Aquaculture research, tập 42, số 4, 2011

Aquaculture Research, 2011, 42, 469^479

doi:10.1111/j.1365-2109.2010.02640.x

Growth and osmoregulation in Atlantic salmon
(Salmo salar ) smolts in response to different
feeding frequencies and salinities
Albert Kjartansson Imsland1,2, Klement Arild VÔge1, Sigurd Olav Handeland1 &
Sigurd Olav Stefansson1
1

Department of Biology, High Technology Center, University of Bergen, Bergen, Norway
Akvaplan-niva, Iceland O⁄ce, Ko¤pavogur, Iceland

2

Correspondence: A K Imsland, High Technology Center, Department of Biology, University of Bergen, N-5020 Bergen, Norway.
E-mail: imsland@vortex

Abstract

Introduction

Groups of Atlantic salmon (Salmo salar) yearling
smolts were reared in duplicate tanks supplied with
freshwater or seawater, and subjected to di¡erent
feeding frequencies, 100% (fed every day), 50% (fed
every other day), 25% (fed every forth day) and 0%
(starved), from 26 May to 26 July. After 8 weeks, all
the groups were re-fed in excess for 6 weeks. Fish

were maintained on their respective a priori salinity
treatments during the 6-week follow-up period. Starvation for a period of 8 weeks in freshwater resulted
in a loss of hypo-osmoregulatory ability when smolts
were challenged with seawater and unfed smolts
maintained in freshwater were unable to adapt to
seawater in mid-July. Ration levels in£uenced the
growth rate and body size signi¢cantly. The overall
growth rate was higher in freshwater than at corresponding rations in seawater. Partial compensatory
growth was observed in the 0 and 25% groups
following re-feeding. Branchial Na1,K1-ATPase
(NKA) activity decreased rapidly in unfed smolts in
freshwater and was the lowest in the starved group,
whereas an initial increase was observed in those
groups reared in seawater. After re-feeding NKA activity di¡erences decreased between the former
feeding groups. Our results suggest that nutritional
factors and/or energy levels are critical for the
maintenance of hydro-mineral balance of salmon
smolts.

In their natural habitat, smolti¢cation in Atlantic
salmon is synchronized to occur during spring by
seasonal changes in the temperature and photoperiod (Hoar 1988; Duston & Saunders 1989; Nilsen,
Ebbeson, Kiilerich, Bj˛rnsson, Madsen, McCormick
& Stefansson 2008; Stefansson, Bj˛rnsson, Ebbesson
& McCormick 2008). Smolt development is normally
accompanied by decreases in lipid reserves and liver
glycogen, even when ¢sh are fed ad libitum (Sheridan
1989; Stefansson, Bj˛rnsson, Sundell, Nyhammer &
McCormick 2003). This has led to the general concept
that wild Atlantic salmon smolts are naturally

‘energy de¢cient’ during the downstream migration.
The ¢ndings of Stefansson et al. (2003) indicated that
smolts also have low energy reserves during the early
marine phase. If salmon smolts are prevented from
reaching seawater, a partial readaptation to freshwater will occur through the abandonment of mechanisms, allowing survival in marine water and a
re-establishment of adaptations to a hypo-saline environment. This process is known as desmolting or
parr-reversion (see Hoar 1988). Earlier research has
focused on the e¡ect of di¡erent environmental factors on the desmolting process and the corresponding loss of seawater tolerance in Atlantic salmon
(e.g. Stefansson, Berge & Gunnarsson 1998; Handeland, Wilkinson, SveinsbÖ, McCormick & Stefansson
2004). However, the e¡ect of food deprivation on osmoregulatory mechanisms during desmolting is
poorly understood. Food deprivation is known to in£uence seawater adaptation in several ¢sh species,

Keywords: Atlantic salmon, ration, food deprivation, osmoregulation, growth, salinity

r 2010 Blackwell Publishing Ltd

469


E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

Materials and methods
Fish stock and rearing conditions
The ¢sh were 1-year-old smolts from the AquaGen
strain, hatched and reared at S×vareid Fish Farms
(601N,Western Norway). The ¢sh were ¢rst fed in February 2002 at 13^14 1C, and later reared under natural light and temperature conditions until April
2003, when they were transferred to the Industrial
Laboratory (ILAB) at the Bergen High Technology
Centre. A total of 1600 smolts were randomly distributed among sixteen 1m2 grey, covered ¢breglass
tanks with a rearing volume of 450 L, and kept in

running pH-adjusted (NaOH, 6.5opHo7.0) freshwater (7.5 L min À 1) until 24 May under a simulated
natural photoperiod (SNP) for Bergen (60125 0 N). At
ILAB, water temperature increased from approximately 6 1C in early May to 8.5 1C in late May (Fig. 1).
From 20 May to 20 July, the temperature increased

470

14

Temperature (°C)

including tilapia (Oreochromis mossambicus, Jˇrss,
Bittorf, V˛lker & Wacke 1984; Kˇltz & Jˇrss 1991) and
rainbow trout (Oncorhynchus mykiss, Jˇrss, Bittorf,
V˛lker & Wacke 1987; Nance, Masoni, Sola & Bornancin1987). In contrast,Triebenbach, Smoker, Beckman
and Focht (2009) found that spring increases in gill
Na1,K1-ATPase (NKA) activity were una¡ected by a
16-week food deprivation in coho salmon (Oncorhynchus kisutch) and Chinook salmon (Oncorhynchus
tshawytscha). Partially in accordance with the results
of Triebenbach et al. (2009) on coho and Chinook salmon, Dickho¡, Mahnken, Zaugg, Waknitz, Bernand
and Sullivan (1989) showed that Atlantic salmon that
were starved in November and December completed
smolti¢cation the following spring and showed good
post-smolt growth during the ¢rst 4 months in seawater. However, in starved Atlantic salmon smolts
reared in freshwater, considerably reduced branchial
NKA have been observed (Virtanen & Soivio 1985).
Furthermore, Stefansson, Imsland & Handeland
(2009) suggested that Atlantic salmon smolts in seawater lose their hypo-osmoregulatory capacity (measured as increased plasma ion levels and reduced
NKA activity) during starvation. Restricted feeding
may, therefore, lead to a disruption of the smolti¢cation process, resulting in reduced hypo-osmoregulatory ability. The present study was performed in order

to describe the consequences of various degrees of
food restriction in Atlantic salmon on growth and
hypo-osmoregulatory ability during the critical early
post-smolt phase.

Aquaculture Research, 2011, 42, 469–479

12
10
8
6
Treatment period
4
May

June

July

Aug.

Sept.

Month

Figure 1 Rearing temperatures for Atlantic salmon post
smolts in seawater (whole line) and freshwater (dotted
line). The temperature pro¢les are based on daily measurements in the experimental tanks.

from 8.5 to 12.0 1C, decreasing to 10 1C in late July,

and remained around 10 1C until termination of the
experiment on 6 September. Fish were fed a standard
dry diet used throughout the experiment (47.0% protein, 18.5% fat, 19.0% carbohydrates, 9.0% ash, Ewos,
Bergen, Norway) from automatic feeders at rates according to the temperature and ¢sh size (Austreng,
Storebakken & —sgard 1987) during the photo phase.

Experimental design
Between 24 and 26 May, water quality was changed in
eight tanks from fresh water (FW) to full-strength sea
water (34.5%, SW) of approximately 8.5 1C. The eight
remaining tanks were maintained on FW. All groups
were reared under an SNP during the entire experimental period. Water £ow was maintained at approximately 7.5 L min À 1 in FW tanks and 8.5 L min À 1 in
SW tanks, maintaining oxygen saturation above
6 mg/L throughout the experiment. All tanks were
checked twice daily and dead ¢sh were removed immediately. On 26 May, concurrent with the establishment
of the two salinity regimes, four feeding regimes were
initiated (0%, 25%,50% and100%) in a 2 Â 4 factorial
design, each group consisting of two replicate tanks.
The degree of food deprivation was obtained by withholding food at di¡erent days in a 4-day cycle, i.e.
100%, fed continuously (control group);50%, fed every
other day; 25%, fed every forth day and 0%, starved.
When fed, all groups were given excess food, with food
distributed at hourly intervals during the photo phase.
After 8 weeks on a di¡erent feed frequency (on 26 July),
restricted frequencies were discontinued, 80 ¢sh from

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479


Aquaculture Research, 2011, 42, 469^479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

each group (i.e. 40 ¢sh from each replicate tank) were
¢n-clipped (adipose ¢n or either pelvic ¢n) to identify
the ration level, transferred to two 2000 L tanks supplied with running FWor SWand fed in excess until 6
September, when the experiment was terminated.
When transferred to larger tanks, ¢sh were maintained on their respective a priori salinity treatments.
The experiment described has been approved by
the local responsible laboratory animal science specialist under the surveillance of the Norwegian Animal Research Authority and registered by the
Authority. The experiment has thus been conducted
in accordance with the laws and regulations controlling experiments in live animals on Norway, i.e. the
Animal Protection Act of 20 December 1974, No. 73,
chapter VI, sections 20^22.

Mortality
Acute mortality was observed among the FW groups
during 7^11 June, in particular in the FW0% (50%
mortality). Mortality was associated with reduced hyper-osmoregulatory ability in the ion-poor FW; hence
on 11 June, a small amount of seawater was mixed
with freshwater to increase the conductivity of the
water from o70 mS cm À 1 to approximately 200^
250 mS cm À 1 (YSI 556,YSI,Yellow Springs, OH, USA),
which remained stable throughout the experiment.
Following this, water treatment mortality ceased immediately. Mortality ranged between10% and 18% in
the other FW groups (13%, FW25%; 10%, FW50%;
18%, FW100%), but was negligible in SW (1%,
SW0%;1%, SW25%; 3%, SW50%;1%, SW100%).

Seawater challenge test (SWCT)

At approximately 14-day intervals between 24 May
and 19 July, the seawater tolerance of 10 ¢sh from
each group in FW was assessed in a 24-h SWCT
(34.5%, Blackburn & Clark 1987). The SWCT was performed in similar types of tanks and with the same
photoperiod (SNP) as in the main experiment. The
temperature during the SWCT was the same as in
the FW tanks.

Sampling procedures
Body weight (g) and fork length (cm) of 40 to 60 ¢sh
in each tank were recorded at approximately 2-week
intervals from 24 May until termination of the ex-

periment on 6 September. The speci¢c growth rate
for the experimental groups between two dates was
calculated as: SGR 5 (lnW2 À lnW1) Â 100/(T2 À T1),
whereW1 andW2 are the mean weights (of all the ¢sh
sampled in each tank) at daysT1 andT2. The condition
factor (CF) was calculated as CF 5100 Â W/L3,
where W is the individual weight and L is the corresponding fork length.
Concurrently, at each sampling, eight ¢sh from
each group were killed by a blow to the head, blood
was sampled from the caudal vein for the determination of plasma Cl À values and the second gill arch
was removed for the determination of NKA activity.
Plasma was separated by centrifugation (10 min at
4 1C and 2500 g) and stored at À 80 1C for subsequent analysis. Plasma Cl À (mM) was determined in
duplicate samples in a Radiometer CMT 10 chloride
titrator (Radiometer, Copenhagen, Denmark). Gill tissue was quickly immersed in ice-cold SEI (Zaugg &
McLain 1982) and frozen at À 80 1C and analysed according to the procedure of Zaugg and McLain (1982),
modi¢ed by Berge, Berg, Fyhn, Barnung, Hansen and

Stefansson (1995).

Statistics
All statistical analysis and graphics were developed
using STATSOFT ^ STATISTICA 8.0. All data sets were
tested for normality using a Kolmogorov^Smirnov
test. To ¢t normality, all data on branchial NKA activity and plasma Cl À levels were log transformed before the statistical tests. A Hartley F-max test was
used to test for homogeneity of variances, while a
two-way ANOVA and a Tukey honest signi¢cance differences (HSD) post hoc test were applied to determine di¡erences between experimental groups. In
all cases, a signi¢cance level (a) of 0.05 was used.

Results
Hypo-osmoregulatory ability
Following a transient increase in the plasma Cl À levels in early June in the seawater-challenged FW0%
group, a signi¢cant reduction in hypo-osmoregulatory ability was observed in SW-challenged ¢sh in
early July, following 6 weeks of starvation (Tukey
HSD test, Po0.05, Fig. 2). A further signi¢cant increase in the plasma Cl À levels was observed 2 weeks
later (Tukey HSD test, Po0.05), with the mean
( Æ SEM) levels reaching 184 Æ 11mM after a 24-h

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

471


Plasma Cl− (mM)

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

195

190
185
180
175
170
165
160
155
150
145
140
135

a
a
b
b
b

a
b
b
b
b
b
b

May

June

Month

July

Figure 2 Plasma chloride levels in Atlantic salmon post
smolts fed di¡erent feeding regimes in freshwater and
challenged in seawater (34.5%) for 24 h. Symbols: 100%
( & ), 50% (D), 25% ( } ), and 0% ( ). Values are given as
means Æ SE (n 5 6^10 in each experimental group). Signi¢cant di¡erences between the di¡erent experimental
groups are given by di¡erent letters (Tukey post hoc test,
Po0.05).

Aquaculture Research, 2011, 42, 469–479

declined in the other three groups. Accordingly, no
signi¢cant di¡erences were seen in NKA activity
among the FWgroups in September (Tukey HSD test,
P40.25, Fig. 3A).
Following transfer to seawater, enzyme activity increased in all groups (Fig. 3B). The activity levels did
not di¡er signi¢cantly among feeding frequencies in
seawater between May and July. Following re-feeding,
NKA activity increased transiently in the SW0%
group to levels signi¢cantly higher than the controls
and SW50% in August (Tukey HSD test, Po0.05),
whereas the ¢nal NKA did not di¡er between the
SWgroups in September (Fig. 3B).



SW challenge. No signi¢cant di¡erences were observed among the other FW groups after seawater

challenge, although the Cl À levels were slightly elevated (albeit not signi¢cantly) after 8 weeks in the
FW25% group.
Plasma ion levels in unchallenged ¢sh grouped according to salinity, with Cl À levels of all the freshwater groups remaining around 135 mM, while all
seawater groups showed levels around 145 mM. No
signi¢cant di¡erences (two-way ANOVA, P40.5) were
found between unchallenged feeding groups in
freshwater or seawater. The exception was in freshwater in early June, when the levels were signi¢cantly reduced in all FW groups to 125 Æ 2 mM,
probably due to the inability of smolts to maintain hyper-osmoregulation in the ion-poor freshwater at this
time. No e¡ect of re-feeding on the plasma Cl À levels
was observed as the values remained around135 mM
in freshwater and around145 mM in seawater during
the entire re-feeding period.

Branchial NKA
Branchial NKA activity was in£uenced both by salinity and by ration (two-way ANOVA, Po0.05, Fig. 3). In
groups remaining in freshwater (Fig. 3A), gill NKA
activity di¡ered signi¢cantly among feeding groups,
with the FW0% showing approximately 50% lower
activity than controls from mid-June onwards (Tukey
HSD test, Po0.05, Fig. 3A). Following re-feeding in
freshwater, NKA increased in the FW0% group and

472

Growth and condition
Body length (Fig. 4) did not change signi¢cantly in
the 0% groups between May and July, while body
weight (Fig. 5) decreased during this period. Body
length (Fig. 4) increased in the other groups at rates
corresponding to the ration levels, while body weight

(Fig. 5) increased in the 50 and 100% fed groups only.
Accordingly, there were signi¢cant di¡erences
among groups within each salinity regime at the
end of the ration period (Tukey HSD test, Po0.05,
Figs 4 and 5). Except for the unfed groups, which did
not di¡er in size after 8 weeks of starvation, body
length (Tukey HSD test, Po0.05, Fig. 4) and weight
(Tukey HSD test, Po0.05, Fig. 5) were signi¢cantly
higher in freshwater groups than in the corresponding seawater groups.
Growth rate was signi¢cantly in£uenced by both
ration and salinity, with an overall negative SGR between May and July in the starved groups (Table 1).
After re-feeding, the growth rate increased in the
groups previously held on reduced ration levels, with
SGR in the previously restricted groups exceeding
SGR of the fully fed controls (Tukey HSD test,
Po0.05,Table 1). However, despite the higher growth
rates in the previously reduced ration groups, body
weight di¡erences persisted during the 6 weeks of
feeding, with the ¢nal body size being larger in the
fully fed groups (Fig. 5).
The CF decreased in all groups between mid-May
and mid-June (Fig. 6). Condition factor remained low
(0.8^0.85) in the 0% groups during July, with no signi¢cant di¡erences between salinities (two-way ANOVA, P40.3, Fig. 6). Condition factor in the fully fed
groups remained relatively constant between 1.0 and
1.05 and was not signi¢cantly di¡erent between sali-

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479


Aquaculture Research, 2011, 42, 469^479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

22
(A) Freshwater
Na+K+ ATPase activity
(µmol Pi mg protein–1 t–1)

20
18
a

16
14

bc
c
cd

ab
b
b

b
b
bc

c
cd
cd

12
10

b
b
b
b

b
b
b

8

c

6
Treatment

b
d

period

4
May

June

d

July

Aug.

Sept.

Month
22

a

(B) Seawater
a

Na+K+ ATPase activity
(µmol Pi mg protein–1 t–1)

20

*

*

a

18
a
a

a
a

16
a
a
a

14
ab
ab
ab

12

a
a
ab
ab
ab
ab
ab

10

a
ab
b
b

8
6
Treatment period

4
May

June

July

Aug.

Sept.

Month

Figure 3 Gill Na1,K1-ATPase activity in Atlantic salmon post smolts fed di¡erent feeding regimes in freshwater (A) and
seawater (B). Symbols: 100% ( & ), 50% (D), 25% ( } ) and 0% ( ). Values are given as means Æ SE (n 5 6^10 in each
experimental group). Signi¢cant di¡erences between the di¡erent experimental groups (all groups in FW and SW tested
simultaneously) are given by di¡erent letters (Tukey post hoc test, Po0.05). ÃA signi¢cant interaction between feeding
regimes and salinity.



nities. Following re-feeding, the CF increased in all
previously reduced ration groups.

Discussion
The restricted feeding frequency signi¢cantly in£uenced the growth rate and CF of Atlantic salmon in

both seawater and freshwater. During the feed deprivation period, the growth rate decreased systematically with the ration level, causing signi¢cant
reductions in body weight compared with the fully
fed controls. Changes in length were less a¡ected by
feed restriction. Positive growth in length (i.e. skele-

tal growth) during periods of starvation and weight
loss (i.e. little or no muscle growth) seems to be a general phenomenon in ¢sh, and may be adaptive in ¢sh
experiencing £uctuating (seasonal) changes in food
abundance (Nicieza & Metcalfe 1997). In ¢sh, muscle
growth occurs by both hyperplasia (cell proliferation
resulting in the generation of new myotubes) and hypertrophy (increase in myotube size) (Rowlerson &
Veggetti 2001; Johansen & Overturf 2006). Kiessling,
Storebakken, Asgard and Kiessling (1991) suggested
that in rainbow trout, periods of rapid growth favoured ¢bre hypertrophy and periods of slow growth
favoured ¢bre recruitment. In Atlantic salmon (Higgins & Thorpe 1990), ¢bre recruitment (hyperplasia)

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

473


Aquaculture Research, 2011, 42, 469–479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

30
(A) Freshwater

Body length (cm)

a
ab
ab

a
a

28
26

a
ab

a
ab

22

a

20

ab
ab
ab

cd

b

a
ab
ab

24

c

c
d

e

b

e

Treatment period

18
May

June

July

Aug.

Sept.

Month
30

*

(B) Seawater

*

*

28
Body length (cm)

b
b
c

26
24

b
b

ab
ab

22

ab

ab

20

ab
b

c
d

ab
b

bc
c

d
e

c
cd
d

c
d

Treatment period

18
May

June

July

Aug.

Sept.

Month

Figure 4 Mean length (cm) in Atlantic salmon post smolts fed di¡erent feeding regimes in freshwater (A) and seawater
(B). Symbols:100% ( & ),50% (D), 25% ( } ) and 0% ( ).Values are given as means Æ SE (n 516^93 in each experimental
group). Signi¢cant di¡erences between the di¡erent experimental groups (all groups in FWand SW tested simultaneously)
are given by di¡erent letters (Tukey post hoc test, Po0.05). ÃA Signi¢cant interaction between feeding regimes and
salinity.



seemed to be most apparent in periods with unfavourable growing conditions, e.g. feed restriction,
which could explain the positive skeletal growth seen
in periods of no muscle growth. Decreased density of
¢bres with increasing size has also been found for
Atlantic salmon, while the estimated total ¢bre number increased simultaneously (Johnston, Manthri,
Smart, Campell, Nickell & Alderson 2003). The intuitive explanation for this is that as the ¢sh grow in
size, more ¢bres are recruited, resulting in a higher
total number of ¢bres, while at the same time, the
¢bres increase in diameter at a higher rate than does
the recruitment of cells, resulting in a lower density
(Johnston et al. 2003). After re-feeding, all previously

474

feed-restricted groups showed an increase in the
growth rates, i.e. compensatory growth (CG). However, only partial weight compensation was achieved
within the 6 weeks of re-feeding in the previously
feed-deprived groups in seawater, whereas full
weight compensation was reached in the FW25%
and FW50% groups. The results of the seawater
groups are comparable with previous experiments,
where partial CG as a result of starvation has been
found, i.e. Arctic charr, Salvelinus alpinus (Jobling,
Jorgensen & Siikavuopio 1993), and Atlantic halibut,
Hippoglossus hippoglossus (Heide, Foss, Stefansson,
Mayer, Roth, Norberg, Jenssen, Nortvedt & Imsland
2006). In contrast, experiments performed on Atlan-

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479


Aquaculture Research, 2011, 42, 469^479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

250
(A) Freshwater
230
210
Body weight (g)

a
a
ab

a
ab
cd

190
a
b
c

170
a
a
a

150
130

ab
ab
ab
ab

a
ab
ab

110

a
b
c

c

e

90
b

70

b
Treatment period

d

e

50
May

June

July

Aug.

Sept.

Month
250

*

(B) Seawater

*

*

230

Body weight (g)

210
b
d
d

190
170
150
130
110

a

ab
ab

ab

90

b
b
b

70

ab
cd
cd

a
ab
ab

b
d
d

ab
b
b

c

e

b

b
Treatment period

e

e

50
May

June

July

Aug.

Sept.

Month

Figure 5 Mean weight (g) in Atlantic salmon post smolts fed di¡erent feeding regimes in freshwater (A) and seawater (B).
Symbols: 100% ( & ), 50% (D), 25% ( } ) and 0% ( ). Values are given as means Æ SE (n 516^93 in each experimental
group). Signi¢cant di¡erences between the di¡erent experimental groups (all groups in FWand SW tested simultaneously)
are given by di¡erent letters (Tukey post hoc test, Po0.05). ÃA signi¢cant interaction between feeding regimes and salinity.



Table 1 Mean speci¢c growth rate (SGR) in Atlantic salmon post smolts fed di¡erent feeding regimes in fresh- and seawater
PeriodÃ

FW100%

FW50%

FW25%

FW0%

SW100%

SW50%

SW25%

SW0%

26 May to 6 June 0.31 (0.03)a À 0.11 (0.05)c
0.02 (0.06)b À 0.31 (0.04)d 0.01 (0.02)b 0.07 (0.04)b À 0.25 (0.04)d À 0.27 (0.04)d
7 June to 21 June 0.20 (0.01)b
0.52 (0.04)a À 0.12 (0.04)c À 0.10 (0.06)c 0.40 (0.04)a 0.16 (0.06)b
0.23 (0.03)b À 0.11 (0.03)c
22 June to 5 July 1.36 (0.06)a
0.34 (0.02)d
1.14 (0.04)b À 0.65 (0.06)f 0.85(0.05)c 0.93 (0.05)c À 0.28 (0.03)e À 0.77 (0.06)f
6 July to 26 July

1.28 (0.04)a
1.01 (0.05)b
0.18 (0.06)e À 0.29 (0.06)f 0.85 (0.05)c 0.33 (0.06)d
0.48 (0.05)d
0.04 (0.06)e
-------------------------------------------------------------------------------------------------27 July to 16
0.94 (0.08)c
1.38 (0.05)b
1.60(0.06)a
1.51 (0.05)a 1.08(0.08)c 1.05 (0.05)c
1.57 (0.03)a
0.91 (0.09)c
August
17 August to 6
0.81 (0.07)e
1.02 (0.05)d
1.27 (0.06)c
1.58 (0.05)b 0.82 (0.06)e 1.04 (0.05)d
1.05 (0.06)d
1.82 (0.06)a
September

ÃThe experimental groups were fed di¡erent feeding rations from 26 May to 26 July, but from 27 July to 6 September, all groups were
re-fed in excess (indicated by a broken line).
Values are given as means (Æ SE) values from the two replicate tanks. Signi¢cant di¡erences between the di¡erent experimental groups
are indicated by di¡erent letters (Tukey post hoc test, Po0.05).

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

475

Aquaculture Research, 2011, 42, 469–479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

1.2

(A) Freshwater

1.1
Condition factor

ab

a
a
a

a
a

ab

a

ab
ab

1.0

ab
ab
b
b

ab
ab
b

0.9

b
b
b
c

0.8

c
c

b
b

c
c

d

Treatment period
0.7
May

June

July

Aug.

Sept.

Month
1.2

*

(B) Seawater

Condition factor

1.1

a
a

a
a

a

a

a

a

*
a
a
a

1.0
ab
ab

a
a
a
b

b
b

0.9

0.8

b

ab

b

c

c

b
c

c

d

Treatment period
0.7
May

June

July

Aug.

Sept.

Month

Figure 6 Condition factor in Atlantic salmon post smolts fed di¡erent feeding regimes in freshwater (A) and seawater (B).
Symbols: 100% ( & ), 50% (D), 25% ( } ) and 0% ( ). Values are given as means Æ SE (n 516^93 in each experimental
group). Signi¢cant di¡erences between the di¡erent experimental groups (all groups in FWand SW tested simultaneously)

are given by di¡erent letters (Tukey post hoc test, Po0.05). ÃA signi¢cant interaction between feeding regimes and salinity.



tic cod, Gadus morhua (Jobling, MelÖy, Dos Santos &
Christiansen 1994), and turbot, Scophthalmus maximus (S×ther & Jobling 1999), displayed a full weight
recovery after restricted feeding similar to the present ¢ndings in freshwater (i.e. FW25% and
FW50%). Overall, the growth rates were higher in
freshwater in line with previous ¢ndings where temporal di¡erences in the growth of Atlantic salmon
smolts and post-smolts reared at di¡erent salinities
are seen during summer (McCormick, Saunders &
MacIntyre 1989; Duston 1994), but less so during autumn, winter and spring. McCormick et al. (1989)
further indicated that ration was more important for

476

smolt growth than di¡erences in salinity, which is
line with the present ¢ndings.
The CF of the control groups in June was similar to
previous observations on cultured smolts (McCormick, Saunders, Henderson & Harmon 1987; Duston
& Saunders, 1989; Handeland et al. 2004). In other
cases where smolts have been reared in freshwater
beyond the peak of smolti¢cation, an increased
CF has been reported (Eriksson & Lundquist 1982;
Hoar 1988), and has been taken as a sign of desmolti¢cation, i.e., the loss of physiological adaptations to
life in seawater. Migrating smolts and post-smolts in
nature may experience periods of restricted food

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

Aquaculture Research, 2011, 42, 469^479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

availability (Stefansson et al. 2003), indicated by the
presence of a large fraction of empty stomachs in
samples of wild post-smolts (Levings, Hvidsten &
Johnsen 1994; Andreassen, Martinussen, Hvidsten
& Stefansson 2001). Arnesen,Toften, Agustsson, Stefansson, Handeland and Bj˛rnsson (2003) further demonstrated that reared post-smolts may take several
weeks to resume feeding following transfer to seawater. The present results suggest that the long-term
consequences of such restricted feeding may involve
a reduction in hypo-osmoregulatory ability.
An initial increase in gill NKA was seen in the SW
groups. Such an increase in NKA is generally accepted as an indicator of good smolt quality, and is
correlated with seaward migration, hypo-osmoregulatory ability and high seawater survival rates in
salmonids (McCormick et al. 1987; Handeland,
BjÖrnsson, Arnesen & Stefansson 2002). Consequently, the salmon smolts in the SW group can
be considered to be fully smolti¢ed as NKA increased signi¢cantly in the initial rearing period,
i.e. 26 May to 6 June.
The SWCTs demonstrated that starvation in FW
led to a severe decline in the hypo-osmoregulatory
capacity (measured as plasma Cl À ), with the most
adverse e¡ect seen at the termination of the starvation period in July. According to Jobling (1996), starvation results in a general decrease in the turnover
of biochemical components in cells, a¡ecting both
the synthesis and the breakdown of tissue, with
growth re£ecting the di¡erence between these two
processes. The negative growth observed in the
starved group indicates that tissue breakdown exceeds the synthesis of new cellular components. This
suggests that the observed increase in the plasma

chloride levels may be caused by increased degradation of osmotic barriers, i.e. external membranes or
swelling (Olsen, Sundell, RingÖ, Myklebust, Hansen
& Karlsen 2008). The initial increase in the plasma
Cl À levels in the FW% group could possibly be linked
to the reduced branchial NKA as measured in vitro
for the FW% group in this period. The explanation
suggested for tilapia (Kˇltz & Jˇrss1991) and rainbow
trout (Jˇrss et al.1987) is that a reduction in branchial
NKA activity would be adaptive during periods of
starvation in order to reduce ATP consumption.
A concurrent decline in NKA activity was seen in
the SW-starvation groups with a rapid 50% increase
in activity levels seen after re-feeding. Stefansson
et al. (2009) reported loss of hypo-osmoregulatory
ability in starved Atlantic salmon post-smolts in seawater, suggesting that this is a general physiological

response of this species, independent of salinity, despite the absence of osmoregulatory disturbances
within the time frame of the present study. Studies of
rainbow trout (Jˇrss, Bittorf, V˛lker & Wacke 1983;
Jˇrss et al. 1987) and tilapia (Jˇrss et al. 1984; Kˇltz &
Jˇrss1991) have shown a reduction in branchial NKA
activity after several weeks of fasting, both in freshwater and in seawater, similar to the present ¢ndings.
Chloride cell numbers, structure and function were
found to change with food deprivation (Kˇltz & Jˇrss
1991), suggesting that structural elements (fatty acids
and/or amino acids) may become limiting after extended periods of food deprivation. In the present
study, starvation had a signi¢cant e¡ect on branchial
NKA activity, with a more pronounced decline in
NKA activity in the freshwater groups. At the termination of the starving period, FW0% had a 64% lower (Po0.01) NKA activity compared with FW100%.
In a comparable trial with rainbow trout (40 g), a

44% reduction in NKA activity was found after 4
weeks of starvation (Jˇrss et al. 1987). The present
¢ndings are also in line with the ¢ndings with those
of Virtanen and Soivio (1985), where an almost total
loss of NKA activity was seen in starving Atlantic salmon smolts, concurrent with a signi¢cant reduction
in circulating cortisol levels. Partly based on own unpublished observations, on the e¡ects of a low-protein diet, Virtanen and Soivio (1985) suggested that
the de¢ciency of some amino acid(s) may inhibit the
pituitary^interrenal axis and thereby the production
of key osmoregulatory hormones such as cortisol.
Considering the key role of cortisol in the di¡erentiation of chloride cells and the regulation of the NKA
activity (McCormick 2001), this suggests a mechanism through which starvation may have a restraining
e¡ect on the NKA enzyme. Upon re-feeding in late
July, the NKA activity increased in the starved group
and the ¢nal NKA in September did not di¡er between the groups within each salinity regime. However, a transient increase in NKA activity to levels
above controls was observed in the SW0% and
SW25% groups after 4 weeks of re-feeding, whereas
no di¡erences were seen at the conclusion of the experiment (September). It is possible that the transient
increase in NKA is due to CG in the re-feeding period
with increased ion in£ux over the intestine epithelium. Sundell, Jutfelt, Agustsson, Olsen, Sandblom,
Hansen & Bjornsson (2003) found that during the
parr-smolt transformation in Atlantic salmon, the intestinal NKA increases in the anterior intestine and
the paracellular permeability appears to decrease in
the posterior intestine. These events correspond with

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

477


E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

the increase in intestinal £uid transport seen during
this developmental stage.Whether the same mechanism can been seen during the CG phase is at present
unknown.

Conclusions
Food deprivation in Atlantic salmon smolts may result in signi¢cant osmotic disturbance. After re-feeding, the previously starved group responded with an
increase in branchial NKA activity. Ration levels in£uenced the growth rate and mean body size signi¢cantly. Partial weight compensation was achieved
within the 6 weeks of re-feeding in the previously
feed-deprived groups in seawater, whereas full
weight compensation was reached in the FW25%
and FW50% groups. The overall growth rate was
higher in fresh water than at the corresponding rations in seawater. Our results suggest that nutritional
factors and/or energy levels are critical for the maintenance of the hydro-mineral balance of salmon
smolts.

Acknowledgements
We thank —se Berge and Gunnar Steinn Gunnarsson
for their assistance during this experiment.

References
Andreassen P.M.R., Martinussen M.B., Hvidsten N.A. & Stefansson S.O. (2001) Feeding and prey-selection of wild
Atlantic salmon post-smolts. Journal of Fish Biology 58,
1667^1679.
Arnesen A.M., Toften H., Agustsson T., Stefansson S.O.,
Handeland S.O. & Bjornsson B.T. (2003) Osmoregulation,
feed intake, growth and growth hormone levels in
01Atlantic salmon (Salmo salar L.) transferred to seawater at di¡erent stages of smolt development. Aquaculture 222, 167^187.
Austreng E., Storebakken T. & —sgard T. (1987) Growth-rate
estimates for cultured Atlantic salmon and rainbowtrout. Aquaculture 60,157^160.

Berge —.I., Berg A., Fyhn H.J., Barnung T., Hansen T. & Stefansson S.O. (1995) Development of salinity tolerance in
underyearling smolts of Atlantic salmon (Salmon salar)
reared under di¡erent photoperiods. Canadian Journal of
Fisheries and Aquatic Sciences 52, 243^251.
Blackburn J. & Clarke W.C. (1987) Revised Procedure for the 24
Hour Seawater ChallengeTest to Masrue SeawaterAdaptability of Juvenile Salmonids. Canadian Technical Report of
Fisheries and Aquatic Sciences, 1515, 35 pp.

478

Aquaculture Research, 2011, 42, 469–479

Dickho¡ W.W., Mahnken C.V.W., Zaugg W.S., Waknitz F.W.,
Bernard M.G. & Sullivan C.V. (1989) E¡ects of temperature
and feeding on smolting and seawater survival of Atlantic
salmon (Salmo salar). Aquaculture 82, 93^102.
Duston J. (1994) E¡ect of salinity on survival and growth of
Atlantic salmon (Salmo salar) parr and smolts. Aquaculture 121, 115^124.
Duston J. & Saunders R.L. (1989) The entrainment role of
photoperiod on hypoosmoregulatory and growth related
aspects of smolting in Atlantic salmon (Salmo salar). Canadian Journal of Zoology 68,707^715.
Eriksson L.O. & Lundqvist H. (1982) Circannual rhythms
and photoperiod regulation of growth and smolting in
Baltic salmon (Salmo salar L.). Aquaculture 28, 113^121.
Handeland S.O., BjÖrnsson B.T., Arnesen A.M. & Stefansson
S.O. (2002) Seawater adaptation and growth of post-smolt
Atlantic salmon (Salmo salar) of wild and farmed strains.
Aquaculture 220, 367^384.
Handeland S.O., Wilkinson E., SveinsbÖ B., McCormick S.D.
& Stefansson S.O. (2004) Temperature in£uence on the

development and loss of seawater tolerance in two fastgrowing strains of Atlantic salmon. Aquaculture 233,
513^529.
Heide A., Foss A., Stefansson S.O., Mayer I., Roth B., Norberg
B., Jenssen M.D., Nortvedt R. & Imsland A.K. (2006) Compensatory growth and ¢llet crude composition in juvenile
Atlantic halibut: e¡ects of short term starvation periods
and subsequent feeding. Aquaculture 261,109^117.
Higgins P.J. & Thorpe J.E. (1990) Hyperplasia and hypertrophy in the growth of skeletal-muscle in juvenile Atlantic
salmon, Salmo-salar L. Journal of Fish Biology 37,505^519.
Hoar W.S. (1988) The physiology of smolting salmonids. In:
Fish Physiology, Vol. 11B (ed. by W.S. Hoar & D.J. Randall),
pp. 275^343. Academic Press, London, UK.
Jobling M. (1996) Fish Bioenergetics. Fish and Fisheries Series,
Vol.13, pp. 194^201. Chapman and Hall, London, UK.
Jobling M., Jorgensen E.H. & Siikavuopio S.I. (1993) The in£uence of previous feeding regime on the compensatory
growth-response of maturing and immature Arctic charr,
Salvelinus alpinus. Journal of Fish Biology 43, 409^419.
Jobling M., MelÖy O.H., Dos Santos J. & Christiansen B.
(1994) The compensatory growth response of the Atlantic
cod: e¡ects of nutritional history. Aquaculture International 2,75^90.
Johansen K.A. & Overturf K. (2006) Alterations in expression of genes associated with muscle metabolism and
growth during nutritional restriction and refeeding in
rainbow trout. Comparative Biochemistry and Physiology
144B,119^127.
Johnston I.A., Manthri S., Smart A., Campell P., Nickell D. &
Alderson R. (2003) Plasticity of muscle ¢bre number in
seawater stages of Atlantic salmon in response to photoperiod manipulation. Journal of Experimental Biology 206,
3425^3435.
Jˇrss K., Bittorf Th., V˛lker Th. & Wacke R. (1983) In£uence
of nutritional biochemical sea water adaptation of rainbow

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479


Aquaculture Research, 2011, 42, 469^479

E¡ect of feeding rations on growth and osmoregulation A K Imsland et al.

trout (Salmo gairdneri Richardson). Comparative Biochemistry and Physiology 75B,713^717.
Jˇrss K., Bittorf Th.,V˛lker Th. & Wacke R. (1984) Biochemical investigations into the in£uence of environmental
salinity on starvation of the tilapia, Oreochromis mossambicus. Aquaculture 40,171^182.
Jˇrss K., Bittorf T.,V˛lker T. & Wacke R. (1987) E¡ects of temperature, food deprivation and salinity on growth, RNA/
DNA ratio and certain enzyme activities in rainbow trout
(Salmo gairdneri Richardson). Comparative Biochemistry
and Physiology 87B, 241^253.
Kiessling A., Storebakken T., Asgard T. & Kiessling K.H.
(1991) Changes in the structure and function of the epaxial muscle of rainbow-trout (Oncorhynchus mykiss) in relation to ration and age. 1. Growth dynamics. Aquaculture
93, 335^356.
Kˇltz D. & Jˇrss K. (1991) Acclimation of chloride cells and
Na/K-ATPase to energy de¢ciency in tilapia (Oreochromis
mossambicus). Zoologische Jahrbucher-Abteilung fur Allgemeine Zoologie und Physiologie derTiere 95, 39^50.
Levings C.D., Hvidsten N.A. & Johnsen B.Ò. (1994) Feeding
of Atlantic salmon (Salmo salar L.) postsmolts in a fjord
in central Norway. Canadian Journal of Zoology 72,
834^839.
McCormick S.D. (2001) Endocrine control of osmoregulation
in teleost ¢sh. American Zoologist 41,781^794.
McCormick S.D., Saunders R.L., Henderson E.B. & Harmon
P.R. (1987) Photoperiod control of parr-smolt transformation in Atlantic salmon (Salmo salar): changes in salinity
tolerance, gill Na1,K1-ATPase activity, and plasma thyroid hormones. Canadian Journal of Fisheries and Aquatic
Sciences 44, 1462^1468.

McCormick S.D., Saunders R.L. & MacIntyre A.D. (1989) The
e¡ect of salinity and ration levels on growth rate and conversion e⁄ciency on Atlantic salmon (Salmo salar) smolts.
Aquaculture 82, 173^180.
Nance J.M., Masoni A., Sola F. & Bornancin M. (1987) The
e¡ects of starvation and sexual maturation on Na1 transbranchial £uxes following direct transfer from freshwater
to seawater in rainbow trout (Salmo gairdneri). Comparative Biochemistry and Physiology 87A, 613^622.
Nicieza A.G. & Metcalfe N.B. (1997) Growth compensation in
juvenile Atlantic salmon: responses to depressed temperature and food availability. Ecology 78, 2385^2400.
Nilsen T.O., Ebbesson L.O.E., Kiilerich P., Bj˛rnsson B.Th.,
Madsen S., McCormick S.D. & Stefansson S.O. (2008) Endocrine systems in juvenile anadromous and landlocked
Atlantic salmon (Salmo salar): seasonal development and
seawater acclimation. General and Comparative Endocrinology 155,762^772.

Olsen R.E., Sundell K., RingÖ E., Myklebust R., Hemre G.I.,
Hansen T. & Karlsen Ò. (2008) The acute stress response
in fed and food deprived Atlantic cod, Gadus morhua L.
Aquaculture 280, 232–241.
Rowlerson A. & Veggetti A. (2001) Cellular mechanisms of
post-embryonic muscle growth in aquaculture species.
In: Muscle Development and Growth, Fish Physiology, Vol.
18 (ed. by I.A. Johnston), pp. 103^140. Academic Press,
San Diego, CA, USA.
Sheridan M.A. (1989) Alterations in lipid metabolism accompanying smolti¢cation and sea water adaptation of
salmonid ¢sh. Aquaculture 82,191^203.
Stefansson S.O., Berge A.I. & Gunnarsson G.S. (1998) Changes
in seawater tolerance and gill Na1,K1-ATPase activity during desmolti¢cation in Atlantic salmon kept in freshwater
at di¡erent temperatures. Aquaculture 168, 271^277.
Stefansson S.O., Bj˛rnsson B.Th., Sundell L., Nyhammer G.
& McCormick S.D. (2003) Physiological characteristics
of wild Atlantic salmon post-smolts during estuarine

and coastal migration. Journal of Fish Biology 63,
942^955.
Stefansson S.O., Bj˛rnsson B.Th., Ebbesson L.O.E. & McCormick S.D. (2008) Smolti¢cation. In: Fish Larval Physiology
(ed. by R.N. Finn & B.G. Kapoor), pp. 639^681. Science
Publishers, En¢eld, NH, USA.
Stefansson S.O., Imsland A.K. & Handeland S.O. (2009)
Food-deprivation, compensatory growth and hydromineral balance in Atlantic salmon (Salmo salar) postsmolts in sea water. Aquaculture 290, 243^249.
Sundell K., Jutfelt F., Agustsson T., Olsen R.E., Sandblom E.,
Hansen T. & Bjornsson B.T. (2003) Intestinal transport
mechanisms and plasma cortisol levels during normal
and out-of-season parr-smolt transformation of Atlantic
salmon, Salmo salar. Aquaculture 222, 265^285.
S×ther B.S. & Jobling M. (1999) The e¡ects of ration level on
feed intake and growth, and compensatory growth after
restricted feeding, in turbot Scophthalmus maximus.
Aquaculture Research 30, 647^653.
Triebenbach S.P., Smoker W.W., Beckman B.R. & Focht R.
(2009) Compensatory growth after winter food deprivation in hatchery-produced coho salmon and Chinook salmon smolts. North American Journal of Aquaculture 71,
384^399.
Virtanen E. & SoivioA. (1985) The patterns of T3,T4 and cortisol and Na1-K1ATPase during smolti¢cation of hatchery-reared Salmo salar and comparison with wild smolts.
Aquaculture 45, 97^109.
ZauggW.S. & McLain L.R. (1982) A simpli¢ed preparation for
adenosine triphosphate determintaion in gill tissue.
Canadian Journal of Fisheries and Aquatic Sciences 39,
215^217.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 469^479

479

Aquaculture Research, 2011, 42, 480^489

doi:10.1111/j.1365-2109.2010.02642.x

Effects of spray-dried blood cell meal with
microencapsulated methionine substituting fish meal
on the growth, nutrient digestibility and amino acid
retention of Litopenaeus vannamei
Huaxin Niu1,2, Jie Chang2,3, Shidong Guo1, Zhongguo Xie1 & Aixia Zhu1
1

State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan University,Wuxi,

China
2
3

School of Animal Science and Technology, Inner Mongolia University for the Nationalities,Tongliao, China
The Key Laboratory of Mariculture (Ministry of Education), Ocean University of China, Qingdao, China

Correspondence: S Guo, State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan
University, 1800, Lihu Road,Wuxi, 214122 Jiangsu, China. E-mail:

Abstract
The e¡ect of substitution of ¢sh meal (FM) by spraydried blood cell meal (SBCM) with microencapsulated DL-methionine supplementation in trial diets
for Litopenaeus vannamei was evaluated. Six isonitrogenous (320 g kg À 1) and isolipidic (85 g kg À 1) diets
were formulated to feed shrimp (2.3 Æ 0.2 g
shrimp À 1) for 56 days. Shrimp were fed with six diets
in which FM protein was gradually replaced by SBCM

protein (0%, 20%, 40%, 60%, 80% and 100% in diets
0^5). Growth performances and feed utilization of
shrimp fed diets containing 0%, 3.5%, 7.0% and
10.5% SBCM protein were not signi¢cantly di¡erent
(P40.05). Growth, feed conversion ratio and protein
e⁄ciency ratio of shrimp fed diets (80 and 100% FM
substitution) were signi¢cantly poorer compared
with other treatments (Po0.05). With increased
levels of dietary SBCM, apparent digestibility coe⁄cient of dry matter, crude protein enhanced from
76.9% to 82.3%, 84.8% to 89.0%, but crude lipid
decreased from 90.6% to 88.3% respectively. The
carcass composition values were not signi¢cantly
(P40.05) a¡ected by the replacement level of FM,
except lipid. There were no signi¢cantly di¡erences
(P40.05) in amino acid retentions among Diets 0^3.
The results suggest that the dietary FM protein could
e⁄ciently be substituted by SBCM up to 60%, without adverse e¡ects on the growth of L. vannamei.

480

Keywords: spray-dried blood cell meal, microencapsulated methionine, growth, digestibility, Litopenaeus vannamei, amino acid retention

Introduction
At present, if the ¢sh and crustacean aquaculture
sectors are to maintain their current growth rates of
8.5% per year (the sector growing over 115-fold from
322765 tonnes in 1950 to 37109751 tonnes in 2006:
FAO 2008), then it follows that the supply of feed ingredients and production will also have to grow at similar rates so as to meet demand. Nowhere is this
supply more critical than with the current dependency of the export-oriented ¢sh and crustacean
aquaculture sector on capture ¢sheries for sourcing

feed demand, including ¢sh meal (FM) (Naylor, Goldberg, Mooney, Beveridge, Clay, Folke, Kautsky, Lubchenco, Primavera & Williams 1998; Tacon & Metian
2008; Hardy 2010). Fish meal is one of the primary
proteins in aquafeeds, which is consumed by marine
shrimp because of its known nutritional and palatability characteristics. Therefore, ¢shmeal’s high cost
and concern about the reliability of future supplies
from plant and animal by-product sources have
prompted e¡orts to identify and develop novel and
relatively cheaper ingredients to function as FM

r 2010 Blackwell Publishing Ltd


Aquaculture Research, 2011, 42, 480^489

substitutes (Davis & Arnold 2000; Forster, Dominy,
Obaldo & Tacon 2003; Amaya, Davis & David 2007;
HernaŁndez, Olvera-Novoa, Aguilar-Vejar, GonzaŁlez-Rodr|¤ guez & Parra 2008; SuaŁrez, Gaxiola, Mendoza, Cadavid, Garcia, Alanis, SuaŁrez, Faillace & Cuzon 2009),
but an equally e¡ective, alternative ingredient has become an on-going research and development goal.
Spray-dried blood cell meal (SBCM) is obtained by
centrifugation of bovine or swine blood, followed by
concentration and dehydration using the spray-dried
technology. The production process includes anti-coagulation, cooling, ¢ltration, separation, cool storage,
quality re¢ning, ultrasonication to disrupt erythrocytes, spray drying and dry powder quick cooling
and elsewhere. It has the advantages of good taste,
high protein content, low ash content, etc. The mild
dehydration process maintains all the functional physico-chemical and biological properties of the product
intact (ToldraØ, Elias, Pare¤s, Saguer & Carretero 2004).
Traditionally, blood meals and powders used in ¢sh
feeds have been processed as £ame-dried or spraydried blood products. Recently, however, because of
the strong commercial markets for spray-dried plasma

(e.g., feeding juvenile pigs), the plasma and cellular
fractions of blood are routinely separated and processed as distinct products. Thus, commercial supplies
of spray-dried animal blood cell meal are available and
not heretofore tested as an ingredient in ¢sh diets.
This product is a high-protein, high-lysine, but lowash, phosphorus content feedstu¡. However, SBCM
(FOB in 2009, $900 tonne À 1) is a more expensive product than £ash-dried blood meals ($700 tonne À 1)
but is cheaper than most FMs (FOB in 2009,
$1100 tonne À 1) in China. Therefore, it is a perfect
substitute of ¢shmeal. Especially, the cell membrane
is totally broken during the processing, improving
the digestibility obviously. This product widely applies
to livestock, poultry and aquatic commercial feeds like
some ¢sh and turtle in China. However, there are no
studies on the e¡ects of replacing FM with SBCM in
shrimp feeds on shrimp growth performance.
Prime-grade SBCM is mostly used in pet and human foods because of its palatability, protein quality
and essential amino acids (EAAs), vitamin and
mineral contents. One obstacle to the broader use of
SBCM in shrimp feeds is the paucity of nutrient digestibility data; most previous studies have not included analyses of test ingredient digestibility or
amino acids (AAs) composition (Brunson, Romaire
& Reigh 1997). Research suggests, however, that FM
may be partially replaced by SBCM and may consequently provide substantial feed cost savings with

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

no loss in shrimp growth performance (Tacon & Metian 2008). In comparison with FM, however, the
methionine content of blood cell meal is very low
and it would be the ¢rst limiting AA of an animal
by-product ingredient for ¢sh (Bureau, Harris & Cho
1999) and shrimp (Akiyama, Dominy & Lawrence

1991). Therefore, the objective of the present study
was to evaluate the e¡ects of substitution of FM by
di¡erent levels of dietary particle SBCM with microencapsulated DL -methionine (MM) on the growth,
feed nutrient digestibility and body compositions of
the Paci¢c white shrimp (Litopenaeus vannamei).

Materials and methods
Materials
Spray-dried blood cell meal (porcine) and DL -methionine (Adisseo) were obtained from Wuxi Dajiang
Group (Wuxi, China) and Ningbo TECH-BANK (Ningbo, China), respectively, and the other ingredients
were obtained from Guangdong Haid Group (Suzhou,
China). Based on the chemical compositions and AA
pro¢le of the protein sources (Table 1), six trial diets
Table 1 Proximate composition and amino acid pro¢le of
Peruvian ¢sh meal (FM) and spray-dried blood cell meals
(SBCM)
FM
Proximate analysis (g kg À 1)
Moisture
78.3
Crude protein
654.4
Crude lipid
75.8
Ash
128.7
Amino acid profile (AA g kg À 1 dry matter)
Aspartic acid
71.1
Glutamic acid

72.4
Serine
25.7
HistidineÃ
16.8
Glycine
51.3
ThreonineÃ
22.7
ArginineÃ
35.3
Alanine
42.5
Tyrosine
21.6
Cysteine
5.4
ValineÃ
32.9
MethionineÃ
17.5
PhenylalanineÃ
29.2
IsoleucineÃ
23.7
LeucineÃ
58.3
LysineÃ
56.1
Proline

26.7

SBCM

61.4
923.8
11.6
35.5
118.3
89.7
38.5
78.8
48.4
27.1
40.1
82.5
21.3
3.0
93.9
2.0
72.9
5.8
136.8
94.8
23.5

ÃEssential amino acid for shrimp (Akiyama et al. 1991; Forster
et al. 2003). Tryptophan was not determined.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489

481


Aquaculture Research, 2011, 42, 480–489

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

were formulated to be isonitrogenous (approximately
326 g kg À 1 crude protein) and isolipidic (approximately 86 g kg À 1 crude lipid). Diet 0 (control) was a
high FM shrimp diet without SBCM and crystalline
AA supplementation. The other ¢ve diets (Diets 1^5)
contained 35, 70, 105, 140 or 175 g kg1 SBCM (Table 2)
in replacement of 20%, 40%, 60%, 80% or 100%, respectively, of FM protein contained in the control diet
(0 g kg1 SBCM). As diets FM content decreased and
SBCM content increased, the test diets were supple-

mented with 6, 8, 10, 12 or 14 g kg À 1 of MM (methionine content: 70% of MM) in order to meet the
methionine requirement of shrimp respectively. Titanium dioxide (TiO2) was used as an inert marker at a
concentration of 1g kg1 of the feed ingredients (Richter, LˇckstÌdt, Focken & Becker 2003). All diets were
formulated to provide approximately crude protein by
adjusting proportions with wheat £our, on a dry matter basis. The formulation and nutrient composition
of the test diets are presented in Table 2. All the ingre-

Table 2. Ingredient composition and proximate analysis of the trial diets
DietsÃ
Diet 0
Ingredients (g kg À 1)
Spray-dried blood cell meal
0

Peruvian fish meal
250
Wheat flour
315
Shrimp shell powder
50
Squid liver meal
50
Soybean lecithin
20
Choline chloride
4
Na alginate
10
Fish oil
20
Ca(H2PO3)2
10
Vitamin mixw
10
Mineral mixz
20
Microencapsulated methionine‰
0
TiO2
1
Proximate analysis (g kg À 1)
Moisture
90
Crude protein

324
Crude lipid
85
Ash
112
Essential amino acid (AA g kg À 1 diet dry matter)z
Histidine
18.1
Threonine
15.2
Arginine
27.5
Valine
21.0
Methionine
14.9
Phenylalanine
24.7
Isoleucine
15.8
Leucine
33.1
Lysine
25.7

Diet 1

Diet 2

Diet 3

Diet 4

Diet 5

35
200
323
50
50
20
4
10
21
10
10
20
6
1

70
150
335
50
50
20
4
10
22
10

10
20
8
1

105
100
347
50
50
20
4
10
23
10
10
20
10
1

140
50
359
50
50
20
4
10
24
10

10
20
12
1

175
0
361
50
50
20
4
10
25
20
10
20
14
1

91
324
84
107

91
327
85
106

89
329
86
106

91
329
89
103

93
329
86
101

18.5
15.2
26.7
21.5
14.2
25.2
14.6
35.6
27.1

18.9
15.3
26.1
21.9
14.6

25.4
14.3
40.0
28.4

19.0
15.5
26.5
23.3
14.0
26.3
13.3
42.8
31.7

19.4
15.5
25.7
25.7
14.7
26.6
11.8
43.6
32.8

19.5
15.7
25.5
28.3
14.2

27.2
11.5
45.1
34.4

ÃNumber in the diet identi¢er indicates the replacement level of ¢sh meal protein with SBCM protein (Diet 0 5 0%, Diet 1 520%, Diet
2 540%, Diet 3 560%, Diet 4 5 80% and Diet 5 5100%).
wVitamin premix contained the following vitamins (kg1 feed): vitamin A (as vitamin A acetate and vitamin A palmitate, 1:1), 5500 IU;
vitamin D3, 1000 IU; vitamin E (as DL-a-tocopheryl acetate), 50 IU; vitamin K3 (asmenadione sodium bisulphite), 10 IU; niacin, 100 mg;
ribo£avin, 20 mg; pyridoxine, 20 mg; thiamin, 20 mg; D-calacium pantothenate, 50 mg; biotin, 0.1mg; foliacin, 5 mg; vitamin B12, 20 mg;
ascorbic acid, 100 mg; inositol, 100 mg.
zMineral premix contained the following minerals (mg kg À 1 feed): NaCl, 257; MgSO4 Á 7H2O, 3855; NaH2PO4 Á 2H2O, 6425; KH2PO4,
8224; FeC6H5O7 Á 5H2O, 642.5; ZnSO4 Á 7H2O, 90.7; MnSO4 Á 4H2O, 41.6; CuSO 4 Á 5H2O, 7.97; CoCl2 Á 6H2O, 0.26; KIO3, 0.77.
‰The optimal process parameters for preparing microencapsulated methionine by spray drying technology were as follows: gelatin and
sodium alginate (wall material), DL -Met (core material), the pressure of spray drying 150 KPa, the inlet and outlet air temperature 155
and 80 1C respectively; the DL -methionine content was 70% of MM.
zEssential amino acid for shrimp (Akiyama et al. 1991; Forster et al. 2003). Tryptophan not determined.

482

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489


Aquaculture Research, 2011, 42, 480^489

dients were ground with a hammer mill and passed
through a 175 mm sieve and mixed in a 30 L kitchen
mixer. Diets were made into sinking pellets (1.5 mm
in diameter) by a simple cold twin-screw extruder,
dried at 65 1C in a forced-air oven for 15 h. The trial

feed pellets were sieved to the appropriate size and
stored at À 20 1C.

Shrimp and feeding trial
Shrimp were obtained from Qingdao Aquafarm in
Jiaozhou (Qingdao, China) and acclimated to indoor
conditions for 10 days in a 1000 L circular tank,
during which shrimp were fed with a commercial
shrimp feed. At the beginning of the feeding trial,
juvenile shrimp (initial mean body weight
2.3 Æ 0.2 g shrimp À 1) were sorted into 18 groups at
30 shrimp per group and were randomly distributed
into the trial 18 tanks, with three replicates for each
diet treatment. Each replicate of 30 juveniles was held
in an indoor 150 L circular ¢breglass tank with
circulating seawater and constant aeration. The
56-day-long trial was designed to determine the
e¡ects of the diets on growth performance. The
shrimp were initially fed at 10% of the biomass of
each tank divided into thrice a day (07:30, 11:30 and
19:30 hours), and the ration was adjusted based on
limiting the amount of unconsumed feed that was
collected. During the trial, the water temperature
was maintained at 28 Æ 1 1C, dissolved oxygen ranged between 6.5 and 7.5 mg L À 1, ammonia was
maintained o0.5 mg L À 1 and salinity was
31 Æ 2 g L À 1.

Sample preparation and digestibility
determination
At the beginning of the trial, 20 shrimp were randomly collected from the remaining acclimated

shrimp for the determination of the initial body composition. At the conclusion of the 56-day period, the
shrimp were feed deprived for 1 day before sampling,
and then counted and batch weighed to determine
[¢nal body wet weight (FBW)] and survival. Ten
shrimp from each tank were randomly collected for
the carcass composition analysis.
Three replicates for the Diet test treatment determined apparent digestibility coe⁄ents (ADCs) of
dry matter, protein and AA in SBCM feedstu¡ for
shrimp and the main feeding trial concurrently.
Each replicate of 30 juveniles was held in an indoor

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

150 L circular ¢breglass tank. The Diet test was on
the basis of 30% substitution of the SBCM in the Diet
0. Apparent digestibility coe⁄ents for dry matter,
protein and AA in SBCM feedstu¡ were determined
using the indicator method described by Forster
et al. (2003).
Shrimp had been adapted to the trial diets for 20
days before faeces collection was begun. They were
fed twice a day at a ¢xed daily ration (10% biomass)
and all unconsumed feed was removed and collected
from the tank 45 min after each feeding. Immediately
after collection, faecal strands were siphoned onto a
mesh using a pipette, gently rinsed with distilled water
several times, transferred to conic tubes and stored
at À 20 1C. Samples were then oven dried at 65 1C,
ground and stored again at À 20 1C until analysis.
Sample collection was performed over a 15-day period

and samples were pooled by replicate for each treatment. Titanium dioxide (1g kg À 1) was added as an external indicator to determine the ADC for dry matter,
crude protein crude lipid and AA of the trial diets.

Chemical analysis
Feed ingredients, formulated diets, faeces and preand post-trial shrimp carcass samples were analysed
in triplicate using standard methods [Association
of Of¢cial Analytical Chemists (AOAC) 1995]. The
crude protein content was calculated from the
nitrogen content by multiplying by 6.25 (AOAC
1995). The crude ash content was obtained by incinerating samples in a mu¥e furnace at 550 1C for
12 h (AOAC 1995). Dry matter was determined by
drying the sample in an oven at 105 1C for 16 h
(AOAC 1995) and weighing to the nearest
0.1mg. Crude lipid was determined gravimetrically
following extraction with chloroform^methanol
(ratio 2:1) (Folch, Lees & Sloane-Stanley 1957). Titanium dioxide in the feed and faeces was determined
following the modi¢ed method described by Richter
et al. (2003).
The AA compositions of ingredients, diets, faeces
and shrimp body were determined using a reversed
phase high-performance liquid chromatograph
(Agilent 1100 Series, Santa Clara, CA, USA), equipped
with a UV detector. Samples for AA analysis were hydrolysed using 6 N hydrochloric acid for 24 h at
115 1C. Cysteine was oxidized with 0.01N sodium hydroxide and determined as cystine. Tryptophan could
not be measured because of its degradation during
acid hydrolysis.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489

483

Aquaculture Research, 2011, 42, 480–489

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

Survival ð%Þ ¼

Final number of shrimp
 100
Initial number of shrimp

Feed conversion ratio ðFCRÞ
¼

Total dry weight feed intake ðgÞ
Weight gain ðgÞ

Protein efficiency ratio ðPERÞ ¼

Weight gain ðgÞ
Protein intake ðgÞ

Apparent digestibility coe⁄cients for dry matter
nutrients of the diets were determined using the
following equations:


%TiO2 in feed
ADC dry matter ð%Þ ¼ 100 1 À

%TiO2 in faeces
ADC nutrients ð%Þ ¼


%TiO2 in feed  % Nutrients in faeces
100 1 À
% Nutrient in feed
The ADCs for dry matter, protein and AA were calculated from the respective digestibility coe¡cients
for Diet 0 and Diet test on the basis of 30% substitution of the SBCM in the Diet 0 (Forster et al. 2003).
ADC of SBCM ingredient ð%Þ
¼

ð70 þ 30Þ ADCDiet test À 70 Â ADCDiet 0
30

where : ADCDiet test ¼ ADC of Diet test; ADCDiet 0
¼ ADC of Diet 0
Amino acid retention ð%Þ ¼


Amino acid gain ðgÞ
100
Amino acid intake ðgÞ

Statistical analysis
Results were presented as mean Æ standard error of
mean of three replicates. Data were analysed by oneway analysis of variance (ANOVA) using the SPSS 17.0
for Windows. When ANOVA identi¢ed di¡erences
among treatments, the di¡erence in means was assessed using Tukey’s multiple range test. Di¡erences
were considered to be signi¢cant at Po0.05.

Linear regression analysis was performed on FBW
against the dietary substitution level of FM by SBCM.

limiting EAA for SBCM was methionine. The crude
protein content for SBCM (923.8 g kg À 1) was very
high compared with FM (654.4 g kg À 1), and the
crude lipid, ash and moisture contents for SBCM were
very low compared with FM respectively. Some EAA
were higher in the SBCM than in FM, except methionine and isoleucine.
The results of growth performance FBWand survival after the 56-day feeding trial are summarized in
Table 3. Survival was in the range of 91.4^98.1% and
there was no signi¢cant di¡erence in the survival of
shrimp among shrimp fed diets (P40.05). The shrimp
that fed Diet 4 and Diet 5 had signi¢cantly lower
FBW than those that fed Diets 0^3 (Po0.05). In addition, regression analysis between FBW and the dietary substitution level of FM protein by SBCM protein
indicated a signi¢cant linear regression (Fig. 1). This
result suggested that increasing dietary SBCM resulted in a decrease in FBW in the white shrimp.
The results of the feed conversion ratio (FCR) and
the protein e⁄ciency ratio (PER) after the 56-day
feeding trial are shown in Table 3. The FCR increased
from 1.42 to 1.80 as SBCM and MM inclusion level increased, but no signi¢cant di¡erence (P40.05) was
found between shrimp fed Diet 0 (0% SBCM), Diet 1
(3.5% SBCM), Diet 2 (7.0% SBCM), Diet 3 (10.5%
SBCM) and Diet 4 (14.0% SBCM) for FCR. However,
Diet 5 (100% substitution of FM with SBCM) showed
signi¢cantly poorer FCR compared with shrimp fed
Diets 0^4 (Po0.05). The PER decreased from 1.89 to
1.58 as SBCM and MM inclusion level increased. No
signi¢cant di¡erence was found in PER among
the 0%, 20%, 40%, 60% and 80% FM diet (Diets

0^4)-fed shrimp. Shrimp fed the 100% FM diet (Diet
5) had a signi¢cantly lower PER compared with the
other dietary treatments. The highest and the lowest
value for PER were obtained with Diet 1 (1.89%) and
Diet 5 (1.58%) respectively.
20
16
FBW (g)

Calculations

12

4
0

Results
The proximate composition and AA pro¢le of the FM
and SBCM varied highly. The composition of the EAA
of the protein sources showed that the principal

484

y = –0.0197x + 14.982
r = 0.751 (P = 0.000<0.001)

8

0

20
40
60
80
100
Percent replacement of FM protein (%)

120

Figure 1. Regression of dietary substitution of ¢sh meal
(FM) protein (%) and ¢nal body weight (FBW) of Litopenaeus vannamei.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489


Aquaculture Research, 2011, 42, 480^489

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

Apparent digestibility coe⁄ents of dry matter,
crude protein, crude lipid and AA of the trial diets
and SBCM ingredient for shrimp are shown in
Table 4. Apparent digestibility coe⁄ents of dry
matter and crude protein increased as the inclusion

Table 3. Mean growth performance and feed utilization of
Litopenaeus vannamei fed diets with increasing levels of
SBCM protein for 56 daysÃ

Diets

IBW (g)

FBW (g)

Suvival
(%)

FCR

PER

Diet
Diet
Diet
Diet
Diet
Diet

2.2
2.4
2.2
2.3
2.4
2.3

14.3a
14.7a
14.9a
14.3a

13.3b
12.5c

98.1
97.2
98.9
95.5
96.7
91.4

1.45a
1.42a
1.46a
1.47a
1.56b
1.80b

1.87a
1.89a
1.86a
1.84a
1.73b
1.58c

0.035
o0.001

0.440
o0.01

0
1
2
3
4
5

ANOVA

Pooled SEM
P-value

0.217
o0.001

1.040
0.283

ÃMeans of triplicates. Values in the same column with di¡erent
superscript are signi¢cantly di¡erent (Po0.05) as determined
using Tukey’s test.
ANOVA, analysis of variance; IBW, initial body weight; FBW, ¢nal
body weight; FCR, feed conversion ratio; PER, protein e⁄ciency
ratio; SEM, standard error of mean; SBCM, spray-dried blood cell
meal.

level of SCDM increased, but ADCs of crude lipid
decreased as the inclusion level of SCDM increased.
Apparent digestibility coe⁄ents of valine, phenylalanine, leucine, lysine and proline increased signi¢cantly as the inclusion level of SCDM increased, but
ADCs of isoleucine decreased as the inclusion level

of SCDM increased. Amino acid apparent digestibility
values of histidine, arginine, lysine, phenylalanine,
alanine, serine, tyrosine and glutamine were high;
however, cysteine, glycine and isoleucine were low
in all diets.
The moisture content of shrimp did not vary significantly among treatments (P40.05), which ranged
from 711.8 to 7116.9 g kg À 1 wet weight (Table 5). The
crude protein and ash content of shrimp fed all diets
had reduced as the inclusion level of SCDM increased, but were not signi¢cantly di¡erent among
treatments (P40.05). Crude lipid content was the
lowest in shrimps fed Diet 4 at 17.8 g kg À 1, and those
fed Diet 4 and Diet 5 were signi¢cantly lower than
that of shrimp fed other diets (Po0.05). All treatments were rich in aspartic glutamic and glycine
acids, and the EAAs leucine and lysine (Table 6).
Percentage of EAA retention (%) in shrimp fed
with trial diets is shown in Fig. 2. Signi¢cant di¡erences (Po0.05) were found among treatments in
EAA retention, except for histidine and isoleucine.

Table 4. Apparent digestibility coe⁄cients of dry matter, crude protein, crude lipid and amino acids of the trial diets and
SBCM for Litopenaeus vannameiÃ

Dry matter
Crude protein
Crude lipid
Aspartic acid
Glutamic acid
Serine
Histidine
Glycine
Threonine

Arginine
Alanine
Tyrosine
Cysteine
Valine
Methionine
Phenylalanine
Isoleucine
Leucine
Lysine
Proline

Diet 0

Diet 1

Diet 2

Diet 3

Diet 4

Diet 5

Pooled SEM

P-value

SBCM

76.9e
84.8c
90.6a
89.1c
94.2
87.4
92.5
89.8
90.7a
91.0
89.2c
89.6ab
81.8
91.1b
92.1
91.3b
89.7a
87.6c
90.3b
90.1b

78.1d
86.3c
90.3a
89.3c
93.3
88.2
92.2
89.6
89.8ab

92.0
89.6b
89.3b
82.1
91.2b
92.5
91.3b
88.7b
90.2b
91.9a
89.3c

80.9c
86.7c
89.7ab
89.6c
93.4
87.1
92.9
90.7
89.5ab
91.7
89.0c
90.6a
82.6
92.1ab
92.9
91.7b
88.0c
92.4a

92.0a
89.6bc

81.8a
87.5b
88.3b
90.8b
92.9
87.6
92.6
90.2
90.6ab
92.1
89.9ab
89.8ab
82.5
93.8a
92.4
93.1a
87.0d
93.3a
92.5a
91.5a

81.5ab
88.8a
88.4b
90.8b
93.9
87.5

92.3
90.2
89.1bc
92.1
90.4a
89.9ab
82.3
93.6a
92.4
93.1a
86.9d
92.8a
92.8a
91.8a

82.3a
89.0a
88.3b
91.2a
92.6
87.8
92.6
90.1
88.5c
91.9
90.2ab
89.5b
82.9
93.9a
92.8

93.4a
85.1e
93.2a
93.0a
92.1a

0.488
0.356
0.263
0.202
0.325
0.233
0.164
0.130
0.208
0.138
0.128
0.124
0.904
0.329
0.091
0.226
0.359
0.512
0.240
0.275

0.000
0.000
0.000

0.000
0.876
0.894
0.574
0.273
0.000
0.049
0.000
0.026
0.333
0.001
0.085
0.000
0.000
0.000
0.000
0.010

84.7
93.2
88.5
90.8
91.2
94.5
95.3
82.7
91.9
94.8
95.8
93.3

84.3
91.7
89.2
92.5
81.4
92.6
95.7
92.9

ÃMeans of triplicates. Values in the same column with di¡erent superscript are signi¢cantly di¡erent (Po0.05) as determined using
Tukey’s test.
SEM, standard error of mean; SBCM, spray-dried blood cell meal.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489

485


Aquaculture Research, 2011, 42, 480–489

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

Table 5 Carcass proximate composition (g kg À 1 wet
weight) of Litopenaeus vannamei fed diets containing di¡erent levels of SBCM for 56 daysÃ
Diet

Moisture

Crude protein

Crude lipid

Ash

Initial
Diet 0
Diet 1
Diet 2
Diet 3
Diet 4
Diet 5

748.7
716.9
715.6
716.2
712.2
714.0
711.8

185.8
219.2
219.9
215.1
214.0
213.2
213.2

20.5
19.3ab

19.5a
19.1abc
18.9abc
17.8c
17.9bc

22.7
17.0
17.2
16.9
16.6
16.8
16.5

0.064
0.063

0.087
0.052

0.020
0.009

0.010
0.381

ANOVA

Pooled SEM
P-value

ÃMeans of triplicates. Values in the same column with di¡erent
superscripts are signi¢cantly di¡erent (Po0.05) as determined
using Tukey’s test.
ANOVA, analysis of variance; SEM, standard error of mean; SBCM,
spray-dried blood cell meal.

Discussion
SBCM used in this study was very high in protein
(923.8 g kg À 1) and lysine (94.8 g kg À 1 protein) content, but had a low ash (35.5 g kg À 1) and methionine
(2.0 g kg À 1 dry matter) content compared with FM.
As methionine is the most limiting AA in diets for
shrimp (Akiyama et al.1991), dietary methionine pro¢les were adjusted analogous that of the control diet,
on the principle of methionine requirement of shrimp
(Millamena, Bautista & Kanazawa 1996)), by methionine supplementation.
Based on the present results, high survival was obtained for all treatments. The high survival during
the feeding trial indicated the good health condition
of the shrimp. No signi¢cant di¡erences in FBW were
found between shrimp fed Diets 0^3. However, FBW
was signi¢cantly lower in shrimp fed Diet 4 and Diet
5 (80% and 100% substitution of FM with SBCM)
compared with the other diets. Lower digestibility,
suppression of feed intake and imbalance of AA could
be the main reasons for the reduced growth performance of shrimp as the dietary plant and animal byproduct protein-level substitution of FM increased
(Forster et al. 2003; Amaya et al. 2007; HernaŁndez
et al. 2008). However, in this study, there was an increasing trend in feed intake and digestibility with
increasing dietary SBCM level. Therefore, in this
study, feeding rate and digestibility could not explain
the reduced growth performance of shrimp with
increasing dietary SBCM level. In addition, the reduced growth was probably due to the imbalance of

486

Table 6. Amino acids expressed in g kg1 of Litopenaeus vannamei carcass after 56 days
AA

Diet 0

Diet 1

Diet 2

Diet 3

Diet 4

Diet 5

Asp
Glu
Ser
His
Gly
Thr
Arg
Ala
Tyr
Cys
Val
Met

Phe
Iso
Leu
Lys
Pro

68.4
118.3
24.5
13.3
62.8
24.1
55.7
48.1
23.2
2.94
36.4
17.0
31.5
31.8
52.6
55.4
48.3

67.3
120.8
25.9
13.7
62.4
23.7

54.4
47.7
22.6
3.2
38.3
17.2
31.3
32.1
52.2
55.1
48.1

67.2
120.5
21.5
13.5
61.4
23.4
54.7
49.7
22.1
2.8
36.6
17.0
30.8
32.0
52.1
52.1
48.9

70.0
122.7
22.7
13.7
58.7
24.0
57.1
50.7
23.3
2.87
37.5
17.4
31.4
31.8
53.4
48.2
43

67.5
119.0
23.6
13.1
58.4
24.5
54.7
51
22.5
3.37
39.5
17.3

29.9
30.2
52.2
46.1
42.4

66.1
118.1
21.8
13.1
58.3
22.8
54.3
47.7
23.2
3.17
35.3
16.5
29.8
29.8
55.0
45.0
41.9

AA in Diet 4 and Diet 5. The dietary ioleucine content
was decreased with increasing dietary SBCM level
and ioleucine was one of the EAA for shrimp (Millamena, Bautista, Reyes & Kanazawa1999). Thus, it can
be concluded that de¢ciency of ioleucine in Diet 4
and Diet 5 is an important limiting factor when a
high SBCM level with MM supplementation was used

as the protein source for shrimp.
No signi¢cant di¡erences were observed for FCR
and PER among shrimp fed Diets 0^5. However, FCR
was signi¢cantly lower and PER was signi¢cantly
higher in shrimp fed Diets 0^3 compared with Diet
4 and Diet 5. Two primary dietary factors may be affected FCR and PER. First, part of the dietary protein
may be metabolized for energy when dietary nonprotein nutrients (carbohydrates and lipids) are not
improved or utilized properly (Hu, Tan, Mai, Ai,
Zheng & Cheng 2008). Second, di¡erent protein
sources with di¡erent protein quality may result in
di¡erent PER values (HernaŁndez et al. 2008). In the
present study, protein sources were di¡erent but the
digestibility of dietary protein was increased with increasing dietary SBCM level. Therefore, di¡erences in
response were attributed to a decrease in digestible
crude lipid of the diets with high levels of SBCM.
Growth and feed e⁄ciency results con¢rmed that L.
vannamei can be fed SBCM with MM reducing 60%
FM level.
Digestibility of a feed is an important factor to consider in determining the utilization of the feed
(Akiyama, Coelho, Lawrence & Robison 1989; Yang,

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489


Aquaculture Research, 2011, 42, 480^489

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

Diet 0

Diet 1

Diet 2

Diet 3

Diet 4

Diet 5

80
EAA retention (%)

70
60
aaaa

50
40
30

aaaa

a
aaa a
b
a

a aa

b

b
aaaaa

a

a a
a a ab

a

a a

b aa
c

aa
ab

a

aa

bb c
bc

ab
b

b
bc
c

a

20
10
0

His

Thr

Arg

Val

Met
EAA

Phe

Iso

Leu

Lys

Figure 2. Percentage of essential amino acid (EAA) retention (%) in Litopenaeus vannamei fed with trial diets containing

di¡erent levels of spray-dried blood cell meal for 56 days. Data with di¡erent superscripts di¡er at Po0.05 as determined
using Turkey’s test.

Zhou, Zhou,Tan, Chi & Dong 2009). Digestibility data
re£ect the percentage of a feed sample that is absorbed from an animal’s intestinal tract (Sales 2009).
High-digestibility diets are especially important under high-density culture conditions, where accumulations of undigested feed can foul the water, increase
the cost of water treatment and the risks of shrimp
disease and mortality. In the present study, ADCs of
dry matter and crude protein of trial diets increased
slightly as SBCM increased. All diets containing a
high level of SBCM were more digestible. The high digestibility of dry matter was probably due to the low
ash content of the diet. The processing of animal
blood meal ingredients appeared to be relatively important when considering the high digestibility coef¢cients recorded for all the diets evaluated. This may
be due to the cellular fractions or other processing of
blood enhanced dry matter digestibility. Diets that
contained a high level of SBCM ingredient were all
highly digestible. The improvement in protein digestibility could be attributable to the total breakdown of
the cell membrane during the processing, improving
the digestibility. Apparent digestibility coe⁄ents of
crude protein trial Diets 0^5 were 84.8%, 85.3%,
85.7%, 87.5%, 88.8% and 89.0% respectively. The
ADCs crude protein of the SBCM observed here was
markedly higher than porcine meat meal in practical
diets (Cruz-SuaŁrez, Nieto-Lo¤pez, Guajardo-Barbosa,
Tapia-Salazar, Scholz & Ricque-Marie 2007; HernaŁndez et al. 2008; Yang et al. 2009). The ADCs of crude
protein of blood meals manufactured using di¡erent
techniques have been shown to be signi¢cantly different. Spray-dried blood products showed consistently high ADCs. The results were in accordance
with the results from a previous study, which showed

that spray-dried blood cell meal (SBCM) was almost

completely digestible (Bureau et al. 1999; El-Haroun &
Bureau 2007). Good results were obtained from both
apparent availability for AAs and apparent protein digestibility for the test ingredients evaluated. Apparent
digestibility coe⁄ents of AA of SBCM ingredient for
the white shrimp ranged from 81.4% to 95.8%. Therefore, ADCs of AA replacement of FM with SBCM were
higher than those of plant protein (Davis, Arnold
& McCallum 2002; Rivas-Vega, Goytortu¤a-Bores,
Ezquerra-Brauer, Salazar-Garc|¤ a, Cruz-SuaŁrez, Nolasco & Civera-Cerecedo 2006; Cruz-SuaŁrez, TapiaSalazar, Villarreal-Cavazos, Beltran-Rocha, NietoLo¤pez, Lemme & Ricque-Marie 2009) and other
animal by-products to replace FM in the diet for
shrimp (Forster et al. 2003;Yang et al. 2009).
Previous studies indicated that there were no signi¢cant di¡erences in body compositions (protein, lipid, ash and moisture) among shrimp fed the diets
with graded levels of rendered animal proteins. However, lipid of shrimp fed high SBCM level diets was affected in the present study, but protein, ash and
moisture were not signi¢cantly di¡erent. The main
reason might be that the dietary SBCM lipid digestibility and availability reduced, which could re£ect
the actual e¡ect of high dietary a SBCM level on lipid
of shrimp. In this study, the relatively steady protein
and EAA contents in shrimp body suggested that
shrimp could keep protein and EAAs to maintain
normal physiological function when exposed to different diet formulations. The carcass ash content
was not in£uenced by the dietary SBCM level, which
was consistent with the results of those previous studies (Tan, Mai, Zheng, Zhou, Liu & Yu 2005; HernaŁndez et al. 2008).

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489

487


E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

There were no signi¢cant di¡erences in AA retentions among Diets 0^3 as a result of balance in the

AA composition of the trial diets, and similar results
were reported by Hu,Wang,Wang, Zhao, Xiong, Qian,
Zhao and Luo (2009), who con¢rmed that when the
dietary AA pro¢le is balanced through supplementation, an improvement in AA retention is probably re£ected in an increase in protein retention. Another
important reason for these results seemed to be that
SBCM protein is a part of shrimp diet. With a high
dietary SBCM level, some EAA and total EAA contents of Diet 4 and Diet 5 increased, which were overabundant for shrimp growth performance and
reduced in AA retentions of shrimp. Therefore, it is
suggested that the usage of a combination of SBCM
and other protein ingredients or AA supplementation to replace FM protein in shrimp diet would provide more economic bene¢ts and nutritional
potential in formulating a nutritionally balanced
and cost-e¡ective diet, and the EAA pro¢le to meet
the nutritional requirement of the tested animals is
the most important factor.
In conclusion, high levels of a combination of
SBCM ingredients can be used to replace FM in the
white shrimp practical feeds with microencapsulated methionine supplementation, in which the protein content and EAA pro¢les of the combination are
formulated similar to FM. However, an optimal replacement of 60% FM protein is suggested at the point of
MM supplementation without negatively a¡ecting
shrimp growth. Inclusion of a relatively high level of
a combination of ingredients in shrimp commercial
diet will result in a signi¢cant reduction in feed costs,
and then make the aquafeeds industry more economically feasible. Therefore, SBCM with MM is an
acceptable alternative animal protein source of nutrients in diets for shrimp.
Acknowledgments
This study has been ¢nancially supported by
Zhejiang, China domestic scienti¢c and technological cooperation and achievement project
(2007d70SA45004): New security special aquatic
feed R&D. The authors thank Ningbo TECH-BANK
and Guangdong Haid Group for providing some ingredients for this study.

References
Akiyama D.M., Coelho S.R., Lawrence A.L. & Robison E.H.
(1989) Apparent digestibility of feedstu¡s by the marine

488

Aquaculture Research, 2011, 42, 480–489

shrimp Penaeus vannamei boone. Nippon Suisan Gakkaishi
55, 91^98.
Akiyama D.M., DominyW.G. & Lawrence A.L. (1991) Penaeid
shrimp nutrition for the commercial feed industry. In:
Marine Shrimp Culture: Principles and Practices (ed. by
A.L. Fast & L.J. Lester), pp. 535^568. Elsevier Science Publishers BV, Amsterdam, the Netherlands.
Amaya E, Davis D.A. & David B.R. (2007) Alternative diets
for the Paci¢c white shrimp Litopenaeus vannamei. Aquaculture 262, 419^425.
Association of O⁄cial Analytical Chemists (AOAC) (1995)
O⁄cial Methods of Analysis of O⁄cial Analytical Chemists
International, 16th edn. Association of O⁄cial Analytical
Chemists, Arlington,VA, USA.
Brunson J.F., Romaire R.P. & Reigh R.C. (1997) Apparent digestibility of selected ingredients in diets white shrimp.
Penaeus setiferus L. Aquaculture Nutrition 3, 9^16.
Bureau D.P., Harris A.M. & Cho C.Y. (1999) Apparent digestibility of rendered animal protein ingredients for rainbowtrout (Oncorhynchus mykiss). Aquaculture 180,
345^358.
Cruz-SuaŁrez L.E., Nieto-Lo¤pez M., Guajardo-Barbosa C., Tapia-Salazar M., Scholz U. & Ricque-Marie D. (2007) Replacement of ¢shmeal with poultry by-product meal in
practical diets for Litopenaeus vannamei, and digestibility
of the tested ingredients and diets. Aquaculture 272,
466^476.
Cruz-SuaŁrez L.E., Tapia-Salazar M., Villarreal-Cavazos D.,
Beltran-Rocha J., Nieto-Lo¤pez M.G., Lemme A. & RicqueMarie D. (2009) Apparent dry matter, energy, protein and

amino acid digestibility of four soybean ingredients in
white shrimp Litopenaeus vannamei juveniles. Aquaculture
292, 87^94.
Davis D.A. & Arnold C.R. (2000) Replacement of ¢sh meal in
practical diets for the Paci¢c white shrimp, Litopenaeus
vannamei. Aquaculture 185, 291^298.
Davis D.A., Arnold C.R. & McCallum I. (2002) Nutritional value of feed peas (Pisum sativum) in practical diet formulations for Litopenaeus vannamei. Aquaculture Nutrition 8,
87^94.
El-Haroun E.R. & Bureau D.P. (2007) Comparison of the bioavailability of lysine in corpuscle meals of various origins
to that of L-lysine HCL for rainbow trout (Oncorhynchus
mykiss). Aquaculture 262, 402^409.
FAO (2008) FAO Fisheries Department, Fishery Information,
Data and Statistics Unit. Fishstat Plus: Universal software
for ¢shery statistical time series. Aquaculture production:
quantities 1950^2006, Aquaculture production: values
1984^2006, Capture production: 1950^2006, Commodities production and trade,1950^2006,Version 2.30.
Folch J., Lees M. & Sloane-Stanley G.H. (1957) A simple method for the isolation and puri¢cation of total lipids from
animal tissues. The Journal of Biological Chemistry 226,
497^509.
Forster I.P., DominyW., Obaldo L. & Tacon A.G.J. (2003) Rendered meat and bone meals as ingredients of diets for

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489


Aquaculture Research, 2011, 42, 480^489

shrimp Litopenaeus vannamei (Boone, 1931). Aquaculture
219, 655^670.
Hardy R.W. (2010) Utilization of plant proteins in ¢sh diets:
e¡ects of global demand and supplies of ¢shmeal. Aquaculture Research 41,770^776.

HernaŁndez C., Olvera-Novoa M.A., Aguilar-Vejar K., GonzaŁlez-Rodr|¤ guez B. & Parra I.A. (2008) Partial replacement
of ¢sh meal by porcine meat meal in practical diets for Paci¢c white shrimp (Litopenaeus vannamei). Aquaculture
277, 244^250.
HuY.,Tan B.P., Mai K.S., Ai Q.H., Zheng S. & Cheng K. (2008)
Growth and body composition of juvenile white shrimp,
Litopenaeus vannamei, fed di¡erent ratios of dietary protein to energy. Aquaculture Nutrition 14, 499^506.
Hu M.H.,WangY.J.,Wang Q., Zhao M., Xiong B.X., Qian X.Q.,
ZhaoY.J. & Luo Z. (2009) Replacement of ¢sh meal by rendered animal protein ingredients with lysine and methionine supplementation to practical diets for gibel carp,
Carassius auratus gibelio. Aquaculture 275, 260^265.
Millamena O.M., Bautista M.N. & Kanazawa A. (1996)
Methionine requirement of juvenile tiger shrimp Penaeus
monodon fabricius. Aquaculture 180, 403^410.
Millamena O.M., Bautista M.N., Reyes O.S. & Kanazawa A.
(1999) Quantitative dietary requirement of postlarval tiger shrimp, Penaeus monodon, for histidine, isoleucine,
leucine, phenylalanine and tryptophan. Aquaculture 179,
69^179.
Naylor R.L., Goldberg R.J., Mooney H., Beveridge M., Clay J.,
Folke C., Kautsky N., Lubchenco J., Primavera J. & Williams M. (1998) Nature’s subsidies to shrimp and salmon
farming. Science 282, 883^884.
Richter H., LˇckstÌdt C., Focken U. & Becker K. (2003) Evacuation of pelleted feed and the suitability of titanium (IV)

E¡ects of spray-dried blood cell meal on L. vannamei H Niu et al.

oxide as a feed marker for gut kinetics in Nile tilapia. Journal of Fish Biology 63,1080^1099.
Rivas-Vega M.E., Goytortu¤a-Bores E., Ezquerra-Brauer J.M.,
Salazar-Garc|¤ a M.G., Cruz-SuaŁrez L.E., Nolasco H. &
Civera-Cerecedo R. (2006) Nutritional value of cowpea
(Vigna unguiculata L. Walp) meals as ingredients in diets
for Paci¢c white shrimp (Litopenaeus vannamei Boone).
Food Chemistry 97, 41^49.

SalesJ. (2009) Prediction of digestible protein and lipid contents
of crustacean feeds. Aquaculture Nutrition 15, doi:
10.1111/j.1365-2095.2009.00685.x.
SuaŁrez J.A., Gaxiola G., Mendoza R., Cadavid S., Garcia G.,
Alanis G., SuaŁrez A., Faillace J. & Cuzon G. (2009) Substitution of ¢sh meal with plant protein sources and energy
budget for white shrimp Litopenaeus vannamei (Boone,
1931). Aquaculture 289,118^123.
Tacon A.G.J. & Metian M. (2008) Global over view on the
use of ¢shmeal and ¢sh oil in industrially compounded
aquafeeds: trends and future prospects. Aquaculture 285,
146^158.
Tan B.P., Mai K.S., Zheng S., Zhou Q., Liu L. & Yu Y. (2005)
Replacement of ¢sh meal by meat and bone meal in practical diets for the white shrimp Litopenaeus vannamei
(Boone). Aquaculture Research 36, 439^444.
ToldraØ M., Elias A., Pare¤s D., Saguer E. & Carretero C. (2004)
Functional properties of a spray-dried porcine red corpuscle cell fraction treated by high hydrostatic pressure. Food
Chemistry 88, 461^468.
Yang Q.H., Zhou X.Q., Zhou Q.C., Tan B.P., Chi S.Y. & Dong
X.H. (2009) Apparent digestibility of selected feed ingredients for white shrimp Litopenaeus vannamei, boone. Aquaculture Research 41,78^86.

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 480^489

489


Aquaculture Research, 2011, 42, 490^498

doi:10.1111/j.1365-2109.2010.02644.x

Role of organic fertilizers in walleye (Sander vitreus)

production in plastic-lined culture ponds
Sarah E Kaatz1, Joseph E Morris1, James B Rudacille2, John Alan Johnson2 & Richard D Clayton1
1

Department of Natural Resource Ecology and Management, Iowa State University, Ames, IA, USA
Iowa Department of Natural Resources, Rathbun Fish Culture Research Facility, Moravia, IA, USA

2

Correspondence: J E Morris, Department of Natural Resource Ecology and Management, 339 Science II, Iowa State University, Ames, IA
50011, USA. E-mail:

Abstract
This study used six 0.04 ha plastic-lined ponds to
compare the e¡ects of a fertilization regime using a
one-time initial application of an organic fertilizer
(alfalfa pellets) with the current regime of weekly applications of organic fertilizers on the abundance and
distribution of aquatic invertebrates and walleye,
Sander vitreus, ¢ngerling production. Walleye, 3^4
days post hatch, were stocked on 1 May 2002 and
harvested on 6^7 June 2002. Throughout the growing season, a ratio of 7:1total nitrate-nitrogen to total
phosphorus was maintained in all ponds regardless
of the treatment. Once fry were stocked, ponds in
Treatment #1 were fertilized weekly with organic
fertilizer (alfalfa pellets; 112 kg ha À 1 week À 1) for a
total of 795 kg ha À 1 pond À 1. Ponds in Treatment
#2 only received an initial application of alfalfa
pellets (112 kg ha À 1). Ponds inTreatment #1had signi¢cantly higher ammonia and nitrate levels as well
as higher chironomid larvae but not zooplankton
compared with the other treatment. At harvest,

walleye in the Treatment #1 ponds were signi¢cantly longer and heavier; however, the survival and
relative weight were not signi¢cantly di¡erent. These
results suggest that weekly applications of organics
are important for the benthic food base and growth
of ¢ngerling walleye reared in plastic-lined ponds.

Keywords: walleye, organic fertilizers, plastic lined,
ponds
Introduction
There are a limited number of peer-reviewed publications on the management of plastic-lined ponds for
¢sh culture (Barkoh 1996; Barkoh, Stacell & Smith
1996; Buurma, Barkoh & Alexander 1996; Sparrow

490

1996; Rogge, Moore & Morris 2003). These studies
have focused on fertilization regimes to promote desirable zooplankton populations and not on the growth
of benthic invertebrate communities. Research is necessary to characterize benthic organisms found in
plastic-lined culture ponds as well as their colonization rates, because benthic invertebrates, especially
chironomid midges, are a potential food source for larval ¢sh (Crowder & Cooper 1982; Fox 1989; Culver &
Geddes 1993; Summerfelt, Clouse & Harding 1993;
Flowers 1996; Mischke 1999; Rogge et al. 2003).
Although there are numerous benthic invertebrates
in ponds, chironomids (Order Diptera) are the most
important food for larval ¢sh (Fox 1989). The life
history of chironomids sheds some light on how
these invertebrates may enter plastic-lined ponds. The
chironomid egg varies in size from170 mm long to over
600 mm (Nolte 1993), and may enter through supply
water depending on the source and the ¢ltration apparatus. However, the colonization of new habitats

by chironomids is conducted predominantly by the
dispersal of fertilized females, and adult movement is
often determined by the wind rather than their ability
to £y (Armitage, Cranston & Pinder 1995). Armitage
et al. (1995) also stated that many colonization studies
focus mainly on the larval stage, and colonization by
adults is often overlooked. In a related study by Kaatz,
Morris, Rudacille and Clayton (2010), they observed
that the primary conduit for chironomids was
airborne and not through the incoming water. As the
dry weight of dipterans is linearly related to the
width of the head capsule (Smock 1980), we measured
chironomid head capsules at the middle of the season
(24 May) and at harvest (6 June), and used Smock’s
regression equation to determine biomass.
Previous work at this facility focused on di¡erent
combinations of fertilizers (Rogge 2002; Rogge et al.

r 2010 Blackwell Publishing Ltd


Aquaculture Research, 2011, 42, 490^498 Role of organic fertilizers in walleye (Sander vitreus) production S E Kaatz et al.

2003).While Rogge (2002) noted that the use of either
organic or inorganic fertilizers did increase walleye
production in plastic-lined ponds, Rogge et al. (2003)
found no signi¢cant di¡erences in ¢sh production
related to fertilizer regimes (organic versus a mix of
inorganic and organic fertilizers) but did recommend
increasing the initial total phosphorus (TP) levels

to 0.10 mg L À 1 before stocking ¢sh. A consequence
of these ¢ndings was a desire to evaluate whether
a fertilization regime of a single application of an
organic fertilizer upon initial pond £ooding would
be as e¡ective as the standardized weekly regime of
a mix of organic and inorganic fertilizers suggested
by Rogge et al. (2003) for walleye culture in plasticlined ponds. The adoption of this limited fertilization
approach would be more economical in terms of
the costs and labour as well as possible improvement
in water quality, e.g., low dissolved oxygen (DO).
The speci¢c objectives of this study were to evaluate
the e¡ects of two treatments, Treatment #1 [based
on Rogge et al. (2003), using recommendations for
weekly applications of a mix of organic and inorganic
fertilizers] and Treatment #2 (only one initial application of an organic fertilizer and weekly applications
of inorganic fertilizers) on the abundance of aquatic
invertebrates in plastic-lined ponds and its concomitant e¡ects on walleye ¢ngerling culture.

Materials and methods
Study site description
This study was conducted at the Rathbun Fish Culture
Research Facility, Moravia, IA, USA. Six 0.04 ha plastic-lined ponds were randomized as replicates in two
treatments. Each pond had a mean depth of ca. 1m, a
bottom slope of 4.6% and a side slope of 2:1. The plastic
lining used was 36-mil-thick black polypropylene
(Geo-Synthetics,Waukesha,WI, USA).
Before this experiment, these ponds were used to
culture channel cat¢sh (Ictalurus punctatus) ¢ngerlings. Upon completion of that culture period, these
ponds were thoroughly rinsed and brushed out
before being £ooded to ensure limited residual e¡ects

of zooplankton inocula and nutrients. Water from
Rathbun Lake was used to ¢ll all the ponds on 18^19
April 2002.Water passed through a rotary drum ¢lter
with a 300 mm screen before entering the ponds
to ¢lter out ¢sh and macro invertebrates. However,
given the screen opening size, some smaller zooplankton species, e.g., Bosmina spp, or early instar stages of
larger cladocerans or chironomids likely entered the

ponds at this time. Lake Rathbun walleye, 3 to 4 days
old, were stocked at a rate of 250 000 ¢sh ha À 1 on 1
May. Fry were counted using an optical fry counter
(Jensorter Model FC2, Jensorter, Bend, OR, USA).

Fertilization
The initial TP levels in all ponds were adjusted on
25 April to 0.10 mg L À 1 at the initiation of the project
using an inorganic fertilizer (12-46-6; 0.04^
0.08 kg pond À 1) for all ponds (Rogge et al. 2003).
Nitrogen (dry urea, 36-0-0) was added to each pond
regardless of the treatment starting 3 May to maintain a nitrate-nitrogen to TP ratio (NO3-N:TP) of 7:1
(Mischke 1999). Although it has been an acceptable
practice to focus on phosphorus as the limiting factor
in freshwater pond production (Boyd 1981, 1997),
later studies have shown it is actually nitrogen
in highly enriched watersheds or in new ponds
(Mischke 1999; Mischke & Zimba 2004; Boyd,
Penseng & Boyd 2008). An additional application
of urea was carried out on 8 May; the total amounts
of urea varied between10.2 and 26.1kg ha À 1 pond À 1
in Treatment #1 and 11.3 and 20.4 kg ha À 1 pond À 1

for Treatment #2. All nitrogen applications ceased
the following week due to increasing ammonia levels
in ponds in Treatment #1.
Organic fertilizer (alfalfa pellets), estimated at
2.55% nitrogen and 0.23% phosphorus (NRC 1993)
and actually measured at 0.07% total ammonia
nitrogen and 0.01% nitrate-nitrogen and 0.45%
phosphorus (Minnesota Valley Testing Laboratories,
New Ulm, MN, USA), was applied to all ponds at a rate
of112 kg ha À 1 on19 April. Thereafter, only the ponds
in the Treatment #1 received weekly applications
of alfalfa pellets at the same rate, resulting in a total
of 795 kg ha À 1 pond À 1 for the culture period. Ponds
in Treatment #2 received only an initial amount of
112 kg ha À 1 pond À 1 (86% lower). Organic fertilizers
were spread along the shallow depths (ca. o0.5 m) to
allow for adequate decomposition.

Sampling methods
Water sampling began on 22 April. Water samples
were collected twice a week using a tube sampler
(Graves & Morrow 1988) to sample the entire water
column. Analysis included ammonia-nitrogen
(NH3-N), nitrite-nitrogen (NO2-N), NO3-N, TP and
bi-weekly alkalinity and hardness. Water chemistry
was assessed using a Hach DR/2010 spectro-

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 490^498

491

Role of organic fertilizers in walleye (Sander vitreus) production S E Kaatz et al.

photometer (Hach, Loveland, CO, USA). Pond-side
water quality testing included DO, pH and temperature readings. A YSI Model 60 pH meter (Yellow
Springs, OH, USA) was used to record pH and
temperature. Standards were used weekly to calibrate the pH meter. Morning (06:00 hours) temperature and DO levels at the bottom, middle and top of
the ponds were taken twice per week using a YSI
Model 55 oxygen meter.
Water samples for chlorophyll a were collected twice
weekly. Samples of 300 mL were ¢ltered through
47 mm glass micro¢bre ¢lters using a vacuum pump
(Barnant Company, Barrington, IL, USA). The ¢lters
were frozen and stored in the dark until analysis in
August and September. Chlorophyll a and phaeophytin levels were analysed at Iowa State University using
the procedures described byAPHA (1998).
Zooplankton was sampled twice weekly with three
oblique tows per pond using an 80 m Wisconsin net
(Wildco Company, Saginaw, WI, USA), pulled a standard length and preserved with a chilled formalin/
sucrose solution (APHA 1998); the total volume
¢ltered was estimated at 68 L. Specimens were
counted and identi¢ed using Pennak (1989). Benthos
was sampled using Hester^Dendy multiple plate
samplers (Hester & Dendy 1962; 786 cm2). Six sets of
eight plates were placed randomly in each pond at a
depth of 0.3 m. One sampler was retrieved each week.
Samples were preserved in 10% bu¡ered formalin.
Samplers were later scraped to quantify organisms
and identify to the family level using Merritt and

Cummins (1996). Aquatic insect adults were sampled
with emergence traps (Edmondson & Winberg 1971).
One trap was placed at the water’s surface in the
shallow end of each pond. Traps were emptied once
a week. Samples were preserved in a chilled formalin/sucrose solution (APHA 1998). Samples were
later identi¢ed to the family level using Merritt and
Cummins (1996).

Aquaculture Research, 2011, 42, 490–498

Fingerlings were harvested on 6^7 June by completely draining the ponds and collecting them from
the catchment basins. The mean length and weight
were determined from 100 ¢sh samples per pond
obtained at harvest. Relative weight (Wr) is a technique that compares ¢sh weight against a standard
weight (Ws) of ¢sh in good condition using published
data (Anderson & Neumann 1996). The formula is
Wr 5 (W/Ws) Â 100. The relative weights for young
of the year walleye were calculated using the equation formulated by Flammang, Olson and Willis
(1999). The Ws equation was log10Ws(g) 5 À 4.8041
2.869 log10 total length (TL, mm). The survival rate
was calculated for each pond by subtracting the estimated harvest number (calculated using the mass
weight divided by the average individual weight)
from the initial stocking number.
Data analyses
Treatment di¡erences in water chemistry (ammonia,
nitrite, nitrate and TP), zooplankton, benthic, emerging aquatic insect populations and ¢sh growth,
survival and biomass were analysed using SAS 9.1 (SAS
Institute, Cary, NC, USA). Water quality and macro-invertebrate data were analysed using the MIXED procedure with an auto regressive covariance structure.
Least-square mean comparisons were made using an
LSD test. Fish production was compared using an unpaired t-test. Signi¢cance was set at P 0.10.

Results
Water chemistry
At the initiation of this study, there were no signi¢cant treatment di¡erences in water quality on the
¢rst sampling date, April 22. Ponds in Treatment #1
had increased levels of NH3-N and NO3-N (Table 1)
over the study period. The signi¢cant interaction

Table 1 Least-square means (Æ standard error) of water quality variables (mg L À 1) in ponds treated in either Treatment #1
(weekly mix of inorganic1organic fertilizers) or Treatment #2 (single application of organic fertilizers and weekly applications of inorganic fertilizers) at the Rathbun Fish Culture Research Facility, Moravia, IA, USA
Water quality
parameter

Treatment #1

Ammonia
Nitrite
Nitrate
Total P
pH

0.267
0.01
0.33
0.110
8.55

Æ
Æ
Æ
Æ

Æ

0.014
0.001
0.023
0.057
0.065

Treatment #2
0.159
0.01
0.21
0.088
8.89

Æ
Æ
Æ
Æ
Æ

0.0140
0.001
0.023
0.0057
0.065

Treatment P4FÃ

Date P4F

Interaction treatment
 time P4F

0.006
0.183
0.020
0.055
0.021

o0.001
o0.001
0.108
o0.001
o0.001

0.062
0.303
0.672
0.457
0.754

ÃRepeated measures using the MIXED model, N 5 4 for each treatment.

492

r 2010 Blackwell Publishing Ltd, Aquaculture Research, 42, 490^498