This item is only available as the following downloads:
RESPONSES MADE BY THE SALT MARSH TELEOST CYPRINODON
VARIEGATUS (ATHERINOMORPHA: CYPRINODONTIDAE) TO LIFE IN A
VARIABLE SALINITY ENVIRONMENT
DENNIS CHARLES HANEY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
/-l /" /
The work described in this dissertation was the focus of my activities for many
years. Throughout this time I have been fortunate to meet and interact with an extraordinary
variety of people, many of whom played integral roles in the successful completion of this
I would first like to acknowledge the assistance of the members of my graduate
committee: Frank Nordlie, Steve Walsh, Brian McNab, Harvey Lillywhite, Ken Sulak, and
Tom Crisman. I would also like to thank a founding member of my committee, Michelle
Wheatly, who, unfortunately for me, left the University of Florida before I was able to
complete my dissertation. All of these people were instrumental in the completion of this
work. In particular, I owe an incredible debt of gratitude to the chair of my committee, Frank
Nordlie. His unfailing friendship, help, and guidance over the years was truly an inspiration
to me, and I can safely say I would not have made it through to the end without him. Steve
Walsh also deserves special recognition for his much-needed assistance. Steve has been a
close friend and colleague for many years, and I am proud to be the first student to complete
a dissertation with Steve as a member of his committee!
So many members of the Zoology Department have helped out that its difficult to
know where to begin. Brent and Sylvia Palmer, John Matter, Frank Hensley (my fusiform
twin!), and Lou Somma were all invaluable in helping me to maintain good mental health
over the years. We spent many hours trying to figure out life, the universe, and everything.
We also had lots of fun looking for unsuspecting fish and herps. Lisa Gregory has been a
friend I could always count on no matter what! John Binello was my trusted field
companion. We spent many days tromping through the salt marsh (and other bodies of
water) in search of numerous fishes including the wily pupfish, sailfin molly, Florida
flagfish, and fat sleeper. Thanks also go to Leo Nico, Adele Hensley, Pam Fuller, Paula
Cushing, Patricia Harrison, Lianna Jarecki, Becky Thompson, Vince DeMarco, Kevin
Baldwin, Ellen Burroughs, Chris Kardish, Mark Hostetler, Doug Weaver, Frank Jordan,
John Anderson, Carol Binello, and everyone else from Zoology!
For the past 4+ years I have worked with many people at the Department of the
Interior's Gainesville laboratory (presently the National Biological Service, formerly U.S.
Fish and Wildlife Service). While working full-time for the past few years certainly slowed
down my progress towards finishing my Ph.D., the extra time was worth it (mostly!). I
formed many new friendships and learned lots of things I wouldn't have otherwise. Jim
Williams showed me that freshwater clams were actually kind of neat. Noel Burkhead
helped remind me that fish were still much cooler! Les Parker, Jayne Brim-Box, and I had
great fun diving in zero visibility water. Howard Jelks, Gary Hill, Ann Foster, Tina Yanchis,
Rob Whiteford, and Bill Stranghoener were all inspiring at one time or another. Special
thanks go to my office mates, Leslie Straub and Cindy Timmerman, for putting up with me
for so long. Cindy especially has been a great friend and confidante.
Last, but certainly not least, I want to thank my family for supporting me all these
years. Mom and Dad never doubted me, though it seemed like I would never finish. Their
encouragement really did help! My brother Scott helped to show me you really can finish a
dissertation and be successful afterwards. For this, and everything else, I can never thank
TABLE OF CONTENTS
LIST OF TABLES ........................................................................................................... vi
LIST OF FIGURES........................................................................................................ vii
ABSTRACT ......................................................................................................................... x
1 INTRODUCTION ................................................................................................. 1
Study Site ............................................................................................................... 1 1
2 INFLUENCE OF ENVIRONMENTAL SALINITY ON ROUTINE
METABOLIC RATE AND CRITICAL OXYGEN TENSION OF
CYPRINODON VARIEGATUS ..................................................................... 12
Introduction ............................................................................................................ 12
M ethods............................................................................................................. 14
Discussion ........................................................................................................... 31
3 INFLUENCE OF SIMULATED TIDAL CHANGES IN AMBIENT
SALINITY ON ROUTINE METABOLIC RATE IN CYPRINODON
VARIEGATUS ............................................................................................... 39
Introduction ......................................................................................................... 39
M ethods................................................................................................................ 40
4 INFLUENCE OF A FLUCTUATING SALINITY REGIME ON
OSMOREGULATION IN CYPRINODON VARIEGATUS.........................55
Introduction ......................................................................................................... 55
M ethods............................................................................................................. 56
Discussion ........................................................................................................... 66
5 INFLUENCE OF ENVIRONMENTAL SALINITY ON BLOOD
OXYGEN LEVELS OF CYPRINODON VARIEGATUS..............................70
Introduction ......................................................................................................... 70
M ethods............................................................................................................. 71
D discussion ........................................................................................................... 79
6 SU M M A RY A N D CO N CLU SIO N S ................................................................ 87
A PPEN D ICES ...................................................................................................................91
1 CRITICAL OXYGEN TENSION FIGURES....................................................91
2 FIELD M EA SUREM ENTS................................................................................102
LITERA TURE CITED ................................................................................................... 108
BIO G RA PHICA L SK ETCH ........................................................................................ 123
LIST OF TABLES
1-1. Phylogenetic classification of the cyprinodontiform fishes (modified from
2-1. Relationships of routine metabolism (RMR), critical oxygen tension (Pc), and
slope in the conformation region at a series of ambient salinities. Values are given
as m means se ........................................................................................................ 19
2-2. Measurements of oxygen concentration (mg L- 1), salinity (ppt), and temperature
(oC) taken at four sites in the Cedar Key area from June 1990 through June 1991
(see text for details on location of sites). Values are given as means, se (in
parentheses), sam ple size........................................................................................... 32
3-1. Acclimation and final salinities used in simulated tidal change study........................42
3-2. Mean routine metabolism (mg 02 h-1) before (acclimation salinity) and
following (final salinity) a simulated tidal change. Values are given as means se.
Groups exhibiting a significant change in metabolism are indicated with an
4-1. Salinity trials used in cyclical salinity study. The group maintained at 30 ppt was
split into three groups following cycle 10 (day 20); groups CD, C, and CQ (see text
4-2. Results of salinity fluctuations experiment. Values in the top row of each cell
represent hematocrit measurements (% erythrocytes), values in bottom row of each
cell represent plasma osmolality measurements (mOsm kg-1). Sample sizes are
n=5 for each cell. All values are expressed as means se. See text for explanation
of group abbreviations............................................................................................... 60
5-1. Hematocrit (Hct), hemoglobin concentration ([Hb]), erythrocyte count (RBC),
mean corpuscular hemoglobin (MCH), mean corpuscular volume (MCV), and
mean corpuscular hemoglobin concentration (MCHC) as a function of salinity for
Cyprinodon variegatus. All values are expressed as means se. See text for
further details on blood indices .................................................................................. 74
LIST OF FIGURES
2-1. Mean adjusted routine metabolic rates (RMR) over a range of salinities in
Cyprinodon variegatus (metabolic rates were mass-adjusted using an analysis of
covariance; bars indicate se; numerical values above the points in the figure
indicate sample sizes at each salinity)......................................................................... 21
2-2. Relationship between mean adjusted routine metabolic rates (RMR) and mean
plasma osmolality over a range of salinities in Cyprinodon variegatus (metabolic
rates were mass-adjusted using an analysis of covariance; bars indicate se;
plasma osmolality data from Nordlie, 1985) .............................................................. 23
2-3. Mean critical oxygen tension (Pc) measurements over a range of salinities in
Cyprinodon variegatus (bars indicate se; numerical values above the points in
the figure indicate sample sizes at each salinity)......................................................... 26
2-4. Relationship between mean adjusted routine metabolic rates (RMR) and critical
oxygen tensions (Pc) over a range of salinities in Cyprinodon variegatus
(metabolic rates were mass-adjusted using an analysis of covariance; bars
indicate se). ......................................................................................................... 28
2-5. Relationship between mean critical oxygen tensions (Pc) and mean plasma
osmolality over a range of salinities in Cyprinodon variegatus (bars indicate se;
plasma osmolality data from Nordlie, 1985). ............................................................. 30
2-6. Relationship between mean adjusted routine metabolic rates (RMR) and mean
plasma osmolality over a range of salinities in Adinia xenica (metabolic rates were
mass-adjusted using an analysis of covariance; bars indicate se; numerical
values above the points in the figure indicate sample sizes at each salinity)................35
3-1. Schematic diagram of respirometry apparatus used for routine metabolism
experiments. See text for detailed description of system............................................44
3-2. Results of metabolic trials where salinity was increased over the course of the
trial. Bars represent groups listed in Table 3-2 for which final salinity was greater
than initial salinity. The height of each bar signifies the magnitude of the salinity
change for each metabolic trial and the asterisk indicates at which of the salinities
(for each metabolic trial) the routine metabolic rate (RMR) was highest. The x axis
has no scale and serves only to visually separate groups........................................... 49
3-3. Results of metabolic trials where salinity was decreased over the course of the
trial. Bars represent groups listed in Table 3-2 for which final salinity was less
than initial salinity. The height of each bar signifies the magnitude of the salinity
change for each metabolic trial and the asterisk indicates at which of the salinities
(for each metabolic trial) the routine metabolic rate (RMR) was highest The x axis
has no scale and serves only to visually separate groups.. ......................................... 51
4-1. Mean plasma osmolality values measured for groups experiencing decreases in
salinity during the course of the experiment. Group designations are as follows:
DI, salinity fluctuated between 30 ppt and 2 ppt; D2; salinity fluctuated between
30 ppt and 10 ppt; D3, salinity fluctuated between 30 ppt and 20 ppt; CD; salinity
constant at 30 ppt for days 0-20, decreased to 2 ppt following subsample on day
20; C, salinity constant at 30 ppt Sample sizes are n=5 for each group. See
text for details of experimental procedure.................................................................. 62
4-2. Mean plasma osmolality values measured for groups experiencing increases in
salinity during the course of the experiment Group designations are as follows:
I1, salinity fluctuated between 30 ppt and 40 ppt; I2; salinity fluctuated between 30
ppt and 50 ppt; 13, salinity fluctuated between 30 ppt and 60 ppt; CI; salinity
constant at 30 ppt for days 0-20, increased to 60 ppt following subsample on day
20; C, salinity constant at 30 ppt. Sample sizes are n=5 for each group. See text
for details of experimental procedure......................................................................... 64
5-1. Mean erythrocyte (RBC) count over a range of salinities in Cyprinodon
variegatus (bars indicate se; numerical values above the points in the figure
indicate sample sizes at each salinity)......................................................................... 76
5-2. Mean hemoglobin concentration ([Hb]) over a range of salinities in Cyprinodon
variegatus (bars indicate se; numerical values above the points in the figure
indicate sample sizes at each salinity)......................................................................... 78
5-3. Mean hematocrit (Hct) over a range of salinities in Cyprinodon variegatus (bars
indicate se; numerical values above the points in the figure indicate sample sizes
at each salinity). .......................................................................................................81
Al-1. Plot indicating the calculation of the critical oxygen tension (Pc) for an
individual Cyprinodon variegatus in water at 0 ppt ..................................................93
A 1-2. Plot indicating the calculation of the critical oxygen tension (Pc) for an
individual Cyprinodon variegatus in water at 50 ppt ................................................95
A 1-3. Plot indicating the calculation of the critical oxygen tension (Pc) for an
individual Cyprinodon variegatus in water at 100 ppt ..............................................97
A 1-4. Generalized critical oxygen tension (Pc) plots at each salinity used in the Pc
experiments. Plots were produced by using the mean Pc, mean routine metabolic
rate (RMR), and mean slope in the conformation region for each salinity group .........99
A2-1. Oxygen concentration (mg L'l), salinity (ppt), and temperature (oC) at four
sites in the Cedar Key area taken between June 1990 and June 1991. Values are
given as means, bars indicate + se. a) Measurements taken on the bottom at site 1;
b) measurements taken on the surface at site 1; c) Measurements taken on the
bottom at site 2; d) measurements taken on the surface at site 2; e) Measurements
taken on the bottom at site 3; f) measurements taken on the surface at site 3; g)
M easurements taken on the surface at site 4 ............................................................ 104
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
RESPONSES MADE BY THE SALT MARSH TELEOST CYPRINODON
VARIEGATUS (ATHERINOMORPHA : CYPRINODONTIDAE) TO LIFE IN A
VARIABLE SALINITY ENVIRONMENT
DENNIS CHARLES HANEY
Chairman: Frank Nordlie
Major Department: Zoology
Cyprinodon variegatus, a common coastal resident of the western Atlantic Ocean
and Gulf of Mexico, lives in ambient salinities ranging from freshwater to 142 ppt. Fish
used in this study were obtained from a Gulf of Mexico salt marsh near Cedar Key, Florida.
In a steady-state experiment, routine metabolic rate (RMR) and critical oxygen tension (Pc)
were determined at salinities ranging from 0 tol00 ppt. Salinities between 0 and 40 ppt had
little influence on RMR or Pc. However, at salinities above 40 ppt, RMR declined, and Pc
increased. The reduction in RMR and rise in Pc correlates with a reduced ability of C.
variegatus to osmoregulate effectively at high salinities. The variations in RMR and Pc at
high salinities suggests that C. variegatus responds by reducing energy expenditures,
effectively increasing the time that individuals can tolerate hypersaline conditions. The
metabolic patterns of C. variegatus as influenced by simulated tidal changes in salinity were
then measured, with RMR unaffected by changes in salinity between 2 and 40 ppt.
However, at extremely high or low salinities metabolism was affected by changes in ambient
salinity. Individuals of C. variegatus responded to fluxes at salinity extremes by reducing
general activity and energy expenditures-essentially waiting for conditions to return to
normal, where they responded by increasing metabolic activity. Cyprinodon variegatus
efficiently regulated plasma osmolality, even when fishes were exposed to large fluctuations
in salinity. However, prior exposure to salinity fluctuations did impart an osmoregulatory
advantage. Fish previously exposed to large salinity fluctuations regulated plasma
osmolality better than fish that previously experienced no or small changes in salinity.
Increasing salinity had a greater impact on osmoregulation than did decreasing salinity.
Salinity also had a significant effect on blood oxygen carrying capacity in C. variegatus,
although differences were only noted at the very highest (60 80 ppt) and lowest (0 ppt)
salinities tested. Oxygen carrying capacity was highest in the group acclimated to 0 ppt.
Erythrocyte count was the most consistent indicator of the influence of salinity on blood
oxygen, with hematocrit the least consistent measure.
Salinity is a crucial physicochemical factor that exerts an important influence on
aquatic life, particularly on estuarine and salt marsh organisms that are exposed to
unpredictable salinity fluctuations diurnally and seasonally. Salt marshes are intertidal beds
of rooted vegetation that are alternately flooded and exposed by rising and falling tides.
Vegetation on higher ground develops a complex network of branching channels through
which water, nutrients, and aquatic organisms move during the tidal cycle. Intertidal salt
marshes are inhospitable yet interesting environments due to their unstable water levels and
to the extreme variability of their physical, chemical, and biological processes (Cooper,
1974; Adam, 1990; Allen and Pye, 1992).
Factors that may influence salinity in salt marshes include precipitation, wind,
frequency and extent of tides, shoreline height, and coastal topography (Remane and
Schlieper, 1971; Wheatly, 1988). Due to the harshness of the physical environment, salt
marshes tend to have low species diversity, but often high abundance of selected species
(McClusky, 1989; Dunson and Travis, 1994).
Whereas variable salinity habitats are also characterized by fluctuations in other
environmental factors, such as temperature, dissolved oxygen, and pH (Vernberg, 1983;
Wheatly, 1988), the distribution and abundance of fishes in these habitats is largely
determined by salinity (McClusky, 1989; Davenport and Sayer, 1993; Gill and Potter,
1993). Not surprisingly, animals respond to fluctuations in salinity in complex ways.
Salinity affects osmoregulation, ventilation, metabolism, acid-base balance, growth,
reproduction, development, and other biological processes (Wheatly, 1988). Some of the
primary responses of teleosts to changes in salinity are reviewed here.
Avoidance is the first line of defense to variable salinity conditions. If behavioral
responses do not sufficiently minimize exposure to variable salinities, aquatic organisms
must rely on physiological and biochemical responses to tolerate environmental changes
(Beitinger and McCauley, 1990). This may be by passive tolerance (osmoconforming) or
active osmoregulation (Truchot, 1987; McClusky, 1989). Only fishes that are capable of
osmoregulation can tolerate wide changes in salinity, and these euryhaline teleosts will be
the focus of the rest of this review.
The basic patterns of osmoregulation in fishes have been extensively reviewed in
recent years (Parvatheswararao, 1970; Eddy, 1982; Evans, 1984; Zadunaisky, 1984;
Karnaky, 1986; Foskett, 1987; Pisam and Rambourg, 1991; Ventrella et al., 1992; Evans,
1993; McCormick, 1994; Wood and Marshall, 1994). Euryhaline teleosts regulate their
blood osmolality at about one-third the concentration of seawater (35 ppt), and thus face
severe osmotic problems whether in freshwater (0 ppt) or seawater. Body fluids of a teleost
in freshwater are hyper-osmotic to the external environment, whereas in seawater they are
hypo-osmotic. Thus, euryhaline fish possess mechanisms for osmoregulating in both
hyper-osmotic and hypo-osmotic conditions.
Teleosts in seawater are susceptible to a loss of body water to the external
environment, and balance water loss by actively drinking large amounts of seawater.
However, both water and salts are absorbed together across the gut. Ingested excess salts
are actively excreted, divalent ions mostly in urine and feces, and monovalent ions by the
Active excretion of salts by the gills takes place via chloride cells (Zadunaisky, 1984;
Kamaky, 1986; Foskett, 1987; Pisam and Rambourg, 1991). Chloride cells in seawater-
acclimated fish are located at the base of the secondary gill lamellae, and are large,
columnar-shaped cells usually extending from the basal epithelium to the external
environment. They are characterized by numerous mitochondria (for this reason they are
often referred to as "mitochondria rich cells") and an extensive tubular reticulum continuous
with the basolateral membrane (Pisam and Rambourg, 1991). The transport enzyme
Na+,K+-ATPase is restricted to this tubular system and to the basolateral membrane. This
ion pump creates a large Na+ gradient (low in the chloride cell cytoplasm) which drives a
NaCI co-transporter by which Cl- enters the cell. The Cl- accumulates sufficiently in the cell
such that it is able to exit to the external environment across the apical membrane, with Na+
following passively down its electrochemical gradient between adjacent chloride cells
(Evans, 1993; McCormick, 1994; Wood and Marshall, 1994).
In freshwater, teleosts continuously face an efflux of salts and an influx of water.
Hence, their osmoregulatory response is to actively transport salts from the external
environment via the gills, to avoid drinking water, and to excrete copious amounts of dilute
urine. Our present understanding of the mechanisms involved in the uptake of Na+ and Cl-
from the environment in freshwater teleosts is somewhat unclear and incomplete. It appears
that Cl- is actively taken up, with Na+ uptake being a passive process. Whether this takes
place via a freshwater-type chloride cell, or through interactions with acid-base balance is
unclear, as evidence exists for both (Wood and Marshall, 1994).
What is unique about osmoregulation in euryhaline fish is not that they possess a
structurally distinct osmoregulatory mechanism, but that they have the ability to function
efficiently in variable salinities. They must not only be able to hypo- and hyper-
osmoregulate, but they must be able to alter their pattern of osmoregulation quickly if they
live where salinity rapidly fluctuates.
Osmoregulation is controlled largely by the endocrine system. In teleosts, hormones
known to exert osmoregulatory effects include prolactin, growth hormone, cortisol,
angiotensin II, arginine vasotocin, atrial natriuretic peptide, thyroid hormones, urotensins,
vasoactive intestinal peptide, insulin, calcitonin, catecholamines and sex steroids (Jenkins,
1981; Bern and Madsen, 1992; Leloup and Lebel, 1993; Sakamoto et al., 1993; Takei, 1993;
McCormick, 1994). The hormone most involved with osmoregulation under freshwater
conditions is prolactin, with arginine vasotocin, urotensin II, catecholamines, and atrial
natriuretic peptide also playing roles (Hirano et al., 1987; Bern and Madsen, 1992; Takei,
1993). Control of osmoregulation in seawater is largely via the effects of cortisol and
growth hormone, together with thyroid hormones, angiotensin II, vasoactive intestinal
peptide, atrial natriuretic peptide, and urotensin I (Balment et al., 1987; Bern and Madsen,
1992; Takei, 1993). Although the exact mechanisms and interactive effects of many of these
hormones are unclear, it is well established that both rapid and long term control of
osmoregulation under hyper-osmotic and hypo-osmotic conditions is mediated via the
Osmoregulation is not the only physiological process affected by salinity. Salinity
adaptation is a complex event that involves a number of physiological and behavioral
responses, including energetic. Many studies have shown that salinity influences
metabolism of fishes, but little agreement exists with respect to the magnitude or direction of
these effects (Nordlie, 1978; Febry and Lutz, 1987; Nordlie et al., 1991; Swanson, 1991;
Morgan and Iwama, 1991; Kirschner, 1993). One reason for a lack of agreement among
studies is that many have focused on the energetic costs associated with osmoregulation,
while ignoring other related physiological and behavioral processes. For example, many
studies have used measured differences in metabolism at different salinities to represent
differences in osmoregulatory costs. The rationale is that osmoregulation is an energetically
costly event, with the energetic costs being related to the osmotic gradient between the fish
and its environment (e.g., Madan Mohan Rao, 1968; Muir and Niimi, 1972; Furspan et al.,
1984; Zadunaisky, 1984). However, this is probably an oversimplification, since other
factors, such as activity, food intake, and permeability changes may also be influenced by
salinity, and in turn, affect metabolism (Swanson, 1991). For this reason metabolic
measurements represent the overall costs associated with living in a particular salinity
environment, with comparisons among salinities reflecting these total costs, not simply the
cost of osmoregulation.
Several attempts have been made to categorize the patterns of metabolic responses
of teleosts to altered salinities (Kinne, 1967; Remane and Schlieper, 1971; Nordlie, 1978;
Morgan and Iwama, 1991). One commonly observed pattern is that the lower rates of
metabolism in response to salinity are associated with environments for which the species
(and life stage) are presumably best adapted for, and in which they are normally found.
Teleosts that inhabit widely variable salinity environments are uniquely characterized by
having metabolic rates that are constant over a wide range of salinities. This allows
euryhaline teleosts to avoid the large metabolic costs normally associated with physiological
adjustment to changing salinity.
Another physicochemical parameter that may act synergistically with salinity is
dissolved oxygen. The dissolved oxygen content of many bodies of water is subject to large
natural fluctuations. This is especially true in shallow water, such as salt marshes, where
chronic and/or periodic hypoxia may be common (Renaud, 1985; Dejours, 1987;
Toulmond, 1987; Graham, 1990). Variations in oxygen can dictate the distribution of some
species in aquatic ecosystems (Boutilier, 1990). Changing environmental salinity may
directly influence respiratory function by affecting both the oxygen solubility in water
pumped over the gills and the solubility of gases dissolved in plasma. Changes in the ionic
composition of the body fluids could also interact with oxygen to influence tolerance to
variable salinity conditions (Truchot, 1987).
Exposure of fish to reduced oxygen tensions initiates physiological responses
mostly directed at increasing the amount of oxygen available to the tissues (Boutilier et al.,
1988). The transfer of oxygen from the environment to the tissues can be characterized as a
series of processes. Gill ventilation is the first step, followed in turn by branchial diffusion,
blood oxygen transport, and diffusion into the tissues (the last of which, due to a lack of
available information, will not be discussed further) (Perry and McDonald, 1993).
Gill ventilation is a function of the frequency and depth of breathing, and is
normally increased when low oxygen tensions are encountered. While ventilation does not
appear to be directly limiting to the uptake of oxygen, substantial cost is involved with
increased ventilatory pumping, which ultimately means that any additional oxygen acquired
is used to fuel the ventilatory apparatus itself (Boutilier et al., 1988; McMahon, 1988;
Cameron, 1989; Perry and McDonald, 1993).
Two processes are largely available to increase branchial oxygen diffusion:
increases in functional gill surface area and increases in the mean water to blood oxygen
partial pressure gradient. This tradeoff is particularly important in regard to salinity, as fish
in waters of low oxygen tension must balance the advantage of maximizing branchial
oxygen diffusion with a disadvantage in osmoregulation due to the accompanying increases
in ion and water exchange (Perry and McDonald, 1993). Increasing blood gas transport is
likely the primary route used by most fish to increase the amount of oxygen delivered to the
tissues. Oxygen transport by the blood in teleosts depends on the respiratory pigment
hemoglobin. Blood oxygen transport is normally increased by increasing the concentration
of hemoglobin, increasing the number of erythrocytes in circulation, and/or by adjusting the
affinity of hemoglobin for oxygen (Davis, 1975; Wells et al., 1989; Jensen et al., 1993;
Perry and McDonald, 1993).
All of the processes described above can be modified to optimize oxygen transport
under a variety of environmental conditions. One additional strategy that can be utilized in
conjunction with the above is the lowering of metabolism in concert with reductions in
oxygen. This potentially minimizes the impact of the lowered oxygen tension, but also
reduces aerobic metabolism and therefore the amount of energy available for physiological
Most fish would be described as metabolic oxygen regulators, as they maintain a
constant metabolic rate over a range of oxygen tensions extending downward from
atmospheric levels to some low level that has been defined as the critical oxygen tension
(Pc). Below the Pc, metabolism is dependent upon oxygen tension, and decreases linearly
with decreases in oxygen. The Pc was probably first documented by Hall (1929), but was
not considered an important index for fishes until formalized by Fry (1947). However, this
is an extremely important variable relating to habitat selection and overall metabolic patterns
of fishes, and calculations of the Pc have since been made for a number of species under a
variety of environmental conditions.
The subject of this study was the sheepshead minnow, Cyprinodon variegatus.
Cyprinodon variegatus is a member of the family Cyprinodontidae, a large and diverse
family containing over 650 species in 80 genera (Parenti, 1981; Parker and Kornfield, 1995;
see Table 1-1). Members of this family are found in fresh, brackish, and salt water, and
distributed pantropically as well as throughout North America. Cyprinodon variegatus is
the type species for both the genus Cyprinodon and the family Cyprinodontidae.
The genus Cyprinodon comprises a group of approximately 30 species of small,
oviparous fishes commonly known as pupfishes. They exhibit remarkable tolerance to
harsh environmental conditions (Miller, 1981) occurring throughout North and Central
America, the Caribbean Sea, and Venezuela (Darling, 1976; Turner and Liu, 1977; Parenti,
1981; Duggins et al., 1983; Barus and Wohlgemuth, 1993). The genus is characterized by
limited genetic divergence and few unique alleles, even in morphologically distinct species
(Kodric-Brown, 1989). Much of the research on the genus has focused on species
inhabiting North American deserts, where the largest concentration of species occurs. Most
of these desert species have small, allopatric distributions (Miller, 1981; Duggins et al.,
1983). Members of this genus also occur in the coastal brackish and marine waters of
eastern North America, where C. variegatus is the dominant pupfish species.
Cyprinodon variegatus is the only pupfish species having an extensive geographic
range. It is found along the Atlantic coast from Massachusetts to the Florida Keys, and
throughout the Gulf of Mexico. A disjunct population occurs along the Yucatan peninsula
(Johnson, 1974; Darling, 1976; Duggins et al., 1983). Populations are also located in the
Table 1-1. Phylogenetic classification of the cyprinodontiform fishes (modified from
Bahamas, West Indies, and Cuba (Duggins et al., 1983). Cyprinodon variegatus has also
been introduced into several areas, where they have negatively impacted native pupfish
species (Echelle and Echelle, 1987; Echelle and Connor, 1989; Kodric-Brown, 1989; Wilde
and Echelle, 1992). Genetic variability within C. variegatus is as large as the amount of
genetic divergence within the entire genus Cyprinodon, with most of this genetic divergence
occurring in populations north of Cape Hatteras-very little genetic divergence is displayed
among southern populations (Darling, 1976; Schwartz et al., 1990).
Cyprinodon variegatus is a numerically dominant and ecologically important
species throughout most of its range, especially in salt marsh and estuarine waters (Kilby,
1955; Simpson and Gunter, 1956; Relyea, 1975; Naughton and Saloman, 1978;
Subrahmanyam and Coultas, 1980; Stout, 1985; Nelson, 1992; Ross and Doherty, 1994).
Due to its importance, C. variegatus has been an important research animal in diverse
disciplines. These include investigations on the species behavior (Itzkowitz, 1974; Itzkowitz,
1978; Mettee and Beckham, 1978; Itzkowitz, 1981, Dwyer and Beulig, 1991), ecology (Doll
and Bast, 1969; Martin, 1970; Martin, 1972; Able, 1976; Harrington and Harrington, 1982;
Fyfe, 1985; Shipley, 1991; Avila et al., 1992; Wright et al., 1993), evolution (Elder and
Turner, 1994), life history (Warlen, 1964; De Vlaming et al., 1978; Berry, 1987; Able, 1990;
Echelle and Echelle, 1994), physiology (Martin, 1968; Karnaky et al., 1976;
Subrahmanyam, 1980; Nordlie, 1985; Barton and Barton, 1987; Nordlie, 1987; Peterson
and Gilmore, 1988; Nordlie and Walsh, 1989; Peterson, 1990; Price et al., 1990; Nordlie et
al., 1991; Dunson et al., 1993), and reproduction (Raney et al., 1953; Warlen, 1964; Berry,
1987; Kodric-Brown, 1987; Conover and DeMond, 1991). The species has also been used
in voluminous toxicology experiments because of its extreme hardiness (e.g., Schimmel and
Hansen, 1975; Hawkins et al., 1984; Battalora et al., 1985; Linton, 1992).
Cyprinodon variegatus has been called "the toughest fish in North America"
(Gunter, 1967) due to its extreme tolerance of harsh environmental conditions. It is found in
waters ranging from freshwater (Ager, 1971; Johnson, 1974; Kushlan, 1980) to salinities of
142 ppt (Simpson and Gunter, 1956), although it typically inhabits brackish water and
coastal salt marshes. Cyprinodon variegatus is tolerant of temperatures ranging from about
1 oC (Berry, 1987), to 41 oC (Strawn and Dunn, 1967), and oxygen levels approaching
anoxia (Odum and Caldwell, 1955). Thus, it is an exceedingly useful experimental subject
for studying how teleost species respond to harsh environmental conditions.
Although certain organisms can withstand greater changes in environmental
conditions than others, the ability to respond to natural environmental changes is a basic
characteristic of all living systems. Unfortunately, the terminology used to describe these
responses is not uniform. Various researchers have attempted to define the terms adaptation,
acclimation, acclimatization, and accommodation (e.g., Prosser, 1955; Kinne, 1962; Prosser,
1975; Smit, 1980; Fontaine, 1993). I will use the term "adaptation" in its broadest sense,
defining it as a modification of the characteristics of an organism that facilitate an enhanced
ability to survive and reproduce in a particular environment. In this way I recognize that
adaptations involve both genetic and physiological (phenotypic) components, while not
attempting to separate these components from one another. The term acclimation will be
used as defined by Prosser (1975), where compensatory changes are measured following
changes in single environmental variables.
This study was designed to examine some of the costs to C. variegatus associated
with living in variable salinity environments. Specifically, I asked the following questions:
(1) What are the metabolic costs associated with different ambient salinities? (2) How does
salinity influence the energetic response at low oxygen tensions? (3) What is the
osmoregulatory response to variable salinity environments? (4) How does salinity influence
blood oxygen levels? These questions were tested by the following: measurement of
metabolism in C. variegatus fully acclimated to a wide range of experimental salinities;
measurement of the critical oxygen tension in C. variegatus fully acclimated to the same
range of salinities; measurement of metabolism prior to, and following, simulated tidal
changes in salinity; monitoring of plasma osmolality in C. variegatus exposed to a group of
different cycling salinity regimes; and measurement of hemoglobin concentration,
erythrocyte count, and hematocrit in C. variegatus acclimated to a wide range of ambient
Fish used in this study were collected from tidal marshes of the Gulf of Mexico
near Cedar Key, Florida. The shore in the Cedar Key area is classified as a zero energy
sector in which wave energy is dampened over the broad, shallow limestone plateau of the
Gulf of Mexico bottom (Stout, 1985). This results in a wide intertidal zone along the coast.
Furthermore, the coastal physiography is extremely diverse due in large part to irregularities
in the shore line of the mainland, to the presence of numerous islands and oyster bars in the
tidal area, and to the maze of intertidal and subtidal creeks and channels (Kilby, 1955). No
significant sediment sources are found in this area, and tides occur on a semi-diurnal basis.
The dominant emergent vegetation in the area is Spartina alterniflora, with the salt marshes
dominated by Juncus roemerianus. Fish communities of the Juncus marsh are dominated
by atheriniforms, with C. variegatus, Fundulus similis, and Poecilia latipinna making up
50-90% of the catch throughout most of the year (Kilby, 1955; Simpson and Gunter, 1956;
Stout, 1985; pers. obs.).
INFLUENCE OF ENVIRONMENTAL SALINITY ON ROUTINE METABOLIC RATE
AND CRITICAL OXYGEN TENSION OF CYPRINODON VARIEGATUS
Most fishes are capable of tolerating only a narrow range of salinities. However,
some fishes live in areas that experience frequent variations in salinity. These euryhaline
species possess important physiological and behavioral mechanisms that enable them to
survive in variable salinity environments. One such fish is the sheepshead minnow,
Cyprinodon variegatus. This species ranges along most of the Atlantic coast of the U.S.,
throughout the Gulf of Mexico, and disjunctly along the Yucatan peninsula (Johnson, 1974;
Darling, 1976; Duggins et al., 1983). It typically inhabits brackish water coastal salt
marshes that undergo frequent salinity fluctuations. Cyprinodon variegatus is capable of
tolerating salinities ranging from 0 ppt (Ager, 1971; Johnson, 1974; Kushlan, 1980) to 142
ppt (Simpson and Gunter, 1956).
This study was designed to examine the metabolic response of C. variegatus over a
variety of environmental salinities. Whereas energetic responses to a number of variables
including temperature, body mass, oxygen, and activity level have been well studied, the
influence of salinity on metabolism of fishes has received less attention. Most previous
studies have found that salinity does affect the energetic of fishes. Unfortunately, the
magnitude and/or direction of these effects are equivocal (Febry and Lutz, 1987; Morgan
and Iwama, 1991; Nordlic et al., 1991; Swanson, 1991). This has led several authors to
categorize the general patterns of metabolic responses to altered salinities (e.g., Nordlic,
1978; Morgan and Iwama, 1991).
One general pattern of fishes that inhabit variable salinity environments is a stable
metabolic rate over a range of salinities, with the range of salinities most commonly tested
between freshwater (0 ppt) and seawater (35 ppt) (Morgan and Iwama, 1991). Physiological
stability enables such fish to have a euryhaline existence that is unfettered by large
metabolic costs associated with adjustment to salinity change. However, there are relatively
few studies of the metabolic response of euryhaline fishes over an even wider range of
salinities that they encounter in their natural habitats.
Fish metabolism is also strongly influenced by the partial pressure of oxygen
(P02). Metabolism of fishes is independent of P02, as they maintain a constant metabolic
rate over a range of P02 extending downward from high atmospheric levels to some lower
level defined as the critical oxygen tension (Pc). Below the Pc (conformation region),
metabolism depends on oxygen tension and decreases linearly with decreases in oxygen
(Fry, 1947). Oxygen consumption declines at the Pc because the gas exchange system can
no longer supply both the extra demands of the respiratory system and the oxygen
demands of the tissues (Hughes, 1964). Since Pc is a useful metabolic parameter,
calculations of the Pc have been made for a number of fishes under a variety of
environmental conditions (e.g., Hall, 1929; Tang, 1933; Ultsch et al., 1978; Ott et al., 1980;
Subrahmanyam, 1980; Donnelly and Torres, 1988; Rantin et al., 1992; Nonnotte et al.,
1993). However, the influence of ambient salinity on the Pc has not been specifically
studied. This is surprising since salinity and oxygen are important abiotic factors that may
act synergistically in affecting metabolism.
This study examined the influence of a wide range of environmental salinities on
routine metabolic rate (RMR) and Pc in C. variegatus. I hypothesized that both RMR and
Pc would be unaffected by alterations in salinity over the range of salinities commonly
encountered in natural habitats of C. variegatus. Salinities outside this range, but within the
range known to be tolerated, were predicted to cause increases in both parameters.
Fish used in this study were obtained by seining canals and ditches in the salt marsh
near Cedar Key, Florida (Gulf of Mexico). Specimens were transported back to the
laboratory in 19 L carboys containing water from the collection site. Upon arrival at the
laboratory, individuals were held overnight in this water with constant aeration. The
following day, fish were transferred into holding tanks (75 to 114 L aquaria) maintained at
the salinity at which fish were captured, and treated prophylactically for 7-14 days in a 5 mg
L"- solution of Acriflavine. Following treatment, groups of approximately 10 fish were
placed into experimental (38 L aquaria) tanks containing water at a salinity within 10 ppt of
that in which they were collected. Both holding and experimental tanks were equipped with
undergravel filtration and constant aeration, and were maintained in rooms on a 12:12
light:dark cycle. Fish were fed Tetramin flake food once each day. All experimental
aquaria were located in a constant temperature environment room that maintained aquaria at
20 1 oC.
Experimental aquaria were used to acclimate fish to salinities ranging from 0 ppt to
100 ppt (0 to 2860 mOsm kg-1). The initial acclimation period was 14 days, after which
fish were either used in a metabolic trial or were transferred to the next higher or lower
salinity in the series. Salinity changes were in steps of 5 ppt, with smaller increments used
to acclimate fish to 0 ppt. This procedure was repeated until determinations had been made
at all experimental salinities. Water used in the freshwater acclimation was obtained from
wells in Alachua and Levy counties, Florida (mean conductivity = 360 ILS cm 1).
Experimental salinities greater than freshwater but less than full seawater were prepared by
diluting filtered Atlantic Ocean seawater (obtained from the C.V. Whitney Laboratory of the
University of Florida, Marineland, Florida) with appropriate quantities of deionized water.
Salinities greater than 35 ppt were produced by supplementing seawater with appropriate
amounts of synthetic sea salt (Instant Ocean). Salinities were monitored daily with an
AO temperature-compensated refractometer, and adjusted as necessary.
Metabolic determinations were carried out in sealed, opaque flasks ranging in
volume from 0.6 to 1.175 L, with the volume selected based on the size of the fish being
tested. Rate of oxygen consumption was used to measure metabolism, with flasks being
used as closed respirometers. Measures of metabolism were considered to be routine
metabolic rates, as animals were sequestered in such a way as to minimize, but not eliminate
activity (Winberg, 1956; Fry, 1957). Experiments were performed with post-absorptive fish
in a resting state, but fish were unconstrained and capable of spontaneous motor activity
(Winberg, 1956; Fry, 1957). For a relatively inactive species such as C. variegatus
(compared to actively swimming salmonids, for example), RMR is perhaps the most
appropriate measure of metabolism, as it more closely reflects normal activity patterns of the
fish than does either active or standard metabolic rate.
Each flask was sealed with a rubber stopper through which two hypodermic needles
(No. 18) were inserted. Each needle was fitted on the inside with catheter tubing, one with a
piece long enough to reach the bottom of the flask, the other half this length. A 25 ml plastic
syringe filled with water at the salinity of the experimental aquarium was inserted into the
needle fitted with the short piece of catheter tubing, while an empty 10 ml syringe was
inserted into the other needle. A 1 ml water sample was drawn into the empty syringe for
each determination of PO2. As each water sample was withdrawn, an equal volume of water
was injected from the filled syringe into the flask. This method of sampling, along with
minor movements by the fish, effectively stirred the water in the flask (Nordlie, pers.
comm.). Determinations of PO2 were made with a Radiometer oxygen electrode
connected to a Radiometer PHM 71 acid-base analyzer.
Measurements of the rate of reduction in P02 were made at 0.5 h intervals, and
continued until fish had depleted the oxygen level to approximately 20 mm Hg (generally 5
to 9 h). Following the final P02 determination, each fish was removed from its flask, damp-
dried and weighed to the nearest 0.01 g. All metabolic determinations were made between
0700 and 1900 hours, and fish were not re-used in other metabolic trials.
To ensure that fish were post-absorptive at the time of testing, food was withheld
from experimental aquaria for 24 h prior to beginning a metabolic reading. Respirometers
were filled with water at the salinity of the experimental aquaria, and placed in a water bath
maintained at 20 1 oC. The entire metabolic apparatus was located in a small, semi-
darkened room in which no other activity took place. In order to allow time for the fish to
adjust to the respirometers, individuals were placed into the flasks (with constant aeration)
12 to 16 h before beginning a trial. At the beginning of the metabolic trial, aerators were
removed and the flask was sealed.
Calculation of oxygen saturation values for the experimental conditions (taking into
account salinity, temperature, relative humidity, and barometric pressure) were made using
the equations of Truesdale et al., 1955 and both RMR (mg 02 h-1) and Pc (mm Hg) were
calculated for each fish. Data used for calculation of metabolic rates were limited to values
obtained while the P02 in the respirometer was greater than 100 mm Hg, in order to ensure
that these calculations were made at oxygen tensions well above the Pc. All data were used
for calculation of the Pc. Determination of the Pc was made using a BASIC program to
calculate the critical point (Yeager and Ultsch, 1989). Following recommendations by
Yeager and Ultsch (1989), data for each fish were first plotted to ensure that the relationship
was a two-step function, following which the midpoint approximation was used to calculate
Oxygen consumption is strongly influenced by body mass, so RMR values were
mass-adjusted using an analysis of covariance (ANCOVA). Log mass-independent RMR
was used as the dependent variable and log mass as the covariate. Least square means
derived from the ANCOVA were used as adjusted RMR values. It was not possible to
perform an ANCOVA for the Pc values, so calculations of Pc were mass-corrected to the
value of the average mass (3.13 g) of all individuals used in this study. The exponent
describing the relationship between mass and metabolism for C. variegatus (Nordlie et al.,
1991), MR = kW-68, was used to correct oxygen consumption rates. Values were corrected
following the relationship MRc = (Wo0.32)(3.13-0.32)(MRo), where MRc is the mass-
corrected oxygen consumption, Wo is the observed mass, and MRo is the observed oxygen
consumption at mass Wo (Ultsch et al., 1978; Cech, 1990). Statistical analyses follow
procedures outlined in Winer et al., (1991) and Sokal and Rohlf (1995). All statistical
analyses were one way tests using the Tukey-Kramer post hoc comparison (p = 0.05), and
values are given throughout as means standard error of the mean (se).
Salt marsh habitats are widely considered to experience unpredictable and
fluctuating abiotic conditions. However, actual physicochemical measurements are
infrequently reported. To address this issue, field measurements were made at four sites in
the Cedar Key area over a one year period. Whenever possible, measurements at each site
were taken both at the surface and on the bottom (generally 1-1.5 m deep). Dissolved
oxygen, salinity, and temperature were measured one to three times each month between
0700 h and 1700 h, for a total of 19 dates between June 1990 and June 1991. Sites 1, 2, and
3 were located deep in the salt marsh where C. variegatus was routinely collected. These
sites were located in close proximity to one another (< 10 m apart), and were interconnected.
Unlike many locations in the salt marsh, these sites were never completely isolated from one
another or from connections to the Gulf of Mexico, even during the lowest tides. Site 4 was
located directly on the Gulf of Mexico, in the town of Cedar Key, approximately five km
from sites 1, 2, and 3. Although C. variegatus is present at site 4, collections were not made
at this location. These field measurements were not intended to indicate the complete ranges
of oxygen, salinity, or temperature experienced by organisms living within the salt marsh.
Thus, the actual ranges of physicochemical conditions experienced by C. variegatus are
most likely greater than reported here. However, these values are a subset of the conditions
experienced by salt marsh inhabitants, and give an indication of some of the variability in
the measured physicochemical parameters.
Mean RMR was calculated for each fish and organized by salinity groups. Mean
values for unadjusted and adjusted RMR (from ANCOVA) are given in Table 2-1, and
adjusted RMR are plotted against ambient salinity in Figure 2-1.
In the range of ambient salinities between freshwater and 40 ppt, adjusted RMR
values were highest at 2 ppt and 40 ppt, being slightly lower and roughly equivalent at the
other measured salinities in this range. At salinities greater than 40 ppt, there was a
progressive decline in adjusted RMR. Overall, adjusted RMR ranged from a maximum of
0.97 mg 02 h-1 at 2 ppt, to a low of 0.64 mg 02 h-1 at 100 ppt, representing a 66%
decline. This decline corresponds with a decreased ability to regulate plasma osmolality at
elevated salinities (Figure 2-2; plasma osmolality data are from Nordlie, 1985).
A multiple linear regression analysis was used to generate a predictive model for
describing the effects of both salinity and mass on metabolism. In this model, Log body
mass (Log W; in g) and salinity (S; in ppt) were used as independent variables, and Log
mass-independent RMR (Log MR; in mg 02 h-1) as the dependent variable. The equation
that best described this relationship is
Log MR = -0.296 + 0.548 (Log W) 0.001 (S) (F2,110 = 85.445; P < 0.0001)
This model described 62% of all variability about the mean, and the random distribution of
the residuals suggests an absence of significant relationships that might have biased the
SC 00 o o0 o o o
S4 0 0 0 6 6 6 6 06
< o (O W W 5 m c
0 d d d d d o d d d 0
g^ o en, 0 05 oa \O N 0 00
SOO Ot O "-! O O I O O O
Sd dd d d d d d o o d
g 41 |1 -H +1 41 +1 41 41 41 41 +1 +-
z00 MH 44 m m m1 i 41 m
Do 6 6 o 6 6 0 o 6 6 0
m 41 m 1 1 + 1 41 41 41 4 4m m
r- 00 Q r S N C C4 0 9
_ c\ rV ci cq t" tn cn ci ci c C4
0 00 00 00 00 00
CS d dddd
U2 M 1 H 4 H-H 4 1 4 H -
c35 l o o *< <) 1 ^* f
E o o c S r o d o o r o
( iq ua3iCxo Su) 2 1)3i;oqu~aW aupnoU paisnfpv
(23iJtusotu) smeiotuso Iu1uSeld
0 0 0 0 0 0 0 0
L) Lr)O 0 0 L O
U-) LIn m mN N
S I v
^ I \ ^ 1
*^ 0 \ i c
. .- d d d d d d
a;lH ai!oqealA auilno^l pa;snlfpV
Critical Oxygen Tension
Measurements of oxygen consumption (mg 02 h 1) for each fish at all time
intervals were mass-adjusted to the mean mass of all individuals used in this study. These
mass-adjusted values were then used to calculate the Pc for each fish. Mean values of Pc
were organized by salinity group and are given in Table 2-1. Plots showing the relationship
between Pc and ambient salinity and between Pc, RMR, and ambient salinity are shown in
Figures 2-3 and 2-4, respectively.
Mean Pc values in the range of ambient salinities from 0 ppt to 40 ppt were not
significantly different from one another (p = 0.95), similar to the pattern exhibited by the
RMR data. Mean Pc values increased at salinities greater than 40 ppt, with the highest levels
recorded at salinities 80 ppt and higher. Pc values ranged from a low of 51.49 mm Hg at a
salinity of 2 ppt, to a high of 79.50 mm Hg at 80 ppt, representing a 45% increase. The rise
in mean Pc values corresponds well with a decreased ability to regulate plasma osmolality,
again similar to the RMR pattern (Figure 2-5; plasma osmolality data are from Nordlie,
Below the Pc, metabolism depends on the oxygen tension and decreases as the P02
decreases. Mathematically, the slope of the resulting line in this conformation region
depends on three factors: the Pc, the RMR at the Pc, and the lethal P02. A comparison of
the slopes in the conformation region reveals that despite the changes noted above in RMR
and Pc values, the P02 at which the fish can no longer survive (under experimental
conditions) is essentially equivalent for all salinities tested. This is reflected in the
increasingly shallow slopes seen at salinities greater than 50 ppt (Table 2-1).
The field measurements of oxygen concentration, salinity, and temperature revealed
very high variability of these physicochemical parameters over the course of the sampling
00 00I -
1~~I P I
(8H tutu) uoisuoL uaxO lu!!J.13
no I I I I I
- o 6 6 o 6 6 6 6
ale- HiloquIaN au!pnol pavsnfpv
(NH tum) uo[sual ua2Sxo luJI!!D
( w/msom) i!elomoso uuSuId
0 0 0 0 0 0 0 0 0
o0 0 u 0o Ln 0 Un 0
S n Ln U) U, e en
*I I I I I a I I 0
-- r -- --
'. I o0
0 0 0 0 0 0 0
0 T Io I ID i *-
(2H ww) uoIsuaj, ua3Cx0 I9!*!3)
period (Table 2-2). For all sites combined, ranges of dissolved oxygen, salinity, and
temperature over the course of the year were 0.6 10.8 mg L-1, 1.0 29.0 ppt, and 9.0 -
38.0 oC, respectively. Even higher salinities (> 40 ppt) were encountered occasionally in the
salt marsh outside the sampling period. Other areas of the salt marsh probably encounter
even greater extremes of these variables. Cyprinodon variegatus was seen on all dates when
physicochemical measurements were made.
In addition to high overall variability, there were large differences among some sites
in close proximity to one another. For example, mean oxygen concentration measurements
taken during the Fall at sites one, two, and three were 2.94 mg L-1, 3.85 mg L-1, and 5.1 mg
L-1, respectively. All three physicochemical parameters also strongly varied temporally
among sampling dates. These data provide good evidence that the Cedar Key salt marsh is
an extremely variable habitat with respect to these physicochemical parameters.
The family Cyprinodontidae is a diverse group of fishes with many species that
tolerate extreme environmental conditions (Lowe et al., 1967; Lotan and Skadhauge, 1972;
Naiman et al., 1976; Stuenkel and Hillyard, 1981; Chung, 1982). Cyprinodon variegatus is
perhaps the most physiologically tolerant member of the family. It has been called "the
toughest fish in North America" (Gunter, 1967) due to its extreme tolerance of severe
abiotic conditions. The species is found in waters ranging in salinity from freshwater (Ager,
1971) to 142 ppt (Simpson and Gunter, 1956), and can reproduce in waters as high as 100
ppt (Martin, 1972). They are known to tolerate temperatures ranging from about 1 C
(Berry, 1987), to temperatures greater than 41 OC (Strawn and Dunn, 1967), and to tolerate
near anoxic conditions (Odum and Caldwell, 1955). Thus, this species is an exceedingly
useful experimental subject for examining how teleosts respond to harsh environmental
T T T t
I I I
I t I
I t I
-a 4 4 4 4 4 4 *
0 o II
I I I
I I I
I I I
a 11 if ^^ cn I I iS 1 i e "1 8 n 11f
- -" c
-I I 4 4. 4-4 4 4
C5 d e
I I I
I I I
F1 a oIn 'IT kn "1 S 4
ST o = l W1i ) 1 o\ qo 1 o\ q o 1. o1 s" .
i r' ci r 1 i 11 11 cn o. 11
D V -I
E W" II
-t I I I I I 4 4
O En V) (
S i8oo 9 i .k 5
E 0'II II o 0 II 00
; 0c a ce !7
Field measurements indicated that the Cedar Key salt marsh varies tremendously,
both spatially and temporally, in dissolved oxygen, temperature, and salinity, thus exposing
organisms living within the salt marsh to extreme environmental conditions. Cyprinodon
variegatus tolerates these conditions very well. Salinities between 0 ppt and 40 ppt have
little effect on the energetic of C. variegatus as measured by RMR and Pc. This species
fits into the Type I metabolic response pattern (species exhibiting no significant change in
metabolic rate over a wide range of environmental salinities) as categorized by both Nordlie
(1978) and Morgan and Iwama (1991). A similar pattern characterizes the related salt marsh
resident, Adinia xenica (D.C. Haney, unpublished data), in which RMR was more variable,
but relatively constant over salinities from 0 ppt to 50 ppt (Figure 2-6).
Cyprinodon variegatus exhibits a decline in RMR only at salinities exceeding 40
ppt. This result was somewhat unexpected, but is nearly identical to the pattern found by
Nordlie et al., (1991) in a study that also examined metabolism of C. variegatus over a wide
range of ambient salinities. Nordlie et al., (1991) concluded that the depression in
metabolism at high salinities is probably related to permeability changes of the gill
membrane and/or integument. The point at which metabolism is reduced corresponds well
with a diminished ability of C. variegatus to osmoregulate efficiently. If osmotic
permeability of the gills is reduced at high salinities to help offset ionic influx and osmotic
efflux, the potential for oxygen uptake may be reduced as well (Kristensen and Skadhauge,
1974; Skadhauge, 1974; Davenport and Sayer, 1993). Evidence for this hypothesis comes
from several recent studies. In the first of these, Kultz and Onken (1993) found that overall
in vitro permeability of the opercular membrane of the cichlid Oreochromis mossambicus
was reduced in hypersaline media, with a simultaneous reduction in passive ion fluxes.
More direct evidence comes from studies by Bindon et al., (1994a) and Bindon et al.,
(1994b), on the rainbow trout, Oncorhynchus mykiss. In these studies the authors
demonstrated that impairment of respiratory gas transfer coincided with chloride cell
proliferation induced by an osmoregulatory challenge.
(2/misou) n!ilouiso tusejld
o aL o L o n 0
Il CI I K LI
0 0 01 0 .0
- c o6 6 c c o o 6 c
a-l ailoqoIaNI au!lnoU poasnrpv
Changes in activity may also account for a depression in metabolism. In a study on
the milkfish, Chanos chanos, Swanson (1991) showed that a depression in metabolism at
an elevated salinity (55 ppt) was highly correlated with a reduction in activity. This may
have been a factor in this study although the method used here for metabolic determinations
minimized activity of fish at all salinities.
The main focus of this study was determination of Pc over a wide salinity range, as
no other study has investigated to any extent the influence of salinity on this parameter.
Critical oxygen tension can be affected by many factors. These include both physical and
biotic parameters such as temperature, salinity, size and activity, reproductive and nutritional
state, and experimental parameters, such as rate of oxygen depletion or manipulation and
disturbance of fish during measurements (McMahon, 1988).
Like other vertebrates, C. variegatus is a good oxygen regulator over a wide range
of oxygen tensions. Critical oxygen tension in C. variegatus was unaffected by changes in
salinity between 0 ppt and 40 ppt, similar to the result for RMR. However, above 40 ppt, Pc
increased with further increases in salinity. This seems reasonable, as fish might be
expected to encounter difficulties obtaining sufficient oxygen at elevated salinities.
Unfortunately, it is difficult to compare my results with those from other groups of
fishes. While the Pc has been measured for a number of freshwater, marine, and intertidal
fishes, only a few authors have considered the effects of salinity in their analyses. Job
(1969) examined oxygen consumption of Tilapia mossambica (= 0. nossambicus) at 0.4
ppt, 12.5 ppt, and 30.5 ppt, and found no effect of salinity on the Pc of small (5 g) fish at
either 15 oC, 30 oC, or 40 oC (Pc approximately 50 mm Hg). Salinity did affect Pc on
larger (80 g) individuals at 15 oC, where the Pc doubled from approximately 50 mm Hg to
100 mm Hg in fish acclimated to 0.4 ppt versus 12.5 ppt or 30.5 ppt However, Pc
calculations made by Job (1969) were extremely rough approximations, and are difficult to
compare with values in this study.
Subrahmanyam (1980) examined the influence of oxygen tension on the metabolic
rate of several salt marsh fishes including C. variegatus, with all measurements made at 25
oC and salinities of 17-21 ppt. All species tested in Subrahmanyam (1980) (C. variegatus,
Poecilia latipinna, Lagodon rhomboides, Leiostomus xanthurus, Fundulus grandis, and
F. similis) were oxygen conformers at oxygen tensions of 85 mm Hg and lower, a
somewhat different response than seen in this study. This difference may be due to the
small sample sizes, and relatively few measurements of oxygen consumption made by
Subrahmanyam (1980). In a study on the golden mullet, Liza aurata, Shusmin (1989),
measured the "oxygen threshold" at salinities ranging from freshwater to 50 ppt. He
showed that the oxygen threshold was relatively constant at all salinities between freshwater
and 40 ppt, with a large increase at 50 ppL Cyprinodon variegatus did not show this type of
response, as the lethal endpoint does not appear to increase at higher salinities in this
Values of the Pc in this study compare fairly well with Pc measurements for other
groups of fishes. Intertidal marine species generally have lower Pc values than C.
variegatus, ranging from 20 to 26 mm Hg in Paraclinus intergripinnis (Congleton, 1980)
to 30 to 40 mm Hg in Gobius cobitus (Bridges, 1988) and Helcogramma medium (Innes
and Wells, 1985; Pelster et al., 1988; Quinn and Schneider, 1991). A number of African
cichlids ( e.g., Oreochromis niloticus, Cichlasoma urophthalamus, Eretmodus
cyanosticus, Dimidiochromis compressiceps) have either slightly lower, or roughly
equivalent Pc values to the approximate value (55 mm Hg) displayed by C. variegatus in
this study at most salinities (Ross and Ross, 1983; Becker and Fishelson, 1986; Palacios
and Ross, 1986; Verheyen et al., 1994). Donnelly and Torres (1988) found Pc values
ranging from 25 to 50 mm Hg for a number of midwater fishes from the eastern Gulf of
Mexico, again values near, or slightly lower than those of C. variegatus. Thus, C.
variegatus has Pc values that are fairly close to those of ecologically and evolutionarily
The variations in RMR and Pc as a function of environmental salinity observed in
this study suggest that C. variegatus responds to high salinities by reducing energy
expenditures. Observations by myself and others indicate that hypersaline conditions are
encountered infrequently, and likely last for short (days) periods of time. Available evidence
suggests that high salinities are metabolically expensive, whether due to an influence on
activity, osmoregulation, or other physiological and/or behavioral processes. Low oxygen
conditions also commonly occur with high salinities. Decreased metabolism in conjunction
with increased Pc reduces energetic expenditures dramatically at elevated salinities. These
responses effectively increase the time C. variegatus can tolerate adverse conditions, albeit
at a cost of a reduction in energetic processes. However, this is an appropriate response to
harsh conditions that appear at variable and unpredictable time intervals, but that are present
for only short periods of time.
The type of metabolic response to salinity in C. variegatus fits the concept of
"scope for survival" as described by Hochachka (1990). This is a pattern of metabolic
response to environmental stressors characterized by a depression in metabolism,
sometimes below maintenance levels. The advantage of such a reduction in energy
expenditure is essentially the slowing of biological time, enabling survival despite the
temporarily imposed physiological stressor.
Cyprinodon variegatus is an extremely successful euryhaline fish that can survive a
large range of salinities, with typically encountered salinities having little effect on normal
metabolism or on the ability of individuals to osmoregulate efficiently. Fish accommodate
extremely high salinities by reducing energetic expenditures. Such a response increases the
amount of time sheepshead minnows can survive hypersaline conditions, enabling them to
"wait out" the difficult conditions.
INFLUENCE OF SIMULATED TIDAL CHANGES IN AMBIENT SALINITY ON
ROUTINE METABOLIC RATE IN CYPRINODON VARIEGATUS
Salt marshes often undergo large and rapid salinity fluctuations, a condition that
may significantly affect the distribution and abundance of organisms within these habitats.
Changes in salinity may dramatically influence the energetic of individual fish. Information
on the metabolic costs associated with salinity fluctuations may be useful in explaining or
predicting distribution patterns of some coastal fish species.
Energetic patterns of fishes have been examined relative to a number of abiotic
factors, most notably temperature and oxygen (e.g., Wells, 1935; Fry, 1957; Beamish, 1964;
Brett and Groves, 1979; Stuenkel and Hillyard, 1981; Johnston and Battram, 1993).
Comparatively fewer studies have been considered the influence of salinity on metabolic
patterns of fishes (e.g., Kinne, 1966; Madan Mohon Rao, 1974; Nordlie et al., 1991;
Swanson, 1991). In these studies, various metabolic responses to salinity have been
reported. Euryhaline fish exhibit one of the few consistent patterns, namely that metabolism
remains relatively unaffected over the range of salinities normally encountered (Nordlie,
1978; Morgan and Iwama, 1991).
Most studies that have examined the influence of salinity on fish energetic have
employed measurements from fish maintained at constant salinities. Data from these
experiments are very useful for understanding the overall metabolic patterns exhibited by
the species tested, as fish were usually fully acclimated to each experimental salinity prior to
testing. However, these experiments provide less information about more ecologically
relevant responses, as fish in their native habitats are often subject to rapid and dramatic
fluctuations in salinity.
Studies of the influence of salinity fluctuations on metabolism of fishes have
usually focused on species that generally experience maximal fluctuations in salinity only
between freshwater (0 ppt) and seawater (35 ppt) (e.g., Davenport and Vahl, 1979; Von
Oertzen, 1984; Moser and Gerry, 1989; Shusmin, 1989; Moser and Miller, 1994).
However, some salt marsh teleosts regularly encounter salinities outside this range. This
study examined the influence of salinity fluxes on routine metabolic rate (RMR) of the salt
marsh teleost, Cyprinodon variegatus, a species that regularly encounters salinities greater
than 35 ppt.
Fish used in this study were fully acclimated to a series of salinities ranging from 0
ppt to 60 ppt, followed by exposure to a simulated tidal change in salinity. The magnitude,
rate, and direction of salinity changes may be important determinants of how salinity
fluctuations affect metabolism. In this study, the direction of the salinity change in
conjunction with the ambient (acclimation) salinity was the primary focus. The magnitude
of the salinity changes were selected to simulate extremes known to occur in the habitat
from which fish were collected, and the rate was chosen to simulate a normal diurnal tidal
cycle. I predicted that C. variegatus would show minimal changes in RMR when the
salinity was changed within the range commonly encountered by the species. Salinity
changes outside this range were hypothesized to lead to depressions in metabolism.
Collections of fish, transportation back to the laboratory, and general laboratory
procedures were performed as described previously. Fish were first sequentially acclimated
to a series of salinities ranging from 0 ppt to 60 ppt (0 to 1715 mOsm kg-1). However,
unlike the previous experiments, these procedures were designed to measure RMR before
and following simulated tidal changes in salinity. Initial salinities (acclimation) and salinities
following the simulated tidal change (final) are shown in Table 3-1.
Metabolic determinations and changes in salinity were carried out in a flow-through
respirometer with an effective volume of 0.875 L (Figure 3-1). The respirometer was
immersed in a thermoregulated reservoir during metabolic trials that served to maintain a
constant temperature within the respirometer. A submersible pump maintained within a
second thermoregulated reservoir (38 L) was connected to the inlet of the respirometer. This
reservoir was vigorously aerated and was used to supply water to the respirometer. The
respirometer outlet emptied back into this main reservoir. A third reservoir (19 L) serving as
a salinity source was connected to the main reservoir via a peristaltic pump. Due to
limitations of the available equipment, the system was closed during oxygen partial pressure
(P02) determinations. Two ports within the respirometer were used to sample water for
determination of PO2. A 10 ml plastic syringe filled with water at the experimental salinity
was inserted into the first port (inlet side), while an empty 10 ml syringe was fitted into the
second port (outlet side). As a 1 ml water sample was drawn into the empty syringe, an
equal volume of water was injected from the filled syringe into the respirometer.
Determinations of PO2 were made with a Radiometer oxygen electrode connected to a
Radiometer PHM 171 acid-base analyzer.
To begin a metabolic trial, the main reservoir and respirometer were filled with water
at the acclimation salinity. Fish were placed into the respirometer 12 h prior to the
beginning of a metabolic determination. This allowed fish sufficient time to adjust to the
experimental setup. The respirometer was then sealed and immersed in the thermoregulation
reservoir, and the submersible pump was turned on at a flow rate of 1 L min-1. The
following morning, the pump was turned off, the system was closed, and P02
measurements begun. Measurements of the rate of reduction in P02 were made at 0.5 to
1.0 h intervals, and continued until fish had depleted the oxygen level to approximately 100
mm Hg (generally 4 to 6 h). Following the final P02 determination, the system was
Table 3-1. Acclimation and final salinities used in simulated tidal change study.
Acclimation Salinity Final Salinity
0 ppt 0 ppt (control)
2 ppt 2 ppt (control)
10 ppt 10 ppt (control)
20 ppt 0 ppt
20 ppt (control)
30 ppt 2 ppt
30 ppt (control)
40 ppt 20 ppt
40 ppt (control)
50 ppt 20 ppt
50 ppt (control)
60 ppt 40 ppt
60 ppt (control)
reopened, and the submersible pump turned on. At this time, either well water (0 ppt) or a
saline solution of variable concentration was placed into the salinity reservoir. The salt
concentration in this reservoir was determined so that its addition to the main reservoir
would change the salinity of the main reservoir and respirometer to the desired final salinity.
The volumes of both reservoirs and the rate of pumping by the peristaltic pump were
adjusted so that the salinity would change at a uniform rate over a 6 h period. The peristaltic
pump was plugged into an automatic timer so that salinity changes took place between 1900
h to 0200 h.
On the second morning the system was re-closed and PO2 measurements were
made at the new (final) salinity. PO2 measurements at the final salinity were thus made 6-
12 h following completion of the salinity fluctuation, a period sufficiently long to be past
the stressful transition period (Von Oertzen, 1984). Following the final P02 determination,
the fish was removed from the respirometer, damp-dried and weighed to the nearest 0.01 g.
All metabolic determinations were made between 0700 h and 1500 h, and fish were not re-
used in other metabolic trials. Fish were thus held within the respirometer for a total of 40
to 44 h (including adjustment period).
Food was withheld from experimental aquaria for 24 h prior to beginning a
metabolic reading to ensure that fish were post-absorptive at the time of testing. Fish were
not fed while in the respirometer. The entire metabolic apparatus was located in a small,
semi-darkened room in which there was no other activity, with determination of metabolic
rates performed as previously described.
Because a metabolic determination for the two salinity treatments was performed on
the same individual, it was not necessary to mass-correct values in order to perform the
statistical analyses. However, to compare data obtained in this study more easily with
published literature, all RMR measurements were mass-adjusted to the value of the average
mass (2.94 g) of all individuals used in this study. Adjusted RMR values were calculated as
previously described. A repeated measures analysis of variance was used to compare
adjusted RMR's between acclimation and final salinities. Statistical analyses follow
procedures outlined in Winer et al. (1991) and Sokal and Rohlf (1995). All analyses of
variance were one way tests using the Tukey-Kramer post hoc comparison (p = 0.05).
Comparison of RMR's at acclimation and final salinities revealed some interesting
patterns (Table 3-2). There were no significant differences (p > 0.05) in RMR between
acclimation and final salinities in any control trials (n = 4 each). Similarly, when both
acclimation and final salinities (2 ppt, 10 ppt, 20 ppt, 30 ppt, and 40 ppt) were in a range that
is typically encountered by C. variegatus, there was a significant change in RMR in only
one trial. Typical salinities in the wild range from 2-5 ppt through 30-35 ppt, with
hypersaline conditions (35-40 ppt) occurring much more frequently than salinities near 0
ppt (pers. obs.). When the acclimation salinity is high (50 ppt and 60 ppt), all groups
exhibited significant elevation of RMR in the lower, more typical, final salinity. The same
general pattern was seen when the final salinity was high, where fish in two of three groups
had depressed RMR at the highest salinities. RMR was depressed when either the
acclimation or final salinity was 0 ppt. The direction of salinity change strongly influenced
the metabolic response. When salinity was increased over the course of the trial (Figure 3-
2), fish were only affected metabolically at the very highest and lowest salinities, where
metabolism was depressed. When salinity was decreased over the course of the trial (Figure
3-3), fish again showed metabolic depression at extreme salinities, but, overall, more groups
exhibited changes in metabolism with changes in salinity than groups in which salinity was
My results demonstrate that C. variegatus maintains a very stable metabolism when
exposed to typical salinity fluctuations seen in the salt marsh habitat at Cedar Key, Florida.
The experimental procedure resulted in no mortalities during or immediately following any
salinity trial. Metabolism was generally unaffected in salinity trials where both acclimation
and final salinities were in the range typically encountered by C. variegatus at Cedar Key. It
Table 3-2. Mean routine metabolism (mg 02 h-1) before (acclimation salinity) and
following (final salinity) a simulated tidal change. Values are given as means se. Groups
exhibiting a significant change in metabolism are indicated with an asterisk.
Acclimation Acclimation Final Final P value Response
Salinity RMR Salinity RMR
(ppt) (mg 02 h-1) (ppt) (mg 02 h-1)
0 ppt 0.466 0.07 20 ppt (n = 5) 0.584 0.09 0.04 Increase
2 ppt 1.242 0.18 20 ppt (n = 5) 1.244 0.23 0.99 No Change
1.024 0.13 30 ppt (n = 5) 0.978 0.06 0.69 No Change
10 ppt 1.084 0.20 30 ppt (n = 5) 1.082 0.19 0.97 No Change
20 ppt 1.348 0.14 0 ppt (n = 5) 0.906 0.09 0.01 Decrease
0.886 0.03 2 ppt (n = 5) 0.478 0.07 0.01 Decrease
0.948 0.08 40 ppt (n = 6) 0.785 0.17 0.33 No Change
1.025 0.11 50 ppt (n = 6) 0.745 0.07 0.01 Decrease
30 ppt 1.028 0.23 2 ppt (n = 5) 0.854 0.10 0.33 No Change
0.628 0.07 10 ppt (n = 5) 0.688 0.11 0.34 No Change
0.728 0.07 50 ppt (n = 6) 0.728 0.12 1.00 No Change
40 ppt 0.704 0.12 20 ppt (n = 5) 0.662 0.09 0.75 No Change
0.626 0.06 60 ppt (n = 5) 0.476 0.03 0.03 Decrease
50 ppt 0.620 0.04 20 ppt (n = 5) 0.904 0.11 0.05 Increase
0.850 0.12 30 ppt (n = 5) 1.076 0.14 0.05 Increase
60 ppt 0.452 0.07 40 ppt (n = 5) 0.578 0.05 0.01 Increase
was only when the acclimation or final salinities were outside the range normally
encountered (i.e., 0 ppt, 50 ppt, and 60 ppt) that changes in metabolism were found. In these
cases metabolism was depressed at each extreme salinity relative to the measurements made
at the more typical salinity.
Figure 3-2. Results of metabolic trials where salinity was increased over the course of the
trial. Bars represent groups listed in Table 3-2 for which final salinity was greater than
initial salinity. The height of each bar signifies the magnitude of the salinity change for each
metabolic trial and the asterisk indicates at which of the salinities (for each metabolic trial)
the routine metabolic rate (RMR) was highest. The x axis has no scale and serves only to
visually separate groups.
RMR's Not Significantly Different (p = 0.05)
RMR's Significantly Different (p = 0.05)
Figure 3-3. Results of metabolic trials where salinity was decreased over the course of the
trial. Bars represent groups listed in Table 3-2 for which final salinity was less than initial
salinity. The height of each bar signifies the magnitude of the salinity change for each
metabolic trial and the asterisk indicates at which of the salinities (for each metabolic trial)
the routine metabolic rate (RMR) was highest. The x axis has no scale and serves only to
visually separate groups.
D RMR's Not Significantly Different (p = 0.05)
SRMR's Significantly Different (p = 0.05)
These results allow for limited comparison because similar studies have examined
the influence of salinity over a much narrower range than examined in this study. The best
comparison is with a study by Wakeman and Wohlschlag (1983) on the red drum,
Sciaenops ocellatus. In their study, test animals were acclimated to a series of salinities
between 1 ppt and 50 ppt, and then abruptly transferred to salinities either 10 ppt higher or
lower than the acclimation salinity. Although Wakeman and Wohlschlag (1983) did not
specifically compare metabolic rates following the transfer with those taken at the
acclimation salinity, there were trends similar to results of this study. Fluctuations in salinity
over a moderate range caused little or no change in metabolism of S. ocellatus on
determinations made 12 to 24 h following the transfer. Only when either the acclimation or
final salinity was 50 ppt were changes in metabolism obvious, with metabolic rates elevated
in 50 ppt compared to 40 ppt. It thus appears that S. ocellatus responds to abrupt salinity
changes somewhat differently than does C. variegatus, as metabolism increased upon
transfer to 50 ppt in S. ocellatus, but decreased in C. variegatus. This difference may be
due to the smaller magnitude of the salinity changes performed by Wakeman and
Wohischlag (1983) (10 ppt vs 20 to 30 ppt in the present study). However, these results
could also be attributable to different sizes and activity patterns of experimental fish
between the two studies.
In a similar study, Shushmin (1989) studied the dynamics of oxygen consumption
in juvenile golden mullet, Liza aurata, following abrupt transfers from 18 ppt to a series of
salinities ranging from 0.4 ppt to 50 ppt. All salinity changes resulted in increased
metabolic rates for one to three days following the transfer, with metabolism generally
returning to near normal levels within three to four days. It is not surprising that salinity
changes are more stressful in the golden mullet, since the species has lower salinity
tolerance and osmoregulatory ability than C. variegatus.
Barton and Barton (1987) examined metabolism in juvenile (< I g) C. variegatus
collected from inland saline lakes of San Salvador Island. They measured metabolic rates of
fish at 10 ppt and 35 ppt following abrupt transfer from an acclimation salinity of 35 ppt.
Metabolic rates in fish measured at 10 ppt were significantly higher than those made at the
acclimation salinity. These results are somewhat different from those obtained in the present
study. However, measurements by Barton and Barton (1987) were made only 3 h following
abrupt transfer to the reduced salinity, whereas measurements in the present study were
made 6-12 h following a gradual change in salinity. Fish in this study thus experienced a
less stressful transition to the new salinity. An ontogenetic effect may also account for the
observed differences, as many investigations have demonstrated that adults and juveniles of
the same species may differ greatly in physiological competence (e.g., Kinne, 1966; Martin,
1968; Oikawa et al., 1991).
Other relevant comparisons can be made with data on juvenile croaker,
Micropogonias undulatus, and juvenile spot, Leiostomus xanthurus (Moser and Gerry,
1989; Moser and Miller, 1994). These studies examined effects on metabolism from
fluctuations in salinity between 0 ppt and 35 ppt. Like C. variegatus, a salinity decrease
from higher salinities to 0 ppt generally led to a decreased metabolism in L. xanthurus.
Similarly, when fish were acclimated to 0 ppt and salinity was increased, metabolism was
elevated at the higher salinities. Contrary to findings for C. variegatus, salinity fluctuations
between 15 ppt and 35 ppt elicited changes in metabolism in both L. xanthurus and M.
undulatus. Thus, these species are influenced to a greater extent by salinity fluctuations than
C. variegatus. However, rates of salinity fluctuation in the above studies were faster than
rates used in the present study (16 or 32 ppt h-1 vs 3.3 or 5 ppt h- 1), and C. variegatus
likely experiences greater fluctuations in salinity than do spot or croaker.
Acclimation state is the most important measured factor influencing the metabolic
response of C. variegatus to simulated tidal changes in salinity. However, direction of the
salinity change also influenced metabolism in C. variegatus. Like the results obtained by
Moser and Miller (1994) on juvenile L. xanthurus, it appears that C. variegatus adjusts to
increasing salinity more effectively than to decreasing salinity, as evidenced by the
responses to salinity changes between 20 ppt and 2 ppt, and between 50 ppt and 30 ppt.
Thus, when salinity is increased over the course of the trial, fish are only affected
metabolically at the very highest and lowest salinities. However, when fish are acclimated to
a high salinity, and salinity is then decreased to a more typical range, individuals take
advantage by increasing metabolism to more normal levels.
Thus, C. variegatus is well adapted to a varying salinity environment Its
metabolism is unaffected by changes in salinity over the typical range encountered, even
when salinity is changed rapidly. Furthermore, this corroborates my hypothesis that C.
variegatus tolerates extremes in salinity by lowering metabolism and decreasing energy
expenditures. Fish appear to wait for conditions to improve, and respond to these more
favorable conditions by returning metabolism to normal levels.
A decrease in energetic expenditures as just described is a potentially adaptive
response for fishes living in variable salinity environments like those of Florida coastal salt
marshes. While few data exist on responses of other salt marsh residents to wide ranges in
salinity, I suspect this may be a general pattern. Cyprinodon variegatus may be unusual
because of its broad tolerances, but is a useful experimental animal because of this as well.
Information gleaned from studies with C. variegatus may indicate areas of examination for
other important salt marsh teleosts which may have more limited salinity tolerance, but
which may follow the same general metabolic patterns seen in C. variegatus.
INFLUENCE OF A FLUCTUATING SALINITY REGIME ON OSMOREGULATION
IN CYPRINODON VARIEGATUS
Few areas in the field of fish physiology have received as much attention as the
study of osmoregulation. The basic patterns of osmoregulation are well understood and are
reviewed extensively by Evans (1984), Karnaky (1986), Ventrella et al. (1992), Evans
(1993), McCormick (1994), and Wood and Marshall (1994). The mechanisms of
osmoregulation in euryhaline fish are not unusual. Euryhaline fishes are unique in having
the ability to osmoregulate efficiently in waters of highly variable salinity. They can
osmoregulate in water both more and less concentrated than their own body fluids, and they
are able to alter their pattern of osmoregulation rapidly if they live in environments where
sudden fluctuations in salinity may occur.
Despite the voluminous research done in this area, few studies have examined
patterns of osmoregulation in fishes that may encounter salinities outside the range from
freshwater (0 ppt) to seawater (35 ppt). Such species tend to be small and of little or no
economic importance, with a notable exception being the milkfish, Chanos chanos (Ferraris
et al., 1988; Swanson, 1991). However, to understand osmoregulation patterns in fishes,
studies need to encompass the range of salinities encountered by species in nature.
One topic that has received little attention concerns the effect of fluctuating salinities
on osmoregulation in telcosts. Salinity is a limiting factor in certain estuarine and salt marsh
habitats, and may widely vary daily and seasonally. Despite this, most investigations of
osmoregulation in fishes involve studies at constant salinities. While such studies provide
important information on overall osmoregulatory patterns, they provide less information
about more ecologically relevant responses.
The aim of the present investigation was twofold. First, I examined the ability of
individuals of the euryhaline teleost, Cyprinodon variegatus, to regulate plasma osmolality
under the influence of a cycling salinity regime. Second I examined a hypothesis proposed
by Goolish and Burton (1988) in a study involving the intertidal copepod Tigriopus
californicus. Goolish and Burton (1988) suggested that species exposed to fluctuating
salinities would be able to respond more rapidly and completely to salinity stress. In other
words, could past exposure to changing salinity result not in improved osmoregulation at
any single salinity, but rather to improved performance immediately following another
salinity fluctuation? These hypotheses were examined by determining plasma osmolality
and hematocrit of individual C. variegatus subjected to fluctuations in salinity over a wide
range of ambient salinities.
Collections of fish used in this study were obtained from tidal creeks in the salt
marsh near Cedar Key, Florida. Specimens were transported back to the laboratory in 128 L
coolers supplied with aeration and filled with water obtained from the collection site. Fish
were obtained in two collections made during September 1994. The salinity of the collection
site was approximately 25 ppt for both collections. This study was conducted at the
Southeastern Biological Science Center (SBSC), National Biological Service, Gainesville,
Florida. Fish were held overnight in the coolers used for transportation (with aeration)
before transfer to a 1.2 m diameter fiberglass holding tank containing water at 30 ppt Fish
were treated prophylactically for 7-14 days in a 5 mg L-1 solution of Acriflavine.
Following treatment, all fish were transferred to experimental aquaria (30 ppt salinity = 860
mOsm kg-1) located within a constant environment room maintained at 20 1 oC and on a
12:12 h lightdark cycle. Both holding and experimental aquaria were equipped with sponge
filters providing continuous aeration, and fish were fed flake food once each day.
Experimental aquaria were used to subject fish to a cycling salinity regime. Seven
salinity trials were performed (Table 4-1). All fish were maintained in experimental aquaria
located on bank 1 for eight days prior to initiating the first salinity change. Salinity changes
were effected by carefully netting fish from the experimental aquaria located on bank 1, and
immediately transferring individuals into the appropriate experimental aquaria located on
bank 2. Fish remained in experimental aquaria located on bank 2 for 24 h before transfer
back to tanks located on bank 1. Fish then remained in bank I experimental aquaria for 24
h before again being transferred to aquaria on bank 2. Thus, one complete cycle began with
24 h in bank 2 aquaria, and ended following 24 h in bank 1 aquaria. The procedure was
repeated so that a total of 10 cycles was performed (20 days). A subsample of fish was
removed for testing from each experimental aquarium just prior to the first change in
salinity (cycle 0), and at the end of cycles 1 (day 2), 5 (day 10), and 10 (day 20). All
subsamples were thus taken following a 24 h period in bank I aquaria (30 ppt).
Following completion of the 10th cycle, fish remaining in all decreasing salinity
groups (groups DI, D2, and D3) were transferred to aquaria at 2 ppt and fish remaining in
all increasing salinity groups (groups Ii, 12, and 13) were transferred to aquaria at 60 ppt
Fish in the aquaria maintained at 30 ppt were split into three groups at that time: one third of
the group was transferred to 2 ppt (group CD), one third transferred to 30 ppt (group C),
and the final third transferred to 60 ppt (group CI). Fish remained in these salinities for 24
h and were then removed for testing (day 21). Salinities were checked daily with a Leica
temperature compensated refractometer and adjusted as necessary.
Hematocrit (Hct) and plasma osmolality were determined at each sampling interval.
Fish were first carefully netted from their experimental aquaria and blotted dry. Blood was
taken by sternal cardiac puncture using heparinized microhematocrit tubes drawn to a fine
point, and fish were weighed and standard length determined. The tubes were then
Table 4-1. Salinity trials used in cyclical salinity study. The group maintained at 30 ppt was
split into three groups following cycle 10 (day 20); groups CD, C, and CI (see text for
Group Direction of Salinity Change Salinity in Bank 1 Salinity in Bank 2
DI Decreasing Salinity (n=25) 30 2
D2 Decreasing Salinity (n=25) 30 10
D3 Decreasing Salinity (n=25) 30 20
C, CD, and Ci No Change (n=35) 30 30
II Increasing Salinity (n=25) 30 40
12 Increasing Salinity (n=25) 30 50
13 Increasing Salinity (n=25) 30 60
centrifuged in a micro-hematocrit centrifuge for 10 minutes to separate plasma from cells.
Hematocrit was read using a micro-capillary reader before plasma was isolated from formed
elements by scoring the tube with a file and retaining only the portion containing plasma.
Plasma osmolality (mOsm kg-1) was determined on 5 pl samples using a Wescor 5500
vapor pressure osmometer. Fish were used without regard to sex, and all blood samples
were taken between 0600 h and 1000 h. Statistical procedures followed Winer et al. (1991)
and Sokal and Rohlf (1995). All statistical analyses were one way tests using the Tukey-
Kramer post hoc comparison (p = 0.05)
All fish entered into the experimental procedure survived the entire duration of the
experiment Fish in all groups ate normally throughout the course of the experiment, and no
discernible changes in behavior were observed during the experimental procedure for any
group. Fish appeared to have no difficulty in tolerating the imposed salinity fluctuations,
even in groups Di and 13, which were subjected to daily changes in salinity of 28 ppt and
30 ppt, respectively. The influence of body mass on both Hct and plasma osmolality was
evaluated using an analysis of covariance, and for no analysis was body mass a significant
covariate. Sizes of fish used in this study ranged from 0.75 g to 5.887 g (2.8 mm to 5.8 mm
standard length, mean 4.1 0.038 mm), with a mean mass of 2.245 0.066 g. Subsamples
of groups were not significantly different in mass.
Hematocrit and plasma osmolality values were compared both between salinity trials
for each time period sampled (i.e., all groups were compared to one another on days 0, 2,
10, 20, and 21), and within a single trial over the time course of the experiment. Groups
experiencing increases in salinity (groups Ii, 12, 13, and CI) were compared separately from
groups experiencing decreases in salinity (groups DI, D2, D3, and CD), with all compared
to the group maintained at a constant salinity (group C). Data from days 0 through 20 are
discussed separately from day 21 data as follows.
Day 0 through 20
Hematocrit values showed no obvious trends either between salinity trials or within
any single trial over the time course of the experiment (Table 4-2). There were no
statistically significant differences (p > 0.05) within any group (between days 0 and 20), or
among groups on days 0, 2, 10, or 20. There was a slight trend towards increased variance
in Hct values over the course of the experiment in all groups, peaking at cycle 5, but
differences were non-significant in all cases.
Plasma osmolality data, unlike the Hct data, exhibited consistent trends. No
significant differences existed among groups on days 0, 2, 10, or 20. However, comparisons
within groups over time revealed a different pattern. All groups, regardless of salinity
regime, showed a trend towards slightly elevated plasma osmolality between days 0 and 20
(Table 4-2; Figures 4-1 and 4-2). For groups C and II, this was statistically significant, with
"a ^ -*
J .5 <^.
E-i (U (
o o u
.*4. 0. c 00
0< i 4.)
v 2 'c
* .- N0
k a p
3 3= u^
\ / 00
', / -
(L- sow) UOtLWJuzO3UOD 3OwsO s Usuld
U > ~
orn il a
U .- 8
o m 15
' S2 ^
~E v- U
c "a. o 2
4 a s
S*c I 0.o
o 0 .rl^
Q a -
Sa .8 0
'5 S. e
SO .-8 j*
( 3)- msoUl) UO!EJ1U3U0D I1O0UISO Bl~US0 d
the values on day 20 significantly higher than values from day 0 and day 1 (p<0.05); all
other groups exhibited the same trend, but differences were statistically non-significant.
This appears to be related to the experimental manipulation of fish, and not to the salinity
regime experienced by each group. All groups regulated plasma osmolality effectively. No
differences in regulatory ability could be seen among any of the groups tested, regardless of
the magnitude or direction of the salinity fluctuation.
The above addresses the question of how effectively C. variegatus regulates plasma
osmolality in the face of salinity fluctuations. Transfer of individuals to either 2 or 60 ppt
following 20 days of exposure to a variety of salinity fluctuations addresses the second
question posed earlier Does past exposure to large salinity fluctuations result in improved
osmoregulatory performance, compared to animals experiencing little or no salinity
fluctuation, immediately following a fluctuation in salinity? All measurements here were
taken 24 h (on day 21) following the final salinity fluctuation.
Hematocrit results showed no consistent trends on day 21 samples. Only one
significant difference was noted, with the Hct value for group C1 being significantly elevated
compared to the value for group C (p<0.05). All other comparisons showed no significant
differences in Hct.
There were significant differences in plasma osmolality for day 21 measurements.
Here groups which had experienced greater fluctuations in salinity over the course of the
experiment showed a much better ability to osmoregulate compared to the control group or
groups which had undergone small salinity fluctuations. Increasing salinity seemed to have
a greater impact on plasma osmolality than did decreasing salinity (Figures 4-1 and 4-2).
Plasma osmolality in groups DI, D2, and D3 all exhibited declines on day 21
relative to day 20, but these differences were not statistically significant. No differences in
regulatory ability among these three groups, or with the control group (group C), could be
discerned. However, group CD, which had been maintained at 30 ppt for the first 20 days of
the experiment, did show a significant decrease in plasma osmolality on day 21 relative to
values measured on days 0 through 20. Furthermore, this group had lower osmolality
values on day 21 than all other groups, although this difference was only statistically
significant when compared to groups C and D3.
More striking differences existed for groups II, 12, 13, and C1. In all cases plasma
osmolality was elevated on day 21 relative to measurements taken from the same group on
day 20 or earlier. Group 13, which had experienced daily fluctuations in salinity of 30 ppt,
showed the smallest increase in osmolality values, being significantly higher than
measurements from day 0 only. Group 12 showed a similar result, with the plasma
osmolality value from day 21 significantly higher than values from day 0 and day 2 only.
Groups Ii and C1 showed the largest changes in plasma osmolality on day 21. Both groups
had osmolality values that were significantly higher than all previous measurements taken
from their respective groups. The value for group I1 was elevated compared to groups 12, 3,
and C, although this increase was statistically significantly only when compared to group C.
Group C1 was significantly elevated when compared to all other groups (including group
The ability to adjust rapidly to altered salinities would be an obvious advantage to
salt marsh organisms. Physiological responses of euryhaline fishes exposed to rapid
changes in salinity can be grouped into two phases (Holmes and Donaldson, 1969): an
adaptive period and a regulatory period. During the adaptive period, plasma osmolality
varies, gradually returning to values approaching original levels. In the regulatory period,
plasma osmolality is more finely controlled as the fish adjusts to the altered salinity and
reaches ionic homeostasis. Fishes which reach the regulatory period quickly (i.e., have short
adaptive periods) should be best able to tolerate alterations in ambient salinity. Although the
length of the adaptive period was not measured in the present study, results nevertheless
suggest that it is relatively short for C. variegatus.
Nordlie (1985) showed that C. variegatus was an excellent osmotic regulator over a
salinity range from 0.3 ppt to 70 ppt. Plasma osmolality values changed only slightly over
this range of salinities in his study, varying by only 40 mOsm kg-1. However, fish in his
study were fully acclimated to each experimental salinity prior to testing. The present study
indicates that C. variegatus is an excellent regulator of plasma osmolality even when fishes
are exposed to large daily fluctuations in salinity. Although fish in all groups in the present
study regulated at slightly higher levels than seen by Nordlie (1985), plasma osmolality
values varied similarly, with differences of less than 40 mOsm kg-1 for all groups on days
0 through 20. The transfer process itself elicited much of this variation, with slight increases
in plasma osmolality seen over the time course of the experiment. A similar trend was seen
in a study by Woo and Wu (1982) on the red grouper, Epinephelus akaara, and the black
sea bream, Mylio macrocephalus.
The influence of a single alteration in salinity on osmoregulatory ability has been
studied for a number of fishes, with changes in plasma osmolality observed in the present
study of similar magnitude to those seen in other euryhaline fishes (Wakeman and
Wohlschlag, 1983; Engel et al., 1987; Ferraris et al., 1988; Mancera et al., 1993; Yoshikawa
et al., 1993; Altimiras et al., 1994; Tort et al., 1994). Few studies have examined how
fluctuating salinities influence osmoregulation of fishes. The best comparison of my results
are with a study on Blennius pholis by Davenport and Vahl (1979). In their study B. pholis
were exposed to alternating periods of freshwater and seawater, with each period lasting 6 h.
Similar to the results from the present study, plasma osmolality did not vary significantly
over the course of the experiment.
No consistent differences in Hct were observed over the course of the experiment.
Nordlie et al., (1995) found that C. variegatus efficiently regulates water content over a
wide range of salinities, with a difference of only 4% observed between 0 ppt and 100 ppt.
These results suggest that there are no large movements of water between the blood and
tissues as a result of fluctuations in salinity over the range tested. The lack of response in
hematocrit in this study may also indicate that within the range of salinities studied C.
variegatus is subject to normal, or tolerance physiological processes, as an increase in
hematocrit values would be expected if C. variegatus were exposed to conditions leading to
resistance processes (Swift, 1982).
It has been hypothesized that fishes occurring in a habitat where salinity often
fluctuates may be able to respond more quickly and completely to alterations in salinity, and
that the limits of tolerance are farther apart if salinity fluctuates periodically (Gunter, 1967;
Spaargaren, 1974; Johnston and Cheverie, 1985). My results provide evidence that prior
exposure to fluctuations in salinity does impart an osmoregulatory advantage. Fishes
previously exposed to large fluctuations in salinity regulated plasma osmolality better than
fishes that had previously experienced no change or small changes in salinity. Increasing
salinity had the greatest impact on regulation of plasma osmolality. Group C1, which had
experienced no prior change in salinity, and group Ii, exposed to the smallest fluctuations in
salinity prior to being transferred to 60 ppt showed large increases in plasma osmolality
compared to the control group and groups which had previously experienced large
fluctuations in salinity. Decreasing salinity groups exhibited the same pattern, but
differences in osmolality were less pronounced. Only group CD, which had not been
previously exposed to fluctuations in salinity, showed a significant decline in plasma
osmolality after transfer to 2 ppt Salinities between 2 ppt and 35-40 ppt are typically
encountered by this population of C. variegatus in its native habitat, with salinities as high
as 60 ppt rarely encountered. Thus, it is not surprising that transfer to 60 ppt elicited greater
changes in regulation of plasma osmolality.
Variations in salinity imposed on fish in this study caused no dramatic effects. No
adverse behavior or mortalities were noted throughout the course of the experiment. Thus,
despite the differences seen in regulation of plasma osmolality between some of the
experimental groups, these differences were not large enough in magnitude to cause
observable distress in the experimental animals. These results indicate that C. variegatus is
well adjusted for life in variable salinity conditions. It would be interesting to compare
results from the present experiments with studies examining the influence of fluctuating
salinity on osmoregulation in fishes that can tolerate changing salinity, but that experience
infrequent salinity variations in their native habitat.
INFLUENCE OF ENVIRONMENTAL SALINITY ON BLOOD OXYGEN LEVELS OF
The sheepshead minnow, Cyprinodon variegatus, is a euryhaline teleost whose
typical habitats are brackish water, coastal salt marshes that experience frequent salinity
fluctuations. Variations in environmental salinity may directly affect the respiratory system
of fishes in at least two ways: by affecting the solubility of oxygen in the water pumped
over the gills, and by affecting the solubility of oxygen dissolved in plasma. Changes in the
ionic composition of bodily fluids may also interact with oxygen to influence tolerance to
variable salinity conditions (Truchot, 1987). Furthermore, fishes in saline water with low
oxygen tension must balance maximizing branchial oxygen diffusion with greater
osmoregulatory demands due to the accompanying increases in ion and water exchange
(Perry and McDonald, 1993). Additionally, the oxygen content of many aquatic habitats is
subject to large natural fluctuations, so oxygen is a potentially limiting factor by itself
(Dejours, 1987; Graham, 1990). This is especially true in shallow waters, where chronic or
periodic hypoxia may be a common phenomenon (Graham, 1990). Salt marsh habitats are
often exposed to hypoxic conditions (Renaud, 1985; Toulmond, 1987).
Cyprinodon variegatus is an extremely competent euryhaline teleost. Over a range
of salinities from freshwater (0 ppt) to 40 ppt, very small changes in metabolism occur
(Nordlic, et al., 1991; this study, Chapter 2). Both oxygen consumption and critical oxygen
tension (Pc) are essentially unaffected by changes in salinity over this range. Salinities
above this range cause metabolic adjustments, with increases in Pc and decreases in
metabolism observed (this study, Chapter 2). The mechanisms involved in maintaining
constant metabolism and Pc at low to moderate salinities, and the alterations of metabolism
and Pc at higher salinities likely involve adjustments in oxygen transport. Not only do high
salinity waters have lowered concentrations of dissolved oxygen, but salinity itself seems to
elicit physiological responses similar to those resulting from lowered levels of oxygen.
Increasing blood oxygen levels with increasing salinity, possibly leveling off at the highest
salinities, could lead to the observed responses in metabolism and Pc in C. variegatus.
Increased blood gas transport is likely the primary mechansim used by most fishes
to increase the amount of oxygen delivered to tissues. Blood oxygen transport in most
teleosts is dependent upon the respiratory pigment hemoglobin. Blood oxygen transport is
normally increased by increasing the concentration of hemoglobin, increasing the number
of erythrocytes in circulation, and/or adjusting the affinity of hemoglobin for oxygen
(Davis, 1975; Wells et al., 1989; Jensen et al., 1993; Perry and McDonald, 1993). I
examined hemoglobin concentration, erythrocyte count, and packed cell volume (hematocrit)
over a wide range of salinities for individuals of C. variegatus to evaluate the influence of
salinity on blood oxygen levels in this extremely euryhaline species.
Collections of fish, transportation back to the laboratory, and general lab procedures
were performed as described previously. Using the same protocol described earlier, fish
were sequentially acclimated to a series of salinities ranging from 0 ppt to 80 ppt (0 to 2285
mOsm kg- 1). At the end of the acclimation period, fish were sacrificed to determine
hemoglobin concentration ([Hb]), hematocrit (Hct), and erythrocyte (RBC) count.
Fish were first carefully netted from their experimental aquaria and blotted dry.
Blood was taken by sternal cardiac puncture using freshly heparinized microhematocrit
tubes drawn to a fine point. Blood from each fish was collected in two microhematocrit
tubes. Once the first tube was filled with a volume > 20 jpl, the blood was immediately
dispensed into a small ceramic crucible. Eppendorf micropipettes were then used to
dispense aliquots of blood from the crucible into test tubes used in the determination of
hemoglobin concentration and erythrocyte count. A second microhematocrit tube was then
filled and used for determination of the hematocrit, and mass and standard length of fish
were then determined. Fish were used without regard to sex, and all blood samples were
taken between 0700 h and 1100 h. Food was withheld from experimental aquaria for 24 h
prior to testing to ensure that all fish were post-absorptive. Statistical procedures followed
Winer et al. (1991) and Sokal and Rohlf (1995). All statistical analyses were one way tests
using the Tukey-Kramer post hoc comparison (p = 0.05).
Hemoglobin concentration was measured spectrophotometrically on 10 ptl samples
using the cyanomethemoglobin method (Brown, 1993). As recommended by Innes and
Wells (1985), the cyanomethemoglobin solutions were centrifuged for 10 min at 5000 g
prior to colorimetric determination to remove erythrocyte debris. This method gives reliable
results when used with fish blood (Blaxhall, 1972; Blaxhall and Daisley, 1973; Coburn,
1973; Innes and Wells, 1985).
Erythrocytes were counted immediately following dilution in test tubes containing
Natt and Herricks solution (Campbell and Murru, 1990). This solution acts as both stain
and dilutent and is routinely used for counting erythrocytes of fish (Campbell and Murru,
1990). A 1:200 dilution was used and cells were counted in an improved Neubaucr
hematocytometer following precautions outlined in Brown (1993).
Hematocrit was measured to determine the packed cell volume of the erythrocytes
contained in the blood. Immediately after filling the second microhematocrit tube one end
was sealed and the tube placed into a micro-hematocrit centrifuge. Tubes were then
centrifuged for 10 min to separate plasma from formed elements. The hematocrit was read
using a micro-capillary reader and expressed as percent erythrocyte.
From the test values obtained, mean corpuscular volume (MCV), mean corpuscular
hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) were
calculated for each fish as follows (Brown, 1993):
Hct x 10' [Hb] (g/ 1) [Hb] (g / dl)
MCV BC MCH B- MCHC- H
RBC/1 'RBC/1 Hct
These erythrocyte indices are used to further define the relationship between hemoglobin
content and size of the erythrocyte.
The various measures of blood oxygen are arranged by salinity group in Table 5-1.
Significant differences over the range of test salinities were found for all parameters except
MCHC. Body mass had no significant influence on any of the measured or calculated
blood oxygen indices.
Salinity exerted the greatest influence on erythrocyte count (Figure 5-1). Values
obtained at 80 ppt and 0 ppt were significantly higher than all other salinities. Erythrocyte
count was next highest at 60 ppt and 70 ppt, being significantly elevated compared to values
in fish acclimated to 2, 10, 20, and 30 ppt Fish acclimated to salinities from 2 ppt through
50 ppt exhibited no significant differences in erythrocyte count.
Measurements of hemoglobin concentration exhibited a similar pattern, although
fewer significant differences were noted (Figure 5-2). Mean values of fishes acclimated to 0
ppt were highest and significantly different from fish acclimated to salinities from 2 ppt
m a C^ 00 ell n m Q cNI
.-: .-: ,. -- 0- 4 ,. i
U -H 1 -H1 1 41 m +4 4H +1 M1
C- J- a' eo 0 '
u t i rt ei Ni 00
> \I N r 0 C z 0t
Sr C- 3 ) 0 o
o1. o -H -H o -
E i n o6 vi \ i 6
H R 4 4 N 0 0 0
.. 0 6 0 -
o S 7 o ooo oo
Q-- --- -- ---
C 1 c>J C-1 N C11
"- aE o 0 o 0 0 .
C C4 on ClC C4 ti)
og o 6 6 6 6 6 6 6 6 o
CJI n- V 00 00 l 0
cso v) ^ f ^ o o o
F;-5 u R I n I I ^ I N I I I I I I
() L) CmJ 1 1-- O0
c ( U_ u01 x) lunoD llJ Poio P9I
( ip 2) uoi1llu;ouoD uiqolgotuiH
through 50 ppt. Hemoglobin concentration was also elevated in fish acclimated to 60 ppt, 70
ppt, and 80 ppt, although these values were only significantly higher than the value of fish
acclimated to 20 ppt.
Hematocrit measurements showed less dependence upon salinity (Figure 5-3).
Mean hematocrit was highest in 0 ppt, and was significantly different than mean values of
fish acclimated to all salinities except 60 ppt and 70 ppt No other significant differences in
hematocrit were noted.
Calculated erythrocyte indices indicated a slightly different pattern. The average
concentration of hemoglobin in the erythrocyte (MCHC) did not vary significantly among
salinity acclimation groups. However, both the average weight of hemoglobin in the
erythrocyte (MCH) and the average volume of the erythrocyte (MCV) were lowest in fishes
acclimated to 80 ppt. MCH was significantly depressed in fish acclimated to 80 ppt when
compared to groups acclimated to 2, 30, 60, and 70 ppt, with MCV values at 80 ppt
significantly lower than values obtained for groups at 10, 20, 30, 50, and 70 ppt.
Changes in environmental salinity can exert profound effects on blood oxygen
transport. Increases in salinity confront fishes with the necessity of satisfying oxygen
requirements under conditions of reduced oxygen availability. Fishes may exploit multiple
strategies to optimize blood oxygen transport. The amount of oxygen delivered to the
tissues by the blood per unit time is a product of the cardiac output, the oxygen tension
difference between arterial and venous blood, and the blood oxygen capacitance coefficient
(Jensen, 1991; Jensen et al., 1993). The capacitance coefficient reflects the hemoglobin's
oxygen transporting properties, and can be adjusted in what have been termed 'qualitative'
and 'quantitative' ways (Jensen, 1991). Regulation of hemoglobin-oxygen (Hb-O2) affinity
represents the primary method for qualitatively altering oxygen carrying capacity, with
control of hemoglobin concentration the primary quantitative mechanism (Jensen, 1991).
0) I I
Blood oxygen carrying capacity can also be increased quantitatively by release of stored
erythrocytes, by accelerating maturation of immature erythrocytes, and/or by production of
new erythrocytes (Murad et al., 1990); release of erythrocytes from storage organs (e.g.,
spleen) appears to be the most likely scenario (Soivio et al., 1980; Wells et al., 1989). Fish
exposed to water of changing salinity would be expected to experience variability in their
blood oxygen capacitance coefficient and blood oxygen carrying capacity (Jensen et al.,
1993). Quantitative mechanisms for adjusting blood oxygen carrying capacity were
examined in this study.
Few studies have examined the influence of salinity on oxygen carrying capacity of
fishes. Guernsey and Poluhowich (1975) examined the blood oxygen capacity of American
eels (Anguilla rostrata) acclimated to 0 ppt, 24 ppt, and 34 ppt. As in C. variegatus,
hematocrit was highest in eels acclimated to 0 ppt. However, while oxygen capacity of
acclimated eels was higher in 0 ppt than 34 ppt, the highest oxygen capacity was seen in
eels acclimated to 24 ppt. In a similar study with the cichlid Oreochromis niloticus, Sun et
al., (1995), observed a similar effect of salinity on measures of blood oxygen, with
hemoglobin concentration significantly higher in 0 ppt than in higher salinities (5 to 20
Other factors may also contribute to variations in oxygen carrying capacity of
fishes. Hall and Gray (1929) were among the first to note that there is a general correlation
between the habits of fishes and the hemoglobin concentration of their blood. More recent
studies have shown that this generalization also applies to erythrocyte count and hematocrit
(e.g., Haws and Goodnight, 1962; Coburn, 1973; Larsson et al., 1976; Putnam and Freel,
1978). In general, highly active fishes and those that regularly encounter hypoxic conditions
have elevated blood oxygen carrying capacity relative to other species. Values determined
for C. variegatus in the present study are comparable to other fishes with similar activity
levels, and indicate that C. variegatus does not possess exceptionally high oxygen carrying
capacity at any salinity tested (Hattingh, 1972; Coburn, 1973; Larsson et al., 1976; Putnam
and Freel, 1978; Smit and Hattingh, 1979; Pelster et al., 1988b).
Salinity had a significant effect on blood oxygen carrying capacity in C. variegatus.
However, differences were seen only at the very highest and lowest salinities tested. As
expected, oxygen carrying capacity increased at high salinities. Elevations were seen at 60
ppt through 80 ppt when compared to measures made at salinities between 2 ppt and 50 ppt.
One finding of interest was indicated by MCV measurements. These results showed that at
80 ppt overall size of erythrocytes had decreased, implying the production and/or release of
smaller, possibly immature erythrocytes. One possible explanation for this is that smaller
erythrocytes have been found to be more effective in oxygen exchange than larger
erythrocytes (Coburn, 1973) due to a larger surface area per unit volume, which may allow a
faster rate of gas exchange. Energetic constraints related to viscosity of the blood may also
play a role. A balance must exist between the advantages for oxygen transport from
increased hematocrit with a disadvantage due to increased viscosity (Wells and Weber,
1991). Hematocrit began to increase as salinity reached 60 ppt, but declined back towards
normal values at 80 ppt. It is possible that at extremely high salinities even small increases
in hematocrit lead to significant viscosity problems. Utilizing smaller erythrocytes at
extreme hypersalinities may help offset this problem.
Unexpectedly, oxygen carrying capacity was highest in the group acclimated to 0
ppt, as indicated by measurements of all blood indices. Although freshwater is tolerated by
C. variegatus, previous work has shown that acclimation to freshwater is difficult and may
result in significant mortality unless decreases in salinity are made in small steps (Nordlie
and Walsh, 1989; Nordlie et al., 1991). Survival in freshwater requires many of the same
responses that are necessary at extremely high salinities, with both freshwater and
hypersaline conditions imposing difficult osmoregulatory problems for C. variegatus. In
both situations, proliferation of mitochondria rich cells on gill epithelia is needed to
maintain ionic balance (Evans, 1984; Evans, 1993; Wood and Marshall, 1994). However,
this extends the surface involved in ionic exchange at the expense of gill epithelia involved
in gas exchange. Recent studies have shown that such proliferation of mitochondria rich
cells does impair respiratory gas transfer (Bindon et al., 1994a; Bindon et al., 1994b).
Mechanisms to increase oxygen carrying capacity of the blood would be expected under
However, freshwater conditions differ significantly from hypersaline conditions in
several ways. Most importantly, metabolism is reduced at extreme hypersalinities (this
study, Chapter 2). In conjunction with elevated Pc, depressed metabolism at these
hypersalinities greatly reduces energetic expenditures, partially alleviating the need for
increased oxygen carrying capacity. Thus, whereas measures of blood oxygen are elevated
at salinities of 60 ppt and higher, increases were moderated by a reduction in overall
energetic expenditures. Fish acclimated to 0 ppt exhibit insignificant reductions in
metabolism, so possible increases in oxygen needs in freshwater can not be compensated
for in this manner. Furthermore, a number of studies have indicated that Hb-02 affinity is
decreased in freshwater conditions (e.g., Benditt et al., 1941; Weber et al., 1976; Woo and
Wu, 1982). Thus, oxygen carrying capacity may have to be increased in quantitative ways if
oxygen affinity of the hemoglobin molecule is decreased in dilute media.
Another finding from the present study was that C. variegatus exhibits little change
in quantitative measures of oxygen carrying capacity over the range of salinities between 2
ppt and 50 ppt. Previous studies have shown that salinities within this range have little
effect on the metabolic rate, Pc, and osmoregulatory ability of C. variegatus (this study,
Chapters 2,3, and 4). However, as oxygen needs would be expected to rise as salinity was
increased over this range, oxygen carrying capacity was expected to increase as well.
Several possibilities may explain why blood oxygen carrying capacity did not
increase over this range of salinities. First, increases in salinity may require little
compensation in oxygen carrying capacity until extreme hypersalinities are reached.
However, it is more likely that the lack of a response may have been due to the fact that
large variations in salinity alone seldom occur under natural conditions. A larger response
would be expected if salinity were varied together with oxygen and/or temperature. Another
possibility is that C. variegatus, like many fish species, may utilize multiple hemoglobins.
Weber (1990) noted that the use of multiple hemoglobins can extend the range of
conditions under which oxygen is transported efficiently, thus enlarging the range of
available habitats. If C. variegatus utilizes such a mechanism, large quantitative changes in
oxygen carrying capacity might not be seen.
Finally, C. variegatus may exhibit mostly qualitative changes in response to salinity,
i.e., the primary response to changing salinity may be better reflected in Hb-02 affinity.
Changes in Hb-02 affinity as a result of environmental changes are extremely common in
fishes (e.g., Johansen and Weber, 1976; Wells et al., 1980; Weber, 1981), although this has
rarely been examined with respect to salinity. However, studies on Anguilla anguilla
(Weber et al., 1976) and Salmo salar (Maxime et al., 1990) did find that increases in
salinity between freshwater and seawater led to increases in Hb-02 affinity. A high affinity
hemoglobin molecule might also be advantageous under hypersaline conditions, although it
may be ineffective during activity (McMahon, 1988). As Hb-02 affinity was not measured
in this study, direct correlation with C. variegatus is purely speculative at this time.
Nevertheless, changes in Hb-O2 affinity would correlate well with the large decreases in
energetic expenditures exhibited by C. variegatus at high salinities. However, other factors
could also be involved. First, C. variegatus may exhibit low salt sensitivity over the normal
range of salinities encountered to reduce dependence of blood oxygen affinity on
environmental salinity (Weber et al., 1976). Second, the primary mechanism used to alter
Hb-02 affinity results from swelling of erythrocytes (Jensen, 1991; Wells and Weber,
1991; Jensen et al., 1993). Swelling of erythrocytes is normally accompanied by a rise in
cell pH and decreased red cell hemoglobin, NTP and ATP concentrations, all of which serve
to increase the oxygen affinity (Wells and Weber, 1991). However, as seen by the MCHC
and MCV values determined in this study, erythrocytes did not swell in response to
increased salinity in C. variegatus. Thus if Hb-O2 affinity is altered with changes in
environmental salinity, it must be changed in some other manner.
This study clearly indicates that salinity does influence the oxygen carrying capacity
of the blood of C. variegatus. Quantitative differences in hemoglobin concentration,
hematocrit, and erythrocyte count were noted in response to changing salinity. As discussed
above, it also seems likely that salinity may influence qualitative changes in blood oxygen
transport in C. variegatus. Further research is needed to better understand the influence of
salinity on blood oxygen levels in euryhaline teleosts.
SUMMARY AND CONCLUSIONS
This study examined costs associated with life of a teleost in a variable salinity
environment, represented here by a salt marsh. Cyprinodon variegatus was used to examine
the influence of salinity on routine metabolic rate (RMR), critical oxygen tension (Pc),
osmoregulation, and blood oxygen carrying capacity. Results are summarized below.
1) Field measurements in the Cedar Key salt marsh indicated that this habitat undergoes
extensive variation in salinity, temperature, and oxygen.
2) RMR was relatively constant over a range of salinities from 0 ppt to 40 ppt. At higher
salinities RMR began to decline, and was significantly depressed under hypersaline
3) Following sequential acclimation to experimental salinities, PC was unaffected by
changes in salinity between 0 ppt and 40 ppt, with Pc increasing at higher salinities.
4) Reduction in metabolism and rise in Pc corresponded well with a reduced ability of C.
variegatus to regulate plasma osmolality efficiently. Osmotic permeability of the gills may
be reduced at high salinities to offset osmotic losses or ionic gains to/from the environment,
indirectly reducing the potential for oxygen uptake as well.
5) Variations in RMR and Pc as a function of environmental salinity observed in this study
suggest that C. variegatus responds to high salinities by reducing energy expenditures.
These responses effectively increases the time C. variegatus can tolerate such conditions,
albeit at a cost of a reduction in energetic processes. This strategy fits the concept of scope
for survival, as described by Hochachka (1990).
6) When C. variegatus was exposed to simulated tidal changes in salinity, RMR was
unaffected in salinity trials where both acclimation and final salinities were in the range
typically encountered by this population in its native habitat. Where the acclimation or final
salinities were extremely high (50 and 60 ppt) or extremely low (0 ppt), RMR was
7) Acclimation state was the most important factor determining the metabolic response to
simulated tidal changes in salinity. However, direction of the salinity change also influenced
metabolism in C. variegatus, with increasing salinity dealt with more efficiently than
8) Simulated tidal experiments corroborate the hypothesis that C. variegatus tolerates
extremes in salinity by lowering metabolism, and hence decreasing energy expenditures.
Following adverse conditions metabolism returns to normal levels.
9) Cyprinodon variegatus is an excellent regulator of plasma osmolality even when
exposed to large fluctuations in salinity within the range of salinities typically encountered.
Daily fluctuations in salinity of up to 30 ppt elicited no significant differences in
osmoregulatory ability when compared to control fish.
10) Prior exposure to fluctuations in salinity does impart an osmoregulatory advantage.
Fishes previously exposed to large fluctuations in salinity regulated plasma osmolality
better than fishes that had previously experienced no or small changes in salinity.
Increasing salinity had a greater impact on regulation of plasma osmolality than did
decreases in salinity.
11) Salinity had a significant effect on blood oxygen carrying capacity in C. variegatus,
although differences were only noted at the very highest (60 to 80 ppt) and lowest (0 ppt)
salinities tested. Oxygen carrying capacity and all blood indices were highest in the group
acclimated to 0 ppt.
12) C. variegatus exhibited little change in oxygen carrying capacity over the range of
salinities between 2 ppt and 50 ppt. Possible reasons for this include: (a) increases in
salinity may require little compensation in oxygen carrying capacity; (b) detectable changes
may only occur when salinity is varied in conjunction with variations in oxygen and/or
temperature; (c) C. variegatus may utilize multiple hemoglobins, and/or; (d) the primary
mechanism to increase oxygen carrying capacity may instead be through adjustment of Hb-
13) Erythrocyte count was the most consistent and hematocrit the least consistent measure
of the influence of salinity on blood oxygen level.
Competition or predation pressure may be less intense in harsh, fluctuating
environments, and that certain species may avoid these pressures by evolution of wide
physicochemical tolerances and the use of such environments (Matthews and Styron,
1981). Cyprinodon variegatus seems to fit this mold well, as this species appears to be a
generalist that very successfully inhabits harsh and variable habitats where it does not have
to be very efficient to compete with other species of fishes (Martin, 1972; Berry, 1987).
This argument may explain why C. variegatus does not invade freshwater in more locales,
and why they are not very abundant in most freshwater systems; except in south Florida
freshwaters where temperature and dissolved oxygen are variable and extreme.
Furthermore, it is believed that C. variegatus has conquered its wide geographic
range by physiological flexibility rather than by local accommodation in physiology or life
history (Berry, 1987). This hypothesis remain to be rigorously tested, although this study
lends further evidence for the physiological flexibility of C. variegatus. Examination of the
physiological ability of other populations of C. variegatus may help to resolve this issue.
The success of a fish species in a stressful environment may depend primarily
upon the proportion of the population that survives the adverse conditions, with even short-
term increases in tolerance significant (Matthews and Styron, 1981). This hypothesis
appears to be relevant to the results of the present study. Cyprinodon variegatus is well
adapted to a varying salinity environment. Its metabolic rate, Pc, and ability to efficiently
osmoregulate is unaffected by changes in salinity over the typical range encountered, even
when salinity is changed very rapidly. Furthermore, C. variegatus appears to tolerate
extremes in salinity by decreasing energy expenditures, waiting for conditions to improve,
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ELIYC40N1_1XLF0E INGEST_TIME 2012-02-20T23:02:44Z PACKAGE AA00009024_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC