narrow (1-3 mg/L) (Plumb 1999). Toxicity
levels have been determined for carp fry as an LC5o ranging from 37.5 to 48 mg/L at
26C and 45 to 37.5 mg/L at 320C, at 24 and 48 hours, respectively (Ghosh and Pal
1969). The pH was maintained between 7.8 and 8.2. Studies have shown that 20 mg/L
KMnO4 is toxic to guppies and 3.2 mg/L KMnO4 is toxic to catfish (Scott and Warren as
cited by Lay 1981), however KMnO4 demand or exposure time is not reported.
Lawrence (1956) reported that toxicity levels of KMnO4 were 3 mg/L for bluegills, 4
mg/L for largemouth bass, and 6 mg/L for goldfish. Finally Lawrence (1956) showed a
toxicity level of 5 mg/L KMnO4 for fathead minnows at a temperature of 680F in a
prolonged bath (no method for removing the KMnO4 was employed), but restocking the
aquaria 24 hours after treatment showed that the water was no longer toxic to fish at that
time. Toxicity of KMnO4 decreases with increasing KMnO4 demand (Tucker and Boyd
1977). Potassium permanganate tolerance is dependent on water quality, exposure time,
and the species of fish.
Potassium permanganate efficacy studies have been conducted with columnaris
and ichthyophthiriasis. Straus and Griffin (2002) reported that a treatment of 1.25 mg/L
KMnO4 was effective against ichthyophthiriasis, meaning that no trophonts were found
on treated channel catfish, Ictaluruspunctatus. In that study, treatments were
administered daily after a complete water change. The pH was maintained at 8.5 +/- 0.1
with a temperature of 170C +/- 2.5 and the study was run until the control fish died. In a
challenge against columnaris, KMnO4 applied at a dose of 2 mg/L as an indefinite bath at
a temperature of 220C, reduced mortality to 69% compared to the 100% mortality in
infected control channel catfish, Ictaluruspunctatus (Thomas-Jinu and Goodwin 2004).
A fish that has been treated with KMnO4 may show signs of tissue damage
depending on the dose ofKMnO4. Darwish et al. (2002) found that channel catfish,
Ictaluruspunctatus, treated with KMnO4 (0.438, 1.315, and 2.190 mg/L) at a pH of 7 +/-
0.2, showed microscopic lesions in the gills using a routine hematoxylin and eosin
histology stain, while liver and trunk kidney showed no lesions. The fish treated with
0.438 mg/L KMnO4 showed mild hypertrophy and spongiosis of the epithelium of the gill
filaments and lamellae. The fish treated with 1.315 and 2.190 mg/L KMnO4 showed
extensive gill epithelial hyperplasia, lamellar fusion, and obliteration of the interlamellar
spaces with inflammatory exudates containing necrotic epithelial cells.
Potassium permanganate is a strong oxidizer and contact with other materials may
cause fire; however by itself it is stable and will keep indefinitely if stored correctly.
Potassium permanganate is incompatible with the following substances: formaldehyde,
powdered metals, alcohol, arsenites, bromides, iodides, phosphorous, sulfuric acid,
organic compounds, sulfur, activated carbon, hydrides, strong hydrogen peroxide, ferrous
or mercurous salts, hypophosphites, hyposulfites, sulfites, peroxides, and oxalates. Care
should also be taken when handling potassium permanganate; it is corrosive and may
cause bums to any area of contact and may be harmful if swallowed or inhaled.
(Mallinckrodt Baker, Inc. 2001)
Potassium permanganate is an important chemical in aquaculture because of its
availability and its various uses. The goal of this study is to determine ifKMnO4 creates
toxicity problems for sailfin mollies held in saltwater compared to cohorts maintained in
freshwater. This study will examine the effect of KMnO4 on the sailfin molly maintained
at different salinity levels and at different KMnO4 concentrations.
Tables and Figures
Figure 1-1. The sailfin molly. A. The gonopodium on this male sailfin molly is
indicated by the arrow. B. A group of sailfin mollies. Photo credit: Chris
Table 1-1. The 96h LC5o levels for potassium permanganate (KMnO4) determined for
juvenile and larval striped bass, Morone saxatilis.
Juveniles (One-month old) Larvae (18-day old)
Salinity Level 96h LC50 Level Salinity Level 96h LC50 Level
g/L mg/L KMnO4 g/L mg/L KMnO4
0 0.96a 0 1.02a
5 3.26b 3 2.11b
10 1.63c 6 1.41ab
15 1.48c 9 1.73ab
*Data taken from Reardon and Harrell (1994). Means with identical superscript are not
significantly different at the 5% level using the SNK test.
Table 1-2. Dosages and target pathogens of KMnO4 based on literature review.
Recommended dose Duration of
of KMnO4 (mg/L) treatment Treatment Use References
1000 10-40 second fungi, protozoa Carpenter et al. 2001;
bath Noga 1996; Lay 1971
100 5-10 minute fish lice Carpenter et al. 2001;
bath Noga 1996
20 1 hour crustaceans, Stoskopf 1993
10 30 minutes fungi, protozoa Lay 1971
5 30-60 minute ectoparasites, Carpenter et al. 2001;
bath gill/skin bacterial Noga 1996
2 prolonged ectoparasites, Bishop 2001; Noga
immersion or gill/skin bacterial 1996; Thomas-Jinu and
bath, infections, Goodwin 2004; Francis-
indefinite, at protozoa, Floyd and Klinger
least 4 hours, oxidation/ 1997; Plumb 1999;
or at least 12 detoxification of Masser and Jensen 1991
hours hydrogen sulfide
2-5 indefinite crustaceans, Stoskopf 1993
2 prolonged counteract Lawrence 1956
2.5 flush for 4 bacterial gill Stoskopf 1993
consecutive disease in
2 flush ectoparasites, Noga 1996
A pilot study was conducted at the University of Florida Department of Fisheries
and Aquatic Sciences wet lab (Gainesville, Florida) to determine KMnO4 treatment levels
to be tested. Six, 37.85L tanks (filled to 30 L), each with one airstone, were set up at a
salinity level of 5 g/L. Salinity was tested using a refractometer (Aquatic Eco-Systems,
Apopka, Florida). Thirty-six sailfin mollies, Poecilia latipinna, (mean weight = 2.3
grams) were received from Segrest Farms (Gibsonton, Florida) and six fish were
arbitrarily placed into each tank. Fish were fed once daily (1% of total biomass of tank) a
flake fish food (Zeigler Tri-color flake, Gardners, Pennsylvania). Fifty percent water
changes were conducted every other day in conjunction with salinity adjustments.
Saltwater was made with Instant Ocean salt mix (Aquarium Systems, Mentor,
Ohio). Salinity was adjusted so that two tanks each were established at 2 g/L, 15 g/L, and
30 g/L, respectively, over a time period of fourteen days. In an effort to meet a deadline,
the salinity level was altered 5 g/L every other day until the desired salinity was reached,
with the exception of the 2 g/L tanks, which were adjusted by an increment of 3 g/L. The
salinity level of 2 g/L was chosen over 0 g/L because captive sailfin mollies prefer to live
in water that contains some salt (Petty, University of Florida, personal communication).
The treatment level of KMnO4 to be tested was 2 mg/L and 6 mg/L; one tank of each
concentration (2 and 6 mg/L) was represented at each salinity (2, 15, and 30 g/L).
The KMnO4 demand of the water was determined to be 0.20 mg/L KMnO4 for 30
g/L salinity water, 0.15 mg/L KMnO4 for 15 g/L salinity water, and 0.10 mg/L KMnO4
for 2 g/L salinity water. Actual concentrations ofKMnO4 were not determined because
chloride interferes with typical spectrophotometric techniques that might be used
(Delfino, The University of Florida, 2004) because it reacts with KMnO4. Each salinity
would require its own standardization curve and the higher salinity water would produce
a less reliable curve because the active concentration of KMnO4 would be changing with
increased contact time (some of the inorganic compounds, including chloride, of
saltwater reduce KMnO4).
The 16th day after fish arrival a fifty percent water change was done on each tank
and one fish was sampled from each tank for histologic assessment of tissue prior to
KMnO4 exposure. The designated KMnO4 treatment (either 2 or 6 mg/L KMnO4) plus
the KMnO4 demand was added to the tanks and behavior and mortality was monitored for
twelve hours. Behavioral observations were recorded to aid in the design of the scoring
system for the larger experiment at time zero (0), 0.08, 0.16, 0.33, 0.66, 1.5, 3, 6, and 12
hours. Each tank was dosed with KMnO4 and then fish were observed for behavioral
alterations for 2 minutes until all tanks had been observed at this observation time. This
required two people to physically monitor the tanks for the first four time periods (zero,
0.08, 0.16 and 0.33 hours). All fish were sampled for histological analysis either at the
time of their death (mortalities were monitored hourly) or following twelve hours KMnO4
exposure. Fish that survived the treatment were euthanized with buffered (sodium
bicarbonate) tricaine methanesulfonate (1 g/L dose) (Finquel, Argent Laboratories,
Redmond, WA) before fixation. Whole fish were fixed in 10% neutral buffered formalin.
After 48 hours of fixation whole fish were separated into two histology cassettes; one
cassette contained complete, whole gill sections and the other cassette contained three
sections of body tissue. After the gills were removed the sample was cut into three
transverse sections (or steaks) cuts were made at the posterior eye, just cranially to the
pectoral fin and at the anus. Tissue was processed according to the schedule listed in
One hundred percent mortality was observed in all tanks at all salinities (2, 15, and
30 g/L) treated with 6 mg/L KMnO4. All fish from the 30 g/L salinity water treated with
6.0 mg/L KMnO4 died by the third hour of observation, while the 2 g/L and 15 g/L
salinity group fish treated with 6.0 mg/L KMnO4 all died by the sixth hour of
observation. Mortality was not observed in the three other tanks treated with 2 mg/L
KMnO4. Behavior observations were recorded and it was determined that fish generally
exhibited the following behaviors in sequential/chronological progression 1) increased
opercular movement; 2) erratic swimming; 3) intermittent loss of equilibrium or lying on
the bottom resting; and 4) the complete loss of equilibrium. Microscopic examination of
control tissue samples were within normal limits for the gills and other internal organs
(e.g. kidney, liver, spleen, intestine). All gill tissue samples collected from KMnO4
treated fish showed mucous cell infiltration and hyperplasia and inflammatory cell
infiltration and hyperplasia. The 15 g/L and 30 g/L salinity treatment at 6.0 mg/L
KMnO4 showed severe inflammation, expansion of the lamellar interstitium by edema
and inflammatory cells with lifting off of the epithelium, lamellar fusion and necrosis of
the gill epithelium. The gill samples offish from the 30 g/L salinity, 2 mg/L KMnO4
treatment were also beginning to show signs of epithelial lifting from edema and lamellar
fusion. The whole body sections of KMnO4 treated fish appeared to be within normal
limits for this species.
The results of this pilot study suggest that KMnO4 caused more severe gill damage
to fish as the salinity increased. A more thorough, replicated experiment will test this
hypothesis. The fish in 30 g/L salinity group treated with 6.0 mg/L KMnO4 died faster
than those in the 2 and 15 g/L salinity treatments. These results suggest that fish in the
30 g/L salinity treatment group were more severely effected by the KMnO4 treatment. A
new experiment will be designed to test levels of KMnO4 that are not toxic to 100%
sailfin mollies (i.e., less than 6 mg/L KMnO4) to determine ifKMnO4 causes more severe
gill damage and possibly higher mortality at higher salinity levels.
MATERIALS AND METHODS
Experimental Fish and Quarantine Procedure
One thousand five hundred sailfin mollies were obtained from Segrest Farms in
Gibsonton, Florida. Upon arrival at the University of Florida Tropical Aquaculture
Laboratory (Ruskin, Florida) fish transport bags were placed in holding tanks for 30
minutes for temperature acclimation. Fish were exposed to a saltwater bath of 25 g/L for
approximately 5 minutes (Stoskopf 1993) to remove ectoparasites. All salt used in this
experiment was made with Instant Ocean salt mix (Aquarium Systems, Mentor, Ohio).
Fish were weighed in groups of approximately 100 (each individual bag with water) on
an electronic scale, netted out, and then placed into one of three holding tanks (680 L
each, 5 g/L salinity water). The bag and water were reweighed after fish were released to
determine an actual weight of fish in the bag. Fish were kept in the holding tanks until
experimental salinity acclimations (described below) were complete. Any sailfin mollies
that died during the salinity acclimation period were necropsied (including microscopic
gill, skin, and fin examination) to ensure that the sailfin mollies were remaining as free of
parasites as possible and for the early detection of other problems.
Fish in the holding tanks were acclimated to water salinities of 2 g/L, 15 g/L, or
30g/L, respectively. A reservoir tank containing no fish, but holding approximately 680L
of 30 g/L salinity water was used for water changes. The water in the reservoir tank was
made with Instant Ocean salt mix and well-water and was aerated for 24 hours prior to
water use (see Appendix B for typical composition). This water was added to the tanks
to adjust the salinities upward. Well water was added to decrease salinity. Salinities
were verified using a refractometer (Aquatic Eco-Systems, Apopka, Florida).
Fish were fed twice a day, five days a week a generic bulk tricolor flake food
(Zeigler Tri-color flake, Gardners, Pennsylvania). The weight of the total fish population
in each holding tank (about 500 fish) was 1247.5 grams (2 g/L salinity water), 1252.5
grams (15 g/L salinity water), and 1249.5 grams (30 g/L salinity water). The
approximate average weight per animal was 2.5 grams. Based on those weights, fish
were fed 12.5 grams (1% of total body weight in holding tank) of food twice per day, five
days a week.
Sailfin mollies were divided into one of three different holding tanks (500 fish per
tank) that were maintained at 5 g/L salinity. Freshwater was considered 2 g/L salinity
and the holding tank being adjusted to that salinity was not adjusted until the seventeenth
day of acclimation. The salinity of the other two holding tanks was raised twice a week
in increments of 5 g/L salinity.
During acclimation an active, air-driven, sponge biofilter was maintained at each of
the salinity levels that the fish would be held (10, 15, 20, 25, and 30 g/L) during the
acclimation period as a replacement or back-up filter. Filters were started 3 weeks before
the sailfin mollies arrived; the filters were fed daily with ammonia until they were needed
in the fish holding tanks. When the salinity was increased in the holding tanks, the "old"
biofilters were replaced with a "new" active biofilter that corresponded with the new
salinity of the holding tank. Seventeen days after fish arrival the desired salinities of 2
g/L, 15 g/L, and 30 g/L, respectively, were reached. Fish were maintained in the holding
tanks for five days more before being distributed to the treatment tanks. Salinity was
monitored daily using a refractometer throughout the acclimation period and salinity was
adjusted as needed.
After acclimation to the appropriate salinity group, the fish were divided into the
thirty-six, 75.7-L treatment tanks filled to 68 liters. Twelve tanks were randomly
assigned to each of the three salinities being tested (2, 15, and 30 g/L). Fish were
monitored in the 75.7-L tanks for three days before being exposed to KMnO4. Each tank
contained an air-driven sponge filter, which also provided aeration. Each tank was
stocked with 37 fish, an average biomass of 86.4 grams per tank (fish were reweighed
with an electronic scale prior to dispersal). Treatments were assigned to individual tanks
using the complete randomized block method (Ott RL and Longnecker M 2001). A
reservoir was available for each of the three salinity treatments so that water changes
could be done when needed.
Tanks were checked twice daily for dead fish and results were recorded.
Behavior was observed at the following times post KMnO4 treatment: time zero, 0.08,
0.16, 0.33, 0.66, 1.5, 3, 6, 12, 24, 48, 96, and 168 hours. Due to physical constraints,
observations for all 36 tanks could not all be made at the same time so a two-minute
delay was instated. Specifically in applying KMnO4, the first tank was dosed at 9:00
a.m., the second at 9:02 a.m., and so on until all tanks had been dosed in succession. Five
trained observers were used to score behavior of the experimental fish. Behavior scores
were categorized as follows:
1. within normal limits fish swimming normally in midwater,
2. slight increase in opercular movement,
3. marked increase in opercular movement, fish behavior mixed between resting and
actively swimming (usually erratically and repeatedly into the bottom and sides of
4. beginning to lose the ability to maintain equilibrium or just laying on the bottom of
the tank, with obvious, labored respiration,
5. loss of equilibrium, floating throughout the tank with the current, very slow
opercular movement, fish on the verge of death.
Tissues for histologic processing were collected at three different time points: one
hour prior to KMnO4 treatment for controls, twelve hours after initiation of treatment,
and seven days post KMnO4 treatment exposure. Six whole fish per tank were taken for
each of the three histological sampling times. Prior to tissue collection, fish were
euthanized using a 1 g/L dose of buffered (sodium bicarbonate) tricaine methanesulfonate
(Finquel, Argent Laboratories, Redmond, WA) dissolved in water of corresponding
salinity, either 2, 15, or 30 g/L. Following euthanasia, a small opening was cut in the
fish's coelom and whole fish were submerged in 10% neutral buffered formalin for at
least 48 hours before processing.
Following fixation, all of the gill tissue was excised from each fish and placed in
individual cassettes for processing. The body of the fixed fish was also cut into three
transverse sections (or steaks, as in the pilot study Chapter 2) and placed in individual
cassettes for processing. All cassettes were then placed in decalcification solution (Cal-
EX, Fisher Scientific, Pittsburgh, PA) to remove any calcified material that could hinder
the microtomy of the tissue. The tissue was rinsed in tap water for four hours prior to
The tissues underwent a fourteen-hour tissue processing including alcohol (7
hours), xylene (3.5 hours), and paraffin wax (3.5 hours) see Appendix A for detailed
processing schedule. The tissue was embedded in paraffin. Tissue in the paraffin block
was cut at 5 micrometers and stained using a standard hematoxylin and eosin stain.
Slides were ready for viewing by light microscopy. Analysis consisted of scoring each
slide by scanning at 100X and examining in detail at 400X looking at approximately 50
gill filaments total using the following system:
1. Normal, includes background environmental damage (inflammation and mucous
cell infiltration) at the tip of the gill
2. Secondary lamellar damage including fusion; mucous cell and inflammatory cell
infiltration and hyperplasia, damage beginning to extend further from the tip of the
3. Expansion of lamellar interstitium by edema and inflammatory cells with lifting off
of epithelium as well as fusion of lamellae
4. Necrosis and expansion of lamellar interstitium by edema and inflammatory cells in
50% or more of the lamellae
To limit observer bias the labels of each slide were covered with a slide label with
an arbitrarily assigned number. After all of the slides were scored and recorded, the
stickers were removed so that the scores could be matched with the slide identification.
The experimental design consisted of exposing sailfin mollies in each of the three
salinities (2, 15, and 30 g/L) to three different concentrations of KMnO4, and a control
group with no KMnO4 treatment (Table 3-1). The dosages listed in Table 3-1 are
calculated doses. Each treatment was replicated three times, making up 36 treatment
Table 3-1. Experimental treatment combinations of KMnO4 concentrations and salinity
Salinity (g/L) KMnO4 Treatment (mg/L)
2 0, 0.5, 1.0, 3.0
15 0, 0.5, 1.0, 3.0
30 0, 0.5, 1.0, 3.0
Salinity was monitored daily using a refractometer. Fish were not fed during the
experiment. During this time 25% water changes were done daily.
Potassium permanganate demand was determined for each salinity level using
water from randomly selected tanks one and two days prior to the KMnO4 treatment. The
resulting six numbers were used to calculate an average KMnO4 demand for all tanks at
the corresponding salinity (Table 3-2).
Table 3-2. Calculated KMnO4 demand for each salinity level. The mean KMnO4
demand was used to calculate the actual KMnO4 dosage for all treatments.
Salinity Level Mean KMnO4 demand
Three days after the fish had been transferred to the treatment tanks, water
chemistry parameters (salinity, pH, TAN, NO2) were tested in each tank. Water
chemistry was measured using a HACH Fish Farming kit (model FF-1A, HACH
Chemical Company, Loveland, CO) for the 2 g/L salinity tanks and a HACH Marine kit
(model FF-3, HACH Chemical Company, Loveland, CO) for the 15 and 30 g/L salinity
tanks. Water temperature was measured with a mercury thermometer. Water hardness
was only tested in make-up water; six 190-L vats of make-up water were made using
Instant Ocean saltmix and well water run through a reverse osmosis system (mixing was
done with aeration).
One hour before KMnO4 treatment, each sponge filter was removed and control
histological tissue samples were collected. Aeration was continued in each tank
throughout the KMnO4 treatment period. Observations were recorded for each replicate
at the following times: time zero, 0.08, 0.16, 0.33, 0.66, 1.5, 3, 6, 12, 24, 48, 96, and 168
hours. Following KMnO4 treatment (12 hours), a 50% water change was done on each
tank, followed by the addition of 5 mL of Kordon AmQuel (San Francisco, CA) to all
tanks (including control tanks) to deactivate the KMnO4 (as indicated by the AmQuel
label). Sponge filters were then replaced. Daily 25% water changes were done
Descriptive statistics of water chemistry parameters were derived using Minitab
version 14.1 (State College, Pennsylvania). A one-way analysis of variance (ANOVA)
was performed on the nitrite and unionized ammonia values by the salinity level followed
by Tukey's HSD multiple comparison procedure. Mortality rate data was analyzed by
salinity and the KMnO4 concentration using a two-way ANOVA. The mortality rate data
was then analyzed by the salinity, KMnO4 concentration combination using one-way
ANOVA followed by Tukey's HSD multiple comparison procedure to find which
treatment concentrations were significant. An arcsin square root transformation of
percentage data was performed prior to analysis. Behavioral and histological scoring
data by the salinity, KMnO4 concentration combination were analyzed using Kruskal-
Wallis tests followed by Dunn's non-parametric multiple comparison test. Minitab
version 14.1 (State College, Pennsylvania) was used for all statistical analysis except
Dunn's multiple comparison test (Hollander and Wolfe 1973). A Type-I error rate (a) of
0.05 was used for all analyses except for Dunn's MCP. A Type-I error rate (a) of 0.15
was used, as recommended by Hollander and Wolfe (1973), for Dunn's MCP because of
the conservative nature of the test.
Water chemistry parameters varied in each treatment tank. Water quality
measurements were taken prior to the KMnO4 application (Table 4-1). The nitrite levels
in the 15 g/L salinity treatment group were found to be significantly higher when
compared to the 2 and 30 g/L salinity treatment groups (F = 19.45, DF = 2, p < 0.001).
The unionized ammonia levels in the 2 g/L salinity treatment group were found to be
significantly higher when compared to the 15 and 30 g/L salinity treatment groups (F =
50.35, DF = 2, P < 0.001). Water hardness levels in the reservoir tanks were 256.6 mg/L
for 2 g/L salinity, 2,716 mg/L for 15 g/L salinity, and 5,480 mg/L for 30 g/L salinity (n =
1 for each salinity level). The temperature was 250C in all tanks (n = 36).
Table 4-1. Select water chemistry values for experimental tanks (n = 12) prior to KMnO4
Salinity pH Unionized Nitrite (mg/L)
(g/L) Ammonia (mg/L)
2 8.33 / 0.0888 0.12 / 0.0435 0.14 / 0.0812
15 8.12/0.0389 0.02/ 0.01084 0.89/0.363
30 8.15 / 0.0522 0.03 / 0.01311 0.36 / 0.361
Data given in the format of mean / standard deviation.
Fish behavior was affected by exposure to KMnO4. There were significant
differences among treatments (H = 30.48, DF = 11, P = 0.001) (Table 4-2). Fish that
were significantly impacted showed signs of distress including increased opercular
movement, erratic swimming, beginning loss of equilibrium, and total loss of
Table 4-2. Kruskal-Wallis analysis of six-hour behavior scores by treatment
combinations (salinity KMnO4 concentration) (N = 3 tanks per treatment).
Treatment Combination: Median (see Average Rank Z
Salinity (g/L) KmnO4 footnote)
2 -0.0 0 9.3 -1.57
2 -0.5 2 13.7 -0.83
2-1.0 3 20.0 0.26
2 -3.0 4 31.5 2.23
15 0.0 0 4.0 -2.49
15 0.5 3 16.3 -0.37
15- 1.0 3 23.8 0.92
15 3.0 4 31.5 2.23
30 0.0 0 5.7 -2.20
30 0.5 2 14.7 -0.66
30-1.0 3 20.0 0.26
30 3.0 4 31.5 2.23
H = 30.48 DF = 11 P = 0.001 (adjusted for ties)
Median key: 0 = within normal limits; 2 = marked increase in opercular movement,
erratic swimming; 3 = beginning loss of equilibrium, periods of rest on bottom of tank;
4 = total loss of equilibrium.
The Dunn's multiple comparison test demonstrated that the 2 g/L salinity, 3.0 mg/L
KMnO4 treatment (mean rank 27.5), the 15 g/L salinity, 3.0 mg/L KMnO4
treatment(mean rank = 27.5), and the 30 g/L salinity, 3.0 mg/L KMnO4 treatment (mean
rank = 27.5) were significantly different indicating that the sailfin mollies in those
treatment groups lost normal equilibrium and were highly effected by the KMnO4
treatment (Table 4-3). All other treatment groups did not show significance, indicated by
a critical value less than 26.3231. Behavior that was affected by exposure to KMnO4
treatment returned to normal within 24 hours after exposure, including fish exhibiting
Table 4-3. Dunn's multiple comparison test for behavior scores shows significant
Treatment Combination: Salinity (g/L) + Difference in Mean Rank
KMnO4 Concentration (mg/L)
2 + 3.0 27.5
15 + 3.0 27.5
30 + 3.0 27.5
*A mean rank above the critical value (26.3231) indicates significance when compared to
the 15 g/L + 0.0 mg/L KMnO4 concentration (the least effected treatment or most
normal). The values were found using Dunn's multiple comparison test. All other
treatments did not show significance.
The slides of affected gill tissue showed secondary lamellar damage including
fusion and mucous and inflammatory cell infiltration and hyperplasia (Figure 4-1). More
severely affected gill tissue showed expansion of the lamellar interstitium by edema and
inflammatory cell infiltration with lifting off of the epithelium as well as secondary
lamellar fusion. The most severely affected gill tissue showed severe necrosis as well as
expansion of the lamellar interstitium by edema and inflammatory cells with lifting off of
the epithelium and secondary lamellar fusion.
There were significant differences among treatments (H=51.06, DF=11, P<0.001)
(Table 4-4). Significance in the 15 g/L salinity, 3.0 mg/L KMnO4 indicates expansion of
the lamellar interstitium by edema and inflammatory cells with lifting off of the
epithelium as well as secondary lamellar fusion (Figure 4-2). Significance in the 30 g/L
salinity, 3.0 mg/L KMnO4 treatment indicates necrosis of the gill tissue as well as
expansion of the lamellar interstitium by edema and inflammatory cells with lifting off of
the epithelium and secondary lamellar fusion (Figure 4-3). Slide scores of treated fish
returned to normal limits by the 168th hour.
Table 4-4. Kruskal-Wallis analysis of the slide score by the salinity/KMnO4 treatment
Treatment Combination N= Median Average Rank Z
Salinity (g/L) KmnO4
2-0 54 0 252.9 -2.66
2-0.5 54 1 268.9 -1.98
2-1.0 54 1 318.7 0.11
2-3.0 54 1 318.7 0.11
15-0 54 1 327.6 0.49
15-0.5 54 1 281.9 -1.44
15-1.0 54 1 348.1 1.37
15-3.0 54 1 391.9 3.20
30-0 54 0 240.5 -3.18
30-0.5 54 1 325.8 0.41
30-1.0 54 1 313.4 -0.11
30-3.0 36 2 446.3 4.41
H = 65.34 DF = 11 P < 0.001 (adjusted for ties)
Median key: 0 = within normal limits; 1 = secondary lamellar damage including fusion,
mucous and inflammatory cell infiltration and hyperplasia; 2 = secondary lamellar
damage including fusion, mucous and inflammatory cell infiltration and hyperplasia.
The 30 g/L salinity, 3.0 mg/L KMnO4 treatment groups shows n=36 due to mortality.
This table includes scores from all tissue collection times (zero, 12 hours, and 168 hours).
Table 4-5. Dunn's multiple comparison test for histology slide scores shows significantly
Treatment (g/L salinity Significantly Different Difference in Critical Value
mg/L KMnO4) From: (g/L salinity mg/L Mean Rank
15- 3.0 2- 0.0 139.0 116.535
15- 3.0 2- 0.5 123.0 116.535
15- 3.0 30-0.0 151.4 116.535
30 3.0 2 0.0 193.4 130.290
30 3.0 2 0.5 177.4 130.290
30 3.0 15 0.5 164.4 130.290
30 3.0 30 0.0 205.8 130.290
30- 3.0 30- 1.0 132.9 130.290
A mean rank above the critical value (116.535) indicates significance when compared to
the treatment indicated. All other treatments did not show significance.
U.4PE- -" U14 7-A1 Zr AU IM KI B
Figure 4-1. Gills from sailfin mollies in 2 g/L salinity water treated with KMnO4. A(200X, 0.0 mg/L
KMn04). Secondary lamellar structure is well-maintained (Slide Score = 0). B(200X, 3.0
mg/L KMnO4). Secondary lamellar fusion, as indicated by the arrow, is commonly seen in
this treatment group (Slide Score = 1). Stained by H&E.
Figure 4-2. Gills from sailfin mollies in 15 g/L salinity water treated with KMnO4. A (200X, 0.0 mg/L
KMnO4). The secondary lamellar structure is well-maintained (Slide score = 0). B(200X)
and C(400X) (3.0 mg/L KMnO4). There is expansion of the lamellar interstitium by edema
and inflammatory cells with lifting off of the epithelium as well as fusion of the lamellae
(Slide score = 2). Stained by H&E.
Figure 4-3. Gills from sailfin mollies in 30 g/L salinity water treated with KMnO4. A.
(200X, 0.0 mg/L KMnO4). Secondary lamellar structure is well-maintained
(Slide Score = 0). B(200X) and C(400X) (3.0 mg/L KMnO4). Necrosis of the
lamellae, fusion of the secondary lamellae, as well as lifting of the epithelial
layer of cells by expansion of the lamellar interstitium by inflammation and
edema was common in this treatment group (Slide score = 3). Stained by
Salinity and KMnO4 concentration were found to have some significantly
different treatment groups (Table 4-6). Certain treatment combinations were also found
to be significantly different (Table 4-7). The 30 g/L salinity, 3.0 mg/L KMnO4 treatment
had significantly higher mortality than all other treatments, with 100-percent mortality
(Figure 4-7). The 15 g/L salinity, 3.0 mg/L KMnO4 treatment had significantly higher
mortality than the 15 g/L salinity, 0.0 and 0.5 mg/L KMnO4 treatments, with 36.33-
percent mortality (Figure 4-6). The 2 g/L salinity, 3.0 mg/L KMnO4 treatment was not
found to be significantly different from other treatments, but that treatment had 11.33-
percent mortality (Figure 4-5). Other treatment groups demonstrated very low, non-
significant mortality over the observation period (168 hours) (see Figures 4.4, 4.5, 4.6,
Table 4-6. Two-way ANOVA of the total percentage mortality by the salinity and the
Source Degrees of Sum of Mean F-test P-value
Freedom Squares Square
Salinity 2 3566.2 1783.08 12.57 < 0.001
KMnO4 3 15327.2 5109.06 36.02 < 0.001
Interaction 6 9085.4 1514.23 10.68 < 0.001
Error 24 3404.0 141.83
Total 35 31382.8
Table 4-7. One-way ANOVA analysis of the total mortality percentage by the treatment
Source Degrees of Sum of Mean F-test P-value
Freedom Squares Square
Treatment 11 6.6463 0.6042 18.62 < 0.001
Error 24 0.7790 0.0325
Total 35 7.4253
0.0 0.5 1.0 3.0
0.0 0.5 1.0 3.0 0.0 0.5 1.0 3.0
Figure 4-4. Cumulative Total Percentage Mortality Over 7 days by Salinity (fresh 2 g/L,
brackish 15 g/L, and salt 30 g/L) and KMnO4 Concentration (0.0, 0.5, 1.0,
and 3.0 mg/L). Connected data points = mean.
560 70 -0.0 mg/L KMnO(sub)4
50 __ -0.5 mg/L KMnO(sub)4
S-- A- 1.0 mg/L KMnO(sub)4
40- -3.0 mg/L KMnO(sub)4
R x 3.0 mg/L KMnO(sub)4
Figure 4-5. The cumulative total percentage mortality shown over time for the 2 g/L
- 0.0 mg/L KMnO(sub)4
- 0.5 mg/L KMnO(sub)4
- A- 1.0 mg/L KMnO(sub)4
- 3.0 mg/L KMnO(sub)4
O 0z Nb bb
Figure 4-6. Cumulative total percentage mortality shown over time for the 15 g/L
- 0.0 mg/L KMnO(sub)4
- -- 0.5 mg/L KMnO(sub)4
- A- 1.0 mg/L KMnO(sub)4
- X- 3.0 mg/L KMnO(sub)4
0- 0t 10 0 0 N 'b\, O O O
Figure 4-7. Cumulative total percentage mortality shown over time for the 30 g/L
K K- -X X -x
DISCUSSION AND CONCLUSIONS
Potassium permanganate use in freshwater aquaculture is common. The
concentration at which toxic effects are seen is affected by a variety of factors such as
water quality and species specific sensitivity. The effectiveness of KMnO4 is related to
the amount of oxidizable material in the water, i.e. organic and inorganic material, which
comprises the KMnO4 demand of the water (Tucker 1984). It has also been suggested
that water of high pH can cause manganese dioxide to precipitate (Stuart 1983 and Noga
1996). The purpose of this study was to determine if KMnO4 creates toxicity problems
for sailfin mollies held in saltwater compared to cohorts maintained in freshwater. This
study examined the effect of KMnO4 on the sailfin molly maintained at different salinity
levels and at different KMnO4 concentrations.
Tanks treated with 2 g/L salinity, 3.0 mg/L KMnO4 demonstrated 11.33%
cumulative mortality (Figure 4-4) throughout the entire 168-hour experiment.
Histological results in this experimental group showed only minimal secondary lamellar
fusion and mucous and inflammatory cell infiltration and hyperplasia in this same
treatment group (Figure 4-1). Sailfin mollies in the 2 g/L salinity water treated with 3.0
mg/L KMnO4 lost equilibrium in the water, demonstrating that they were negatively
affected by the KMnO4 treatment. The 2 g/L treatments also had a high mean ammonia
level. Mortality rate in this group may have been affected by the added stress from the
high ammonia levels in the water. The 3.0 mg/L KMnO4 concentration is higher than
that of the usual recommended dose of 2.0 mg/L KMnO4. It is possible that the 3.0 mg/L
KMnO4 dose is above the safety range of KMnO4 treatment for sailfin mollies kept in 2
g/L salinity water causing minimal mortality.
Tanks treated with 3 mg/L KMnO4 compared to control tanks treated with 0.0
mg/L KMnO4, were visually different due to KMnO4 concentration (the higher the
KMnO4 concentration the more purple the color of the water). This may cause observers
to approximate the KMnO4 concentration level, potentially creating bias. The behavior
scoring system was made in an attempt to allow the observer to judge the fish solely on
their behavior without factoring in the color of the water. The combined results of
behavior, mortality rate, and histological analysis suggests that KMnO4 treatment may
not be a safe application for the 2 g/L salinity, 3.0 mg/L KMnO4 even though it is not
significantly different from other treatments.
Use of KMnO4 in 15 g/L salinity water is a higher risk treatment than it would be in
2 g/L salinity water. The histology scoring system revealed expansion of the lamellar
interstitium by edema and inflammatory cells with lifting off of the epithelium as well as
fusion of the lamellae. This damage was consistently observed in this treatment group
(Figure 4-2). The lifting of the epithelium away from the blood vessels presumably
results in decreased efficiency of gas exchange and ammonia excretion, and therefore
increases the risk of mortality. As the epithelial layer of the lamellae is moved further
from the blood vessels more space is created and it is more difficult for gas (oxygen,
CO2) and other metabolites to move through the epithelium into the blood vessel and vice
versa. This effect will at least slow gas exchange and at severe levels may stop gas
exchange thus killing the fish.
Sailfin mollies in the 15 g/L salinity, 3.0 mg/L KMnO4 demonstrated 36.33%
mortality. Mortality variability was high (minimum = 0 %, maximum = 79 %),
indicating some inconsistency between populations housed in different tanks within the
same treatment group. This may indicate water chemistry problems (high nitrite levels in
the 15 g/L salinity treatments) or that there is variability in how individual fish react to
the chemical. A 3 mg/L KMnO4 dose is higher than the recommended dose possibly
contributing to toxicity. Further testing with increased replication would be necessary to
clarify why there was so much variation in mortality among fish populations in the 15
g/L salinity water, 3.0 mg/L KMnO4 treatment group.
As in the 2 g/L salinity, 3.0 mg/L KMnO4, the 15 g/L salinity, 3.0 mg/L KMnO4
treatment groups demonstrated loss of equilibrium during the treatment period. In
conjunction with high mortality and the gill trauma observed, this suggests that the 15
g/L salinity water, 3.0 mg/L KMnO4 treatment holds a higher risk for sailfin mollies.
KMnO4 should probably not be considered as a treatment choice in 15 g/L salinity water
without conducting a small-scale animal safety test on the species being considered.
The 30 g/L salinity, 3.0 mg/L KMnO4 treatment was not safe for sailfin mollies in
this experiment. As in the 2 and 15 g/L salinity water, 3.0 mg/L KMnO4 treatment
groups, the 30 g/L salinity water, 3.0 mg/L KMnO4 treatment group fish lost equilibrium
during the treatment. The maximum behavior score of 4 indicated a loss of equilibrium,
floating throughout the tank with the current with very slow opercular movement. Loss
of equilibrium at this point may indicate that the fish is directing all of its energy towards
respiration, letting equilibrium take second priority, or is so severely debilitated and has
lost the ability to maintain its equilibrium.
Histological changes to the gill in the 30 g/L salinity water, 3.0 mg/L KMnO4
treatment were severe and given the maximum histological score of 3. Lesions included
necrosis and expansion of lamellar interstitium by edema and inflammatory cells in 50%
or more of the lamellae. Gill damage observed in this group was much more severe than
that observed in other treatment groups. The response of the gill to KMnO4 treatment, as
described above, may have reduced gas exchange to the point of death by suffocation in
fish from this treatment group.
Mortality of sailfin mollies in the 30 g/L salinity water, 3.0 mg/L KMnO4 treatment
group was significantly different from all other treatment groups tested in this
experiment. One-hundred percent mortality occurred within 6 hours of exposure to 3.0
mg/L KMnO4 in all replicate groups. The behavior, histology, and mortality results from
this experiment indicate that KMnO4 is toxic to sailfin mollies from the 30 g/L salinity,
3.0 mg/L KMnO4 treatment.
Water Chemistry Parameters
Water chemistry analyses among treatment groups were highly variable. Salinity
change can cause biological filtration activity to decrease or even discontinue (Hovanec
and DeLong 1996). Although an attempt was made to maintain consistent water quality
by adding a conditioned biofilter to each of the different salinities encountered by the
fish, water quality was variable, especially at the lower salinities. The differences in the
nitrification process may have been due to the differences in salinity affecting the rates of
bacteria infiltration of the filter. However, the highest mean UIA levels were observed in
the 2 g/L salinity and the highest NO2 levels were observed in the 15 g/L salinity. This
may have contributed to mortality data collected. The stress of coping with the water
chemistry problems added to the caustic KMnO4 treatment may have caused mortality in
some fish creating variability in the results.
Although the 15 g/L salinity, 3.0 mg/L KMnO4 treatment group had 36.33%
mortality, the highest mortality was seen in the 30 g/L salinity, 3.0 mg/L KMnO4
treatment. The 30 g/L salinity treatment group exhibited the most normal water quality
(in reference to the UIA and NO2 levels), showing that mortality in this treatment group
was probably not related to UIA or NO2 levels.
In this experiment the mean pH was 8.33 for 2 g/L salinity, 8.12 for 15 g/L salinity,
and 8.15 for 30 g/L salinity, demonstrating low variability. Previous studies (Noga 1996,
Stuart 1983) suggested that the high pH of saltwater was responsible for mortality of fish
because manganese dioxide may precipitate onto the gills. In this study the lack of
variation in pH between treatment tanks does not support this hypothesis.
Ideally pH should have been measured electronically instead of colorimetrically.
The colorimetric tests lack precision, leading to increased variability. In the future more
comprehensive water quality analyses and other methods for parameter detection should
be used that do not involve colorimetric tests.
In this experiment only a calculated dose of KMnO4 was reported. The actual dose
of KMnO4 was not tested because the spectrophotometric methods available are not
reliable when testing water containing chloride or saltwater (Delfino, University of
Florida, personal communication). There was no feasible option to determine the actual
KMnO4 concentration in all three salinities tested (2, 15, and 30 g/L, respectively).
Technology or methodology should be developed to determine the actual dose of KMnO4
in water regardless of salinity.
The objective of this experiment was to determine toxicity of KMnO4 (0.0, 0.5, 1.0,
and 3.0 mg/L) at varying salinity levels (2, 15, and 30 g/L respectively) using the sailfin
molly as a model. The study demonstrated that a concentration of 3.0 mg/L KMnO4 in
water of low organic content was not safe for the sailfin molly in 2, 15 or 30 g/L salinity
water as indicated by mean mortality rates of 11.33, 36.33 and 100 percent, respectively.
This data suggests that KMnO4 is not safe for sailfin mollies at concentrations of 3.0
mg/L. The data in this experiment also suggests that KMnO4 is a higher risk treatment in
water with salinity of > 15 g/L.
The cause of observed KMnO4 related mortality can be speculated to include an
unknown reaction between KMnO4 and the salinity, oxidation-reduction by-products that
are toxic, or attributes of the sailfin molly that may not be present in other fish. Many
elements other than sodium and chloride are present in saltwater. Potassium
permanganate may be interacting with one of these elements present in Instant Ocean salt
mix (Aquarium Systems, Mentor, Ohio) producing a new compound that causes a toxic
reaction in the fish. As indicated by Mallinckrodt Baker (2001), KMnO4 is incompatible
with bromides, iodides, ferrous salts, and arsenites, all of which are contained in Instant
Ocean salt mix (Aquarium Systems, Mentor, Ohio) in various forms (see Appendix B).
The consequences of these interactions are unknown in fish, but could be potential
hazards. This should be looked into further to rule out other chemical reactions that
would result in the fish being exposed to another compound other than KMnO4.
One example of a chemical interaction that could be taking place in saltwater
involves KMnO4 and bromide. KMnO4 causes an oxidation-reduction reaction, similar to
that of ozone (Camel and Bermond 1998). When ozone is used in saltwater systems it
can produce hypobromous acid, hypobromite ion, and then bromate through the oxidation
of bromide (Tango and Gagnon 2003). These compounds may be toxic to fish in the
water by way of a pH change. It is possible that the oxidation-reduction reaction
produced by KMnO4 could have similar byproducts that cause toxicity. In future
experiments, pH and oxidation-reduction potential should be measured throughout to
The sailfin molly used in this experiment is a euryhaline species that is able to
adapt to differing salinities well. The sailfin mollies' ability to adapt to fluctuating
salinities and ion concentrations may have affected the results of this experiment. The
addition of KMnO4 may change the ionic balance of the water compared to the blood of
the fish, impacting osmoregulation. Osmotic homeostasis must be maintained in the fish
to maintain good health. Sailfin mollies may be better suited to living in fresh to brackish
water environments as that is more of their natural distribution. It is possible that at
higher salinities even though the sailfin molly is able to survive, it is not able to flourish.
A sailfin molly in saltwater may be under more environmental stress. Therefore, when
challenged with a treatment such as KMnO4, the fish is less able to cope with the added
stress. Experiments should be completed to see if this possible cause of mortality and gill
damage is unique to KMnO4 treatment or if it encompasses other types of chemical
treatment or stressors as well.
Gill damage such as expansion of lamellar interstitium by edema and inflammatory
cells with lifting off of epithelium as well as fusion of lamellae and in severe cases
necrosis was seen in the 15 and 30 g/L salinity, 3.0 mg/L KMnO4 treatment groups.
Mortality was 11.33% in the 2 g/L salinity, 3.0 mg/L KMnO4 treatment group, 36.33%
in the 15 g/L salinity treatment group, 3.0 mg/L KMnO4 treatment group, and 100% in
the 30 g/L salinity, 3.0 mg/L KMnO4 treatment suggesting that KMnO4 at the highest
concentration tested was not safe for sailfin mollies in this experiment. Results also
suggest that as the salinity of the water increases, the toxicity of KMnO4 to sailfin mollies
PROCESSING SCHEDULE FOR THE SHANDON EXCELSIOR AUTOMATIC
Table A-1. Processing schedule for the shandon excelsior automatic tissue processor
Reagent Used: Time in hours:
Alcohol Ethyl 50-75% 1:00
Alcohol Ethyl 1:00
Alcohol Ethyl 1:00
Alcohol Ethyl 1:10
Alcohol Ethyl 1:10
Alcohol Ethyl 75-100% 1:30
Wax Paraffin 1:00
Wax Paraffin 1:15
Wax Paraffin 1:15
TYPICAL COMPOSITION OF INSTANT OCEAN SALT
Table B-1. Typical composition of instant ocean salt
Solution at Approximate Salinity of 35ppt
Instant Ocean Seawater*
ate 200 126
less than 0.04 0.014
less than 0.04 less than 0.001
less than 0.025 less than 0.001
less than 0.015 less than 0.001
less than 0.015 less than 0.001
less than 0.015 less than 0.001
less than 0.015 less than 0.001
less than 0.015 less than 0.001
less than 0.015 less than 0.002
less than 0.015 less than 0.001
less than 0.01 0.01
less than 0.006 less than 0.001
less than 0.005 less than 0.001
less than 0.004 0.002
less than 0.002 less than 0.001
* Data for seawater values taken from An Introduction to the Chemistry of the Sea.
1998. M.E.Q. Pilson
LIST OF REFERENCES
American Public Health Association, American Water Works Association, Water
Environment Federation: Standard Methods for the Examination of Water and
Wastewater, ed 20, Washington, D.C., 1998.
Balsano JS, Rasch EM, and Monaco PJ. 1989. The evolutionary ecology of Poecilia
Formosa and its triploid associate. In: Meffe GK and Snelson FF, editors. Ecology
and evolution of livebearing fishes (Poeciliidae). Englewood Cliffs : Prentice Hall.
Bishop Y, ed. 2001. The Veterinary Formulary. 5th ed. London : Pharmaceutical Press.
Boyd CE. 1979. Water Quality in Warmwater Fish Ponds. Auburn: Auburn University.
Camel V and Bermond, A. (1998). The use of ozone and associated oxidation processes
in drinking water treatment. Water Research 32 : 3208-3222.
Carpenter JW, Mashima TY, and Rupiper DJ. 2001. Exotic Animal Formulary. 2nd ed.
Philadelphia : W.B. Saunders Company. 423p.
Cams Chemical Company. 2004. Cams The World Leader in Potassium
Permanganate Technology. http://www.camschem.com Accessed 10 May 2004.
Collins MT, Gratzek JB, Dawe DL, Nemetz TG. 1975. Effects of Parasiticides on
Nitrification. Canadian Journal ofFisheries and Aquatic Sciences 32 : 2033-2037.
Courtenay WR and Meffe GK. 1989. Small fishes in strange places: a review of
introduced poeciliids. In: Meffe GK and Snelson FF, editors. Ecology and
evolution of livebearing fishes (Poeciliidae). Englewood Cliffs : Prentice Hall. p
Darwish AM, Griffin BR, Straus DL, and Mitchell AJ. 2002. Histological and
hematological evaluation of potassium permanganate exposure in channel catfish.
Journal ofAquatic Animal Health 14 : 134-144.
Engstrom-Heg R. 1971. Direct Measurement of potassium permanganate demand and
residual potassium permanganate. New York Fish and Game Journal 18 : 117-122.
Environmental Protection Agency. 1985. Pesticide Fact Sheet Number 80. Washington
D.C. : Office of Pesticides and Toxic Substances. 3p.
Francis-Floyd R and Klinger R. 1997. Use of potassium permanganate to control
external infections of ornamental fish. University of Florida Institute of Food and
Agricultural Sciences Extension Fact Sheet FA-37.
Ghosh AK and Pal RN. 1969. Toxicity of four therapeutic compounds to fry of Indian
major carps. Fishery Technology VI: 120-123.
Gustafson DL. 1981. The influence of salinity on plasma osmolality and routine oxygen
consumption in the sailfin molly, Poecilia latipinna (Lesueur), from a freshwater
and an estuarine population. Gainesville (FL) : The University of Florida. 24p.
Hollander M and Wolfe DA. 1973. Nonparametric statistical methods. John Wiley and
Sons, New York.
Hovanec TA and DeLong EF. 1996. Comparative analysis of nitrifying bacteria
associated with freshwater and marine aquaria. Applied and Environmental
Microbiology 62 : 2888-2896.
Lawrence JM. 1956. Preliminary results on the use of potassium permanganate to
counteract the effects of rotenone on fish. The Progressive Fish-Culturist 18 : 15-
Lay BA. 1971. Applications for potassium permanganate in fish culture. Transactions
of the American Fisheries Society 100 : 813-816.
Levine G and Meade TL. 1976. The Effects of Disease Treatment on Nitrification in
Closed System Aquaculture. In: World Mariculture Society, editor. Proceedings
of the seventh annual meeting, World Mariculture Society. San Diego, California,
January 25-29, p. 483-491.
Mallinckrodt Baker, Inc. 2001. Material Safety Data Sheet Potassium Permanganate.
Mallinckrodt Chemicals / J.T. Baker website.
http://www.itbaker.com/msds/englishhtml/p6005.htm. Accessed 2004 May 10.
Marking L and Bills T. 1975. Toxicity of potassium permanganate to fish and its
effectiveness for detoxifying antimycin. Transactions of the American Fisheries
Society 104 : 579-583.
Masser MP and Jensen JW. 1991. Calculating treatments for ponds and tanks. Southern
Regional Aquaculture Center Publication No. 410.
Meffe GK and Snelson FF. 1989. An ecological overview of poecilid fishes. In: Meffe
GK and Snelson FF, editors. Ecology and evolution oflivebearing fishes
(Poeciliidae). Englewood Cliffs: Prentice Hall. p 13-32.
Noga EJ. 1996. Fish Disease: diagnosis and treatment. St. Louis : Mosby-Year Book,
Nordlie FG, Haney DC, and Walsh SJ. 1992. Comparisons of salinity tolerance and
osmotic regulatory capabilities in populations of sailfin molly (Poecilia latipinna)
from brackish and fresh waters. Copeia 1992 : 741-746.
Ott RL and Longnecker M. 2001. An introduction to statistical methods and data
analysis. Pacific Grove: Duxbury. 1152p.
Parenti LR and Rauchenberger M. 1989. Systematic overview of the poeciliines. In:
Meffe GK and Snelson FF, editors. Ecology and evolution of livebearing fishes
(Poeciliidae). Englewood Cliffs : Prentice Hall. p 3-12.
Petrovicky I. 1988. Aquarium fish of the world. London : The Hamlyn Publishing
Group Limited. 499p.
Plumb JA. 1999. Health maintenance and principal microbial diseases of cultured fishes.
Ames : Iowa State University Press. 328p.
Reardon IS and Harrell RM. 1994. Effects of varying salinities on the toxicity of
potassium permanganate to larval and juvenile striped bass, Morone saxatilis
(Walbaum). Aquaculture and Fisheries Management 25 : 571-578.
Robins R. 2003 Oct 20. Biological profile of the sailfin molly. Florida Museum of
Natural History website.
Accessed 2004 May 6.
Spotte S. 1992. Captive seawater fishes : science and technology. New York : John
Wiley & Sons, Inc. 942p.
StoskopfMK. 1993. Fish Medicine. Philadelphia: W.B. Saunders Company. 882p.
Straus DL and Griffin BR. 2002. Efficacy of potassium permanganate in treating
ichthyophthiriasis in channel catfish. Journal ofAquatic Animal Health 14 : 145-
Stuart N. 1983. Treatment of Fish Diseases. The Veterinary Record 112 : 173-177.
Tango MS and Gagnon GA. 2003. Impact of ozonation on water quality in marine
recirculating systems. Aquaculture Engineering 29 : 125-137.
Thomas-Jinu S and Goodwin AE. 2004. Acute columnaris infection in channel catfish,
Ictaluruspunctatus (Rafinesque): efficacy of practical treatments for warmwater
aquaculture ponds. Journal ofFish Diseases 27 : 23-28.
Tucker CS. 1984. Potassium permanganate demand of pond waters. Progressive Fish-
Culturist 46 : 24-28
Tucker CS. 1989. Method for estimating potassium permanganate disease treatment
rates for channel catfish. The Progressive Fish-Culturist 51 : 24-26.
Tucker CS and Boyd CE. 1977. Relationship between potassium permanganate
treatment and water quality. Transactions of the American Fisheries Society 106 :
Emily N. Marecaux grew up in Ashland, Maine. After graduating from high school
she attended college at The University of Findlay in Findlay, Ohio. While at The
University of Findlay she pursued a Bachelor of Science degree majoring in biology, pre-
veterinary science, with a minor in chemistry. After graduating in May of 2002, Emily
came to The University of Florida to work on a Master of Science degree from the
Department of Fisheries and Aquatic Sciences with a focus on fish health. Emily has
accepted a position at the University of Arkansas at Pine Bluff as a fish health extension