Aquaculture 198 Ž2001.

41–54
www.elsevier.nlrlocateraqua-online

Swim bladder malformation in hatchery-reared

striped trumpeter Latris lineata žLatridae/
A.J. Trotter a,b,) , P.M. Pankhurst a , P.R. Hart b
a
School of Aquaculture, Tasmanian Aquaculture and Fisheries Institute and CooperatiÕe Research Centre for
Aquaculture, UniÕersity of Tasmania, Locked Bag 1-370, Launceston, Tasmania 7250, Australia
b
Marine Research Laboratories, Tasmanian Aquaculture and Fisheries Institute and CooperatiÕe Research
Centre for Aquaculture, UniÕersity of Tasmania, Nubeena Crescent, Taroona, Tasmania 7053, Australia

Received 17 July 2000; received in revised form 4 December 2000; accepted 4 December 2000

Abstract

Swim bladder malformation is common in both larvae and later life stages of cultured striped
trumpeter Latris lineata. This study used histology and gross morphology of whole larvae to
describe the progression of abnormal development that proceeded initial liquid dilation of the
primordial swim bladder. In addition, radiography was used to compare swim bladder morphology
of cultured juveniles with wild-caught specimens. The histomorphology of the swim bladder prior
to lumenal dilation was typical of transient physostome larvae reported in the literature. A distinct
swim bladder lumen present in larvae between 5.2–5.7 mm standard length ŽSL. was assumed to
be liquid dilated and coincided with mouth opening. Initial gaseous inflation was first apparent
when larvae attained 5.7–6.2 mm SL, after the resorption of the yolk sac and oil globule and the
onset of first feeding Ž- 5.5 mm SL.. Successful gaseous inflation ranged between 0% and 75% in
the cohorts of larvae examined. When gaseous inflation failed to occur, abnormal development
proceeded in which the liquid dilated swim bladder collapsed, occluding the lumen. Hypertrophy
of the swim bladder epithelium and hyperplasia of the rete mirable ensued. Radiography revealed
that wild-caught striped trumpeter had a euphysoclist Ždual-chambered. swim bladder, in which
the chambers are separated by a diaphragm. In comparison, cultured juveniles and sub-adults
displayed highly varied swim bladder morphologies, including apparently normal, malformed, and
non-functional swim bladders in which a gaseous lumen was entirely absent. Kyphosis was
inflicted in extreme cases of swim bladder malformation in which a distended single chamber both
displaced the viscera and pushed the spine upwards. It was concluded that striped trumpeter larvae

)
Corresponding author. Marine Research Laboratories, Tasmanian Aquaculture and Fisheries Institute and
Cooperative Research Centre for Aquaculture, University of Tasmania, Nubeena Crescent, Taroona, Tasmania
7053, Australia. Tel.: q61-3-62-277-234; fax: q61-3-62-278-035.
E-mail address: atrotter@utas.edu.au ŽA.J. Trotter..

0044-8486r01r$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 4 - 8 4 8 6 Ž 0 0 . 0 0 5 9 4 - 9
42 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

can be afflicted with swim bladder malformations consistent with cultured larvae of other transient
physostomes; however, swim bladder malformation of later life stages of striped trumpeter is
atypical in comparison to other literature reports and may be unrelated to the larval malformation
described. q 2001 Elsevier Science B.V. All rights reserved.

Keywords: Swim bladder malformation; Teleost larvae; Latris lineata; Gas bladder; Kyphosis

1. Introduction

Striped trumpeter Latris lineata ŽLatridae., a demersal species inhabiting the South-
Eastern Australian and New Zealand coastlines, is being investigated as a potential
species for commercial aquaculture. Significant advances have been made towards
commercial production of striped trumpeter; however, larval rearing protocols that can
routinely promote normal swim bladder development have not been defined.
Teleosts display two basic swim bladder morphologies: physoclistous and physosto-
mous. Physoclists regulate buoyancy by secretion of gas into the swim bladder via the
highly vascularised rete mirabile and swim bladder gas gland ŽSteen, 1970.. In contrast,
physostomes gulp air at the water surface, then pass gas into the swim bladder via the
pneumatic duct that connects the gut and swim bladder ŽSteen, 1970.. An intermediate
group, to which striped trumpeter belongs ŽGoodsell et al., 1996., is physostomous as
larvae and physoclistous as adults Žtransient physostomes. ŽBailey and Doroshov, 1995..
A further group has a swim bladder only during the larval stage Žfilled via a pneumatic
duct., which is followed by total regression of the swim bladder during metamorphosis
Ži.e. Dover sole, Solea solea, Boulhic and Gabaudan, 1992; turbot, Scophthalmus
maximus, Padros and Crespo, 1995.. Adult striped trumpeter is further defined as
euphysoclistous, having a swim bladder with dual-chambers separated by a diaphragm
ŽGoodsell et al., 1996..
Transient physostomes have an essential requirement for surface access during the
brief period in which initial swim bladder inflation occurs. The mechanism of gas
passage to the swim bladder during this period is poorly understood and may be species
specific ŽDoroshev et al., 1981; Chatain, 1990.; however, a recent study suggested that a
surfactant from the common bile duct aids in the passage of bubbles through the
pneumatic duct for a brief period during ontogeny when these structures are in close
association ŽMarty et al., 1995b.. Subsequent development of the muscular pyloric
sphincter imposes a physical barrier between the bile duct and pneumatic duct, ending
the inflation window ŽMarty et al., 1995b.. When initial inflation fails, swim bladder
development is normally arrested, and subsequent abnormal cellular proliferation of the
swim bladder epithelium and associated structures often ensues ŽWeppe and Bonami,
1983; Bennett et al., 1987; Chatain, 1990.. However, later inflation of the swim bladder
via secretory mechanisms has been reported in juvenile fish, although it is typically
associated with lordosis ŽChatain, 1994; Kitajima et al., 1994.. Because initial swim
bladder inflation generally coincides with depletion of endogenous energy reserves
ŽBulak and Heidinger, 1980; Battaglene and Talbot, 1990; Battaglene et al., 1994.,
metabolic sparing provided by buoyancy regulation from the swim bladder is critical to
larval survival, and failure to successfully inflate the swim bladder can have a severe
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 43

effect upon larval mortality ŽChatain, 1990; Battaglene and Talbot, 1992.. In addition to
increased mortality, non-functional swim bladders have been associated with spinal
deformities ŽPaperna, 1978; Chatain, 1994; Kitajima et al., 1994. and reduced growth
rates, the latter again due to greater metabolic demand associated with the requirement
to achieve buoyancy via swimming ŽChatain, 1990; Battaglene and Talbot, 1992;
Kitajima et al., 1994; Marty et al., 1995a..
Swim bladder malformation is a common problem in hatchery-reared transient
physostomes, having been reported in at least 26 species worldwide ŽKitajima et al.,
1994; Battaglene, 1995; Egloff, 1996.. The incidence of malformation is often high
Ž70–100%. in species where larval rearing protocols are under development ŽBattaglene,
1995; Egloff, 1996.. This is the case for striped trumpeter, in which initial swim bladder
inflation was extremely variable; often 0–10% and only occasionally attained levels of
65–95% ŽRuwald et al., 1991; Goodsell et al., 1995.. Larval rearing continues to be a
major bottleneck for commercial production of this species. Mortality events coincide
with both failure to inflate the swim bladder and a high prevalence of jaw deformity in
post-flexion larvae. Furthermore, juveniles and adults that have been successfully
cultured appear to exhibit negative buoyancy, suggesting that swim bladder malforma-
tion or malfunction continue past the larval stages.
The aim of the present study was to examine swim bladder morphology immediately
prior to and proceeding the onset of initial swim bladder inflation and to describe normal
and abnormal developmental events. In addition, swim bladder morphologies of cultured
striped trumpeter juveniles and adults were compared with wild-caught counterparts.

2. Methods
2.1. LarÕal rearing
Three cohorts of striped trumpeter larvae were examined. Larvae were produced
during MayrJune 1998, AugustrSeptember 1998, and AprilrMay 1999 Žcohorts 1–3,
respectively. from captive wild-caught broodstock held at the Marine Research Labora-
tories ŽMRL., Tasmanian Aquaculture and Fisheries Institute ŽTAFI., University of
Tasmania. Larvae stocked at 40–50 ly1 were reared in either 250-l Žcohort 2. or 1000-l
Žcohorts 1 and 3. dark blue cylindro-conical fibreglass tanks in recirculating seawater.
Temperature was 13.08C Žcohorts 1 and 2. or 14.78C Žcohort 3., salinity; 35‰;
photoperiod was 12:12 light:dark Žcohort 1 and 2. or 16:8 light:dark Žcohort 3., and light
intensity was 1.2 mmol sy1 my2 Ž; 200 lux. at the water surface during the light phase.
First feeding larvae were fed rotifers Brachionus plicatilis enriched with either Tahitian
Isochrysis sp., Tetraselmis suecicia or the commercial enrichment product DHA Selcoe
ŽINVE Aquaculture. twice daily, at a density of 5–10 mly1 . Artemia sp. nauplii were
added twice daily at a density of 1–2 mly1 from day 13 post-hatching. Surface
skimming of oily surface-films commenced prior to the onset of first-feeding.
2.2. External morphology and histology
Larvae were sampled daily from hatching Ž n s 10, cohorts 1 and 2 and n s 40 cohort
3 until 19, 30 and 20 days post-hatching for cohorts 1, 2 and 3, respectively. Larvae
.
44 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

were anaesthetised in 0.04% 2-phenoxyethanol ŽSigma. and examined under a stereo

microscope. A lateral view of each larva was captured using a Sony video camera
CCD-IRISrRG and Scion software to record external swim bladder morphology and
standard length ŽSL-length from the rostral tip to the end of the notochord.. Larvae were
then grouped according to external morphology and SL: ŽStage A. primordial swim
bladder Ž; 4.8–5.2 mm SL.; ŽStage B. lumenal dilation of the swim bladder Žliquid
filled. Ž; 5.2–5.7 mm SL.; ŽStage C. initial inflation window-onset of initial swim
bladder inflation to a distinct plateau in the proportion of larvae achieving initial swim
bladder inflation Ž; 5.7–6.2 mm SL.; ŽStage D. post-inflation window Ž; 6.2–7.0 mm
SL.. External morphologies of larvae in groups C and D included individuals in which
the swim bladder was inflated Žreflective gas present. and non-inflated Žno gas present..
A subsample of 70 larvae was subsequently processed for histological examination
ŽStage A: n s 8; stage B: n s 19; Stage C: n s 20; Stage D: n s 23.. Larvae were fixed
overnight at 48C in a solution of 5% gluteraldehyde in 0.1 M sucrose–phosphate buffer
and stored in 70% ethanol at 48C until processing. Fixed larvae were dehydrated in an
ascending ethanol series, embedded in JB-4 plastic resin ŽAgar Scientific. and serially
sectioned in the sagittal plane at 2 mm. Sections were stained using a polychrome stain
prior to examination by light microscopy.

2.3. Radiography of juÕenile and adult fish

In September 1998, cultured striped trumpeter were sampled from two populations
bred in 1994 Žcohort A, n s 9-entire population. and 1997 Žcohort B, n s 10-random
sample. at the MRL, TAFI. A further five wild-caught fish were sampled and X-radi-
ography was used to compare morphology of the swim bladder and axial skeleton. Fish
were anaesthetised in 0.02% 2-phenoxyethanol, weighed, fork length measured ŽFL-
measured from the rostral tip to the caudal fork., and X-rayed using an Atomscope 100P
portable X-ray machine Ž60 kvp; 30 mA; 0.02–0.04 s exposure; 65 cm FFD.. The swim
bladder volume Žmm3 . was calculated using the equation for a prolate spheroid from
Blaxter and Hempel Ž1963.: Vy s pr6 LH 2 Žvolume of a prolate spheroid, where L is
the length and H is the height of the swim bladder.. Swim bladder morphologies were
variable and are only approximately prolate spheroid. Notwithstanding the variation in
shape, the equation has been previously applied under similar situations ŽBlaxter and
Hempel, 1963; Butler and Percy, 1972; Avila and Juario, 1987., and may therefore be
useful for comparisons among species. Relationships between swim bladder volumes
and weight were analysed by regression using the JMP 3.0 statistical package.

3. Results

3.1. External morphology

On the day of hatching, striped trumpeter larvae were poorly developed, the eyes
were non-pigmented and presumed non-functional, the mouth was not formed and the
primordial swim bladder was not visible. Both the yolk sac and oil globule were almost
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 45

Table 1
Mean SL Žmm."SE and age post-hatching Žin brackets. of three cohorts of L. lineata larvae coincident with
developmental events. Cohorts 1 and 2 Ž ns10., cohort 3 Ž ns 40.
Developmental event Cohort 1 Cohort 2 Cohort 3
Hatching 3.62"0.01 Ž0. 3.72"0.03 Ž0. 3.60"0.03 Ž0.
Mouth opening and swim bladder dilation 5.03"0.02 Ž8. 5.32"0.02 Ž9. 5.41"0.01 Ž7.
Depletion oil globule and yolk sac 5.01"0.02 Ž10. 5.44"0.04 Ž10. 5.49"0.02 Ž9.
First feeding 5.01"0.02 Ž10. 5.44"0.04 Ž10. 5.52"0.03 Ž10.
Initial swim bladder inflation – – 5.74"0.03 Ž11.

completely resorbed by first feeding Ž; 5.0–5.5 mm SL. ŽTable 1.. The primordial swim
bladder was first visible as a translucent spherical sac between the notochord and the
digestive tract in larvae between; 4.8 and 5.2 mm SL. Coinciding with the opening of
the mouth, swim bladders were noticeably enlarged and appeared to be liquid dilated
ŽTable 1.. Dilated swim bladders displayed varying morphologies ranging from spherical
to ellipsoid. Initial swim bladder inflation Žreflective gas bubble visible. occurred in
1.25% of the larvae from cohort 3 on day 11 post-hatching Ž5.74 " 0.03 mm SL..
Inflation increased to 76.25% at day 14 post-hatching Ž6.12 " 0.05 mm SL., and was
80% at day 20 post-hatching when sampling terminated. No larvae from cohorts 1 and 2
achieved initial swim bladder inflation ŽTable 1..

3.2. LarÕal swim bladder histology

In larvae of stage A Ž; 4.8–5.2 mm SL., histological examination confirmed a

non-inflated swim bladder primordia located adjacent to the pancreas, between the
digestive tract and the notochord and immediately dorsal to the junction of the
oesophagus and the foregut. The lumen, if present, was small Ž; 8.5 mm in diameter.
and the epithelium of the primordial swim bladder comprised dark staining, cuboidal
cells ranging in height from ; 8–14 mm. The swim bladder primordia was bound by a
connective tissue sheath, the presumptive tunica externa ŽFig. 1a.. The primordial
pneumatic duct was discernible at the anterior margin of the swim bladder primordia
Žnot visible in micrograph..
Larvae defined as having non-gaseous lumenal dilation of the swim bladder Ž; 5.2–5.7
mm SL. had elongated swim bladder lumens with varying degrees of dilation, ranging
from 35 to 69 mm in height and 62 to 170 mm in length. Cells of the swim bladder
epithelium remained of irregular shape, the height of which decreased progressively
during the non-gaseous dilation phase ŽFig. 1b and c.. The pneumatic duct was now
visible entering the posterior pole of the swim bladder. The connective tissue of the
tunica externa decreased in depth with age, gas gland cells developed at the anterior pole
of the swim bladder and the developing rete mirabile became evident beneath the swim
bladder. Organic debris of unknown origin was observed occasionally in the lumen
during this stage ŽFig. 1c..
Larvae sampled during the initial swim bladder inflation window Ž; 5.7–6.2 mm SL.
displayed three distinct histological morphologies; liquid dilation, collapsed and gas-in-
46 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

flated Žliquid dilation and collapsed swim bladders could not be separated by external
observation.. Liquid dilated swim bladders had a similar morphology to those larvae of
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 47

the previous stage. In swim bladders of all three morphologies, the rete mirabile
increased in size and structural complexity, with many capillaries and erythrocytes
evident. Gas-inflated swim bladders were ellipsoid in shape with a very low profile
epithelium Ž; 4 mm. and a pronounced area of gas gland cells at the anterio-ventral
region of the swim bladder ; 18 mm in height ŽFig. 1d.. The epithelium of apparently
collapsed swim bladders appeared as two layers of cuboidal cells in the location where
the lumen had been. No folding of the epithelium was yet apparent ŽFig. 1e.. The
pneumatic duct was still present during this stage in swim bladders of all three
morphologies. Bacteria were observed in the swim bladder lumen of one larva with a
gas-inflated swim bladder ŽFig. 1f. and the digestive tract lumen of a number of larvae
with either non-inflated or inflated swim bladders.
Larvae sampled after the initial swim bladder inflation window Ž; 6.2–7.0 mm SL.
were designated as either gas-inflated or non-inflated from external appearance. Histo-
morphology of the former revealed swim bladder morphology similar to the gaseous
inflated swim bladders described during the initial inflation window, but with greater
expansion ŽFig. 1g.. In some larvae with gas-inflated swim bladders, the pneumatic duct
remained open up to and including termination of sampling Žday 20 post-hatching.. In
contrast, the structure of the non-inflated swim bladders had changed markedly. The
epithelial cells were enlarged Žhypertrophied. and folding of the epithelium was appar-
ent. Both the cytoplasm and the nucleus of the epithelial cells of these larvae had
increased in size, with the staining of the nucleus intensifying ŽFig. 1h.. In some larvae,
the membrane of epithelial cells appeared to be degenerating. The capillaries of the rete
mirabile had proliferated and the pneumatic duct remained open.

3.3. Swim bladder morphology in juÕeniles and adults

Radiography confirmed that wild-caught striped trumpeter were found to have a

euphysoclist swim bladder. The presumptive secretory Žanterior. and resorptive Žpost-
erior. chambers were separated by a diaphragm. The swim bladder occupied the full
length of the body cavity between the viscera. All wild-caught specimens displayed
similar swim bladder morphologies; cylindrical with tapering at each pole and in which
the diaphragm position varied according to the current secretory–resorptive state ŽFig.
2a.. Mean swim bladder volume was 25.59 cm3, and the mean weights and lengths of
the specimens were 2082.50 g and 49.42 cm FL, respectively ŽTable 2..
Swim bladders were present in all of the 4-year-old hatchery-reared fish Žcohort A.;
however, the morphologies were highly variable and all were presumed to be abnormal

Fig. 1. Photomicrographs of swim bladder development of larval L. lineata. ŽA. Stage A-primordial swim
bladder; ŽB. Stage B-early liquid dilation; ŽC. Stage B-late liquid dilation; ŽD. Stage C-initial inflation window
Žinflated.; ŽE. Stage C-initial inflation window Žcollapsed.; ŽF. Stage C-initial inflation window Žinflated:
bacterial invasion.; ŽG. Stage D-post inflation window Žinflated.; ŽH. Stage D-post inflation window
Žnon-inflated.. b: bacteria, cl: collapsed lumen, fg: foregut, gg: gas gland, n: notochord, od: organic debris, oe:
oesophagus, p: pancreas, pd: pneumatic duct, sbe: swim bladder epithelium, sbl: swim bladder lumen, rm: rete
mirabile, te: tuncia externa, Scale bars: Ž1. s80 mm ŽA, B, C, D, E and H.; Ž2. s100 mm ŽG..
48 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

Fig. 2. Radiographs of normal and malformed swim bladders in striped trumpeter: ŽA. wild-caught-normal; ŽB.
Type 1-severe malformation Žalso displaying kyphosis.; ŽC. Type 2-intermediate malformation; ŽD. Type
3-minor malformation; ŽE. Type 4-non-inflated; ŽF. Type 5-reduced volume; ŽG. Type 6-presumed normal. sb:
swim bladder, d: diaphragm. Swim bladder types are fully described in the text. Scale bar s 5 cm.

compared to those of the wild-caught fish ŽTable 2.. The following are the three general
swim bladder types.
Type 1: The diaphragm was absent and the shape of the swim bladder was more
spherical than cylindrical. These swim bladders were highly distended, protruding
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 49

Table 2
Morphometrics and swim bladder morphologies of three populations of L. lineata juveniles and adults;
4-year-old cultured fish Žcohort A, ns9., 2-year-old cultured fish Žcohort B, ns10. and wild-caught captive
fish Ž ns 5.
Parameter Cohort A Cohort B Wild-caught
Mean weight"SE Žg. 910.55 Ž77.80. 101.00 Ž10.50. 2082.50 Ž28.60.
Mean fork length"SE Žcm. 37.94 Ž0.95. 19.55 Ž0.64. 49.42 Ž0.55.
Mean swim bladder volume"SE Žcm3 . 45.42 Ž5.21. 2.77 Ž1.16. 25.59 Ž5.95.
Swim bladder presence Ž%. 100 80 100
Normal swim bladder Ž%. 0 30 100
Kyphosis Ž%. 33 0 0
Vertebrae fusion Ž%. 20 0 0
Diaphragm present Ž%. 60 30 100

ventrally into the normal location of the viscera and in some cases this was coincident
with pronounced kyphosis Žhumpback curvature of the spine.. In addition, the poles of
the swim bladder were rounded ŽFig. 2b..
Type 2: Swim bladders were more elongate and the general malformation was less
severe than the above, occurring both with and without diaphragms. Spinal deformities
in this group included kyphosis and fusion of vertebrae. Fused vertebrae were sometimes
accompanied by abnormal orientation and morphology of neural spines and ribs ŽFig.
2c..
Type 3: Swim bladder morphology was most similar in appearance to the wild-caught
fish; however, the posterior chamber was often greatly enlarged and displayed individual
variation in size and shape ŽFig. 2d..
Kyphosis was observed in 33% of the fish from cohort A ŽTable 2.. Mean weight of
the cohort was 910.55 g, and the mean swim bladder volume was 45.42 cm3. Weight
was not positively correlated to swim bladder volume Ž P ) 0.05, r s 0.66, df s 8.
within the cohort.
Two-year-old hatchery-reared fish Žcohort B. also exhibited a wide range of swim
bladder morphologies. Three additional swim bladder types.
Type 4: Gas absent from the swim bladder Žnon-functional. ŽFig. 2e..
Type 5: Swim bladders of greatly reduced volume Ž- 0.35 cm3 . when compared to
the remainder of the cohort Žswim bladder volumes ) 4.5 cm3 . without diaphragms
ŽFig. 2f..
Type 6: Presumptive normal swim bladders, which most closely approximated those
of the wild-caught fish ŽFig. 2g..
Swim bladders were deemed non-functional ŽType 4. in 20% of the fish from cohort
B, a further 40% had Type 5 swim bladders and presumptive normal swim bladders
were observed in 30% of the cohort ŽTable 1.. One fish displayed a Type 1 swim
bladder; however, pronounced kyphosis was not observed in any of the fish. A
significant relationship was found between weight and swim bladder volume Ž P - 0.01,
r s 0.82, df s 9. within the cohort.
50 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

4. Discussion

Histomorphology and external morphology of the swim bladder of striped trumpeter

larvae during development of the swim bladder primordia was typical of other larvae of
transient physostomes examined with respect to orientation within the abdominal cavity
and cellular structure ŽBulak and Heidinger, 1980; Doroshev et al., 1981; Yamashita,
1982; Chatain, 1990; Boulhic and Gabaudan, 1992; Makino et al., 1995; Sarasquete et
al., 1995.. Swim bladder development of striped trumpeter larvae to this stage appeared
normal and consistent with previous work on this species by Goodsell et al. Ž1995,
1996..
In all cohorts of striped trumpeter larvae examined, the phase of initial swim bladder
dilation without gas inclusion Žliquid dilation. occurred concomitant with mouth opening
and well before the onset of initial gas inflation observed in cohort 3. Dilation of the
swim bladder is described as an event which precedes initial swim bladder inflation in
Pagrus major, ŽYamashita, 1982. and Lateolabrax japonicus ŽMakino et al., 1995.. The
origin of the liquid is unknown; however, after rotifers and algae were observed in some
swim bladders, it was suggested that the lumen is filled with sea water passed from the
mouth through the pneumatic duct to the swim bladder ŽYamashita, 1966; Takashima et
al., 1980; Battaglene, 1995.. Alternatively, the liquid may be secreted by the swim
bladder epithelium. Yamashita Ž1966, 1982. and Makino et al. Ž1995. suggested that
liquid dilation of the swim bladder is a normal event in transient physostomes that
occurs prior to initial gas inflation; however, this is subject to debate. Battaglene Ž1995.
postulated that this event may be a normal event or a precursor to malformation of the
swim bladder. Takashima et al. Ž1980. discussed a possible link between the delayed
drainage of the lumenal liquid and subsequent abnormal pathology. All striped trumpeter
larvae examined in this study displayed liquid dilation, whether they were from cohorts
1 and 2 Žtotal failure of gas inflation., or cohort 3 Ž85% gas inflation at day 20
post-hatching., suggesting that this may be a normal event in this species. Understanding
of the transition from this stage to initial gaseous inflation is not complete, and it is
likely that this is where malformation first occurs.
From; 5.7–6.2 mm SL, striped trumpeter inflated their swim bladders, displayed
abnormal cellular development Žnon-inflated swim bladder., or remained liquid dilated.
The latter morphology remained only briefly and was not observed after this stage. It is
widely accepted that gaseous initial inflation must occur if normal development of the
swim bladder is to proceed ŽChatain, 1990; Battaglene, 1995.. While it is generally
accepted that the columnar epithelium present in the primordial swim bladder differenti-
ates into a squamous epithelium at the time of initial gaseous inflation, the process of
transformation from one to the other is unclear. It is possible that this developing
epithelium behaves in a similar fashion to the epithelium lining the urinary bladder of
mammals Žknow as transitional epithelium. that transforms rapidly to a greatly extended
state to accommodate filling ŽGriepp and Robbins, 1983; Geneser, 1986.. Nevertheless,
the ellipsoid swim bladder of striped trumpeter achieving initial swim bladder inflation,
with a low profile epithelium, pronounced area of gas gland cells at the anterio-ventral
region of the swim bladder and an open pneumatic duct, is typical of normal swim
bladder development described in other transient physostomes Ž Morone saxatilis, Doro-
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 51

shev et al., 1981; P. major, Yamashita, 1982; Sparus aurata, Soares et al., 1994; L.
japonicus, Makino et al., 1995; Stizostedion Õitreum, Marty et al., 1995b..
The collapsed swim bladder in some striped trumpeter is likely to denote the onset of
malformation and appeared to result from the drainage of fluid from the lumen without
replacementrdisplacement with gas. In affected fish, the epithelium is initially closely
apposed and the lumen occluded. Hypertrophy and hyperplasia of swim bladder
structures ensue only after lumenal collapse. The onset of malformation during this
period is consistent with reports of swim bladder malformation in M. saxatilis,
Doroshev et al. Ž1981.; Dicentrarchus labrax and S. aurutus Ž S. aurata., Chatain
Ž1990. S. maximus, Padros ´ et al. Ž1993.; S. Õitreum, Marty et al. Ž1995b.. Hypertrophy
and hyperplasia of the gas gland cells are commonly reported when initial gas inflation
fails ŽSoares et al., 1994; Padros and Crespo, 1995; Makino et al., 1995; Sarasquete et
al., 1995..
After the initial inflation window, hyperplasia of the rete mirabile was also observed
in larvae with non-inflated swim bladders, which is consistent with the pathology
described for striped trumpeter by Goodsell et al. Ž1996., and other species including red
sea bream, Chrysophrys major Ž P. major . ŽTakashima et al., 1980. and S. maximus
ŽPadros and Crespo, 1995.. Hypertrophy of the swim bladder epithelium and concomi-
tant folding of the swim bladder epithelium increased progressively with time. The
cytoplasm and nuclei of the epithelium were greatly enlarged and degeneration of the
membrane of the swim bladder epithelium suggests the onset of necrosis ŽKerr et al.,
1995.. Whether these malformations lead to necrosis and severe fibrosis of the swim
bladder epithelium, as described in M. saxatilis ŽBennett et al., 1987., to total regression
of the swim bladder as reported in S. aurata ŽPaperna, 1978., or preclude late secondary
inflation mechanisms as reported in S. aurutus ŽChatain, 1994. and P. major, ŽKitajima
et al., 1994., were not determined in this study. Nevertheless, successful inflation of the
swim bladder is not a prerequisite for survival in striped trumpeter, highlighted by fish
of cohort B with non-functional swim bladders ŽType 4.. This has been reported in
similar species including S. aurutus ŽChatain, 1994. and P. major ŽKitajima et al.,
1994.. A consequence of non-inflation of the swim bladder is emphasised by the
significant relationship between weight and swim bladder volume in this cohort, likely
due to the greater metabolic cost of buoyancy control achieved via swimming, as
reported by Chatain Ž1990. and Battaglene and Talbot Ž1992..
Larvae with gas-inflated swim bladders after the initial inflation window were
presumed normal and were consistent with the description provided by Goodsell et al.
Ž1996. for this species and comparable species, including C. major ŽTakashima et al.,
1980., M. saxatilis, ŽDoroshev et al., 1981. and L. japonicus ŽMakino et al., 1995..
When larvae attain normal development, the problematic events of swim bladder
ontogeny should have passed and juveniles with a normal functional swim bladder could
be expected ŽChatain, 1990.. However, this does not seem to be the case for previous
cohorts of cultured striped trumpeter in which a high incidence and wide variation of
swim bladder malformation was observed and may reflect events occurring after initial
swim bladder inflation. It is possible that swim bladder malformation in later life stages
is related to a broader issue of developmental abnormality in the species, such as the
onset of jaw malformation that is prevalent from metamorphosis ŽCobcroft et al.,
52 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

unpublished data.. In support of this, the deformities of the vertebral column of striped
trumpeter are atypical of those reported in conjunction with swim bladder malformation
in other transient physostomes in which an oblique swimming mode is adopted to
compensate for negative buoyancy. The resulting abnormal stress on the axial skeleton
results in both lordotic deformity and fusion of vertebrae ŽChatain, 1994; Kitajima et al.,
1994.. In contrast, kyphosis observed in 4-year-old Žcohort A. striped trumpeter appears
to result from mechanical deformation of the vertebral column by an enlarged, spherical,
Type 1 swim bladder, whereas fusion of the vertebrae and associated skeletal elements
appear unrelated to any one specific swim bladder morphology.
Variable initial swim bladder inflation success in striped trumpeter, highlighted by
total failure in cohorts 1 and 2 in this study, likely reflects sub-optimal husbandry
practices during early larval rearing, and perhaps suggests striped trumpeter is more
susceptible than other transient physostomes to this condition. Biotic factors including
nutrition and parental stock have been reported to influence initial swim bladder
inflation success ŽHarel et al., 1992; Kitajima et al., 1994; Bailey and Doroshov, 1995;
Tandler et al., 1995.; however, higher levels of swim bladder non-inflation have been
primarily linked to sub-optimal environmental parameters such as temperature, salinity,
and lighting ŽAl-Abdul-Elah et al., 1983b; Hadley et al., 1987; Battaglene and Talbot,
1990, 1993; Kitajima et al., 1994; Bailey and Doroshov, 1995.. Further to this, physical
barriers to the air–water interface, in particular excessive water turbulence Ždue to flow
rate andror aeration., and oily surface-films, have been reported to reduce initial swim
bladder inflation in some transient physostomes and result in total failure in others
ŽAl-Abdul-Elah et al., 1983a; Chatain, 1990; Battaglene and Talbot, 1993; Kitajima et
al., 1994.. In addition, bacterial invasion of the swim bladder has been implicated in
non-inflation of the swim bladder as a result of inflammation of the pneumatic duct
blocking the passage of air ŽMarty et al., 1995b. and may be linked to swim bladder
malformation in the present study. Future research should focus on defining environmen-
tal optima for larval rearing and investigate swim bladder ontogeny during metamorpho-
sis.

Acknowledgements

The authors extend thanks to David Morehead for rearing of striped trumpeter larvae
and critical review of the manuscript, Stephen Battaglene for critical review of the
manuscript and general advice, and Barry Munday and Barbara Nowak for assistance
with the interpretation of histological sections. Alan Beech, Debbie Gardner, Greg
Goodchild and Bill Wilkinson are thanked for technical support of larval rearing.
Michael Eland and Jenny Cobcroft are thanked for assistance with radiography. This
study was supported by the Cooperative Research Centre for Aquaculture and Tassal.

References
Al-Abdul-Elah, K., Akatsu, S., Downing, N., Al-Ameeri, A.-A., Al-Marzouk, A., El-Zahr, C., 1983a. Studies
towards the improvement of larval rearing and fingerling production of sobaity Ž Acanthopagrus cuÕieri .:
 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54 53

III. Factors influencing swim bladder inflation in early larvae. 1982 Annual Research Report, Kuwaiti
Institute for Scientific Research, pp. 61–64.
Al-Abdul-Elah, K., Akatsu, S., Teng, S.K., 1983b. Effects of water temperature and water turbulence on the
growth, survival and occurrence of initial swimbladder inflation in sobaity Ž Acanthopagrus cuÕieri . larvae.
1983 Annual Research Report, Kuwaiti Institute for Scientific Research, pp. 46–48.
Avila, E.M., Juario, J.V., 1987. Yolk and oil globule utilization and developmental morphology of the
digestive tract epithelium in larval rabbitfish, Siganus guttatus ŽBloch.. Aquaculture 65, 319–331.
Bailey, H.C., Doroshov, S.I., 1995. The duration of the interval associated with successful inflation of the
swimbladder in larval striped bass Ž Morone saxatilis .. Aquaculture 131, 135–143.
Battaglene, S.C., 1995. Induced ovulation and larval rearing of four species of Australian marine fish. PhD
thesis, University of Tasmania, Australia, 215 pp.
Battaglene, S.C., Talbot, R.B., 1990. Initial swim bladder inflation in intensively reared Australian bass larvae,
Macquaria noÕemaculeata ŽSteindachner. ŽPerciformes: Percichthyidae.. Aquaculture 86, 431–442.
Battaglene, S.C., Talbot, R.B., 1992. Induced spawning and larval rearing of snapper Pagrus auratus ŽPisces:
Sparidae., from Australian waters. N. Z. J. Mar. Freshwater Res. 26, 179–183.
Battaglene, S.C., Talbot, R.B., 1993. Effects of salinity and aeration on survival of and initial swim bladder
inflation in larval Australian bass. Prog. Fish. Cult. 55, 35–39.
Battaglene, S.C., McBride, S., Talbot, R.B., 1994. Swim bladder inflation in larvae of cultured sand whiting,
Sillago ciliata Cuvier ŽSillaginidae.. Aquaculture 128, 177–192.
Bennett, R.O., Kraeuter, J.N., Woods III, L.C., Lipsky, M.M., May, E.B., 1987. Histological evaluation of
swim bladder non-inflation in striped bass larvae Morone saxatilis. Dis. Aquat. Org. 3, 91–95.
Blaxter, J.H.S., Hempel, G., 1963. The influence of egg size on herring larvae Ž Clupea harengus L... J. Cons.
Perm. Int. Explor. Mer. 28, 211–240.
Boulhic, M., Gabaudan, J., 1992. Histological study of the organogenesis of the digestive system and swim
bladder of the Dover sole, Solea solea ŽLinnaeus 1758.. Aquaculture 102, 373–396.
Bulak, J.S., Heidinger, R.C., 1980. Developmental anatomy and inflation of the gas bladder in striped bass,
Morone saxatilis. Fish. Bull. 77 Ž4., 1000–1003.
Butler, J.L., Percy, W.G., 1972. Swimbladder morphology and specific gravity of myctophids off Oregon. J.
Fish. Res. Bd. Can. 29, 1145–1150.
Chatain, B., 1990. Problems related to the lack of functional swimbladder in intensive rearing of Dicentrar-
chus labrax and Sparus auratus. Adv. in Trop. Aquacult. AQUACOP IFREMER Actes de Colloque 9,
699–709.
Chatain, B., 1994. Abnormal swimbladder development and lordosis in sea bass Ž Dicentrarchus labrax . and
sea bream Ž Sparus auratus .. Aquaculture 119, 371–379.
Doroshev, S.I., Cornacchia, J.W., Hogan, K., 1981. Initial swim bladder inflation in the larvae of physoclistous
´ Cons. Int. Explor. Mer. 178, 495–500.
fishes and its importance for larval culture. Rapp. P.-V. Reun.
Egloff, M., 1996. Failure of swim bladder inflation of perch, Perca fluÕiatilis L. found in natural populations.
Aquat. Sci. 58 Ž1., 15–23.
Geneser, F., 1986. Textbook of Histology. Munksgaard, Copenhagen, 831 pp.
Goodsell, A., Wikeley, D., Searle, L., 1995. Histological, histochemical and morphological development of
striped trumpeter larvae and the effect of larval diet. Final Report to the Fisheries and Research
Development Corporation. Marine Research Laboratories, Taroona, Tasmanian Department of Primary
Industry and Fisheries, 42 pp.
Goodsell, A., Wikeley, D., Searle, L., 1996. Histological investigation of swim-bladder morphology and
inflation in cultured larval striped trumpeter Ž Latris lineata. ŽTeleostei, Latridae.. Mar. Freshwater Res.
47, 251–254.
Griepp, E., Robbins, E., 1983. Epithelium. In: Weiss, L. ŽEd.., Histology: Cell and Tissue Biology. 4th edn.
Elsevier Biomedical, New York, pp. 109–138.
Hadley, C.G., Rust, M.B., Van Eenennaam, J.P., Doroshov, S.I., 1987. Factors influencing initial swim
bladder inflation by striped bass. Am. Fish. Soc. Symp. 2, 164–169.
Harel, M., Tandler, A., Kissil, G.Wm., Applebaum, S., 1992. The kinetics of nutrient incorporation into body
tissues of gilthead seabream S. aurata females and the subsequent effects on egg composition and egg
quality. Isr. J. Aquacult. Bamidgeh 44, 127.
Kerr, J.F.R., Gobe, G.C., Winterford, C.M., Harmon, B.V., 1995. Anatomical methods in cell death. In:
54 A.J. Trotter et al.r Aquaculture 198 (2001) 41–54

Schwartz, L.M., Osborne, B.A. ŽEds.., Methods in Cell Biology, vol. 46. Academic Press, San Diego, pp.
1–27.
Kitajima, C., Watanabe, T., Tsukashima, Y., Fujita, S., 1994. Lordotic deformation and abnormal development
of swim bladders in some hatchery-bred marine physoclistous fish in Japan. J. World Aquacult. Soc. 25
Ž1., 64–77.
Makino, N., Uchiyama, M., Iwanami, S., Tohyama, T., Tanaka, M., 1995. Differentiation and development of
the swimbladder in larvae of the Japanese sea bass Lateolabrax japonicus. Nippon Suisan Gakkaishi 61
Ž2., 143–150 Žin Japanese, with English abstract..
Marty, G.D., Hinton, D.E., Cech, J.J., 1995a. Oxygen consumption by larval Japanese medaka with inflated or
uninflated swim bladders. Trans. Am. Fish. Soc. 124, 623–627.
Marty, G.D., Hinton, D.E., Summerfelt, R.C., 1995b. Histopathology of swimbladder noninflation in walleye
Ž Stizostedion Õitreum. larvae: role of development and inflammation. Aquaculture 138, 35–48.
Padros, F., Crespo, S., 1995. Swimbladder pathology during larval development of turbot Ž Scophthalmus
maximus L... ICES Mar. Sci. Symp. 201, 159–162.
´ F., Minkoff, G., Sala, R., Crespo, S., 1993. Histopathological events throughout the development of
Padros,
turbot Ž Scophthalmus maximus L.. larvae. J. Comp. Pathol. 109, 321–334.
Paperna, I., 1978. Swimbladder and skeletal deformations in hatchery bred Sparus aurata. J. Fish Biol. 12,
109–114.
Ruwald, F.P., Searle, L.D., Oates, L.A., 1991. A preliminary investigation into the spawning and larval rearing
of striped trumpeter, Latris lineata. Technical Report No. 44, Department of Primary Industry, Tasmania,
Division of Sea Fisheries, Sea Fisheries Research Laboratories, Crayfish Point, Taroona, 17 pp.
´
Sarasquete, M.C., Polo, A., Yufera, M., 1995. Histology and histochemistry of the development of the
digestive system of larval gilthead seabream, Sparus aurata L. Aquaculture 130, 79–92.
˜
Soares, F., Dinis, M.T., Pousao-Ferreira, P., 1994. Development of the swim bladder of cultured Sparus
aurata L.: a histological study. Aquacult. Fish. Manage. 25, 849–854.
Steen, J.B., 1970. The swim bladder as a hydrostatic organ. In: Hoar, W.S., Randall, D.J. ŽEds.., Fish
Physiology vol. IV. Academic Press, New York, pp. 413–443.
Takashima, F., Arai, Y., Nomura, M., 1980. Abnormal development of the swimbladder in hatchery-reared red
sea-bream, Chrysophrys major. J. Tokyo Univ. Fish. 67 Ž1., 67–73 Žin Japanese, with English abstract..
Tandler, A., Harel, M., Koven, W.M., Kolkovski, S., 1995. Broodstock and larvae nutrition in gilthead
seabream Sparus aurata—new findings on its mode of involvement in improving growth, survival and
swimbladder inflation. Isr. J. Aquacult. Bamidgeh 47 Ž3–4., 95–111.
Weppe, M., Bonami, J.R., 1983. Non-inflation of the swimbladder in hatchery-reared sea bass and sea bream:
a significant problem in marine aquaculture. Bull. Eur. Ass. Fish Pathol. 3 Ž4., 59–60.
Yamashita, K., 1966. Fundamental studies for the culture of Chrysophrys major: IV. On disease of larval and
young fish Ž2.. Abnormal expansion of the swimbladder. Bull. Jpn. Soc. Fish. 32 Ž12., 1006–1014 Žin
Japanese, with English abstract..
Yamashita, K., 1982. Differentiation of the swimbladder structure in larvae of the red seabream Pagrus major.
Jpn. J. Ichthyol. 29 Ž2., 193–202 Žin Japanese, with English abstract..