Swim Bladder Function and Buoyancy Control in Pink Snapper and Mulloway

Fish Physiol Biochem (2014) 40:335346
DOI 10.1007/s10695-013-9846-y
Swim bladder function and buoyancy control in pink

snapper (Pagrus auratus) and mulloway (Argyrosomus
japonicus)
John Stewart Julian M. Hughes
Received: 4 June 2013 / Accepted: 13 August 2013 / Published online: 25 August 2013
Springer Science+Business Media Dordrecht 2013
J. Stewart J. M. Hughes
Cronulla Fisheries Research Centre of Excellence,
Cronulla, Australia
seawater (1.026 g/ml), whereas mulloway were significantly denser than seawater. Pink snapper secreted
gas into their swim bladders at a rate of
0.027 0.005 ml/kg/min (mean SE), almost 4
times faster than mulloway (0.007 0.001 ml/kg/
min). Rates of swim bladder gas resorption were 11
and 6 times faster than the rates of gas secretion for
pink snapper and mulloway, respectively. Pink snapper resorbed swim bladder gas at a rate of
0.309 0.069 ml/kg/min, 7 times faster than mulloway (0.044 0.009 ml/kg/min). Rates of gas
exchange were not affected by water pressure or water
temperature over the ranges examined in either
species. Pink snapper were able to acclimate to
changes in hydrostatic pressure reasonably quickly
when compared to other marine teleosts, taking
approximately 27 h to refill their swim bladders from
empty. Mulloway were able to acclimate at a much
slower rate, taking approximately 99 h to refill their
swim bladders. We estimated that the swim bladders
of pink snapper and mulloway ruptured after decreases
in *2.5 and 2.75 times the hydrostatic pressure to
which the fish were acclimated, respectively. Differences in buoyancy, gas exchange rates, limitations to
vertical movements and acclimation times between
the two species are discussed in terms of their differing
behaviour and ecology.
J. Stewart (&) J. M. Hughes

Fisheries NSW, Sydney Institute of Marine Science,
Chowder Bay Road, Mosman, NSW 2088, Australia
e-mail: John.Stewart@dpi.nsw.gov.au
Keywords Swim bladder Buoyancy control

Snapper Mulloway Pagrus Argyrosomus
Hyperbaric chamber
Abstract Physoclist fish are able to regulate their

buoyancy by secreting gas into their hydrostatic organ,
the swim bladder, as they descend through the water
column and by resorbing gas from their swim bladder
as they ascend. Physoclists are restricted in their
vertical movements due to increases in swim bladder
gas volume that occur as a result of a reduction in
hydrostatic pressure, causing fish to become positively
buoyant and risking swim bladder rupture. Buoyancy
control, rates of swim bladder gas exchange and
restrictions to vertical movements are little understood
in marine teleosts. We used custom-built hyperbaric
chambers and laboratory experiments to examine
these aspects of physiology for two important fishing
target species in southern Australia, pink snapper
(Pagrus auratus) and mulloway (Argyrosomus japonicus). The swim bladders of pink snapper and
mulloway averaged 4.2 and 4.9 % of their total body
volumes, respectively. The density of pink snapper
was not significantly different to the density of
123
336
Introduction
Many teleosts possess a gas-filled swim bladder that
acts primarily as a hydrostatic organ and, in some
species, secondarily for sound production (Harden
Jones 1951; Alexander 1959; Hallacher 1974). Teleosts may be either physoclists, having a closed swim
bladder, or physostomes, having a swim bladder which
is connected to their oesophagus via a duct, allowing
uptake and release of gas through the mouth. Two
simple laws of physics affect buoyancy control in
species with gas-filled swim bladders. Firstly, Pascals
law states that hydrostatic pressure increases with
water depth, such that in the marine environment
pressure increases by *1 atm for every 10 m of
depth. Secondly, Boyles law states that at constant
temperature the volume of a gas is inversely proportional to the pressure: for example, when pressure is
halved, the volume of the gas doubles. As fish ascend
through the water column, the decrease in hydrostatic
pressure causes the swim bladder to increase in
volume and the fish to become positively buoyant.
Conversely, as fish descend through the water column,
the increase in hydrostatic pressure causes the swim
bladder to decrease in volume and the fish will become
negatively buoyant. Physoclist fish secrete gas into the
swim bladder in response to negative buoyancy and
resorb gas from the swim bladder in response to
positive buoyancy, but both processes take time
(Tytler and Blaxter 1973; Harden Jones and Scholes
1985). At any point in time, the volume of the swim
bladder determines the depth of neutral buoyancy
(Stensholt et al. 2002). Around this depth of neutral
buoyancy, there exists a zone through which fish can
move freely, termed free vertical range (FVR),
compensating for non-neutral buoyancy by swimming
(Harden Jones and Scholes 1985; Stensholt et al. 2002;
Strand et al. 2005). FVR is therefore governed by
swim bladder volume, the rate of gas exchange and the
pressure under which the fish is exposed (i.e. water
depth). It has been hypothesized that physoclistous
fish approach neutral buoyancy only at the top of their
FVR (Davenport 1999), therefore minimizing the risk
of an uncontrollable rapid ascent causing the swim
bladder gas to expand at a rate faster than gas can be
resorbed and risking swim bladder rupture. However,
the energetic costs of having to swim constantly in
order to maintain depth when negatively buoyant must
outweigh the risks associated with potential swim
123
bladder rupture. In fact, the position within a fishs

FVR that it prefers to occupy will be species-specific
and determined by ecologically driven behavioural
patterns (e.g. Lea et al. 1999; Parker et al. 2006;
Hannah and Matteson 2007). There exists very little
information concerning the FVR of fish, but for cod
(Gadus morhua) FVR has been estimated experimentally (Harden Jones and Scholes 1985) and confirmed
via acoustic profiles (Stensholt et al. 2002) to be
approximately between a 25 % pressure decrease and
a 50 % pressure increase, from the depth at equilibrium. The FVR in terms of distance therefore increases
with increasing pressure as a result of Pascals law.
Knowledge of buoyancy control in fish, including
the FVR and the rate at which it changes, is important
in understanding restrictions to vertical movements
and in developing hypotheses regarding fish behaviour, including diel migrations for activities such as
feeding and spawning (Holbrook and de Perera 2011).
Such understanding is of particular importance for
species which are important to fisheries, both in terms
of knowing when and where fish may be susceptible to
different fishing gears, but also in interpreting acoustic
survey data to estimate relative abundance (Stensholt
et al. 2002). Another emerging area of research that
requires information on the rates at which fish are able
to adjust swim bladder volume to achieve neutral
buoyancy (termed acclimation) is that of assessing
decompression injuries (barotrauma) in fish, particularly in manipulative experiments where fish need to
be acclimated to certain water pressures to simulate
capture from depths. Many of these experiments use
custom-built hyperbaric chambers in which fish are
acclimated to different pressures before being depressurized to simulate capture and their survival rates
quantified (e.g. Gaspin et al. 1978; Harden Jones and
Scholes 1985; Rummer and Bennett 2005; Parker et al.
2006; Pribyl et al. 2009). Clearly, acclimation times
need to be known if experiments are to accurately
simulate the effects on fish of capture from any
particular depth.
Physoclists secrete gas from the blood into the
swim bladder through the multitude of parallel capillaries of the rete mirabile contained within the gas
gland which contacts the swim bladder lumen (Wittenberg et al. 1964). The arterial flow of blood through
the counter-current arrangement of the rete mirabile
multiplies blood gas tensions in the gas gland so that it
may exceed that of the arterial blood several-fold and
allows gas to be secreted against large pressure

gradients into the swim bladder (Strand et al. 2005).
Gas is removed via passive diffusion into the blood
through a capillary bed known as the resorption
chamber or oval (Ross 1979). Unfortunately, there
have been relatively few studies that have attempted to
quantify the rates of swim bladder gas secretion and
resorption and, with the exception of only a couple of
species (cod: Harden Jones and Scholes 1985; black
rockfish Sebastes melanops and China rockfish S.
nebulosus: Parker et al. 2006), this aspect of teleost
physiology is largely unknown.
The majority of studies in this field have estimated
fish acclimation times from observations of fish
behaviour whilst pressurized in hyperbaric chambersfish with positive buoyancy tend to be oriented
head-down and rise when motionless, fish with neutral
buoyancy tend to hover with a horizontal orientation
and fish with negative buoyancy tend to have a headup orientation and sink when motionless (Gaspin et al.
1978; Harden Jones and Scholes 1985; Rummer and
Bennett 2005; Parker et al. 2006; Pribyl et al. 2009).
The resulting acclimation rates are given in units of
pressure per time (e.g. atm/hParker et al. 2006) or
simply as days to acclimate to a particular depth, e.g.
Gaspin et al. (1978). Wittenberg et al. (1964) used a
syringe to extract all gas from the swim bladders of a
range of species and reported the time taken in h for
the swim bladders to fully re-inflate. However, fish
size is related to gas secretion rate as bigger fish have a
larger gas gland surface area and gas secretion rates
are therefore more usefully reported as a mass-specific
rate (e.g. per unit of body weight: Harden Jones and
Scholes 1985). In addition, one of the most comprehensive studies on swim bladder gas exchange rates
(Harden Jones and Scholes 1985), and the only one to
report actual gas exchange rates in a marine teleost,
may have misreported these exchange rates because
their calculations were based on the volume of the
swim bladder in cod being 5 % of the total fish volume
when at equilibrium. Davenport (1999) has since
shown this to be inaccurate and that in cod the swim
bladder occupies on average less than 4 % of fish
volume.
The sparid Pagrus auratus (pink snapper) and the
sciaenid Argyrosomus japonicus (mulloway) are two
of the most highly prized recreational and commercial
fishing targets in coastal waters of southern Australia.
Both species use estuaries as nurseries with adults
337
moving offshore and being common to depths of

100 m (Kailola et al. 1993). Scientific assessments for
these species indicate that both are overfished in some
parts of their distributions and that recovery programmes need to be developed (Rowling et al. 2010).
In the state of New South Wales (NSW), Australia,
management arrangements for pink snapper and
mulloway include minimum legal lengths and recreational bag limits. These management arrangements
result in the subsequent release of [50 % of each
species captured (Henry and Lyle 2003) and consequently research into the survival of these released fish
is paramount. Limited work has been done to investigate the effects of barotrauma on these species
(Stewart 2008; Lenanton et al. 2009; Butcher et al.
2012) with field-based assessments proving difficult.
The research presented here was done as a precursor
to a series of experiments that utilized custom-built
hyperbaric chambers to simulate various catch and
release scenarios and to investigate subsequent rates of
mortality in pink snapper and mulloway. The objective
of the present study was therefore to examine buoyancy control and estimate rates of swim bladder gas
exchange for pink snapper and mulloway. We directly
measure changes to swim bladder volume under
different water pressures and times, thus enabling, for
the first time, direct calculation of gas exchange rates.
Methods
Experimental fish collection
Pink snapper
Pink snapper used in these experiments were captured
using hook and line from shallow (\10 m) water
within the Port Hacking estuary (348040 2400 S,
1518070 4300 E) during 2010. They were taken from
shallow water to minimize any barotrauma injuries
and maintained in a 35,000-l flow-through aquarium
for between 6 and 9 months prior to being used in
experiments.
Mulloway
Mulloway used in these experiments were purchased
from an aquaculture facility (Clearwater Marine
Farms, Pty Ltd) to ensure that they had no history of
123
338
barotrauma. They were maintained in a 1-million l

flow-through seawater pond for between 6 and
9 months prior to being used in experiments. The fish
averaged 45.2 cm in total length (TL) (range
41.151.5 cm) and were considered an ideal size
because mulloway must be legally released in NSW if
\45 cm TL (their minimum legal length).
Swim bladder size and fish buoyancy
The relationships between swim bladder volume and
various fish metrics were estimated using subsamples
of experimental fish which were housed at sea level.
Fish were euthanized using a lethal dose of anaesthetic
(Aqui-S New Zealand, Ltd) before being measured
(fork lengthFL for snapper and TL for mulloway) to
the nearest mm, patted dry with absorbent paper and
weighed to the nearest 0.1 g. Fish had their volumes
estimated by weighing the amount of seawater
displaced when they were fully immersed and dividing
this value by the density of seawater (1.026 g/ml).
Fish density was calculated as body weight/fish
volume. The volume of gas within the swim bladders
of fish was estimated using an 18-gauge hypodermic
needle and syringe to extract the gas. This method
appeared to provide accurate measurements of swim
bladder gas volume as no leakage of gas was apparent
once the syringe plunger had been drawn to its
stopping point (see also Wittenberg et al. 1964).
Gas secretion rates
Swim bladder gas secretion rates were estimated using
two custom-built hyperbaric chambers. The chambers
consisted of a 1000-l capacity fibreglass pool filter
(SMD1050; Waterco, Ltd), with the internal plumbing
modified (Fig. 1). The chambers were capable of
simulating depths of up to 70 m (8 atm). Seawater was
pumped into the chambers from a flow-through
seawater pond using submersible borehole pumps
(J725D-3; Davey Water Products, Pty Ltd) capable of
pumping water at a flow rate of 22.540.0 l/min under
pressures of 80 atm, respectively. Internal pressures
in the chambers were controlled by diaphragm valves
Valves, Inc.) which restricted
(50 mm PVC; GEMU
the flow of water out of the chamber. Internal pressures
were measured using digital pressure gauges (XP2iAX; Crystal Engineering Corp.).
123
Manipulative experiments were done in the chambers to quantify the amount of gas secreted into the
swim bladders of pink snapper and mulloway at
different water temperatures, water pressures and
times of exposure to these variables. The experimental
procedure used was based on the understanding that
under increased pressure, and subsequent reduction in
swim bladder volume resulting from Boyles law, fish
will secrete gas into their swim bladders until it
occupies the same volume as when at equilibrium.
Fish were placed in the hyperbaric chambers and the
pressure increased to the desired level at a descent rate
of *1 m/s.
Pink snapper were exposed to two different pressures: 2 and 3 atm (equivalent to the pressures at 10
and 20 m water depth, respectively). Duration at
pressure varied between 1 and 4 h before the pressure
was decreased to sea level (1 atm) at an ascent rate
of *1 m/s, the fish were then euthanized using a
lethal dose of anaesthetic as described above and
removed from the chamber. The fish were immediately measured, patted dry and weighed, and the
volume of gas within the swim bladder extracted using
a hypodermic needle and syringe as described above.
It was assumed that no gas escaped from the swim
bladder prior to this extraction. This was done for 10
fish during winter water temperatures (1516 C) and
11 fish during summer water temperatures (1920 C).
Mulloway were exposed to three different pressures: 2, 3 and 4 atm (equivalent to the pressures at 10,
20 and 30 m water depth, respectively). Duration at
pressure varied between 2 and 22 h before the pressure
was decreased to sea level, and the fish treated in the
same manner as described above for pink snapper.
This was done for 8 fish during spring water temperatures (1517 C) and 12 fish during late summer
water temperatures (2023 C).
The quantity of swim bladder gas within an
individual fish at equilibrium was estimated using
the fish weight to swim bladder gas relationships
determined for each species (see above). The quantity
of gas secreted during the experiment was calculated
as the volume of gas extracted using the syringe
following time at pressure in the chamber, adjusted for
its expansion between the pressure at which it was
secreted and sea level, minus the estimated volume of
gas at sea level equilibrium. Gas secretion rates were
reported as ml/kg fish body weight/min (Harden Jones
and Scholes 1985).
339
Fig. 1 Line drawing of the

hyperbaric chamber
Pressure gauge
Lid. Oval 415 x 310 mm
Sight glass
100 mm
Inflow
1775 mm
Diaphragm valve
Outflow
Sight glass
100 mm
Drain plug
1045 mm
A linear model (LM) was used to test the effects of

water pressure and water temperature, and their
interaction on gas secretion rate (ml/kg/min).
The model was calculated using the freeware
statistical package R (R Development Core Team
2006). The significance of each variable to the model
was tested using the null hypothesis that they were
significantly different from 0 using partial z tests.
Gas resorption rates
Swim bladder gas resorption rates were estimated
using the hyperbaric chamber setup described above.
Manipulative experiments were done to quantify the
amount of gas resorbed from the swim bladders of
pink snapper and mulloway at different water temperatures and water pressures. Gas resorption rates were
reported as ml/kg fish body weight/min (Harden Jones
and Scholes 1985).
The experimental procedure used was to place
individual fish in the chamber and increase the
pressure to a desired level (P1) at a descent rate
of *1 m/s. The fish were maintained at P1 for a
period that allowed them to secrete sufficient gas into

their swim bladders to achieve neutral buoyancy.
These periods were determined using the rates of gas
secretion estimated previously. Following acclimation, the pressure in the chamber was decreased to a
level whereby the fish were observed to be positively
buoyant but not uncontrollably so (P2). The fish were
maintained at P2 for a period estimated to be shorter
than the period necessary to resorb all excessive gas
from the swim bladder based on the understanding that
swim bladder gas resorbs substantially (up to 5 times)
faster than it is secreted (Parker et al. 2006). The
chamber pressure was then decreased to sea level
(1 atm) at an ascent rate of *1 m/s, the fish were then
euthanized using a lethal dose of anaesthetic as
described above and removed from the chamber.
The fish were immediately measured, patted dry and
weighed, and the volume of gas within the swim
bladder extracted using a hypodermic needle and
syringe as described above.
The volume of gas within an individual fishs swim
bladder at equilibrium was estimated using the fish
weight to swim bladder gas volume relationships
123
340
determined above for each species. This volume was then

adjusted according to Boyles law to estimate the volume
it would occupy at sea level if changed from P1. The
volume (at sea level) of gas resorbed was then calculated
from the difference between this estimate and the quantity
extracted using the syringe, adjusted for the pressure at
which it was resorbed according to the volume it would
have occupied at P2, again using Boyles law.
Pink snapper were acclimated to pressures between
2 and 2.3 atm (equivalent to 10 and 13 m water
depths, respectively). Our estimates of gas secretion
rates suggested that pink snapper would acclimate
from sea level to a depth of 10 m after 14 h on average
and *16 h using the mean secretion rate minus 1 SE.
The fish were subsequently left in the chamber at P1
for between 18 and 22 h before being depressurized to
P2 which was either 1.25 or 1.5 atm (equivalent to 2.5
or 5 m water depth, respectively). Pink snapper were
maintained at P2 for between 30 and 40 min before
being depressurized at an ascent rate of *1 m/s to sea
level. This was done for 12 pink snapper during winter
water temperatures (1516 C) and 7 pink snapper
during summer water temperatures (1920 C).
Mulloway were acclimated to pressure between 2
and 2.25 atm (equivalent to 10 and 12.5 m water
depths, respectively). Our estimates of gas secretion
rates suggested that mulloway would acclimate from
sea level to a depth of 10 m after 46 h on average
and *53 h using the mean secretion rate minus 1 SE.
The relatively protracted time needed to acclimate
mulloway meant that it was possible to only do 1 run
per week and so to ensure that the mulloway were fully
acclimated they were left in the chamber at P1 for at
least 7 days (between 190 and 209 h) before being
depressurized to P2. P2 was either 1.3 or 1.6 atm
(equivalent to 3 or 6 m water depth, respectively). The
mulloway were maintained at P2 for between 4 and
6 h before being depressurized to sea level. This was
done for 8 mulloway during winter water temperatures
(18 C) and 7 mulloway during summer water temperatures (2223 C).
A LM was used to test the effect of water pressure
and water temperature, and their interaction on gas
resorption rate (ml/kg/min).
Free vertical range
The FVR of pink snapper and mulloway was estimated
in terms of limitations to their vertical ascent due to
123
swim bladder rupture. The increase in swim bladder

gas volume that occurred before the swim bladder
ruptured was investigated for pink snapper. Pink
snapper were euthanized, patted dry with absorbent
paper and weighed to the nearest 0.1 g. The body wall
of the gut cavity of each fish was then cut away so that
the swim bladder could be seen, being careful not to
interfere with its structural integrity. A syringe with an
18-gauge hypodermic needle was inserted through the
side of the fish into the swim bladder and air forced in
using pressure on the syringe plunger. The quantity of
air inserted before swim bladder rupture (determined
by either explosive bursting or obvious gas leakage)
was recorded. The relationship between swim bladder
volume and fish weight determined above was used to
estimate the amount of gas already in the swim bladder
and the relative increase in pressure before rupture
calculated from these data.
The decrease in water pressure before swim
bladder rupture occurred in mulloway was investigated using the hyperbaric chambers, rather than by
forcing air into the swim bladder using a needle and
syringe. The mulloway swim bladder was found to
be so flexible and strong that it was impossible to
force in sufficient air to cause rupturethe air
continually escaped through the small hole made by
the needle. Three mulloway were therefore acclimated to each of 10, 15, 20 and 25 m water depths
before being depressurized at an ascent rate
of *1 m/s to sea level. The fish were euthanized
as described above and dissected to examine whether
swim bladder rupture had occurred.
Results
Pink snapper
Pink snapper sampled varied between 10.8 and
31.0 cm FL. The mean ( SE) per cent of fish volume
that was swim bladder gas was 4.2 0.1 % (n = 10).
The mean ( SE) density of pink snapper was
1.025 0.005 g/ml which was not significantly different to the density of seawater (1.026 g/ml) (one
sample t test, p = 0.79, n = 23). The relationship
between fish volume (ml) and body weight (g) for pink
snapper was linear:
Fish volume 0:97 body weight 0:85;

r 2 0:997; n 23
The linear relationship between swim bladder gas
volume (ml) and body weight (g) (Gas volume = 0.05 9 body weight0.05, r2 = 0.975,
n = 26) was a better fit than the linear relationship
with fish volume (ml) (Gas volume = 0.03 9 fish
volume ? 1.32, r2 = 0.890, n = 10).
Mulloway
Mulloway sampled varied between 39.4 and 55.5 cm
TL. The mean ( SE) per cent of fish volume that was
swim bladder gas was 4.9 0.1 % (n = 14). The mean
( SE) density of mulloway was 1.038 0.002 g/ml
and was significantly different to the density of seawater
(1.026 g/ml) (one sample t test, p \ 0.001, n = 26).
The relationship between fish volume (ml) and body
weight (g) for mulloway was linear:
Fish volume 0:94 body weight 19:78;
r 2 0:999; n 26
The linear relationship between swim bladder gas
volume (ml) and body weight (g) (Gas volume =
0.04 9 body weight ? 2.79, r2 = 0.973, n = 16)
was a similar fit to the linear relationship with fish
volume (ml) (Gas volume = 0.04 9 fish volume ?
1.23, r2 = 0.973, n = 14).
Gas secretion rates
The method of quantifying swim bladder gas secretion
rates by using a hypodermic needle and syringe to
extract gas proved successful. The assumption that no
swim bladder gas escaped prior to extraction proved
valid. No fish had obviously ruptured swim bladders
when dissected following euthanasia and gas extraction, no gas was observed escaping from the fish, and
no external signs of barotrauma, other than swollen
abdomens, were observed during these trials.
Pink snapper
The rate of gas secretion into the swim bladders of
pink snapper was not significantly affected by water
pressure, water temperature or interactions between
these factors (p [ 0.05). The mean ( SE) rate of pink
341
snapper swim bladder gas secretion was therefore

calculated to be 0.027 ( 0.005) ml/kg/min.
Mulloway
The rate of gas secretion into the swim bladders of
mulloway was not significantly affected by water
these factors (p [ 0.05). The mean ( SE) rate of
mulloway swim bladder gas secretion was therefore
Gas resorption rates
Pink snapper
The rate of gas resorption from the swim bladders of
pink snapper was not significantly affected by water
these factors in the LM (p [ 0.05). However, when the
factors were analysed separately, the rate of gas
resorption was negatively affected by water temperature (p \ 0.05). The mean ( SE) rate of pink
snapper swim bladder gas resorption was therefore
Mulloway
The rate of gas resorption from the swim bladders of
mulloway was not significantly affected by water
these factors in the LM (p [ 0.05). However, when the
factors were analysed separately, the rate of gas
resorption was positively affected by water temperature (p \ 0.001). The mean ( SE) rate of mulloway
swim bladder gas resorption was therefore calculated
to be 0.044 ( 0.009) ml/kg/min.
Change in pressure before swim bladder rupture
Twelve pink snapper were used to estimate the change
in pressure required before the swim bladder ruptures.
The average ( SE) increase in internal swim bladder
pressure before rupture was 2.49 ( 0.14) times the
pressure at equilibrium. This change in pressure
indicates that any fish from greater than the pressure
equivalent to 14.9 m water depth will rupture the swim
bladder if brought directly to the surface.
123
342
Discussion
body tissue of mulloway must be denser than that of

pink snapper. The difference in buoyancy between
pink snapper and mulloway may therefore be attributable to their differing behaviours and mechanisms
for maintaining their preferred position in the water
column (e.g. Jarvis and Lowe 2008). The energetic
costs to pink snapper of maintaining position in the
water column will be less than those for mulloway;
however, the advantage to being negatively buoyant
for mulloway likely relates to some aspect of behaviour, potentially related to a greater FVR than if
neutrally buoyant.
Swim bladder gas secretion
Pink snapper have a swim bladder volume that is on

average 4.2 % of the total volume of the fish. In
contrast, mulloway have a relatively larger swim
bladder, averaging 4.9 % of their total body volume.
Despite the general approximation that in marine
teleosts the swim bladder occupies *45 % of the
volume of the fish (Harden Jones and Scholes 1985),
the relative volume of the swim bladder to total fish
volume is clearly species-specific and can vary
from *1 % (Harden Jones 1951) to [5 % (Parker
et al. 2006) in marine teleosts. If the main function of
the swim bladder is to provide buoyancy, then its
volume relative to total fish volume will depend on the
overall density of fishs tissue and the buoyancy
required for normal energy-efficient behaviour. There
are some advantages to having a relatively small swim
bladder, including a lower rate of change in buoyancy
with change in depth, and also a smaller acoustic
signal and so lower detectability by predators which
use echolocation to hunt (e.g. dolphins; Davenport
1999). However, these advantages are weighed against
a greater energy cost in maintaining position in the
water column whenever negatively buoyant (Alexander 1990; Strand et al. 2005). In addition to controlling
buoyancy using the swim bladder, fish may maintain
position in the water column by generating hydrodynamic lift (upwards or downwards) using their fins and
by tilting their bodies (Blake 1979). The density of
pink snapper was not significantly different to the
density of seawater, indicating that pink snapper at
equilibrium are neutrally buoyant. In contrast, mulloway were denser than seawater and are therefore
negatively buoyant at equilibrium. Given the greater
relative volume of the swim bladder in mulloway, the
The rate of swim bladder gas secretion for pink snapper

was nearly 4 times faster than that for mulloway. Swim
bladder gas secretion rate was not significantly affected
by the water pressures or water temperatures experienced by either species in the current study. These
findings suggest that gas is secreted at a constant rate
whenever a certain level of pressure differential exists
between the gas gland and the swim bladder. Harden
Jones and Scholes (1985) reported a weak positive
relationship (p \ 0.01) between gas secretion rate and
water pressure for the cod Gadus morhua, whereas
Strand et al. (2005) found filling the swim bladder to
be radically slower at greater depths. Rummer and
Bennett (2005) reported acclimation rates of 68, 98 and
174 h to depths of 30, 50 and 110 m in hyperbaric
chambers for red snapper Lutjanus campechanus.
Using Boyles law, these acclimation rates equate
to *91 h to refill the swim bladder at 30 m, 118 h to
refill the swim bladder at 50 m and 189 h to refill the
swim bladder at 110 m, suggesting gas secretion (in
terms of ml/h) slowed markedly with increasing
pressure. Rummer and Bennett (2005) did not quantify
gas secretion rates directly; rather they used observations of fish behaviour to estimate when they were fully
acclimated. It is clear that more work is needed to
understand the process of swim bladder gas secretion
and how it is affected by water pressure.
Swim bladder gas secretion is known to be an active
physiological process (Parker et al. 2006) and is
therefore likely to occur faster with increased metabolism. It would seem logical then that swim bladder gas
secretion rates would increase with greater temperatures, and Harden Jones and Scholes (1985) presented
strong evidence of this in cod. It is possible that the
None of the mulloway acclimated to either 10 or

15 m water depth ruptured their swim bladders when
depressurized to sea level in the hyperbaric chambers.
However, all the mulloway acclimated to pressures
equivalent to 20 and 25 m water depth ruptured their
swim bladders. The average increase in pressure
before swim bladder ruptured was therefore 2.75
times the pressure at equilibrium (equivalent to the
pressure change from 17.5 m water depth to sea level).
123
temperature ranges over which we did our experiments

(*56 C) were too narrow to detect any significant
differences given the inherent variability between
individual fish and measurement error. Harden Jones
and Scholes (1985) detected the effect of temperature
over a much greater temperature range of *18 C. In
addition, McNabb and Mecham (1971) also found that
the swim bladders of the freshwater sunfish Lepomus
macrochirus inflated faster at higher temperatures.
Our results confirm swim bladder gas secretion to
be a slow process in marine teleosts (Strand et al.
2005). Being one of the few studies to quantify rates of
swim bladder gas exchange, we were able to estimate
acclimation rates for fish moving between different
depths (Tables 1, 2). Wittenberg (1958) and Wittenberg et al. (1964) used syringes to empty the swim
bladders of 10 species of physoclist fish and reported
the time taken to refill the swim bladder. Tytler and
Blaxter (1973), Gaspin et al. (1978), Harden Jones and
Scholes (1985), Rummer and Bennett (2005) and
Parker et al. (2006) all reported observed times for
different species to appear acclimated to various water
pressures within hyperbaric chambers. We used Boyles law to convert these rates into times needed to
completely refill empty swim bladders for comparison
between species (Table 3). The results showed that
swim bladder gas secretion rates are hugely variable,
even between closely related species. Whilst much of
the variation reported may be due to the subjective
nature of determining when a fish appears to behave as
if fully acclimated within hyperbaric chambers, Parker
et al. (2006) showed that two Sebastes species have
extremely different gas secretion rates and were able to
demonstrate that this was related to their behaviour
and vertical movements within the water column.
Likewise Wittenberg et al. (1964) related the extremely rapid gas secretion rate in bluefish (Pomatomus
saltatrix) to the species habit of rapidly moving
between the sea floor and the surface when feeding.
The finding that pink snapper are able to secrete gas
into their swim bladders 4 times faster than mulloway
is therefore likely to enable them to more quickly and
efficiently maintain near-neutral buoyancy whilst
moving into deeper waters.
Swim bladder gas resorption
Very few studies have attempted to quantify the rate of
swim bladder gas resorption. We found that the rate of
343
Table 1 Estimated times for mulloway (40 cm TL) and pink
snapper (35 cm FL) to acclimate to different water depths from
sea level
Water depth
(m)
Mulloway
Time to acclimate
(hours)
Pink snapper
Time to acclimate
(hours)
10
49
13
20
30
66
74
18
20
40
79
22
50
82
22
60
85
23
70
86
23
80
88
24
90
89
24
100
90
24
Table 2 Estimated times for mulloway (40 cm TL) and pink

snapper (35 cm FL) to acclimate to sea level from different
water depths
Water depth
(m)
Mulloway
Time to acclimate
(hours)
Pink snapper
Time to acclimate
(hours)
10
16
20
30
31
47
5
7
40
63
50
79
12
60
94
14
70
110
17
80
126
19
90
142
21
100
157
24
swim bladder gas resorption for pink snapper was *7

times faster than that for mulloway. These rates of gas
resorption were also *11 and 6 times faster than the
rates of gas secretion in each species, respectively.
Both species are therefore able to maintain nearneutral buoyancy whilst ascending through the water
column much faster than when descending. The swim
bladder gas resorption rates were not significantly
affected by the water pressures or water temperatures
experienced by either species in the current study
when these factors were analysed together in the LM.
However, when the factor water temperature was
123
344
Table 3 Comparative acclimation times for various marine teleosts

Species name
Common name
Approximate time
to refill empty swim
bladder (hours)
References
Pomatomus saltatrix
Bluefish
Wittenberg et al. (1964)
Lagodon rhomboides
Pinfish
49
McCutcheon (1962)
Gadhus morhua
Cod
10
Scholander et al. (1956)
Stenotomus versicolor
Scup
1012
Wittenberg (1958)
Anguilla sp.
Eels
1018
Wittenberg (1958)
Gadhus morhua
Cod
20
Harden Jones and Scholes (1985)
Opsanus tau
Toadfish
1824
Wittenberg (1958)
Tautoga sp.
Tautog
24
Wittenberg (1958)
Pagrus auratus
Pink snapper
27
Present study
Sebastes melanops
Black rockfish
44
Parker et al. (2006)
Pollachius virens
Prionotus carolinus
Saithe
Searobin
48
48
Tytler and Blaxter (1973)

Wittenberg (1958)
Prionotus evolan
Searobin
48
Wittenberg (1958)
Fundulus heteroclitus
Mummichog
48
Copeland (1952)
Lutjanus campechanus
Red snapper
91
Rummer and Bennett (2005)
Sciaenidae
Croaker
93
Gaspin et al. (1978)
A. japonicus
Mulloway
99
Present study
Tetraodontidae
Toadfish
150
Leiostomus xanthurus
Spot Croaker
183
Sebastes nebulosus
China rockfish
236
Parker et al. (2006)
Morone saxatilis
Striped bass
240
Fundulus heteroclitus
Mummichog
244
Fundulus majalis
Striped killifish
266
Morone americana
White perch
312
Acclimation rates are presented as the time needed to secrete sufficient gas to fully inflate the swim bladder
analysed in isolation, it had a significant negative

effect on the rate of swim bladder gas resorption in
snapper and a significant positive effect on the rate of
swim bladder gas resorption in mulloway. Swim
bladder gas is removed via passive diffusion into the
blood through the resorption chamber (Ross 1979),
and there are, therefore, no obvious reasons why such
passive diffusion would be substantially affected by
temperature or water pressure. However, in one of the
only studies to attempt to quantify swim bladder gas
resorption rates, Harden Jones and Scholes (1985)
reported much faster resorption rates with increased
pressure in cod. An examination of these results,
however, shows that these authors did not account for
the proportional increases in volume of swim bladder
gas being equal to the proportional reductions in water
123
pressure, regardless of the water pressure to which the

fish were acclimated. In addition, Fig. 3 (Buoyancy
adjustment when subject to 25 % reductions in
pressure every 30 min) presented in Harden Jones
and Scholes (1985) does not show any difference in
acclimation times with equal decreases in pressure.
We suggest that the conflicting results on the effect of
water temperature in isolation on the rate of swim
bladder gas resorption in the present study (negative
for snapper and positive for mulloway) are erroneous
and probably the result of individual variability,
measurement errors and small sample sizes. We
conclude that whilst further work is needed to fully
understand the mechanisms controlling the rates of
swim bladder gas resorption, based on the linear model
incorporating both water temperature and water
The findings that mulloway and pink snapper have

different densities when at equilibrium and very
different rates of swim bladder gas exchange suggest
that they utilize the water column in different ways.
Snapper most often occur in schools up off the seafloor
with schooling occasionally occurring at the surface
(Kailola et al. 1993). The energetic costs of maintaining position in the water column are therefore reduced
by the neutral buoyancy of the semi-pelagic snapper.
In contrast, the negatively buoyant mulloway are most
often found in close proximity to the sea floor (Kailola
et al. 1993). Pink snapper also possessed much faster
rates of swim bladder gas exchange and would
therefore be able to more quickly change their position
in the water column whilst maintaining near-neutral
buoyancy than mulloway. Mulloway, having relatively slow rates of swim bladder gas exchange, will
be more restricted in their FVR and in the time taken to
acclimate when ascending or descending through the
water column than snapper.
Mulloway resorb swim bladder gas *6 times faster
than they can secrete it, indicating that they can
acclimate faster when ascending through the water
column than when descending. Mulloway have a
thick, flexible swim bladder that takes up a large
volume within the visceral cavity. We hypothesize
that this swim bladder can contain enclosed gases
under substantial pressure prior to rupture, therefore
allowing mulloway to ascend through the water
column whilst minimizing the risks of losing buoyancy control or suffering swim bladder perforation.
However, our hyperbaric chamber trials indicated
swim bladder rupture occurred after a decrease in
pressure ratio of *2.75, similar to that observed for
pink snapper. Similarly, the finding that pink snapper
resorb swim bladder gas at *11 times faster than it is
secreted indicates that they too acclimate faster when
ascending through the water column than when
descending. In contrast though, pink snapper have a
much smaller swim bladder that is surrounded by the
ribs and muscle tissue, except where it is exposed to
the visceral cavity. In addition, it is made of apparently
quite thin, non-elastic material, and we found that the
Depth (m) before swim bladder rupture
Implications for vertical range
Pink snapper
0
5
10
RUPTURE
15
20
25
30
35
40
45
50
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95100
Acclimated depth (m)
B
Depth (m) before swim bladder rupture
pressure, neither factor significantly effects swim

bladder gas resorption rates in pink snapper and
mulloway.
345
Mulloway
0
5
10
RUPTURE
15
20
25
30
35
40
45
50
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95100
Acclimated depth (m)
Fig. 2 Vertical ascent ranges for A pink snapper and B mulloway before swim bladder rupture
swim bladders burst when subjected to, on average, a

decrease in pressure of just *2.5 times. It is important
to note that cutting away some body wall to view the
swim bladder of pink snapper may have affected the
structural integrity of the swim bladder and so underestimated pressure change before rupture. We believe
that any such bias was unlikely given our observations
that rupture often occurred ventrally, nowhere near the
cut away section, and that our estimates of pressure
change before rupture were consistent.
The vertical range (in terms of m) before swim
bladder rupture occurs increases with increasing depth
at equilibrium (Fig. 2). This is due to the decline in
relative pressure increase with increasing depth (in
terms of m). A pink snapper could potentially ascend
from 30 m water depth to 7 m depth before swim
bladder rupture (a pressure change ratio of *2.4 but a
vertical distance of 23 m); however, a pink snapper at
123
346
100 m water depth could ascend to a depth of 35 m

before swim bladder rupture (a pressure change ratio
of *2.4 but a vertical distance of 65 m) (Fig. 2A). A
mulloway could potentially ascend from 30 m water
depth to 5 m before swim bladder rupture (a pressure
change ratio of2.7 and a vertical distance of 25 m);
however, a mulloway from 100 m water depth could
ascend to a depth of 30 m before swim bladder rupture
(a pressure change ratio of2.7 and a vertical distance
of 26 m) (Fig. 2B).
Acknowledgments The research presented in this paper was
funded by the New South Wales Saltwater Recreational Fishing
Trust. Dr Matthew Ives provided statistical advice and reviewed
the paper prior to submission.
References
Alexander RM (1959) The physical properties of the swimbladders of fish other than Cypriniformes. J Exp Biol
36:347355
Alexander RMN (1990) Size, speed and buoyancy adaptations
in aquatic animals. Am Zool 30:189196
Blake RW (1979) The energetics of hovering in the mandarin
fish (Synchropus picturatus). J Exp Biol 82:2533
Butcher PA, Broadhurst MK, Hall KC, Cullis BR, Raidal SR
(2012) Assessing barotrauma among angled snapper
(Pagrus auratus) and the utility of release methods. Fish
Res 127128:4955
Copeland DE (1952) The histophysiology of the teleostean
physoclistous swimbladder. J Cell Comp Physiol 40:317
Davenport J (1999) Swimbladder volume and body density in an
armoured benthic fish, the streaked gurnard. J Fish Biol
55(3):527534
Gaspin JB, Wiley ML, Peters GB (1978) Experimental investigations of the effects of underwater explosions on
swimbladder fish. II. 1975 Chesapeake Bay tests. Sept 76.
Rept. No. NSWC/WOL/TR-76-61. In : Report Bibliography. Defense Documentation Center, Defense Logistics
Agency, Cameron Station, Alexandria, VA
Hallacher LE (1974) The comparative morphology of extrinsic
gasbladder musculature in the scorpion fish genus Sebastes
(Pisces: Scorpaenidae). Proc Calif Acad Sci, 4th series,
40(3): 5986
Hannah RW, Matteson KM (2007) Behavior of nine species of
Pacific rockfish after hook-and-line capture, recompression, and release. Trans Am Fish Soc 136:2433
Harden Jones FR (1951) The Swimbladder and the vertical
movement of teleostean fishes. J Exp Biol 28:553566
Harden Jones FR, Scholes P (1985) Gas secretion and resorption in
the swimbladder of cod Gadus morhua. J Comp Physiol
155:319331
Henry GW, Lyle JM (2003) The National Recreational and
Indigenous Fishing survey. FRDC Project 99/158. NSW
Fisheries Final Report Series No. 48. AFFA: Canberra.
Web access: http://www.affa.gov.au/corporate_docs/
publications/pdf/ fisheries/final_recsurvey_report.pdf
123

Holbrook RI, de Perera TB (2011) Fish navigation in the vertical
dimension: can fish use hydrostatic pressure to determine
depth?. Fish Fish 12(4):370379
Jarvis ET, Lowe CG (2008) The effects of barotrauma on the
catch-and-release survival of southern California nearshore
and shelf rockfish (Scorpaenidae Sebastes spp.). Can J Fish
Aquat Sci 65:12861296
Kailola PJ, Williams MJ, Stewart PC, Reichelt RE, McNee A,
Grieve C (1993) Australian Fisheries Resources. Bureau of
Resource Sciences, Department of Primary Industry and
Energy, and the Fisheries Research and Development
Corporation, Canberra, Australia. 422 pp
Lea RN, McAllister RD, Ventresca DA (1999) Biological
aspects of nearshore rockfishes of the genus Sebastes from
Central California. Calif Dep Fish Game Bull 177:3109
Lenanton R, Wise B, St John J, Keay I, Gaughan D (2009) Maximising survival of released undersize west coast reef fish.
Final FRDC Report (Project 2000/194). Department of
Fisheries, North Beach, Western Australia, Australia. 126 pp
McCutcheon FH (1962) Swimbladder volume control in the pinfish, Lagodon rhomboides L. J Cell Comp Physiol 59:203
McNabb AR, Mecham AJ (1971) The effects of different
acclimation temperatures on gas secretion in the swimbladder of the bluegill sunfish, Lepomis macrochirus.
Comp Physiol Biochem A40:609616
Parker SJ, McElderry HI, Rankin PS, Hannah RW (2006)
Buoyancy regulation and barotrauma in two species of
nearshore rockfish. Trans Am Fish Soc 135:12131223
Pribyl AL, Schreck CB, Kent ML, Parker SJ (2009) The differential response to decompression in three species of
nearshore Pacific rockfish. North Am J Fish Manag
29(5):14791486
Ross LG (1979) The haemodynamics of gas resorption from the
physoclist swimbladder: the structure and morphometrics of
the oval in Pollachim virens (L). J Fish Biol 14(3):261266
Rowling K, Hegarty A, Ives M (2010) Status of fisheries
resources in NSW 2008/09. Industry and Investment NSW,
Cronulla 392 pp
Rummer JL, Bennett WA (2005) Physiological effects of swim
bladder overexpansion and catastrophic decompression on
red snapper. Trans Am Fish Soc 134:14571470
Scholander PF, van Dam L, Enns T (1956) The source of oxygen
secreted into the swimbladder of cod. J Cell Comp Physiol
48:517
Stensholt BK, Aglen A, Mehl S, Stensholt E (2002) Vertical
density distributions of fish: a balance between environmental and physiological limitation. ICES J Mar Sci
59:679710
Stewart J (2008) Capture depth related mortality of discarded
snapper (Pagrus auratus) and implications for management. Fish Res 90:289295
Strand E, Jorgensen C, Huse G (2005) Modelling buoyancy
regulation in fishes with swimbladders: bioenergetics and
behaviour. Ecol Model 185:309327
Tytler P, Blaxter JHS (1973) Adaptation by cod and saithe to
pressure changes. Neth J Sea Res 7:3145
Wittenberg JB (1958) The secretion of inert gas into the swimbladder of fish. J Gen Physiol 41:783
Wittenberg JB, Schwend MJ, Wittenberg BA (1964) The
secretion of oxygen into the swim-bladder of fish III. J Gen
Physiol 48(2):337355

Swim Bladder Function and Buoyancy Control in Pink Snapper and Mulloway

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Swim Bladder Function and Buoyancy Control in Pink Snapper and Mulloway

Hochgeladen von

Copyright:

Verfügbare Formate

Fish Physiol Biochem (2014) 40:335346