North American Journal of Aquaculture

This article was downloaded by: [Department Of Fisheries]
On: 25 September 2012, At: 23:59

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
North American Journal of Aquaculture

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/unaj20
Parental Male Effects on Landlocked Fall Chinook

Salmon Progeny Survival
Matthew M. Wipf
a b
& Michael E. Barnes
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop,
Spearfish, South Dakota, 57783, USA
b
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop,
Spearfish, South Dakota, 57783, USA
Version of record first published: 22 Aug 2012.
To cite this article: Matthew M. Wipf & Michael E. Barnes (2012): Parental Male Effects on Landlocked Fall Chinook Salmon
Progeny Survival, North American Journal of Aquaculture, 74:4, 443-448
To link to this article: http://dx.doi.org/10.1080/15222055.2012.681105
PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to
anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses should
be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,
proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly in
connection with or arising out of the use of this material.
North American Journal of Aquaculture 74:443448, 2012

C American Fisheries Society 2012
ISSN: 1522-2055 print / 1548-8454 online
DOI: 10.1080/15222055.2012.681105
ARTICLE
Parental Male Effects on Landlocked Fall Chinook Salmon

Progeny Survival
Matthew M. Wipf*1 and Michael E. Barnes
Downloaded by [Department Of Fisheries] at 23:59 25 September 2012
South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearfish,
South Dakota 57783, USA
Abstract
The Lake Oahe, South Dakota, population of landlocked fall-run Chinook salmon Oncorhynchus tshawytscha is
maintained entirely by hatchery propagation and exhibits relatively poor egg survival during hatchery incubation.
This study was undertaken to determine the influence of male gametes on embryo survival. Eggs from an individual
female were subdivided and subsequently fertilized with milt from four discrete males. This was repeated with three
additional females using the milt from the same four males. This entire procedure was then replicated three times,
using four new females and four new males each time, for a total of 16 males and 16 females. The eggs from each
unique cross were then incubated discretely. There was no significant effect of spawning males on subsequent embryo
survival to the eyed stage of egg development. Swim-up fry length and weight were also not significantly affected by
male parentage. In contrast, there was a significant maternal effect on eyed egg survival, and swim-up fry length and
weight, which varied significantly among progeny from individual females. These results suggest that the relatively
poor survival exhibited by Lake Oahe landlocked fall Chinook salmon eggs during hatchery incubation is largely a
function of initial egg quality from spawning females.
Eggs from landlocked fall-run Chinook salmon Oncorhynchus tshawytscha from Lake Oahe, South Dakota exhibit
poor, and extremely variable, survival during hatchery incubation (Barnes et al. 1999b, 2000, 2001). This high egg mortality
has been attributed to poor egg quality (Barnes et al. 2001),
and research has primarily focused on possible maternal effects
(Barnes et al. 2003b). The possible effects of spawning male
contributions on egg survival have historically been controlled
by the use of pooled milt (a spawning ratio of multiple males to
one female; Barnes et al. 2003a).
In salmonids the reported effects of male contributions
on subsequent embryo survival have generally been variable
(Smoker 2000, 2004). Spawning male age and size can influence sperm potency (Wedekind et al. 2007), suggesting that
the use of pooled milt may be beneficial (Billard et al. 1996).
In addition, the use of pooled milt can increase the speed and
efficiency of spawning procedures (Billard et al. 1996; Withler and Beacham 1994). Anderson (1994) and Alcock (2005)
both suggested that offspring quality can be improved by using
milt from multiple males during spawning, with subsequent increases in overall offspring fitness. The use of pooled milt was
recommended for spawning of many fish species by Piper et al.
(1982).
However, the use of pooled milt has produced mixed results
in a variety of salmonids, such as Chinook salmon (Withler
1988; Withler and Beacham 1994), rainbow trout O. mykiss
(Gile and Ferguson 1995; Babiak et al. 1998), pink salmon O.
gorbuscha (Gharrett and Shirley 1985), coho salmon O. kisutch
(Fleming and Gross 1994), and sockeye salmon O. nerka (Foote
et al. 1997). The potentially negative genetic consequences of
using pooled milt during spawning procedures led Simmons and
Kotiaho (2002) and Campton (2004) to recommend spawning
*Corresponding author: matt.wipf@state.sd.us

1
Present address: South Dakota Game, Fish, and Parks, McNenny State Fish Hatchery, 19619 Trout Loop, Spearfish, South Dakota 57783,
USA.
Received December 5, 2011; accepted March 25, 2012
443
444
WIPF AND BARNES
ratios of one male to one female. Likewise, Withler (1988)

stated that in small populations of fish, eggs and milt should be
added at a one female to one male ratio so as to combat genetic
homogeneity and reduce the chances of genetic bottlenecking.
There has been no prior research examining paternal influences on landlocked fall Chinook salmon embryo characteristics. Thus, the objective of the current study was to determine the
influence of male gametes on the subsequent embryo survival
to the eyed egg stage, as well as the fry lengths and weights at
complete yolk sac absorption.
METHODS
Spawning occurred during October, 2008 at Whitlocks
Spawning Station, Lake Oahe, using randomly selected ripe fish
that had ascended the fish ladder. After each fish was euthanized
using carbon dioxide, milt was manually stripped from 16 males
(Piper et al. 1982), split into 16 discrete tubes per male, and then
placed on ice. Eggs were pneumatically spawned from ripe females with compressed oxygen at a pressure of approximately
0.56 kg/cm2 injected via a needle into the coelomic cavity to express the eggs into a plastic pan (33 38 cm, 13 cm deep). The
entire egg contribution from an individual female was divided
into four similar-sized subgroups, each placed into a separate
container (15 15 cm, 5 cm deep). Each of these subgroups
was fertilized by milt from one of four discrete males. Using
the milt from the same four males, this process was repeated
with the subdivided egg contributions from three additional females. This entire procedure was then replicated three times,
using four new females and four new males each time, for a
total of 16 males and 16 females in the study. No spawn containing noticeably overripe eggs was used (Barnes et al. 2000;
2003a). Weights (g) and total lengths (mm) were recorded from
all of the fish after spawning.
After placement of the milt from discrete males into the eggs
from discrete females, sperm was activated using 16 C lake
water for 1 min. The fertilized eggs were then washed with
lake water, put into discrete, lake water-filled plastic bags, and
transported 4 h to McNenny State Fish Hatchery, rural Spearfish,
South Dakota (Barnes et al. 1999a). Upon arrival at McNenny,
the eggs were disinfected in a 100-mg/L active iodine solution

for 10 min and inventoried using the water displacement method
(Piper et. al. 1982).
Eggs were then placed into discrete vertical-flow upwelling
incubator trays and incubated in 11 C well water (total hardness
as CaCO3 , 360 mg/L; alkalinity as CaCO3 , 210 mg/L; pH, 7.6;
total dissolved solids, 390 mg/L). At the eyed egg stage (incubation day 32), all dead eggs were removed, eyed eggs were
reinventoried, and returned to discrete incubator trays. If survival in the egg group was extremely low (<1%) the remaining
eggs and possible subsequent fry were considered too few to be
representative, the egg group was not retrayed. Percent survival
to the eyed stage was calculated: 100[1 (egg mortality/initial
number of eggs)]. At complete yolk sac absorption, fry lengths
and weights were recorded.
Because percentage data are not normally distributed (Ott
1984), egg survival data were arcsine-transformed prior to analysis of variance (ANOVA). Percent survival data were analyzed
using a randomized complete block design, the eggs from each
female acting as a block. Two-way ANOVA for incomplete
blocks was conducted on swim-up fry length and weight data.
Analysis was completed using the General Linear Model, Univariate Analysis function of the SPSS, version 9.0, statistical
analysis program (SPSS, Inc., Chicago, Illinois). Significance
was set at = 0.05.
RESULTS AND DISCUSSION

Mean spawning female lengths (702 mm) and postspawn
weights (3,027 g) were not significantly different among the
replicates (Table 1) nor were male lengths (731 mm) and weights
(3,946 g). There was no significant effect of male parentage on
embryo survival (Table 2). However, spawning females did have
a significant effect on egg survival. There was no significant interaction between spawning males and females on egg survival.
Fry lengths and weights at complete yolk sac absorption were
also significantly different among the female contributions, but
showed no paternal influence (Table 3). Although mean survival
to the eyed egg stage in this study was very low and only ranged
TABLE 1. Mean postspawning length (mm) and weight (g) of male and female fall Chinook salmon broodfish from Lake Oahe, South Dakota.
Mean SE by group
Gender and size
Female
Length (mm)
Weight (g)
Male
Length (mm)
Weight (g)
a
Overall
721 3
2,965 129
713 18
3,229 340
696 13
3,111 146
678 16
2,803 218
702 8
3,027 108
731 28
3,880 469
755 11
4,166 272
725 23
3,795 361
713 13
731 10
3,946 202
Weights not obtained due to mechanical error.
445
MALE EFFECTS ON PROGENY SURVIVAL
TABLE 2. Percent survival to the eyed egg stage of development of progeny from 16 discrete pairs of autonomous male and female fall Chinook salmon
broodfish in each of four different groups and overall means for each male and female.
Female
Male
1
2
3
4
Mean SE
65.5
81.9
77.4
78.7
75.9 0.8
55.8
52.1
49.6
51.1
52.2 0.4
5
6
7
8
Mean SE
5
15.1
16.2
11.4
7.4
12.5 1.1
6
33.3
14.9
42.7
35.2
31.5 2.1
9
10
11
12
Mean SE
9
0.5
0.9
5.0
3.6
2.5 1.4
10
10.5
7.6
10.5
12.5
10.3 0.6
13
14
15
16
Mean SE
13
4.4
6.6
4.6
11.5
6.8 1.3
14
59.3
46.9
47.3
42.4
49.0 1.0
19.7
10.9
4.1
8.7
10.9 2.0
31.9
20.3
20.3
27.8
25.1 1.2
7
49.7
39.9
43.2
41.0
43.5 0.7
Mean SE
Group 1
43.2
41.3
37.8
41.6
3.2
3.3
5.3
4.7
8
52.4
50.4
29.8
44.3
44.2 1.5
37.6
30.3
31.8
32.0
2.8
3.2
2.7
3.0
11
0.7
8.0
9.7
11.5
7.5 1.7
12
6.8
3.4
2.2
2.1
3.6 1.2
4.6
5.0
6.8
7.4
2.3
1.5
1.5
2.0
15
3.2
3.4
7.6
3.7
4.5 1.0
16
0.2
0.6
0.2
1.0
0.5 0.6
16.8
14.4
21.8
19.0
6.9
5.8
5.6
5.0
Group 2
Group 3
Group 4
from 4.6% to 43.2%, these values are within the normal range
for this salmon population (Barnes et al. 1997).
Egg survival is a function of both fertilization and embryo
development. This study did not differentiate between these two
components but rather looked at their combined effect on survival to the eyed stage of development. It is possible that paternal
effects may occur at either fertilization or during development,
and that these effects may be masked by just focusing on overall
egg survival.
The results of this study indicate minimal paternal effects on
progeny survival. The differences in reproductive success could
be attributed almost entirely to the spawning female. Similarly
Nagler et al. (2000) found that female rainbow trout had a significant effect on embryo survival and that male contributions
to embryo survival were negligible. Both Smith and Fretwel
(1974) and Einum and Fleming (2000) found that increases in
embryo survival and growth were due to maternal fitness. Maternal effects are obvious with respect to egg size, which can be an
indicator of maternal fitness and a predictor of embryo survivability (Einum et al. 2002; Mousseau and Fox, 1998). However,
previous studies have stated embryo survival and development
can be independently related to parental donations whether paternal (Aas et al. 1991) or maternal (Springate et al. 1984).
The use of pooled milt was historically used in Pacific salmon
hatcheries (Billard et al. 1996) and is the established protocol
for salmon spawning in South Dakota (Barnes et al. 2003a).
Withler and Beacham (1994) suggest using pooled milt to maximize spawning efficiency and potential fertilization success.
Spawning time is particularly important in large production
hatcheries or remote facilities where personnel travel times must
be considered. However, using pooled milt may reduce the effective number of breeders and therefore reduce genetic diversity and effective population sizes (Babiak 1998; Withler 1988;
Campton 2004; Wedekind et al. (2007)).
Although not directly addressed in this study, using pooled
milt may be influencing the reproductive success of Lake Oahe
446
WIPF AND BARNES
TABLE 3. Lengths (mm) and weights (g) of fry progeny from 16 discrete pairs of autonomous male and female fall Chinook salmon broodfish in each of four
different groups and overall means for each male and female. Dashes indicate unsuccessful pairing.
Female
Male
1
2
3
4
Mean SE
33.3
33.8
33.7
30.3
32.8 0.8
5
6
7
8
Mean SE
5
33.2
33.1
34.1
32.7
33.3 0.3
9
10
11
12
Mean SE
13
14
15
16
Mean SE
13
32.8
33.3
33.6
32.6
33.1 0.2
1
2
3
4
Mean SE
5
6
7
8
Mean SE
Lengths (mm) for group 1

33.0
31.8
32.8
31.9
32.2
30.9
32.9
32.4
32.7 0.2
31.7 0.3
6
7
34.0
31.5
34.9
32.0
34.4
33.6
35.5
32.9
34.7 0.3
32.5 0.5
10
11
33.5
34.1
33.7
33.9
33.8 0.1
Mean SE
32.3
32.4
32.3
31.8
0.5
0.6
0.6
0.6
8
31.5
32.0
32.3
32.4
32.0 0.2
32.6
33.0
33.6
33.4
0.6
0.7
0.5
0.7
12
33.2
31.3
32.8
32.7
32.5 0.4
33.3
32.7
33.2
33.3
0.2
1.4
0.4
0.6
31.1
30.9
32.4
31.7
31.5 0.3
16
33.3
33.1
33.4
32.8
0.5
0.4
0.1
0.5
1
0.29
0.30
0.28
0.28
0.29 0.01

14
15
32.7
34.3
32.3
33.7
33.2
33.5
32.1
33.6
32.6 0.2
33.8 0.2
Weights (g) for group 1
2
3
0.25
0.25
0.25
0.24
0.25
0.26
0.25
0.24
0.25 0.01
0.25 0.01
4
0.25
0.26
0.26
0.26
0.26 0.01
0.26
0.26
0.26
0.26
0.01
0.01
0.01
0.01
5
0.29
0.29
0.29
0.28
0.29 0.01

6
7
0.27
0.25
0.29
0.25
0.28
0.26
0.29
0.26
0.29 0.01
0.26 0.01
8
0.24
0.24
0.24
0.25
0.24 0.01
0.26
0.27
0.27
0.27
0.01
0.01
0.01
0.01
447
MALE EFFECTS ON PROGENY SURVIVAL

TABLE 3.
Continued.
9
9
10
11
12
Mean SE
13
14
15
16
Mean SE
13
0.25
0.25
0.25
0.25
0.25 0.01

10
11
0.26
0.26
0.27
0.26
0.26 0.01

14
15
0.23
0.29
0.22
0.29
0.23
0.28
0.22
0.29
0.22 0.01
0.29 0.01
salmon population during hatchery rearing by decreasing the

genetic health of the spawning population. The relatively small
size and genetic isolation (Barnes et al. 2000) of this unique
population, coupled with the use of pooled milt, may lead to
potential inbreeding, which has been linked to poor embryo
survival (Klug et al. 2007). The genetic status of Lake Oahe
salmon is unknown, and there is no genetic spawning protocol
for this population. Because there was no significant paternal
effect on egg survival during pairwise spawning, this may be an
effective technique to negate possible homogeneity, potentially
improve genetic diversity, and possibly increase embryo survival
in Lake Oahe fall Chinook salmon, unless there are overriding
spawning efficiency and timing concerns.
REFERENCES
Aas, G. H., T. Refstie, and B. Gjerde. 1991. Evaluation of milt quality of Atlantic
salmon. Aquaculture 95:125132.
Alcock, J. 2005. Animal behaviour: an evolutionary approach, 8th edition.
Sinauer, Sunderland, Massachusetts.
Anderson, M. 1994. Sexual selection. Princeton University Press, Princeton,
New Jersey.
Babiak, I., J. Glogowski, M. Luczynski, K. Goryczko, S. Dobesz, and H.
Kuzminski. 1998. The effect of individual male potency on fertilization ability of fresh and cryopreserved milt of rainbow trout, Oncorhynchus mykiss
(Walbaum). Aquaculture Research 29:337340.
Barnes, M. E., R. J. Cordes, and W. A. Sayler. 1997. Use of formalin during
incubation of eyed eggs of inland fall Chinook salmon. Progressive FishCulturist 59:303306.
Barnes, M. E., R. P. Hanten, R. J. Cordes, W. A. Sayler, and J. Carreiro. 2000.
Reproductive performance of inland fall Chinook salmon. North American
Journal of Aquaculture 62:203211.
Barnes, M. E., R. P. Hanten, J. J. P. Lott, and M. Gabel. 2001. Environmental
influences on landlocked fall Chinook salmon reproductive characteristics.
North American Journal of Aquaculture 63:5865.
Barnes, M. E., J. P. Lott, W. A. Sayler, and R. J., Cordes. 1999b. Practical
observations on the use of eggs from electroshocked females during spawning of inland fall Chinook salmon. North American Journal of Aquaculture
61:162166.
12
0.24
0.25
0.25
0.25
0.25 0.01
16
0.25
0.25
0.25
0.25
0.01
0.01
0.01
0.01
0.26
0.25
0.25
0.26
0.02
0.02
0.02
0.02
Barnes, M. E., W. A. Sayler, and R. J. Cordes, 1999a. Transportation influences

on inland fall Chinook salmon egg survival. North American Journal of
Aquaculture 61:2733.
Barnes, M. E., W. Sayler, J. C. Cordes, and R. P. Hanten. 2003a. Potential
indicators of egg viability in landlocked fall Chinook salmon spawn with or
without the presence of overripe eggs. North American Journal of Aquaculture
65:4955.
Barnes, M. E., M. H. Zehfus, J. A. Schumacher, K. S. Stock, F. Farrokhi, and
R. L. Nutter. 2003b. Thiamine influences on landlocked fall Chinook salmon
reproductive characteristics. Prairie Naturalist 35:113116.
Billard, R., J. Jorgen, and O. T. Jensen. 1996. Gamete removal, fertilization
and incubation. Pages 291364 in W. Pennel and B. A. Batron, editors.
Developments in aquaculture and fisheies science volume 29: principles of
salmonid culture. Elsevier, Amsterdam, The Netherlands.
Campton, D. E. 2004. Sperm competition in salmon hatcheries: the need to
institutionalize genetically benign spawning protocols. Transactions of the
American Fisheries Society 133:12771289.
Einum, S., and I. A. Fleming. 2000. Highly fecund mothers sacrifice offspring
survival to maximize fitness. Nature 405:565567.
Einum, S., A. P. Hendry, and I. A. Fleming. 2002. Egg-size evolution in aquatic
environments: does oxygen availability constrain size? Proceedings of the
Royal Society London Biological Sciences 269:23252330.
Fleming, I. A., and M. R. Gross. 1994. Breeding competition in a Pacific salmon
(coho: Oncorhynchus kisutch): measures of natural and sexual selection.
Evolution 48:637657.
Foote, C. J., C. S. Brown, and C. C. Wood. 1997. Spawning success of males
using alternative mating tactics in sockeye salmon, Oncorhynchus nerka.
Canadian Journal of Fisheries and Aquatic Sciences 54:17851795.
Gharrett, A. J., and S. M. Shirley. 1985. A genetic examination of spawning
methodology in a salmon hatchery. Aquaculture 47:245256.
Gile, R. S., and M. M. Ferguson. 1995. Factors affecting male potency in pooled
gamete crosses of rainbow trout, Oncorhynchus mykiss. Environmental Biology of Fishes 42:267275.
Klug, W. S., M. R. Cummings, and C. A. Spencer. 2007. Essentials of genetics,
6th edition. Pearson Prentice Hall, Upper Saddle River, New Jersey.
Mousseau, T. A., and C. W. Fox. 1998. Maternal effects as adaptations. Oxford
University Press. New York.
Nagler, J. J., J. E. Parsons, and J. G. Cloud. 2000. Single pair mating indicates
maternal effects on embryo survival in rainbow trout, Oncorhynchus mykiss.
Ott, L. 1984. An introduction to statistical methods and data analysis. PWS
Publishers, Boston.
448
WIPF AND BARNES
Piper, R. G., I. B. McElwain, L. E. Orme, J. P. McCaren, L. G. Fowler, and J. R.

Leonard. 1982. Fish hatchery management. U.S. Fish and Wildlife Service,
Washington, D.C.
Simmons, L. W., and J. S. Kotiaho. 2002. Evolution of ejaculates: patterns
of phenotypic and genotypic variation and condition dependence in sperm
competition traits. Evolution 56:16221631.
Smith, C. C., and S. D. Fretwell. 1974. The optimal balance between size and
number of offspring. American Naturalist 108:499506.
Smoker, W. W., A. J. Gharrett, M. S. Stekoll, and S. G. Taylor. 2000. Genetic
variation of fecundity and egg size in anadromous pink salmon Oncorhynchus
gorbuscha Walbaum. Alaska Fishery Research Bulletin 7:4450.
Smoker, W. W., I. A. Wang, A. J. Gharrett, and J. J. Hard. 2004. Embryo survival
and smolt to adult survival in second-generation outbred coho salmon. Journal
of Fish Biology 65:254262.
Springate, J. R. C., N. R. Bromage, J. A. K. Elliot, and D. L. Hudson. 1984. The

timing of ovulation and stripping and their effects on the rates of fertilization
and survival to eying, hatch and swim-up in the rainbow trout (Salmo gairdneri
R.) Aquaculture 43:313322.
Wedekind, C., G. Rudolfsen, A. Jacob, D. Urbach, and R. Muller. 2007. The
genetic consequences of hatchery-induced sperm competition in a salmonid.
Biological Conservation 137:180188.
Withler, R. E. 1988. Genetic consequences of fertilizing Chinook salmon
(Oncorhynchus tshawytscha) eggs with pooled milt. Aquaculture 68:
1525.
Withler, R. E., and T. D. Beacham. 1994. Genetic consequences of the simultaneous or sequential addition of semen from multiple males during hatchery spawning of Chinook salmon (Oncorhynchus tshawytscha). Aquaculture
126:1123.


Koi Goldfish Hybrid Females Produce Triploid Progeny

when Backcrossed to Koi Males
a
Boris Gomelsky , Kyle J. Schneider & Debbie A. Plouffe
Aquaculture Research Center, Kentucky State University, 103 Athletic Road, Frankfort,
Kentucky, 40601, USA
b
AquaBounty Canada, Inc., 718 Bay Fortune, Souris, Prince Edward Island, C0A 2B0, Canada
Version of record first published: 04 Sep 2012.
To cite this article: Boris Gomelsky, Kyle J. Schneider & Debbie A. Plouffe (2012): Koi Goldfish Hybrid Females Produce
Triploid Progeny when Backcrossed to Koi Males, North American Journal of Aquaculture, 74:4, 449-452


DOI: 10.1080/15222055.2012.676014
COMMUNICATION
Koi Goldfish Hybrid Females Produce Triploid Progeny

when Backcrossed to Koi Males
Boris Gomelsky* and Kyle J. Schneider
Aquaculture Research Center, Kentucky State University, 103 Athletic Road, Frankfort, Kentucky 40601, USA
Debbie A. Plouffe
AquaBounty Canada, Inc., 718 Bay Fortune, Souris, Prince Edward Island C0A 2B0, Canada
Abstract
Hybrids of koi (an ornamental variant of the common carp
Cyprinus carpio) and goldfish Carassius auratus auratus were produced by artificial spawning. All 3-year-old F1 hybrid males examined were sterile, whereas some F1 hybrid females were fertile and
produced eggs after hormonal injection. Backcross progeny were
obtained by using intact koi sperm to inseminate eggs from F1 hybrid females; gynogenetic progeny were obtained by inseminating
eggs from F1 hybrid females with koi sperm that was genetically
inactivated by ultraviolet irradiation. Flow cytometric analysis of
DNA content indicated that the backcross progeny were triploid,
while the gynogenetic progeny, pure koi, pure goldfish, and F1 hybrids were all diploid. The triploidy of backcross progeny obtained
without application of any treatment to the eggs demonstrates that
the koi goldfish hybrid females produce diploid eggs.
The common carp Cyprinus carpio and goldfish Carassius

auratus auratus are nonnative cyprinid species that naturally reproduce in North America (Panek 1987; Schofield et al. 2005).
Ornamental forms of common carp (i.e., koi) and goldfish are
popular decorative fish in many countries throughout the world,
including the United States. Goldfish are also raised in the
United States for use as bait fish and as forage in fish hatcheries
(Schofield et al. 2005).
Common carp naturally hybridize with goldfish in their native range and in many areas where both species have been
introduced (Trautman 1957; Bardach et al. 1972; Taylor and
Mahon 1977; Pullan and Smith 1987; Schofield et al. 2005).
Cases of hybridization between koi and goldfish have also
been described in the aquarium fish and koi hobbyist literature
(Hemdal 2003; Muha 2007). Hybrids of common carp or koi
and goldfish have been produced artificially for different types
of studies (e.g., Suzuki 1962; Hedrick et al. 2006). The morphometric and meristic characteristics of common carp (or koi)
goldfish hybrids have been widely reported (Taylor and Mahon

1977; Pulan and Smith 1987; Schofield et al. 2005). In contrast
to many interspecies fish hybrids, the hybrids of common
carp and goldfish are not sterile, and the backcrossing of hybrids with parental species has been reported (Trautman 1957;
Bardach et al. 1972).
Previous studies on hybridization of common carp with two
other subspecies of C. auratus (Japanese crucian carp C. auratus
cuvieri and silver crucian carp C. auratus gibelio) revealed that
F1 hybrid females produced triploid progeny when backcrossed
to males of parental species (Ojima et al. 1975; Cherfas et al.
1994). Triploidy of backcross hybrids resulted from the diploidy
of eggs produced by hybrid females. The present study was
performed to investigate whether koi goldfish hybrid females
have the same feature. For this purpose, the ploidy of backcross
progeny obtained from hybrid females was determined.
METHODS
Fish spawning and rearing were conducted at the Aquaculture
Research Center, Kentucky State University, Frankfort. Koi
goldfish hybrids were produced by artificial spawning in spring
2008. To induce final oocyte maturation and ovulation in females
and spermiation in males, broodfish were given intramuscular
injections of carp pituitary extract (Sigma Chemical, St. Louis,
Missouri) at a concentration of 3 mg/kg of body weight. Males
received a single injection approximately 16 h before stripping
of sperm. Females received two injections (10% and 90% of the
total dose) 12 h apart. After injections, broodfish were kept in
tanks at a water temperature of 21.5 C; ovulation was observed
1112 h after the resolving injection. Eggs and sperm were collected from broodfish by hand stripping. A mixture of eggs from
six koi females of different colors (whitered, whiteyellow, and
*Corresponding author: boris.gomelsky@kysu.edu

Received November 11, 2011; accepted February 18, 2012
449
450
GOMELSKY ET AL.
solid white) was fertilized with a mixture of sperm taken from

two solid-red goldfish males. Eggs were artificially inseminated
in plastic bowls according to standard techniques for common
carp breeding (Horvath et al. 2002) and were treated with a solution of water : pasteurized cows milk (volumetric ratio = 8:1)
to remove adhesiveness. Embryos were incubated in McDonald
hatching jars.
The resulting F1 hybrids were raised in 20-m3 outdoor tanks
during their first summer and then were transferred to earthen
ponds, where they were reared for an additional 2.5 years. The
functional fertility of 3-year-old hybrids was investigated in
spring 2011. For this purpose, 10 selected hybrid females and
20 hybrid males were injected with carp pituitary extract by
using the dosages and methodology described above for koi and
goldfish breeders.
After hormonal injection, hybrid males did not release any
sperm. To determine whether the testes of hybrid males contained any viable spermatozoa, we sacrificed the fish, removed
their testes, and thoroughly dissected and washed the testes with
a saline solution (0.85% NaCl). The resulting suspension was
used for insemination of koi eggs.
Five koi goldfish hybrid females produced ovulated eggs
after hormonal injection and were used in crosses. Two separate groups of backcross progeny were obtained by crossing
F1 hybrid females with koi males. Backcross progeny group
1 was obtained by using sperm from a single koi male to inseminate a mixture of eggs collected from four hybrid females.
Backcross progeny group 2 was produced by crossing a single hybrid female with a single koi male. Gynogenetic progeny
were also obtained by using genetically inactivated (by ultraviolet [UV] irradiation) koi sperm to inseminate eggs from the
same F1 hybrid female that was used to produce backcross
progeny group 2. Sperm was irradiated with a UV crosslinker
(FB-UVXL-1000; Fisher Scientific) at 4,000 J/m2. This dosage
of UV irradiation was chosen based on previously performed
experiments involving the induction of gynogenesis in koi
(Alsaqufi 2011). Before irradiation, the sperm was diluted (1:9)
in a saline solution (0.85% NaCl). The backcross progeny
groups and the gynogenetic progeny were obtained without ap-
plication of any treatment to the eggs. After transition to active

feeding, the larvae were stocked in 20-m3 outdoor tanks for
rearing.
Ploidy of 4-month-old backcross progeny (mean total
length SD = 13.8 0.85 cm) and gynogenetic progeny
(12.8 0.75 cm) was determined by flow cytometric analysis
of DNA content. The analyses were performed at AquaBounty
Canada, Inc., Prince Edward Island, by using a Becton Dickinson (BD) FACSCalibur flow cytometer. The ploidy of some
koi, goldfish, and F1 hybrids was also analyzed for comparison.
Instrument quality control for DNA quantitation was performed
using CellQuest Pro software and DNA QC particles (BD, catalog number 349523) to assess resolution and linearity. Blood
samples were drawn from each fish by caudal venipuncture and
were collected in 3.0-mL Vacutainer tubes containing lithium
heparin (BD, catalog number 366667). For each sample, two
drops of heparinized blood from a syringe were collected in
500 L of sheath fluid (BioSure, catalog number 1019). An
80-L quantity of the bloodsheath fluid mixture was stained
in 500 L of propidium iodide solution containing a detergent
for lysing (BioSure, catalog number 1021) along with 40 L of
chicken red blood cells (BioSure, catalog number 1005) as an
internal staining control. Samples were incubated in propidium
iodide solution in the dark for 10 min prior to analysis. For each
sample analyzed, 10,000 events were recorded and the relative
DNA content was determined as the ratio of sample fluorescence
peak intensity to internal standard (i.e., chicken red blood cells)
fluorescence peak intensity.
RESULTS
The koi goldfish hybrids exhibited dark coloration that
is typical of wild-type common carp and goldfish. After eggs
from female koi were mixed with the suspension of dissected
testes from F1 hybrid males, examination of eggs under a
dissecting microscope indicated that none of the eggs was
fertilized.
The results of ploidy analysis for backcross progeny, gynogenetic progeny, the parental species, and the F1 hybrids are
TABLE 1. Ploidy of koi, goldfish, F1 hybrids, and backcross and gynogenetic progeny groups as determined by flow cytometric analysis. The relative DNA
content was calculated as the ratio of sample fluorescence peak intensity to internal standard fluorescence peak intensity.
Relative DNA content

Group
Koi
Goldfish
F1 ( koi goldfish)
Backcross progeny group 1 ( F1 koi)
Backcross progeny group 2 ( F1 koi)
Gynogenetic progeny ( F1 koi [inactivated sperm])
Number of fish analyzed
Mean (SD)
Range
Fish ploidy
4
2
10
15
12
12
2.56 (0.04)
2.52
2.56 (0.05)
3.75 (0.08)
3.84 (0.07)
2.51 (0.04)
2.532.62
2.512.52
2.492.60
3.613.84
3.733.96
2.432.59
2n
2n
2n
3n
3n
2n
COMMUNICATION
presented in Table 1. The mean relative DNA content was similar

for koi, goldfish, F1 hybrids, and gynogenetic progeny and varied from 2.51 to 2.56. The mean relative DNA content was 3.75
in backcross progeny group 1 and 3.84 in backcross progeny
group 2. The ratio of DNA content in backcross progeny to the
DNA content in the other groups was approximately 1.5, indicating that the backcross progeny were triploid (3n) fish, whereas
the gynogenetic progeny, parental species, and F1 hybrids were
all diploid (2n).
DISCUSSION
The results of the present study reveal the functional genetic
sterility of F1 males produced by the hybridization of koi and
goldfish. These males developed testes, but their gonads did
not contain spermatozoa with the capability of fertilizing eggs.
Sterility of goldfish common carp hybrid males was also
reported by Yamaha et al. (2003).
Flow cytometric analysis indicated that the DNA content in
koi and goldfish was similar. These results concur with data
reported by Ohno et al. (1967), who indicated that common
carp and goldfish had comparable amounts of DNA comprising 5052% of the typical DNA content in placental mammals. Similar mean values of relative DNA content were observed in F1 hybrids, whereas backcross progeny appeared to be
triploid.
The triploidy of backcross progeny obtained from hybrid females (without application of any treatment to eggs) indicates
that these females produced diploid eggs. Diploidy of eggs from
hybrid females resulted in diploid gynogenetic progeny, which
were produced by use of genetically inactivated spermatozoa
chromosomes but without application of any treatment to the
eggs. Hybrid females obtained by crossing common carp males
with females of two other C. auratus subspecies (Japanese crucian carp and silver crucian carp) have previously been shown
to produce diploid eggs (Ojima et al. 1975; Cherfas et al. 1994).
The results of the present study demonstrate that the same phenomenon is observed in hybrids obtained from two popular
ornamental fishes, the koi and goldfish. Specifically, the hybrid
females producing diploid eggs were obtained by crossing koi
females with goldfish males.
The ability of interspecies F1 hybrids to produce diploid
eggs has been described for several other fish taxa in addition
to Carassius and Cyprinus; these include hybrids of the brown
trout Salmo trutta and Atlantic salmon S. salar (Johnson and
Wright 1986; Galbreath and Thorgaard 1995) and hybrids of the
pumpkinseed Lepomis gibbosus and green sunfish L. cyanellus (Dawley et al. 1985; Dawley 1987). Cherfas et al. (1994)
and Shimizu et al. (2000) showed that generation of unreduced
diploid eggs by hybrid females results from the occurrence of
premeiotic endomitosis (i.e., doubling of chromosomes without
cytokinesis) in early oogenesis; the resulting tetraploid oocytes
undergo two normal, consecutive meiotic divisions.
451
The koi goldfish hybrids had a color pattern that is typical

of wild-type common carp and goldfish, which indicates that the
alleles causing melanin depigmentation in parental forms are not
expressed in the hybrids. The analysis of color inheritance in
hybrids will be presented in a separate publication.
ACKNOWLEDGMENTS
Support for this study was provided by Kentuckys Regional
University Trust Fund to the Aquaculture Program as Kentucky
State Universitys Program of Distinction.
REFERENCES
Alsaqufi, A. S. 2011. Application of microsatellite DNA markers in studies
on induced gynogenesis in ornamental (koi) carp. Masters thesis. Kentucky
State University, Frankfort.
Bardach, J. E., J. H. Ryther, and W. O. McLarney. 1972. Aquaculture: the
farming and husbandry of freshwater and marine organisms. Wiley, New
York.
Cherfas, N. B., B. I. Gomelsky, O. V. Emelyanova, and A. V. Recoubratsky.
1994. Induced diploid gynogenesis and polyploidy in crucian carp, Carassius auratus gibelio (Bloch), common carp, Cyprinus carpio L., hybrids.
Aquaculture and Fisheries Management 25:943954.
Dawley, R. M. 1987. Hybridization and polyploidy in a community of three
sunfish species (Pisces: Centrarchidae). Copeia 1987:326335.
Dawley, R. M., J. M. Graham, and R. J. Schultz. 1985. Triploid progeny of
pumpkinseed green sunfish hybrids. Journal of Heredity 76:251257.
Galbreath, P. F., and G. H. Thorgaard. 1995. Sexual maturation and fertility
of diploid and triploid Atlantic salmon brown trout hybrids. Aquaculture
137:299311.
Hedrick, R. P., T. B. Waltzek, and T. S. McDowell. 2006. Susceptibility of
koi carp, common carp, goldfish and goldfish common carp hybrids to
cyprinid herpesvirus-2 and herpesvirus-3. Journal of Aquatic Animal Health
18:2634.
Hemdal, J. F. 2003. Aquarium fish breeding. Barrons, Hauppauge, New York.
Horvath, L., G. Tamas, and C. Seagrave. 2002. Carp and pond fish culture, 2nd
edition. Blackwell Scientific Publications, Oxford, UK.
Johnson, K. R., and J. E. Wright. 1986. Female brown trout Atlantic salmon
produce gynogens and triploids when backcrossed to male Atlantic salmon.
Muha, L. 2007. Spring ponds part 2: the good, the bad and the ugly. Tropical
Fish Hobbyist (April):6467.
Ohno, S., J. Muramoto, L. Christian, and N. B. Atkin. 1967. Diploid-tetraploid
relationship among old world members of the fish family Cyprinidae. Chromosoma 23:19.
Ojima, Y., M. Hayashi, and K. Ueno. 1975. Triploidy appeared in the backcross offspring from funa-carp crossing. Proceedings of Japanese Academy
of Science 51:702706.
Panek, F. M. 1987. Biology and ecology of carp. Pages 115 in E. L. Cooper,
editor. Carp in North America. American Fisheries Society, Bethesda, Maryland.
Pullan, S., and P. J. Smith. 1987. Identification of hybrids between koi (Cyprinus
carpio) and goldfish (Carassius auratus). New Zealand Journal of Marine and
Freshwater Research 21:4146.
Schofield, P. J., J. D. Williams, L. G. Nico, and M. R. Thomas. 2005. Foreign nonindigenous carps and minnows (Cyprinidae) in the United Statesa
guide to their identification, distribution, and biology. U.S. Geological Survey,
Scientific Investigations Report 20055041.
Shimizu, Y., N. Shibata, M. Sakaizumi, and M. Yamashita. 2000. Production of diploid eggs through premeiotic endomitosis in the hybrid medaka
452
GOMELSKY ET AL.
between Oryzias latipes and O. curvinotus. Zoological Science 17:951

958.
Suzuki, R. 1962. On the behavior of carp-goldfish hybrids. Japanese Journal of
Ichthyology 10:1315.
Taylor, J., and R. Mahon. 1977. Hybridization of Cyprinus carpio and Carassius
auratus, the first two exotic species in the lower Laurentian Great Lakes.
Environmental Biology of Fishes 1:205208.
Trautman, M. B. 1957. Fishes of Ohio with illustrated keys. Ohio State University Press, Columbus.
Yamaha, E., M. Murakami, K. Hada, S. Otani, T. Fujimoto, M. Tanaka,
S. Sakao, S. Sato, and K. Arai. 2003. Recovery of fertility in male
hybrids of a cross between goldfish and common carp by transplantation of PGC (primordial germ cell)-containing graft. Genetica 119:121
131.


Demonstration of Blue Crab Culture in Inland LowSalinity Waters of West Alabama

a
Luke A. Roy , Gregory N. Whitis & William C. Walton
Department of Fisheries and Allied Aquacultures, Auburn University, 203 Swingle Hall,
Auburn, Alabama, 368495419, USA
To cite this article: Luke A. Roy, Gregory N. Whitis & William C. Walton (2012): Demonstration of Blue Crab Culture in Inland
Low-Salinity Waters of West Alabama, North American Journal of Aquaculture, 74:4, 453-456


DOI: 10.1080/15222055.2012.676006
COMMUNICATION
Demonstration of Blue Crab Culture in Inland Low-Salinity

Waters of West Alabama
Luke A. Roy,* Gregory N. Whitis, and William C. Walton
Department of Fisheries and Allied Aquacultures, Auburn University, 203 Swingle Hall, Auburn,
Alabama 368495419, USA
Abstract
The decline in natural stocks of blue crabs Callinectes sapidus
in U.S. coastal waters has created more interest in the aquaculture
of this species. The 2010 Deepwater Horizon oil spill threatened the
health of blue crab populations in the Gulf of Mexico and spurred
interest in alternative sources of blue crab harvest. Inland lowsalinity waters (ILSW) of west Alabama have a proven potential
for the culture of euryhaline crustaceans. This study sought to
evaluate the potential for blue crab culture in ILSW for commercial
aquaculture and as a potential nursery for restoration efforts. A
21-d bioassay was conducted using four different pond waters of
differing ionic composition and salinity (2.05.7). No significant
differences in survival, weight gain (%), and final weight were
observed among different pond waters. An on-farm demonstration
was conducted in a 1.5-acre low-salinity pond (5.0) at a shrimp
farm in west Alabama. Crabs grew from 2.0 to 202.0 g in less than
120 d when offered a varied diet of commercial shrimp feed, catfish
offal, and fresh whole catfish. Results from these trials suggest that
there is excellent biological potential for culture of blue crabs in
ILSW of west Alabama.
The fishery for blue crab Callinectes sapidus is the most

valuable fishery in several states, including North Carolina,
Maryland, Virginia, and Mississippi, having an estimated value
approaching US$200 million/year (Eggleston et al. 2009). The
decline in capture harvests of blue crab has increased interest in the aquaculture of this species, particularly in the four
states previously mentioned. Along the Gulf Coast, the 2010
BP Deepwater Horizon Oil Spill posed an additional threat to
natural populations of blue crab and other commercial species,
and spurred further interest in culture of this species within the
region.
The Alabama inland shrimp industry is concentrated about
150 mi north of Mobile and the Gulf of Mexico, where inland artesian groundwater salinity ranges from 2 to 10
(Roy et al. 2010). Inland low-salinity water (ILSW) suitable for
mariculture has been identified in several counties in west Alabama (Boyd et al. 2009). Inland shrimp farmers have been
growing the Pacific white shrimp Litopenaeus vannamei in
west Alabama since 1999 and are currently producing 250,000
lb of shrimp annually, which has a farm gate value of over
$1 million (D. Teichert-Coddington, Alabama Inland Shrimp
Producers Association, personal communication). Farmers in
west Alabama have also successfully cultured a number of
other marine species in ILSW, including Gulf killifish Fundulus grandis (Phelps et al. 2010; S. McNulty, Aquatic Innovations LLC, personal communication) and pinfish Lagodon
rhomboides (McNulty, personal communication). Elsewhere in
the United States, ILSW has also been used to culture red drum
Sciaenops ocellatus (Miranda and Sonski 1987). Thus, the viability of ILSW has been well established for rearing euryhaline
marine species that are capable of effective osmotic and ionic
regulation in low-salinity environments.
In order to effectively grow marine species in ILSW of west
Alabama, farmers must supplement potassium and magnesium
fertilizers to their pond water. The ILSW in west Alabama
is naturally deficient in these two ions (Saoud et al. 2003).
Deficiencies in potassium and magnesium, as well as high
sodium : potassium ratios, have resulted in reduced survival and
osmoregulatory capacity of marine species cultured in ILSW
(Fielder et al. 2001; Saoud et al. 2003; Roy and Davis 2010).
West Alabama shrimp farmers typically counteract low levels
of potassium and magnesium with fertilizers rich in these two
ions, including muriate of potash (agricultural grade potassium
chloride) and K-Mag (potassium magnesium sulfate; Davis
et al. 2004). While blue crabs are excellent osmoregulators
capable of tolerating and thriving in extremely low salinities
(Henry 1988; Henry 2005), the effects of ILSW with altered
ionic levels and ratios differing from dilute seawater has not
yet been examined in regards to blue crab culture.
*Corresponding author: royluke@auburn.edu

Received October 4, 2011; accepted February 14, 2012
453
454
ROY ET AL.
TABLE 1. Pond water ionic composition for 21-d bioassays conducted with blue crabs using four different low-salinity shrimp pond waters.
Ions
Control
TC S3
TC S4
DO 1
DO 7
Crab pond
K
Mga
Na
Caa
Na:K
Salinity ()
Total hardnessa
53.99
646.62
1,629.15
121.40
30.18
4.60
768.02
69.52
52.03
778.98
47.07
11.20
5.80
99.10
30.94
47.07
819.47
101.58
26.49
5.40
148.65
73.03
201.67
2,459.07
205.63
33.67
2.10
407.30
16.40
282.43
2,023.87
141.22
123.39
2.10
423.65
35.45
180.86
2,003.63
178.38
56.53
5.00
359.23
Expressed as mg/L calcium carbonate.
The overall goal of this study was to assess the aquaculture

potential of the blue crab as an alternative crop for existing
shrimp producers, and to test the potential of ILSW of west Alabama as a nursery for restoration efforts. To meet this objective,
growth and survival of blue crabs reared in low-salinity pond
waters with a range of salinities and ionic profiles was evaluated
in a 21-d bioassay and commercial pond demonstration study.
METHODS
A 21-d bioassay was conducted in a static aquarium system
located at the Alabama Fish Farming Center (AFFC) in Greensboro, Alabama. Two months prior to starting the bioassay, pond
water was collected from four ponds located at two different lowsalinity shrimp farms (DO and TC) in Greene County, Alabama.
Three of the four ponds (TC S3, TC S4, DO 1) had received
potassium and magnesium fertilizers (muriate of potash and KMag) at rates commonly applied annually to shrimp ponds prior
to the production season to raise aqueous potassium and magnesium to levels optimal for shrimp culture (Roy et al. 2010).
One of the ponds (DO 7) had not yet been fertilized and had
levels of potassium and magnesium that would be considered
suboptimal for shrimp culture (Roy et al. 2010). Eighty gallons
of water were collected from each pond using a bilge pump
and transported to the AFFC in hauling tanks. Water from each
pond was transferred to four randomly selected replicate 20-gal
aquariums. Additionally, to serve as a control, four replicate
aquariums were filled with dechlorinated tap water. The salinity
of the control treatment was raised to 4.6 using reconstituted
seawater (Crystal Sea Salt, Baltimore, Maryland). The aquarium system consisted of 20 aquariums, each equipped with an
air stone and supplied with air from a regenerative blower. Water
samples were taken at the beginning of the experiment for ion
analysis (Table 1). Sodium and potassium were measured using
a Cole Parmer digital flame photometer (Vernon Hills, Illinois;
model 265500) according to Roy et al. (2007). Magnesium and
calcium were measured according to Eaton et al. (2005).
Blue crabs were obtained from the Gulf Coast Research
Laboratory crab hatchery in Ocean Springs, Mississippi. Crabs
were transported from Mississippi to the AFFC in coolers and
stocked into the aquaria system, one crab per aquarium. The
initial weight of the crabs was 2.04 0.53 g (mean SD).
Throughout the experiment, the crabs were fed daily in the mornings and evenings. The diet consisted of fillets of fresh channel
catfish Ictalurus punctatus or shrimp. Feed was placed in the
aquaria, and crabs were allowed to feed for 60 min. Leftover feed
was removed from the tank to limit ammonia and nitrite buildup
in the tanks. Aquaria were checked for molts 23 times/d, and
throughout the 21-d trial each crab molted twice with the exception of one crab that only molted once. Once detected, all molts
were removed immediately from the experimental tanks. Dissolved oxygen, pH, and temperature were measured daily, while
salinity was measured weekly. Total ammonia nitrogen was analyzed weekly according to Nesslers method (APHA et al. 1989),
whereas nitrite was measured weekly according to Parsons et al.
(1985). At the end of the bioassay, crabs were harvested and individually weighed to determine growth and survival.
Statistical analyses were performed using SAS (version 9.2;
SAS Institute, Cary, North Carolina). All data were analyzed
using one-way ANOVA and StudentNewmanKeuls multiplerange test to determine if significant differences (P < 0.05)
existed among treatment means (Steel and Torrie 1980).
On July 7, 2010 (the same day that the 21-d bioassay began),
4,187 blue crabs were stocked into a 1.5-acre pond (5.0) at a
low-salinity shrimp farm in Greene County, Alabama. The initial
weight of the crabs at stocking was 2.04 0.53 g (mean
SD). The crabs were transported from Mississippi as previously
described and allowed to acclimate to temperature for 1 h prior
FIGURE 1. Growth of blue crabs in a 1.5-acre low-salinity (5.0) pond

fertilized with potassium and magnesium fertilizers in Greene County, Alabama,
from July 2010 to June 2011. [Figure available in color online.]
455
COMMUNICATION
TABLE 2. Growth performance of blue crabs in static 20-gal aquaria with pond water from four different shrimp ponds for 21 d. Values represent the mean
SD. No significant differences were observed among treatments.
Variable
TC S3
Initial weight (g)

2.09
Survival (%)
100.0
Final weight (g)
12.93
Weight gain (%) 516.64
Growth (g/week)
3.61
TC S4
0.19
1.77
0.0
100.0
2.09
9.85
45.86 461.60
0.64
2.70
DO 1
0.25
2.47
0.0
100.0
0.70
14.30
45.51 490.91
0.17
3.94
DO 7
1.02
2.16
0.0
100
5.78
13.55
67.05 529.42
1.59
3.80
Control
0.16
1.71
0.0
100.0
0.61
9.53
28.95 438.40
0.16
2.60
PSEa
P-value
0.37
0.255 0.2620
0.0
5.06
1.790 0.2518
222.86 54.390 0.7490
1.59
0.526 0.2656
PSE = pooled SE.
to stocking in the pond. The demonstration pond was fertilized

(Table 1) several weeks prior to stocking crabs with muriate of
potash and K-Mag at application rates typically used by west
Alabama inland shrimp farmers (Roy et al. 2010). Crabs were
fed a variety of different feeds during the trial, including floating
catfish feed, sinking shrimp feed, whole shrimp, whole catfish,
and catfish offal from a local fish processing plant. Whole shrimp
and whole catfish were fed following die-offs in the farmers
shrimp and catfish ponds.
Crabs were sampled monthly (collected by trapping) until
October when water temperatures began to drop (Figure 1).
From November to March, no crabs were trapped, presumably
due to colder water temperatures. Although crabs were up to
harvest size by the end of October, the crabs were not harvested
in 2010 due to decreased water temperatures. Crabs were sampled again in April 2011 and then harvested in June and July by
using two crab pots (18 18 18 in). The traps were baited
with frozen shrimp. A catfish seine was utilized on two occasions in unsuccessful attempts to harvest the crabs before the
traps were utilized. After the traps no longer successfully captured crabs, the pond was completely drained and the remaining
crabs were counted.

After 21 d, there were no differences in survival, final weight,
or percent weight gain in crabs reared in the four different lowsalinity pond waters or the control (Table 2). Water quality
parameters remained acceptable throughout the course of the
experiment (Table 3). No mortalities were observed in any of
the treatments. Mean final weight and weight gain (%) ranged
from 9.53 to 14.30 g and 438.40529.42%, respectively. The
lowest mean final weights and weight gains were observed in
the control water and TC S4, although the differences were not
significant. Crabs grew 2.63.93 g/week, growth rates of less
than 3 g/week being observed in the control and TC S4.
The sodium : potassium ratio (Na:K) in seawater is approximately 28:1 (Roy et al. 2010), whereas inland saline waters
of west Alabama are typically low in potassium and have high
Na:K ratios, typically greater than 100:1 (Saoud et al. 2003;
McNevin et al. 2004; Roy et al. 2010). Past experiments with
penaeid shrimp in west Alabama revealed that as Na:K ratios
were reduced closer to 28:1, by means of potassium fertilizers,
survival and growth improved (Roy et al. 2006, 2007, 2010; Roy
and Davis 2010). The Na:K ratio of DO 7 was 123.39:1, which
in our experience with shrimp would result in poor growth and
survival. The Na:K ratio of the other treatments ranged from
33.67:111.20:1, which would be ideal for Pacific white shrimp
culture in west Alabama. However, the blue crabs that were
reared in DO 7 pond water actually had the highest weight gain
and second-highest final individual weight in the 21-d bioassay,
albeit not statistically significant. This suggests that blue crabs
might be more tolerant of high Na:K ratios than Pacific white
shrimp and that west Alabama farmers might be able to reduce
their fertilizer application rates and still successfully grow blue
crabs.
The pond demonstration yielded a total of 1,100 crabs, for
26.3% survival at harvest with an average weight of 212.19
46.51 g (mean SD). The average carapace width at harvest
was 151.7 mm and ranged from 125 to 175 mm. In a pond study
TABLE 3. Water quality parameters for a 21-d bioassay with blue crabs in static 20-gal aquaria with pond water from four different low-salinity shrimp ponds.
Values represent the mean SD.
Pond
Control
DO 1
DO 7
TC S3
TC S4
Dissolved
oxygen (mg/L)
7.34
7.30
7.32
7.35
7.31
0.07
0.04
0.02
0.05
0.05
Temperature
( C)
26.9
26.8
26.9
26.8
26.8
0.12
0.11
0.02
0.09
0.17
7.9
8.2
8.1
8.5
8.6
pH
Salinity ()
4.6
5.8
5.4
2.1
2.1
0.21
0.12
0.10
0.04
0.02
0.18
0.13
0.59
0.06
0.08
Total ammonia
nitrogen (mg/L)
4.13
1.92
2.29
2.04
2.17
1.78
0.35
0.42
0.64
0.24
Nitrite
nitrogen (mg/L)
1.77
0.42
1.60
0.36
0.78
0.98
0.26
1.54
0.19
0.46
456
ROY ET AL.
conducted in North Carolina, Eggleston et al. (2009) reported

survivals of blue crab ranging from 2.9% to 18.8%. Eggleston
et al. (2009) also reported that hatchery-reared crabs reached
market size in 77 d when reared in ponds ranging in salinity from
0.4 to 0.5. Although the crabs in our study were not harvested
before winter as low water temperatures made harvest impractical, the crabs reached market size by the end of 3 months.
Taken together, the 21-d bioassay and pond demonstration
suggest that blue crabs can grow well in ILSW of west Alabama.
Techniques utilized by the inland shrimp industry to raise levels
of potassium and magnesium in pond waters appear to also be
effective for the culture of blue crabs. In addition to aquaculture
potential and given the blue crabs ability to effectively acclimate
to low and high salinities, inland saline waters of west Alabama
might also be valuable as a nursery for blue crab restoration
efforts.
ACKNOWLEDGMENTS
The authors would like to extend their thanks to those who
have taken the time to critically review this manuscript as well
as those who helped in supporting this research. This research
was supported by the AFFC and Odom Farms, Eutaw, Alabama.
Mention of a trademark or proprietary product does not constitute an endorsement of the product by Auburn University and
does not imply its approval to the exclusion of other products
that may also be suitable.
REFERENCES
APHA (American Public Health Association), American Water Works Association, and Water Pollution Control Association. 1989. Standard Methods for
the examination of water and waste water, 17th edition. APHA, Washington,
D.C.
Boyd, C. A., P. L. Chaney, C. E. Boyd, and D. B. Rouse. 2009. Distribution
of ground water suitable for use in saline-water aquaculture in central and
west-central Alabama. Journal of Applied Aquaculture 21:228240.
Davis, D. A., T. M. Samocha, and C. E. Boyd. 2004. Acclimating Pacific
white shrimp, Litopenaeus vannamei, to inland, low salinity waters. Southern
Regional Aquaculture Center, Publication 2601, Stoneville, Mississippi
Eaton A. D., L. S. Clesceri, E. W. Rice, and A. E. Greenberg. 2005. Standard
methods for the examination of water and waste water, 21st edition. American
Public Health Association, Washington, D.C.
Eggleston, D. B., G. Plaia, and H. Daniels. 2009. Blue crab stock enhancement:
further progress in freshwater pond rearing. Final Report to the North Carolina
Blue Crab Fishery Grant Program, 07-STOK-01, Raleigh.
Fielder, D. A., W. J. Bardsley, and G. L. Geoff. 2001. Survival and growth of
Australian snapper, Pagrus auratus, in saline groundwater from inland New
South Wales, Australia. Aquaculture 201:7390.
Henry, R. P. 1988. Subcellular distribution of carbonic anhydrase activity in the
gills of the blue crab, Callinectes sapidus. Journal of Experimental Zoology
245:18.
Henry, R. P. 2005. Critical salinity, sensitivity, and commitment of salinitymediated carbonic anhydrase induction in the gills of two euryhaline species of
decapod crustaceans. Journal of Experimental Zoology 303:4556.
McNevin, A. A., C. E. Boyd, O. Silapajarn, and K. Silapajarn. 2004. Ionic
supplementation of pond waters for inland culture of marine shrimp. Journal
of the World Aquaculture Society 35:460467.
Miranda, L. E., and A. J. Sonski. 1987. Survival of red drum fingerlings in
fresh water: dissolved solids and thermal minima. Proceedings of the Annual Conference of Southeastern Association of Fish and Wildlife Agencies
39(1985):228237.
Parsons, T. R., Y. Maita, and C. M. Lalli. 1985. A manual of chemical and
biological methods for seawater analysis. Pergamom Press, New York.
Phelps, R. P., W. H. Daniels, N. Sansing, and T. W. Brown. 2010. Production of
Gulf killifish in the Black Belt region of Alabama using saline groundwater.
Roy, L. A., and D. A. Davis. 2010. Requirements for the culture of the Pacific
white shrimp, Litopenaeus vannamei, in low salinity waters: water modification and nutritional strategies for improving production. Pages 6178 in
L. E. Cruz-Suarez, D. Ricque-Marie, M. Tapia-Salazar, M. G. Nieto-Lopez,
D. A. Villareal-Cavazos, and J. Gamboa-Delgado, editors. Avances en nutricion acucola X. Memorias del decimo simposio internacional de nutricion
acucola. [Advances in aquatic nutrition X. Memoirs of the 10th international
symposium on aquatic nutrition]. Universidad Autonoma de Nuevo Leon,
Monterrey, Nuevo Leon, Mexico.
Roy, L. A., D. A. Davis, and I. P. Saoud. 2006. Effect of lecithin and cholesterol supplemented to practical diets for Litopenaeus vannamei reared in low
salinity water. Aquaculture 257:446452.
Roy, L. A., D. A. Davis, I. P. Saoud, C. A. Boyd, H. J. Pine, C. E. Boyd.
2010. Shrimp culture in inland low salinity waters. Reviews in Aquaculture
2:191208.
Roy, L. A., D. A. Davis, I. P. Saoud, and R. P. Henry. 2007. Effects of varying
levels of aqueous potassium and magnesium on survival, growth, and respiration of the Pacific white shrimp, Litopenaeus vannamei, reared in low salinity
waters. Aquaculture 262:461469.
Saoud, I. P., D. A. Davis, and D. B. Rouse. 2003. Suitability studies of inland
well waters for Litopenaeus vannamei culture. Aquaculture 217:373383.
Steel, R. G. D., and J. H. Torrie. 1980. Principles and procedures of statistics: a
biometrical approach. McGraw-Hill, New York.


Effects of Diets with 28% or 32% Protein and Meat and

Bone Meal or Corn Gluten Feed on Performance of
Golden Shiners in Pools
a
R. T. Lochmann & H. Phillips
Aquaculture/Fisheries Center of Excellence, The University of Arkansas at Pine Bluff, 1200

North University Drive, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA
To cite this article: R. T. Lochmann & H. Phillips (2012): Effects of Diets with 28% or 32% Protein and Meat and Bone Meal or
Corn Gluten Feed on Performance of Golden Shiners in Pools, North American Journal of Aquaculture, 74:4, 457-462


DOI: 10.1080/15222055.2012.676009
ARTICLE
Effects of Diets with 28% or 32% Protein and Meat

and Bone Meal or Corn Gluten Feed on Performance
of Golden Shiners in Pools
R. T. Lochmann* and H. Phillips
Aquaculture/Fisheries Center of Excellence, The University of Arkansas at Pine Bluff,

1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
Alternative diets made without marine proteins and with more plant ingredients are being used in channel catfish
Ictalurus punctatus with the goal of maintaining profitability. The potential to use similar diets for baitfish may be
greater because they consume natural foods throughout production. We conducted a feeding trial in outdoor pools
with golden shiners Notemigonus crysoleucas using four practical diets with 28% or 32% protein in formulas with
porcine meat and bone meal (MBM) or corn gluten feed (CGF) and no animal protein. Groups of 200 fish with a total
initial weight of 37.7 0.9 g (mean SE) were stocked into each of four 4.1-m3static pools per treatment and fed
daily to apparent satiation for 8 weeks. Diet effects were assessed by measuring growth, survival, feed conversion,
condition index, and body composition. Chlorophyll a and zooplankton were sampled to gauge natural productivity.
Individual weight and total length of golden shiners were greater in fish fed diets with 32% protein than in those fed
diets with 28% protein and in fish fed diets with CGF than in those fed diets with MBM. However, the differences
were not commercially relevant as all fish would grade into the small crappie minnow category. Mean individual
weight gain (based on group initial and final weights), feed intake, feed conversion, and survival were similar in golden
shiners fed diets with 28% or 32% protein and MBM or CGF. However, golden shiners fed diets with MBM had more
body fat, higher Fultons K index, and higher relative weight than those fed diets with CGF. These traits may indicate
greater robustness, which is more important in baitfish than rapid growth. The 28%-protein diet with MBM was the
least expensive option that increased body fat and condition of golden shiners. Natural productivity was positively
correlated with fish growth, enhancing the potential to use less expensive diets in golden shiners in outdoor systems.
The production of golden shiner Notemigonus crysoleucas

for bait occurs primarily in Arkansas, and the total farm gate
value of baitfish sold in the USA in 2007 was approximately
US$21.5 million (NASS 2009). Baitfish production has declined
slightly since 2002 but not as dramatically as food fish such as
channel catfish Ictalurus punctatus. In contrast to food fish, diet
cost is a lower percentage of variable costs in baitfish production.
Nevertheless, the increased cost of diets over the past few years
has reduced the profitability of baitfish production. Attempts
to cope with higher diet costs by reducing the feeding rate or
frequency also reduces yield and profits (Pounds et al. 1992).
Therefore, other solutions are needed to restore the profitability
of the baitfish industry. Alternative diets with mostly or only
plant ingredients have been tested recently in channel catfish

(Robinson and Li 2008, Li et al. 2010, 2011) with the goal
of maintaining profitability while using lower-cost diets. The
potential to use similar diets for baitfish may be higher because
they consume natural foods throughout production (Lochmann
and Phillips 1996).
One way to reduce diet cost while maintaining nutritional
quality is to use animal protein sources other than marine fish
meal. Porcine meat, bone, and blood meal (MBM) is a byproduct of the rendering industry that has a good balance of
most essential amino acids (NRC 2011), is available in large
quantities year-round, and costs less than half the price of fish
meal. Meat and bone meal has been used successfully in diets for
*Corresponding author: rlochmann@uaex.edu

Received October 21, 2011; accepted February 14, 2012
457
458
LOCHMANN AND PHILLIPS
channel catfish (Hedrick et al. 2005), tilapia Oreochromis spp.

(Nguyen et al. 2009), and gibel carp Carassius auratus gibelio
(Yang et al. 2004; Hu et al. 2008). However, no studies have
specifically tested MBM as a diet ingredient for golden shiners.
Plant protein sources often cost less than animal protein
sources. However, plant ingredients may also contain more
antinutritional factors (trypsin-inhibitor, phytic acid, fiber,
etc.) than animal products, and nutrients in plants may be less
bioavailable than those in animal ingredients (Francis et al.
2001). Many plant feedstuffs also have inferior amino acid profiles relative to animal feedstuffs (NRC 2011). Ingredients with
incomplete or imbalanced amino acid contents are likely to reduce growth and feed conversion efficiency and can compromise
fish health (Glencross et al. 2007). However, it is possible to
formulate all-plant diets with a mixture of ingredients that minimize the potential negative effects of plant ingredients while
decreasing diet cost and maintaining fish performance (Wu et al.
1999; Li et al. 2003; Lochmann et al. 2004; Sink et al. 2010).
Corn gluten feed (CGF) is one of the coproducts of corn
that remains after starch and other components are removed
for ethanol or corn syrup production (Schroeder 2010).
Traditionally, CGF has been used in livestock and pet foods,
but supplies are increasing due to the growth of the biofuels
industry. Corn gluten feed has been tested as a feed ingredient
for Nile tilapia Oreochromis niloticus (Wu et al. 1995) and
channel catfish (Robinson et al. 2001) with positive results, but
CGF has not been tested in baitfish.
We conducted a feeding trial in outdoor pools with golden
shiners using four practical diets with 28% or 32% protein in
formulas with MBM or no animal protein and CGF. In addition
to different ingredients, two different total dietary protein levels
were also tested. Most commercially available diets for baitfish
contain 28% or 32% protein, and performance of golden shiners
is similar over this range of protein (Lochmann and Phillips
2009). However, diets with 32% protein generally cost more
than those with 28% protein. Diet effects were assessed by
measuring growth, survival, feed conversion, condition index,
and body composition.
METHODS
Feeding trial.An 8-week feeding trial was conducted with
golden shiners in plastic-lined, 4.1-m3 outdoor pools at the
University of Arkansas at Pine Bluff. The pools were filled
with reservoir water (4,062 L) and maintained as static systems
during the trial to encourage plankton blooms. Groups of 200
fish with a total initial weight of 37.7 0.9 g (mean SE) and
a calculated individual weight of 0.19 g per fish ( = 0.4 lbs/1000
size) were stocked into each of four pools per treatment.
Four experimental diets were formulated (Table 1) to contain
either 28% or 32% protein using different ingredients. The
traditional diets contained 5% MBM, while the alternative diets
contained 20% CGF. The diet with 32% protein and MBM was
the control because it is most similar to traditional commercial
formulas, which contain a small amount of animal protein. We

did not use a commercial diet as a control because the ingredient
composition would be proprietary, limiting our ability to explain
diet effects. All diets were designed to meet or exceed the
nutrient requirements of channel catfish (NRC 2011), which are
similar to those of golden shiners (Lochmann and Phillips 2009).
Analyzed proximate composition was provided by the manufacturer (ARKAT, Dumas, Arkansas) except for crude fiber,
which was analyzed at the University of Arkansas at Pine Bluff
according to the ANKOM filter bag technique (AOAC 2005).
The gross energy of the diets was estimated from the analyzed
protein, lipid, and carbohydrate content (nitrogen-free extract)
of the diet (Serrano et al. 1992), because digestible energy values for individual diet ingredients are not available for golden
shiners. At the time of manufacture, the diets cost $409/ton
(28-MBM), $399/ton (28-CGM), $431/ton (32-MBM), and
$421/ton (32-CGF). Diets were extruded as 0.625-cm floating
pellets, then crumbled in a blender and screened to retain
particles of 12 mm in diameter prior to feeding.
The fish were fed twice on weekdays and once on weekends.
Preweighed feed equivalent to 68% of body weight daily
(with larger amounts used early in the study when the fish
were smaller) was offered to the fish until they fed to apparent
satiation, which occurred in 30 min or less. Due to the small
particle size, it was not possible to recover feed that was offered
but was not consumed. The remaining feed from the preweighed
ration was weighed and subtracted from the total to estimate
feed consumption. Subsamples of 50 fish per pool were counted
and weighed every 2 weeks to track growth. Temperature and
dissolved oxygen (DO) were measured in the morning and
afternoon daily in each pool using a DO meter (YSI 55 DO, Yellow Springs Instruments, Yellow Springs, Ohio). Chlorophyll
a content (corrected for pheophytin a) was determined in each
pool at weeks 2.5 and 5. Chlorophyll a analysis followed the
American Public Health Association (APHA; 2005), except that
ethanol was used as a solvent (Nusch 1980). Major zooplankton
groups (rotifers, copepod nauplii, and adult copepods) were also
determined in each pool at week 3. Six, 1-L water samples were
obtained with a tube sampler that encompassed the entire water
column. The samples were concentrated by straining them
through a70-m Wisconsin plankton net and then preserving the
resulting concentrate in 70% isopropyl alcohol. Samples were
identified and quantified using a Sedgwick-Rafter cell and a
microscope. The pH (pH meter, Denver Instruments, Colorado),
total ammonia nitrogen (salicylatecyanurate method), and nitrite were measured in each pool at weeks 2.5 and 5 with a Hach
DR/890 colorimeter test laboratory (Hach company, Loveland,
Colorado). Un-ionized ammonia was calculated from the results
of the total ammonia nitrogen and pH measurements. Water
quality data are summarized in Table 2, and all parameters were
within acceptable limits for golden shiners (Stone et al. 1997).
At the end of the trial all fish were counted and weighed
in groups. One group of 50 fish per pool was retained for
measurements of lengths and weights of individual fish to
459
ALTERNATIVE DIETS FOR GOLDEN SHINERS
TABLE 1. Ingredient (%, as-fed) and analyzed composition of diets with 28% or 32% protein with or without animal protein fed to golden shiners in outdoor
pools for 8 weeks. Traditional diets contained porcine meat, bone, and blood meal (MBM), while alternative diets contained corn gluten feed (CGF).
Diet 1
(28-MBM)
Ingredienta
Meat, bone, blood meal, pork (65%)
Soybean meal (48%)
Cottonseed meal (41%)
Corn gluten feed
Corn
Wheat middlings
Lysine-HCI
Dicalcium phosphate
C-free vitamin premixb
Trace mineral premixb
Poultry fatc
Crude proteind
Lipidd
Ashd
Fiberd
Nitrogen-free extracte
Moisture
Energy: proteinf
5.0
32.6
10.0
29.6
20.0
0.1
0.6
0.1
0.1
2.0
30.0
6.0
6.8
4.7
49.9
5.0
52.0
Diet 2
(28-CGF)
34.2
10.0
20.0
20.0
12.4
0.2
1.0
0.1
0.1
2.0
29.4
6.8
6.3
5.6
48.4
6.8
52.9
Diet 3
(32-MBM)
5.0
44.4
10.0
27.8
10.0
0.6
0.1
0.1
2.0
33.9
5.4
6.9
4.3
46.1
6.8
45.4
Diet 4
(32-CGF)
46.5
10.0
20.0
20.1
0.1
1.1
0.1
0.1
2.0
33.1
5.7
6.3
5.2
45.7
8.2
46.2
All diets were supplemented with 0.02% Stay-C 35 (DSM Nutritional Products, Basel, Switzerland), which provided an active vitamin C level 50mg/kg in finished diets.
Met requirements for channel catfish (NRC 2011) and presumably adequate for golden shiners (Chen et al. 2003, 2004).
c
Sprayed on finished diets.
d
Analyzed composition (percent, dry basis).
e
Nitrogen-free extract (NFE), calculated as NFE = 100 (protein + lipid + moisture + ash + fiber), is an estimate of soluble carbohydrate.
f
Kilojoules of gross energy per gram of protein. Estimated energy content of the diets was based on values of 16.7, 16.7, and 37.7 kJ/g for carbohydrate, protein, and lipid, respectively
(Serrano et al. 1992).
a
TABLE 2. Water quality parameters in a feeding trial with golden shiners

fed diets with 28% or 32% protein with or without animal protein in pools for
8 weeks. Diet did not have significant effects on water quality parameters, so
data (except for zooplankton) was analyzed by RM ANOVA with time as the
independent variable. Zooplankton was identified and quantified only once.
Parameter
Mean SE
Temperature
(morning, C)
Temperature
(afternoon, C)
Dissolved oxygen
(morning, mg/L)
Dissolved oxygen
(afternoon, mg/L)
Chlorophyll a (g/L)
pH
Ammonia (mg/L)
Nitrite (mg/L)
Rotifers/L
Copepods/L
Nauplii/L
28.6 0.1
<0.0001 (fluctuated)
32.3 0.1
<0.0001 (fluctuated)
6.5 0.0
<0.0001 (decreased)
9.0 0.0
<0.0001 (increased)
33.8
8.6
0.03
0.03
434.2
16.8
5.4
P-value (time)
7.1
0.0003 (increased)
0.1
<0.0001 (increased)
0.01
0.03 (increased)
0.01
0.02 (increased)
282.0
5.7
3.3
determine condition factor. A single pooled sample of five

individuals from each group of 50 was homogenized and used
for proximate analysis. Protein was analyzed using the Kjeldahl
procedure, and dry matter and ash content were determined
according to standard methods (AOAC 1995). Total lipids were
extracted and quantified gravimetrically (Folch et al. 1957).
Statistical Analysis.The experimental design was a 2 2
factorial. Percentage data were arcsine square root transformed
prior to statistical analysis. Performance data and body composition were analyzed with a two-way analysis of variance
(ANOVA) in Statview (SAS Institute, Cary, North Carolina)
with protein amount (28% or 32%) and protein source (CGM
or MBM) as the independent variables. Water quality data were
initially analyzed with a two-way ANOVA, but there were no
diet effects (P > 0.05). Therefore, data (except for zooplankton
abundance) were combined by sampling date and analyzed
using repeated measures ANOVA in Statview. Treatment means
were considered different at P < 0.05, and specific differences
in means were identified using Fishers least significant difference test. Correlations between weight gain and individual
types of plankton were determined with a z-test and results were
considered significant at P < 0.05. Water quality data were also
460
TABLE 3. Mean individual final weight, length, weight gain, Fultons K index, relative weight, feed conversion, and survival of golden shiners fed diets with
28% or 32% protein with or without animal protein for 8 weeks. Traditional diets contained porcine meat, bone, and blood meal (MBM), while alternative diets
contained corn gluten feed (CGF). Individual final weight, length, Fultons K index, and relative weight are based on measurements of 50 individual fish per
replicate (200 per treatment). Mean individual weight gain, feed conversion, and survival are means of four replicates consisting of 99200 fish each.
Diet
and statistics
28-MBM
28-CGF
32-MBM
32-CGF
Pooled SE
P (Protein amount)
P (Protein type)
P (Protein amount
type)
a
Mean
individual
final weight
(g)
Mean
individual
final length
(cm)
1.3
1.5
1.4
1.7
0.4
4.9
5.2
5.1
5.4
0.3
Fultons
K index
(g)
1.05
1.00
1.03
0.96
0.15
Two-way ANOVA
0.16
Relative
weight
(Wr)
Mean
individual
weight gain
(g)a
Feed
conversionb
Survival
(%)
127.6
118.6
122.8
113.6
19.1
0.80
0.56
0.70
0.95
0.16
1.9
2.5
2.6
2.0
0.4
94.0
95.1
83.6
97.4
6.0
0.39
0.82
0.51
0.98
0.92
0.24
0.15
0.12
0.31
0.001
<0.0001
0.07
(32>28)
(32>28)
<0.0001
<0.0001
0.006
0.001
(CGF>MBM) (CGF>MBM) (MBM>CGF) (MBM>CGF)
0.57
0.71
0.79
0.97
Weight gain was calculated as follows: Final group weight/final number of fish initial group weight/initial number of fish.
Dry feed weight/fish weight gain.
analyzed using descriptive statistics to obtain overall means

and standard errors for each parameter for the entire trial.
RESULTS
Growth, Feed Utilization, and Influence of Natural
Productivity
Mean individual final weight and length of golden shiners
were greater in fish fed diets with 32% protein than 28% protein
and in fish fed diets with CGM than MBM (Table 3). Fultons
K index and relative weight were similar in fish fed diets with
28% or 32% protein but greater in fish fed diets with MBM
than fish fed diets with CGF (Table 3). Apparent mean SE
individual feed intake was 1.4 0.1, 1.5 0.0, 1.8 0.5, and
2.0 0.7 g for diets 14, respectively, and was not different
among diets. However, feed intake was probably overestimated
slightly due to the difficulty in observing fish consuming small,
sinking particles in water with low transparency. Mean individual weight gain, feed conversion, and survival were similar in
golden shiners fed diets with 28% or 32% protein and MBM
or CGM (Table 3). Survival in one pool fed the 32-MBM diet
was only 49.5%, which greatly increased the variability in that
treatment. The mortality did not appear to be diet related, as
survival in the other pools fed the 32-MBM treatment was
94%. The water quality was good, similar among pools, and did
not explain the mortality. There were also no signs of disease
observed in fish in the pool. Although the cause of poor survival is unknown, excluding the pool from statistical analysis
did not produce any additional diet effects. There were no interactive effects between protein amount and source on general
performance criteria. Although chlorophyll a concentration and

zooplankton abundance did not differ by diet, final weight gain
of golden shiners was positively correlated with the abundance
of rotifers (correlation = 0.891, P < 0.0001), copepods (correlation = 0.848, P < 0.0001), and nauplii (correlation = 0.588,
P = 0.02). The correlation between chlorophyll a and weight
gain was weaker (0.353, p = .0468) but also significant.
Whole-body protein, lipid, dry matter, and ash were similar
in fish fed diets with 28% or 32% protein (Table 4). Whole-body
protein, dry matter, and ash also did not differ in fish fed diets
with MBM or CGF (Table 4). Whole-body lipid was higher
TABLE 4. Mean proximate composition (%, wet) of whole golden shiners
fed diets with 28% or 32% protein with or without animal protein for 8 weeks.
Each mean represents four pooled samples consisting of five individual fish
each. MBM = meat and bone meal; CGF = corn gluten feed.
Diet
and statistics
28-MBM
28-CGF
32-MBM
32 -CGF
Pooled SE
P (Protein amount)
P (Protein type)
P (Protein amount
type)
Protein
Lipid
13.9
9.9
14.5
8.5
13.8
9.2
13.2
8.5
0.4
0.4
Two-way ANOVA
0.06
0.47
0.96
0.03
(MBM>CGF)
0.09
0.40
Dry
matter
Ash
27.5
26.4
26.4
27.3
0.9
2.3
2.3
2.1
2.7
0.2
0.87
0.90
0.44
0.13
0.28
0.11
ALTERNATIVE DIETS FOR GOLDEN SHINERS
in fish fed diets with MBM than in those fed diets with CGF
(Table 4). There were no interactive effects of protein amount
and source on body composition.
DISCUSSION
Weight and length data from individual fish revealed more
diet differences than weight gain data obtained from initial and
final group weights of groups divided by the number in the
group.
Higher-protein diets often support better growth of small
fish, but there are no documented benefits of using 32%-protein
diets over 28%-protein diets in golden shiners, especially in
outdoor systems (Lochmann and Phillips 2009). Data from
individual fish may reflect true biological responses to diet
more accurately than bulk responses to pooled data, because
the latter obscure individual variability. The greater length and
weight of golden shiners fed diets with CGF was unexpected, as
CGF is low in lysine and other essential amino acids (Schroeder
2010). The diets with CGF also had more fiber than diets
with MBM, which reduces the digestible energy in the diet
(Robinson et al. 2001). Digestible energy values are unknown
for specific feed ingredients for golden shiners, but digestible
energy is always lower than gross energy due to the loss of fecal
energy from indigestible components (NRC 2011). Therefore,
the slightly higher gross energy-to-protein ratios calculated
for the diets with CGF in this study do not conflict with other
studies showing that CGF reduces the overall available energy
from the diet. Goldfish Carassius auratus fed diets formulated
with kaolin to dilute energy density compensated for the lower
dietary energy by increasing their feed intake (Rozin and Mayer
1961). In rainbow trout Oncorhynchus mykiss, feed intake
was inversely related to digestible energy content of diets with
different protein concentrations and sources (Morales et al.
1994). Feed consumption by small golden shiners in pools is
difficult to quantify, but there were no apparent differences in
feed form, flotation, or feed intake among diets in this study.
There were also no differences in natural productivity in the
pools among treatments, so there is no obvious explanation
for the enhanced growth of golden shiners fed diets with CGF.
However, most of the fish in this study would grade into the
small crappie minnow category (1416 grader size), which
all have the same retail price (J. Anderson, Anderson Minnow
Farm, personal communication). Therefore, the differences in
individual weights and lengths would have little commercial
significance.
Perhaps more importantly, golden shiners fed diets with
CGF also had less body fat than those fed diets with MBM. A
similar effect occurred in channel catfish fed diets with CGF
(Robinson et al. 2001), resulting in higher carcass yield and
potentially higher product quality. As mentioned previously,
this effect is likely due to the lower available energy content of
the CGF diets because they contain more indigestible fiber than
the MBM diets. In contrast to food fish, baitfish must withstand
461
multiple stressors after harvest, such as repeated grading,

transport, and retail display, when feed is often withheld.
Golden shiners fed diets with MBM (and presumably more
available energy than CGF) had more body fat, higher relative
weight, and higher condition factor, which might improve their
resilience postharvest compared to leaner fish. Hardiness is
valued over fast growth in the baitfish industry, where fish must
be maintained in good condition once they reach market size
until they are sold and reach their final destination.
Aside from the prepared diets, the plankton clearly contributed to the weight gain of golden shiners. Pools with more
natural productivity had larger fish than other pools of fish fed
the same prepared diet, and phytoplankton (as indicated by
chlorophyll a concentrations) as well as zooplankton (rotifers,
copepods, and nauplii) abundance correlated positively with
fish weight gain. This confirms past findings of the large contribution of natural foods to golden shiner growth even when
nutritionally complete diets are fed (Lochmann and Phillips
1996). Although natural productivity is difficult to predict and
control, the additional nutrients supplied by plankton may allow
greater use of less expensive alternative diet formulas in pondraised golden shiners without reduced fish performance. For
instance, zooplankton contain all essential amino acids (Aragao
et al. 2004; Mitra et al. 2007), which are often suboptimal in
alternative protein sources. Live foods also contain substantial
amounts of essential fatty acids, vitamins, and minerals. In
addition, live foods can enhance digestion in young fish by
contributing exogenous sources of enzymes such as proteases,
lipases, and amylases (Mitra et al. 2007) and by stimulating
beneficial gut microflora (Conceicao et al. 2010). Therefore,
good performance of pond-reared baitfish using diets with
marginal concentrations of essential nutrients is likely due to coconsumption of natural foods throughout the production cycle.
Partial budget analysis could not be conducted in this study
because there were no differences in yield among treatments. All
diets supported similar weight gain, feed conversion efficiency,
and survival of golden shiner. However, the 28-MBM diet also
increased whole-body lipid, relative weight, and condition index
of golden shiner. The 28-MBM diet cost $10 per ton more than
the 28-CGF diet, but the 28-MBM diet enhanced more desirable production traits in golden shiners at a lower cost than either
diet with 32% protein. Additional information is needed to verify whether or not fish with more body fat and higher condition
index are more resilient to postharvest events, and to assess the
economic impact of these factors on the baitfish industry.
ACKNOWLEDGMENTS
We thank Anderson Minnow Farm for donating fish for this
study. Sam Harrison, Emily Goodwin, and Ruguang Chen provided technical assistance. We thank Nathan Stone, Peter Perschbacher, and Hugh Thomforde for reviewing this manuscript.
462
REFERENCES
AOAC (Association of Official Analytical Chemists). 1995. Official methods of
analysis, 16th edition. AOAC International, Arlington, Virginia.
AOAC (Association of Official Analytical Chemists). 2005. Official methods of
analysis, 18th edition. AOAC International, Gaithersburg, Maryland.
APHA (American Public Health Association). 2005. Standard methods for the
examination of water and wastewater, 21st edition. American Public Health
Association, Washington, D. C.
Aragao, C., L. E. Conceicao, M. T. Dinis, and H. Fyhn. 2004. Amino acid pools
of rotifers and Artemia under different conditions: nutritional implications
for fish larvae. Aquaculture 234:429445.
Chen, R., R. Lochmann, A. Goodwin, K. Praveen, K. Dabrowski, and K.-J.
Lee. 2003. Alternative complement activity and resistance to heat stress in
golden shiners (Notemigonus crysoleucas) are increased by dietary vitamin
C levels in excess of requirements for prevention of deficiency signs. Journal
of Nutrition 133:22812286.
Chen, R., R. Lochmann, A. Goodwin, K. Praveen, K. Dabrowski, and K.-J. Lee.
2004. Effects of dietary vitamins C and E on alternative complement activity, hematology, tissue composition, vitamin concentrations and response to
heat stress in juvenile golden shiner (Notemigonus crysoleucas). Aquaculture
242:553569.
Conceicao, L. E. C., M. Yufera, P. Makridis, S. Morais, and M. T. Dinis. 2010.
Live feeds for early stages of fish rearing. Aquaculture Research 41:613640.
Folch, J., M. Lees, and G. H. Sloane-Stanley. 1957. A simple method for
the isolation and purification of total lipids from animal tissues. Journal of
Biological Chemistry 226:497509.
Francis, G., H. P. S. Makkar, and K. Becker. 2001. Antinutritional factors in
plant-derived alternate fish ingredients and their effects in fish. Aquaculture
199:197227.
Glencross, G. D., M. Booth, and G. L. Allan. 2007. A feed is only as good as
its ingredientsa review of ingredient evaluation strategies for aquaculture
feeds. Aquaculture Nutrition 13:1734.
Hedrick, R. L., T. J. Popma, and D. A. Davis. 2005. Pond production and fatty
acid profiles of the fillets of channel catfish reared on diets with different
protein sources. North American Journal of Aquaculture 67:304311.
Hu, M., Y. Wang, Q. Wang, M. Zhao, B. Xiong, X. Qian, Y. Zhao, and
Z. Luo. 2008. Replacement of fish meal by rendered animal protein ingredients with lysine and methionine supplementation to practical diets for gibel
carp, Carassius auratus gibelio. Aquaculture 275:260265.
Li, M. H., E. H. Robinson, D. F. Oberle, and P. M. Lucas. 2010. Effects of various
corn distillers by-products on growth, feed efficiency, and body composition
of channel catfish, Ictalurus punctatus. Aquaculture Nutrition 16:188193.
Li, M. H., E. H. Robinson, B. G. Bosworth, D. F. Oberle, and P. M. Lucas.
2011. Use of corn gluten feed and cottonseed meal to replace soybean meal
and corn in diets for pond-raised channel catfish. North American Journal of
Li, M. H., D. J. Wise, B. B. Manning, and E. H. Robinson. 2003. Effect of dietary
total protein and animal protein on growth and feed efficiency of juvenile
channel catfish Ictalurus punctatus and their response to Edwardsiella ictaluri
challenge. Journal of the World Aquaculture Society 34:223228.
Lochmann, R., and H. Phillips. 1996. Stable isotopic evaluation of the relative
assimilation of natural and artificial foods by golden shiners Notemigonus
crysoleucas in ponds. Journal of the World Aquaculture Society 27:
168177.
Lochmann, R., and H. Phillips. 2009. Nutrition and feeding of baitfish. University of Arkansas at Pine Bluff, Cooperative Extension Program ETB256-PD2-09RV, Pine Bluff.
Lochmann, R., N. Stone, and H. Phillips. 2004. Evaluation of 36%-protein diets with or without animal protein for rearing tank-hatched golden shiner
Notemigonus crysoleucas fry in ponds. North American Journal of Aquaculture 66:271277.
Mitra, G., P. K., Mukhopadhyay, and S. Ayyappan, 2007. Biochemical composition of zooplankton community grown in freshwater earthen ponds: nutritional implication in nursery rearing of fish larvae and early juveniles.
Morales, A. E., G. Cardenete, M. de la Higuera, and A. Sanz. 1994. Effects of
dietary protein source on growth, feed conversion and energy utilization in
rainbow trout, Oncorhynchus mykiss. Aquaculture 124:117126.
NASS (National Agricultural Statistics Service). 2009. Census of agriculture
(2007), United States, summary and state data, volume 1, geographic area
series part 51. U.S. Department of Agriculture, NASS AC-07-A-51, Washington, D.C.
Nguyen, T. N., D. A. Davis, and I. P. Saoud. 2009. Evaluation of alternative
protein sources to replace fish meal in practical diets for juvenile tilapia,
Oreochromis spp. Journal of the World Aquaculture Society 40:113121.
NRC (National Research Council). 2011. Nutrient requirements of fish and
shrimp. National Academy Press, Washington, D.C.
Nusch, E. 1980. Comparison of different methods for chlorophyll and phaeopigment determination. Archiv fur Hydrobiologie Beiheft Ergebnisse der Limnologie 14:1436.
Pounds, G. L., C. R. Engle, and L. W. Dorman. 1992. Economic effects of intensification of baitfish production. Journal of the World Aquaculture Society
23:6476.
Robinson, E. H., and M. H. Li. 2008. Replacement of soybean meal in channel
catfish, Ictalurus punctatus, diets with cottonseed meal and distillers dried
grains with solubles. Journal of the World Aquaculture Society 39:521527.
Robinson, E. H., M. H. Li, and B. B. Manning. 2001. Evaluation of corn
gluten feed as a dietary ingredient for pond-raised channel catfish Ictulurus
punctatus. Journal of the World Aquaculture Society 32:6871.
Rozin, P., and J. Mayer. 1961. Regulation of food intake in the goldfish.
American Journal of Physiology 201:968974.
Schroeder, J. W. 2010. Corn gluten feed. Composition, storage, handling, feeding and value. North Dakota State University, Extension Publication AS-1127,
Fargo.
Serrano, J. A., G. R. Nematipour, and D. M. Gatlin III. 1992. Dietary protein
requirement of the red drum (Sciaenops ocellatus) and relative use of dietary
carbohydrate and lipid. Aquaculture 101:283291.
Sink, T. D., R. Lochmann, and N. Kinsey. 2010. Growth and survival of channel
catfish Ictalurus punctatus fry fed diets with 36 or 45% total protein and allplant or animal-protein sources. Journal of the World Aquaculture Society
40:124129.
Stone, N., E. Park, L. Dorman, and H. Thomforde. 1997. Baitfish culture
in Arkansas: golden shiners, goldfish, and fathead minnows. University of
Arkansas at Pine Bluff Cooperative Extension Program, Publication MP386,
Pine Bluff.
Wu, Y. V., R. Rosati, D. J. Sessa, and P. Brown. 1995. Utilization of corn gluten
feed by Nile tilapia. Progressive Fish-Culturist 57:305309.
Wu, Y. V., K. W. Tudor, P. B. Brown, and R. R. Rosati. 1999. Substitution of
plant proteins or meat and bone meal for fish meal in diets of Nile tilapia.
Yang, Y., S. Xie, Y. Cui, W. Lei, Y. Zhu, Y. Yang, and Y. Yu. 2004. Effect
of replacement of dietary fish meal by meat and bone meal and poultry byproduct meal on growth and feed utilization of gibel carp, Carassius auratus
gibelio. Aquaculture Nutrition 10:289294.


Evaluation of Hydrogen Peroxide and Temperature

to Control Mortality Caused by Saprolegniasis and to
Increase Hatching Success of Largemouth Bass Eggs
a
Michael D. Matthews , Joshua C. Sakmar & Nick Trippel
Florida Fish and Wildlife Conservation Commission, Florida Bass Conservation Center, 3583
County Road 788, Webster Florida, 33597, USA
b
Florida Fish and Wildlife Conservation Commission, Eustis Fisheries Research Laboratory,
601 West Woodward Avenue, Eustis, Florida, 32726, USA
To cite this article: Michael D. Matthews, Joshua C. Sakmar & Nick Trippel (2012): Evaluation of Hydrogen Peroxide and
Temperature to Control Mortality Caused by Saprolegniasis and to Increase Hatching Success of Largemouth Bass Eggs, North
American Journal of Aquaculture, 74:4, 463-467


DOI: 10.1080/15222055.2012.676608
ARTICLE
Evaluation of Hydrogen Peroxide and Temperature to

Control Mortality Caused by Saprolegniasis and to Increase
Hatching Success of Largemouth Bass Eggs
Michael D. Matthews* and Joshua C. Sakmar
Florida Fish and Wildlife Conservation Commission, Florida Bass Conservation Center,
3583 County Road 788, Webster Florida 33597, USA
Nick Trippel
Florida Fish and Wildlife Conservation Commission, Eustis Fisheries Research Laboratory,
601 West Woodward Avenue, Eustis, Florida 32726, USA
Abstract
We evaluated the use of hydrogen peroxide (H2 O2 ), temperature, or both, to control mortality presumptively
caused by fungus Saprolegnia in Florida largemouth bass Micropterus salmoides floridanus eggs in a flow-through
hatchery system. Four treatmentsambient water (AW; control), ambient water and H2 O2 (AWHP), heated water
(HW), and both heated water and H2 O2 (HWHP)were tested on each of 30 replicate spawns. Four 8 cm 5 cm
sections of spawning mat, one for each treatment, were cut from each of the 30 selected spawns. Egg counts on each
cut mat section were recorded. Water temperature in all AW and HW trial tanks ranged from 17.9 C to 19.2 C
and from 22.1 C to 23.6 C, respectively, during treatment. The water temperature difference between treatments
averaged 4.4 C. Hydrogen peroxide trial tanks received two, 100 mg/L H2 O2 treatments 8 h apart, until hatch. The
7.5 L/min incoming water flow was not reduced during treatment. The addition of HW, H2 O2 , or both, significantly
increased mean percent hatch (79, 79, and 91%, respectively) over that of the untreated controls (49%). Significant
differences were found in mean percent hatch levels between AW and the other three treatment groups (P < 0.05).
The combination of higher incubation temperature and H2 O2 administration also resulted in significantly greater
hatch than did either higher incubation temperature or increased H2 O2 administration alone; percent hatch did
not differ significantly between eggs that either were only incubated at the higher temperature or received only
H2 O2 administration. Increasing incubation water temperature to 2223 C instead of using only water at ambient
temperature (1819 C) or adding 100 mg/L H2 O2 twice daily in a flow-through system significantly increased hatching
percentage of Florida largemouth bass eggs.
Fungi from the family Saprolegniaceae commonly infest

fish and fish eggs at temperatures below 25.5 C (Tucker and
Robinson 1990). Eight genera reportedly cause infestions, but
only Saprolegnia sp., Achlya sp., and Aphanomyces sp. are of
concern in freshwater aquaculture (Woo and Bruno 1999). Outbreaks of Saprolegnia regularly cause chronic losses in fish
populations and can cause rapid increases in egg mortality.
Koeypudsa et al. (2005) reported that temperatures suitable for
growth of Saprolegnia ranged from 5 C to 30 C, but they observed no growth at temperatures above 30 C.
Marking et al. (1994) tested 21 different antifungal chemicals
on rainbow trout Oncorhynchus mykiss eggs and determined
that salt (sodium chloride), formalin, and H2 O2 were the most
suitable and effective at controlling fungal infestations. Rach
et al. (2005) showed that H2 O2 concentrations of 1,000 mg/L
and formalin concentrations of 1,667 mg/L effectively increased
*Corresponding author: michael.matthews@myfwc.com

Received May 3, 2011; accepted March 11, 2012
463
464
MATTHEWS ET AL.
survival of lake trout Salvelinus namaycush eggs up to the eyed

egg stage. Egg hatch increased in walleye Sander vitreus, white
sucker Catostomus commersonii, and paddlefish Polyodon
spathula when treated with 283, 565, and 1,130 mg/L H2 O2
for 15 min every other day (Gaikowski et al. 2003). Barnes and
Gaikowski (2004) found that a daily 15-min treatment with
1,000 mg/L H2 O2 was suitable for use on Chinook salmon O.
tshawytscha eggs.
Saprolegnia control with H2 O2 has been proven equally effective in warmwater species. In three trials conducted by Small
and Wolters (2003), povidoneiodine treatment at 100 mg/L,
followed by 15 min of treatment with H2 O2 at 250 mg/L, increased the hatching success of channel catfish Ictalurus punctatus by 26%. Hatching success of channel catfish eggs treated for
15 min with 250 mg/L H2 O2 was 30% more than for povidone
iodine-treated eggs and formalin-treated eggs. Treatment of egg
masses with 70 mg/L hydrogen peroxide in a flow-through
trough increased survival by 48% over nontreated controls.
Mitchell et al. (2009) increased channel catfish hatch rates by
using 125, 250, and 500 mg/L H2 O2 , but cautioned against using
concentrations above 500 mg/L in hatching troughs. According
to Mitchell et al. (2010), H2 O2 was as effective as copper sulfate
pentahydrate, diquat bromide, and formalin at limiting fungal
outbreaks on catfish eggs. Use of hydrogen peroxide to control fungi due to Saprolegniaceae has been approved for use on
warmwater finfish eggs at 7501,000 mg/L for 15 min/d in a
continuous-flow system on consecutive or alternate days until
hatch (USFDA 2007). Despite good raceway sanitation practices and ozone-treated recycled water supply, fungal outbreaks
still caused high mortality of bass eggs in incubation raceways
at the Florida Bass Conservation Commission (FBCC). Van
West (2006) reported that ozone was effective in fungal control in water supplies, but could not be used to treat affected
fish. Saprolegnia-induced mortality on Florida largemouth bass
Micropterus salmoides floridanus eggs has accounted for unacceptable losses in total egg production in years past at the
FBCC. Hydrogen peroxide treatments of 750 and 1,000 mg/L
controlled fungus on eggs at the FBCC, but both concentrations
caused moderate to high levels of mortality on newly hatched fry.
Rach et al. (1998) noted that high concentrations of H2 O2 (3 g/L
or higher) reduced hatching success of warmwater species.
The FBCC produces all of the Phase I (total length >25 mm
but <75 mm) Florida largemouth bass in Florida. Wild-caught
broodfish are spawned indoors in 25-m-long concrete raceways
on Spawntex mats as substrate. Spring spawning of largemouth
bass occurred naturally from February to March at 1822 C. Fall
largemouth bass spawned at water temperatures between 19 C
and 24 C in late September through October after temperature
and photoperiod manipulation. High levels of fungal-induced
egg mortality documented at the FBCC extended spawning
seasons longer than needed. Rather than testing different concentrations of H2 O2 as reviewed in past studies, we tested
elevated temperature, H2 O2 , or both combined to reduce egg
loss caused by fungus. Increasing temperature may be a strategy
to reduce fungal-related mortality by decreasing the duration of

egg incubation. McCormick (1978) demonstrated that increasing temperature on white bass Morone chrysops eggs decreased
hatching time. Hardy (1978) reported similar results for largemouth bass, finding hatch at 10 C after 1321 d; at 17.7 C,
hatch occurred in 34 d, and eggs kept at 22.2 C hatched in
2 d. We found no studies that used the temperature and H2 O2
combination against fungal growth. Our objective was to find an
effective yet practical way in a large-scale indoor production setting to reduce largemouth bass egg loss to fungal growth, using
temperature manipulation, H2 O2 , or a combination of both.
METHODS
During the 2010 spring spawning season at the FBCC, an
experiment was designed to test fungal control using elevated
temperature and H2 O2 on largemouth bass eggs. Three 25-mlong raceways were stocked with 20 pairs of male and female
Florida largemouth bass. Each raceway received 20 Spawntex
mats for spawning substrate. Spawns were collected each morning with mats immediately replaced. A total of 30 spawns were
randomly selected from the 109 spawns collected during the
20-d spawning period. The four treatments were heated water (HW), H2 O2 and ambient water (AWHP), heated water and
H2 O2 (HWHP), and ambient water (AW), which was used for
the control. Four 8 cm 5 cm sections, one for each treatment,
were cut from each of the 30 randomly selected mats. Each mat
section needed to include approximately 100 eggs. Eggs were
visually counted, noting the initial number of live and dead eggs.
All spawning mats were collected less than 6 h after completed
spawning.
Each of four 378-L trial tanks, 1.8 m 0.5 m 0.5 m in
size, was assigned one of the four treatments. Each trial tank
was separated into three sections with screens, thus allowing for
three simultaneous replicates per tank. This setup was repeated
10 times over the 20-d spawning period to achieve 30 test replicates per treatment. To avoid temperature shock, the eggs on all
mat sections were counted and moved at ambient water temperatures. Tanks assigned to the AW treatment received no antifungal agents or any other changes from the conditioned 1819 C
well water supply and were used as the control. Heated water
and HWHP tanks were maintained at 2223 C. Boilers supplied
continuous heated water for this experiment. Eggs (<6 h old)
were tempered for 2 h to the increased temperature. Hydrogen
peroxide (35% hydrogen peroxide, w/w; Eka Chemicals Inc.,
Marietta, Georgia, trade name 35% PEROX-AID, specific gravity 1.1317) was added to the inlet of the tank twice daily at
100 mg/L to tanks labeled AWHP and HWHP at 0800 hours
and again at 1600 hours until hatch. Calculation for a 100 mg/L
concentration of 35% H2 O2 was as follows (see manufacturers
label):
[(target conc., mg/L)/396,100 mg/L PEROX-AID)]
[(tank vol, L) (1,000 mL/L)] = mL of H2 O2
CONTROLLING SAPROLEGNIASIS ON LARGEMOUTH BASS EGGS
Mat sections were suspended by hanging each section from

a wooden dowel in the respective trial tanks. All test tanks
had a constant 7.5 L/min flow rate and received conditioned
well water from one source. Constant aeration was applied to
each tank for homogeneous mixing of tank water and H2 O2 if
applicable. Dose verification was recorded in five heated and five
ambient tanks that received H2 O2 treatments using the following
equation (USFWS AADAP 2008):
H2 O2 , mg/L = [(a M 85.025)/V ] 1,000

where a = mL of potassium permanganate (KMnO4 ), M = 0.02
molar concentration of KMnO4 , 85.025 is a constant, V = volume of sample in mL, and 1,000 is a constant. Concentrations
of H2 O2 between 85 mg/L and 115 mg/L were considered acceptable. Water samples were taken to check for residual H2 O2
before the second dose of H2 O2 was added to these same 10
tanks. Water temperatures were recorded daily in each tank until hatch. Daily dissolved oxygen and pH levels were monitored
on five HW and five AW treatment tanks. Temperature and dissolved oxygen were measured with a YSI model 550A (Yellow
Springs, Ohio), and pH was measured with an Oakton pHTestr
30 (Vernon Hill, Illinois). All dissolved oxygen readings were
taken before H2 O2 administration if applicable. Total ammonia nitrogen, total alkalinity, and total hardness were monitored
daily in the source water. We used the salicylate method to determine total ammonia nitrogen (Hach, Loveland, Colorado); total
alkalinity and total hardness were measured by titration using
the LaMotte AQ-2 test kit (Chestertown, Maryland).
Fry were collected after hatch with a siphon and counted.
Percent hatch was calculated from the number of fry collected
divided by the number of initial eggs (minus dead eggs). Fungal coverage on each mat section was ranked on a 010 scale,
0 indicating no fungal growth and 10 denoting complete mat
coverage. All evaluations were completed by the same person
throughout the entire study. This score was compared to percent
hatch to view graphically how fungal growth affects hatch rate.
A natural logarithmic transformation was used to meet the
assumptions of equal variances. A generalized linear mixed
model (SAS PROC GLIMMIX) was investigated to determine
whether the independent variable treatment assignment explained the variation in the probability of dependent variable
percent hatch. The adjusted least-squares means (Scheffes
adjustment) were calculated to identify differences in the probability of hatch between treatments; differences were declared
statistically significant if P < 0.05.
RESULTS
The H2 O2 concentrations determined during dose verification ranged from 86 to 96 mg/L. Hydrogen peroxide was not
detected in water samples taken immediately before administration of the second daily H2 O2 dose. Oxygen levels never
fell below 8 mg/L, and pH ranged from 7.7 to 8.1. Total
465
ammonia nitrogen remained negligible throughout the study,

and total hardness and alkalinity consistently remained at 340
and 360 mg/L as CaCO3 , respectively.
The mean SE numbers of largemouth bass eggs per mat
section were 94 6.5, 91 5.8, 86 5.2, and 87 5.4
for AW, AWHP, HW, and HWHP treatments, respectively. The
mean number of dead eggs per mat ranged from 2 to 4 on day
zero for all treatment groups. The initial egg viability mean
SE were 97 0.7, 97 0.6, 96 1.2, and 96 0.8%, for AW,
AWHP, HW, and HWHP treatments, respectively. Treatment
explained a significant amount of the variation in egg hatch,
Type III test of fixed effects (GLIMMIX: F = 52.70, df =
4, P < 0.0001). All HW- and HWHP-treated eggs hatched in
4048 h while all AW- and AWHP-treated eggs hatched in 84
96 h. Mean percent hatches in the AWHP, HW, and HWHP
treatment groups (79, 79, and 91%) were significantly greater
than the 49% in the AW group (Figure 1). The addition of HW,
H2 O2 , or both, decreased fungus levels observed on hatching
mats relative to AW controls. The mean SE of observed
fungus levels for each treatment were 6.5 0.4, 3.5 0.3,
2.9 0.4, and 1.1 0.3 for AW, AWHP, HW, and HWHP,
respectively (Figure 1). Significant differences in percent hatch
were identified between all treatment comparisons (P < 0.05)
except for HW versus AWHP. The odds ratios determined to
describe the probability of egg hatch indicated that eggs treated
with HP or HW were four times more likely to hatch than
the untreated (AW) eggs and eight times more likely to hatch
when HP and HW were used in combination. The combination
treatment (HWHP) significantly increased the probability of egg
hatch compared with that of eggs administered HW or HP alone.
the eggs assigned to the HWHP being about twice as likely to
hatch as the single-treatment eggs.
DISCUSSION
This experiment focused on Florida largemouth bass eggs
treated with 2223 C heated water, 100 mg/L H2 O2 , or both. In
lieu of evaluating different H2 O2 concentrations, we extended
treatment duration and number of daily H2 O2 treatments (two)
at a concentration 7.510 times less than that approved for use
in warm water. The 7.5 L/min continuous flow rate created a
turnover of the complete 378-L tank volume in 50 min. The
increased time of the ever-decreasing H2 O2 concentration
decreased fungal-induced mortality at ambient temperatures.
Published data on largemouth bass eggs treated with H2 O2 in
flow-through systems are sparse, possibly because bass spawning generally occurs in outdoor ponds, not indoors in raceways.
We compared our results with a published study on warmwater
channel catfish eggs that used one daily H2 O2 treatment added
at the tank inlet (trial 1: H2 O2 additions of 0, 100, 200, and
300 mg/L; trial 2: H2 O2 additions of 0, 35, 70, and 100 mg/L) at
a continuous flow rate similar to our study. The results showed
that the use of 100 mg/L and 70 mg/L H2 O2 in the first and
second trial significantly improved hatching success compared
466
MATTHEWS ET AL.
FIGURE 1. Mean (+ SE, n = 30) hatching success (%) versus fungus level (scored 0 = no growth to 10 = total fungal coverage) for largemouth bass eggs
treated twice daily until hatch with 100 mg/L hydrogen peroxide at a constant 7.5 L/min flow rate with continuous aeration. All treatments were added at the inlet
of the tanks. Different letters (a, b, and c) denote a significant difference between the two treatments.
with no treatment (0 mg/L H2 O2 ), the 70 mg/L H2 O2 treatment

yielding the higher success between the two trials (Small and
Wolters 2003). Our results similarly showed improved hatching
success with the use of lower concentrations of H2 O2 . Mitchell
et al. (2009) also found that H2 O2, applied at 125, 250, and
500 mg/L for 6 d, significantly reduced fungal growth on channel catfish eggs under continuous flow rates of two exchanges
per hour.
The 100 mg/L concentration increased hatching success, and
preliminary observations provided evidence that mortality of
newly hatched bass fry did not increase. Further investigation is
required to determine lethal concentrations of H2 O2 on newly
hatched bass fry to verify our observations. However, target
Animal Safety Study CAP-97-00048-08 reported zero mean
percent mortality of largemouth bass fry after 60- and 180-min
treatments with 121 and 103 mg/L H2 O2 (USFDA 2007). Small
and Wolters (2003) noticed that using H2 O2 concentrations of
500 mg/L on channel catfish eggs in a 15-min bath caused poor
hatching success; 200 and 300 mg/L concentrations caused premature hatching in a flow-through system. We found no evidence of premature hatching of bass eggs at 100 mg/L when
we compared the hatch times of AW to AWHP and of HW to
HWHP.
Heated water without H2 O2 was also effective at decreasing
fungal mortality by reducing duration until hatch. The duration of largemouth bass egg incubation is reduced by nearly 2 d
by increasing water temperatures from 17.7 C to 22.2 C (Hardy
1978). We recorded similar results in our study; eggs in all tanks
receiving HW or HWHP treatments hatched 3648 h sooner than
those receiving AW and AWHP treatments. Kelly (1968) stated
that temperatures between 12.7 C and 23.9 C did not directly

cause mortality in largemouth bass eggs and reported generally
poor egg survival of nonacclimated eggs at temperatures of 10,
26.7, and 29.4 C. The poor survival was hypothesized to be
due to critical periods in early embryo development when the
embryo may be more sensitive to rapid temperature changes.
Acclimation was cited as another reason for poor egg survival,
but eggs not acclimated to temperatures between 12.7 C and
23.9 C showed no difference in survival compared with acclimated eggs (Kelly 1968). White bass eggs incubated at 26 C
hatched in 24 h, but incubation at 14 C took 108 h (McCormick
1978). Implementing heated water in incubation tanks can have
the added value of maintaining a stabilized water temperature,
thus adding the benefit of consistency in incubation duration.
The negative aspect of heating the incubation water is that if the
broodstock are spawned at a water temperature different from
the incubation temperature, the eggs may need to be tempered
to incubation temperature, which can be very time-consuming.
Past experience found the lack of tempering can lead to high
egg mortality, in contrast to Kellys (1968) findings. Heated
water decreased incubation duration and bass eggs hatched in
less time. The H2 O2 increased hatching percentage by visibly
reduced fungal growth. Results from this study support both of
these statements and show that, in combination, heated water
and H2 O2 have a synergistic effect on hatching success.
We did not directly compare the 100 mg/L concentration with
a 750 or 1,000 mg/L H2 O2 concentration and are not claiming
that the lower dose improves hatching success over that at the
currently approved H2 O2 concentrations. Our results show that
the 100 mg/L H2 O2 dose significantly improves hatch rate over
CONTROLLING SAPROLEGNIASIS ON LARGEMOUTH BASS EGGS
no treatment. The proposed benefit of the lower H2 O2 concentration allows spawnings from multiple days to be treated in
the same tank without the increased fry mortality noted at the
approved dosage rate (USFDA 2007). From a hatchery management perspective, raceway space limitations mandate that
multiple day collections of spawns be placed together. Adding
approved concentrations of H2 O2 (7501,000 mg/L, continuous
water flow) to the raceways decreased egg mortality on mats
hung on days 1 and 2, but caused unacceptable fry losses from
eggs collected on days 1 and 2 when mats from a third day
were added and treated accordingly. Due to observed increases
in fry mortality, 7501,000 mg/L H2 O2 concentrations limited
use to prehatch conditions only. Without treatment, the mats
from day 3 had high levels of fungal-induced egg mortality.
Hydrogen peroxide concentrations of 100 mg/L are effective at
reducing fungal growth and have the added benefit of not causing mortality in the newly hatched fry. In fall 2010 we tested
the 100 mg/L H2 O2 concentration administered twice daily in
4,920-L tanks stocked with three consecutive days of spawns
(up to 40 spawns/raceway). This resulted in zero spawns lost
to fungus from 205 spawns treated. No increased fry mortality
was observed. Water temperatures averaged 23 C.
This study revealed some options for production managers
faced with site-specific restrictions. Heating water or the use of
chemicals may not be practical or allowed at some facilities, but
implementing just one of the three treatments significantly increased hatch rates. Another benefit from this treatment method
resulted from not needing to cut water flow to the treatment
tank. This completely removed the risk of egg loss to lower
water levels and to anoxic conditions that occur when using
other chemical treatments when water flow is not promptly restored. Higher egg survival equates to requiring fewer spawns,
less broodstock, less hatchery space, and less time spent on bass
production.
We recommend that approval status for lower levels of H2 O2
be investigated for fungus control on warmwater eggs. Potentially this could be completed with some additional efficacy and
with target animal safety studies that show improved egg hatch
and fry survival rates at reduced H2 O2 concentrations and increased treatment duration. This should also provide the data
to justify that the proposed two treatments at 100 mg/L would
have less chemical discharge and be no less safe than one treatment at 1,000 mg/L. Potential benefits from this will be lower
amounts of H2 O2 needed, lower risk of fry mortality, and lower
concentrations of chemical in hatchery effluent.
ACKNOWLEDGMENTS
We thank Rick Stout and the Florida Bass Conservation Center staff for their assistance with mat collection. We also thank
Bill Pouder and Jon Fury for their comments on this manuscript.
We also acknowledge Dr. Jesse Trushenski for her insightful
467
review of this manuscript. Finally, we thank Mark Gaikowski

for his help with data analysis.
REFERENCES
Barnes, M. E., and M. P. Gaikowski. 2004. Use of hydrogen peroxide during incubation of landlocked fall Chinook salmon eggs in vertical-flow incubators.
Gaikowski, M. P., J. J. Rach, M. Drobish, J. Hamilton, T. Harder, L. A. Lee,
C. Moen, and A. Moore. 2003. Efficacy of hydrogen peroxide in controlling mortality associated with saprolegniasis on walleye, white sucker, and
paddlefish eggs. North American Journal of Aquaculture 65:349355.
Hardy, J. D. Jr. 1978. Aphredoderidae through Rachycentridae. U.S. Fish and
Wildlife Service FWS/OBS-78/12:volume 3.
Kelly, J. W. 1968. Effects of incubation temperature on survival of largemouth
bass eggs. Progressive Fish-Culturist 30:159163.
Koeypudsa, W., P. Phadee, J. Tangtrongpiros, and K. Hatai. 2005. Influence
of pH, temperature, and sodium chloride concentration on growth rate of
Saprolegnia sp. Journal of Scientific Research at Chulalongkorn University
30:123130.
Marking, L. L., J. J. Rach, and T. M. Schreier. 1994. Evaluation of antifungal
agents for fish culture. Progressive Fish-Culturist 56:225231.
McCormick, J. H. 1978. Effects of temperature on hatching success and survival
in the white bass. Progerssive Fish-Culturist 40:133137.
Mitchell, A. J., A. A. Radomski, D. L. Straus, and R. Carter. 2009. The effect of hydrogen peroxide on the hatch rate and Saprolegnia sp. infestation of channel catfish eggs. North American Journal of Aquaculture 71:
276280.
Mitchell, A. J., D. L. Straus, B. Farmer, and R. Carter. 2010. Comparison
of percent hatch and fungal infestation in channel catfish eggs after copper
sulfide, diquat bromide, formalin, and hydrogen peroxide treatment. North
American Journal of Aquaculture 72:201206.
Rach, J. J., M. P. Gaikowski, G. E. Howe, and T. M. Schreier. 1998. Evaluation
of the toxicity and efficacy of hydrogen peroxide treatments on eggs of warmand coolwater fishes. Aquaculture 165:1125.
Rach, J. J., S. Redman, D. Bast, and M. P. Gaikowski. 2005. Efficacy of hydrogen peroxide versus formalin treatments to control mortality associated with
saprolegniasis on lake trout eggs. North American Journal of Aquaculture
67:148154.
Small, B. C., and W. R. Wolters. 2003. Hydrogen peroxide treatment during
egg incubation improves channel catfish hatching success. North American
Tucker, C. S., and E. H. Robinson. 1990. Channel catfish farming handbook.
Van Nostrand Reinhold, New York.
USFDA (U.S. Food and Drug Administration). 2007. Freedom of information
summary: original new animal dug application. NADA 141255 35%
PEROX-AID hydrogen peroxide liquid solution. USFDA, Washington,
D.C. Available: http://www.fda.gov/downloads/AnimalVeterinary/Products/
ApprovedAnimalDrugProducts / FOIADrugSummaries / UCM051418 . pdf.
(June 2012).
USFWS AADAP (U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program) 2008. Procedures to verify concentrations of hyR
drogen peroxide (35% PEROX-AID
) solutions. USFWS AADAP, Standard
Operating Procedure Miscellaneous 240.1, Bozeman, Montana. Available:
http://www.fws.gov/fisheries/aadap/National%20Aquaculture%20Drug%20
Research%20Forum/SOPs/SOP%20240-1%20analysis%20of%20H2O2.pdf.
(June 2012).
Van West, P. 2006. Saprolegnia parasitica, an oomycete pathogen with a fishy
appetite: new challenges for an old problem. Mycologist 20:99104.
Woo, P. T. K., and D. W. Bruno. 1999. Fish diseases and disorders, volume 3.
Viral, bacterial, and fungal infections. CAB International, New York.


New Programmable Electrofishing Device for Use in

Aquaculture
a
Fulgencio Soto , Manuel Jimnez , Antonio Mateo , Jos A. Villarejo , Esther de Jdar
& Jacinto Jimnez
Departamento de Tecnologa Electrnica, Universidad Politcnica de Cartagena, Campus

Muralla del Mar, s/n 30202 Cartagena, Spain
To cite this article: Fulgencio Soto, Manuel Jimnez, Antonio Mateo, Jos A. Villarejo, Esther de Jdar & Jacinto Jimnez
(2012): New Programmable Electrofishing Device for Use in Aquaculture, North American Journal of Aquaculture, 74:4,
468-476


DOI: 10.1080/15222055.2012.690828
ARTICLE
New Programmable Electrofishing Device for Use

in Aquaculture
Fulgencio Soto, Manuel Jimenez,* Antonio Mateo, Jose A. Villarejo,
Esther de Jodar, and Jacinto Jimenez
Departamento de Tecnologa Electronica, Universidad Politecnica de Cartagena,

Campus Muralla del Mar, s/n 30202 Cartagena, Spain
Abstract
The new equipment described here was developed in response to a shortcoming in the technology available for
electrofishing marine species. The system supplies a need that exists in commercial operations, and also allows
research into the effect of the waveform on the quality of the meat. The equipment is capable of generating any
kind of waveform, identifying the environment the electrodes are in, and logging the voltage and current values that
are applied to kill the fish. Although it has been designed in the context of electrofishing for tuna, it is perfectly
compatible with electrofishing in freshwater. The system is composed of two interacting parts: a software section and
a hardware section. The software is capable of determining the relationship of each of the electrical wave parameters
with the final quality of the meat. The electrical hardware main constituent is a full bridge with an inductor-capacitor
(LC) filter. The effectiveness of two different types of control for this kind of converter is compared: one being a
feedback control, and the other a feedforward control. This equipment has been tested under natural conditions
using three different waveforms, selected on the basis of our previous tuna electrostunning experience. These are:
high-frequency pulsed DC, low-frequency pulsed DC, and low-frequency, decreasing exponential AC. Of the tested
waveforms, low-frequency, decreasing exponential AC showed the best results as far as flesh quality was concerned.
The system showed, in all cases, a high degree of reliability and safety.
Throughout the world, various species of tuna are farmed in

cages. Since 1995, the farming of red tuna Thunnus thynnus has
expanded in the Mediterranean Sea (Miyake et al. 2003). The
main objective is to obtain the best possible quality in order to
achieve the highest possible return. The price of red tuna meat
can vary between US$80 and $280 per kilogram depending on
its quality, and this is strongly influenced by the slaughtering
method.
Usually, farmed tuna have been slaughtered by shooting them
underwater using a power head (Lupara), shooting to the head
from above the water, coring or spiking, or a blow on the head
with a club (Kavatic et al. 2003; EFSA 2009). Those techniques
cause a high degree of stress to the tuna, leading to a loss in
the quality of the meat (De la Gandara 2003; Gregory 2008). To
avoid these negative effects, consideration has been given to the
*Corresponding author: manuel.jimenez@upct.es

Received January 11, 2012; accepted April 24, 2012
468
use of electrical discharge for slaughter (Van de Vis et al. 2003;

Gregory 2008).
The use of electricity for the slaughter of freshwater fish
species has been known for nearly 100 years since the first
official record in 1917, the year in which the first electrofishing
system was patented. The basic method followed in freshwater
electrofishing is shown in Figure 1. The electrodes of the source
of power are introduced into the water, a voltage is applied, and
the electrical field generated stuns or kills the fish.
This method cannot be used in seawater with tuna or other
marine species because of the high level of conductivity of
the water. One cubic meter of freshwater is equivalent to an
electrical resistance of 100 , whereas 1 m3 of seawater is
equivalent to an electrical resistance of only 0.2 . Under such
conditions, to generate a potential gradient capable of stunning
NEW PROGRAMMABLE ELECTROFISHING DEVICE
469
fish are very similar. The most important ones are hemorrhages
and bloodspots, injuries to the nerves and internal tissues
(McMichael 1993), and even more serious damage such as
spinal injuries (Dalbey et al. 1996) (see Figure 2).
Owing to the shortcomings in the technology available for
electrofishing of marine species our research group (Electronic
Engineering and Systems Division) has been working for the last
few years in the development of marine electrofishing devices.
After some preliminary experiences carried out with simple selfmade equipment (Soto et al. 2006), the aim of developing a flexible and programmable electrofishing device was established.
METHODS
FIGURE 1. Basic field of effect on fish of electrofishing method in freshwater.
[Figure available in color online.]
or slaughtering tuna in seawater would require an unreasonable

amount of electric power (Kolz 1989; Lines et al. 2004). Because
of this, the electricity is applied by means of a harpoon that acts
as an electrode inside the fish (Garca 2002).
Many studies regarding the use of electricity in freshwater
can be found in the literature (McMichael 1993; Sharber et al.
1994; Reynolds 1996; Lines 2003, 2005). However, experience
with electrofishing in seawater is much more limited and focused
on small-sized species (Roth 2004; Lambooij 2008; Nordgreen
et al. 2008). The use of electricity for slaughtering large, farmed
red tuna is almost nonexistent. The causes of injury in fish induced by electrofishing are similar in freshwater and seawater,
with the most important factor being the waveform and frequency used.
Despite the differences in method and in the size and weight
of the fish (rainbow trout Oncorhynchus mykiss or salmon,
in comparison with red tuna), the injuries caused by the
application of the electrical discharge within the body of the
Experimental Conditions
Laboratory conditions.The electrofishing device was first
tested in the laboratory using the equipment described below.
Various preliminary tests were performed simulating field conditions such as water, fish or air presence at the output of the
system (harpoon).
The AC power supply was provided by a 220-V (root mean
square), 50-Hz transformer bank connected to the AC main
supplies. Two rheostats were used as load: the first one (110
, 3 A) was used to simulate the fish, and the second one (11
, 3 A) acted as the water. The test in air was simulated with
no load. Tektronix P5200 high-voltage differential probes were
employed to measure voltage. Current measurements were performed using A6302 current probes connected to AM 503B
amplifiers, both from Tektronix. To capture waveforms, a Yokogawa DL1640 color oscilloscope was selected, which provided
four 200-MS/s channels.
Field conditions.The tests described were performed at a
tuna farm located in the southeast of Spain (coast of Region of
Murcia). Each cage of this farm was 30 m deep and 50 m long. At
the beginning of the harvest time each cage contains about 1,000
FIGURE 2. Spinal injuries caused by the application of the electrical discharge in (a) rainbow trout (source: Lamarque 1990) and (b) red tuna using our
equipment. [Figure available in color online.]
470
SOTO ET AL.
FIGURE 3. Example of the slaughtering method used in large tuna tanks for
commercial aquaculture.
specimens. The tests were carried on from November 2008 to

January 2009. At this time, the average water temperature was
16.2 C and conductivity was 55 mS/cm.
The slaughtering process is as follows. Two scuba divers
enter the cage (see Figure 3). One carries a harpoon gun whose
FIGURE 4.
harpoon is connected by a wire to the power converter, and

the other carries an underwater switch. The diver carrying the
harpoon gun selects a tuna for slaughter and fires at it. The
diver controls the power converter and hence the timing of the
discharge with the underwater switch. Although this process
could be automated, the scuba divers feel safer if they have
control of the process.
In order to electroslaughter the tuna, a potential gradient has
to be generated between the inside of its body and the water.
When the harpoon penetrates the tuna (see Figure 4), the current
will seek pathways of lower impedance to the water, and so
the impact area will influence the results. The best point of
impact is located in the lateral body wall, similar to that shown
in Figure 4. The possible routes that the current could follow
inside the tuna are shown in Figure 4. Despite the importance
of the point of impact (location and depth), it is very difficult
to control these variables. Depth is controlled with the harpoon
limiter; however, the point of impact will depend on the divers
experience.
After the discharge to the tuna, the fish is hoisted on
deck. This is done with a crane using a rope that the scuba
divers tie to the tunas tail. On deck, the tuna is bled and
a metallic rod is inserted through the vertebrae to rapidly
kill the fish. This process is also useful in detecting serious spinal injuries, since when such injuries occur, there is a
Harpoon details and current pathways for electroslaughtering tuna.
misalignment of successive vertebrae that blocks the path of

the rod.
The test protocol followed to evaluate each waveform is described below. The tests began using a high voltage (or current)
to achieve a fast electroslaughtering in order to avoid unsafe
situations for the scuba divers. If an optimal (fast and safe) electroslaughtering was not achieved with the first discharge (which
may be a sign of a badly stuck harpoon), the same voltage and
current values were used for the next tuna. If the waveform
continued to be ineffective (i.e., it did not slaughter the tuna
outright) in the subsequent tests, it was discarded and a new
one tested. Otherwise (waveform effective), depending on the
scuba divers comments and the on-deck spinal rod examination, the voltage was either maintained if no spinal injuries were
observed or decreased so as to reduce them. In some cases, the
efficiency of the tests was so low that it was not possible to kill
the tuna, in which case that waveform was quickly rejected. In
other cases, the waveform was rejected because spinal injuries
were too serious and this could involve financial loss for the
company. The size of the tuna tested and the number of samples
for each test day (not for each waveform type) were dependent
on customer orders.
The waveforms that we first tested with the TEFSYS (Tuna
ElectroFishing System) equipment were selected on the basis of
our previous tuna electrostunning experience (Soto et al. 2006).
The waveforms used and their main parameters are shown in
Figure 5. These are high-frequency pulsed DC (HF-PDC), lowfrequency pulsed DC (LF-PDC), and low-frequency, decreasing
exponential AC (LF-DEAC). The first two have been widely
used in freshwater electrofishing (Lamarque 1990; Dalbey 1996;
Ainslie 1998) and previously used by us with ad hoc systems
specifically designed for each waveform.
The third waveform (LF-DEAC), not tested so far, was selected to match the good results previously obtained with lowfrequency waveforms and the regulated AC (Soto et al. 2006).
We also added an exponential decreasing component in order
to achieve a high initial value to stun the tuna followed by an
attenuation with the aim of reducing the damage to the fish.
To assess the waveforms, the following procedure was performed. Besides the first examination carried out on deck, definitive results were obtained and recorded at the company factory.
There, the tuna were filleted in four parts; these were inspected
by an expert who rated the meat quality and the spinal injuries.
The meat quality score (Core) varies from 0 to 5 (5 is the best).
The meat quality index is influenced not only by the color of the
meat but also by the presence of burns, blood clots, and bloodspots. The spinal injury score (RC) varies also from 0 to 5 (0
indicates no spinal injuries). Spinal injuries have also been observed to affect the final quality, since the flesh located around an
injury is affected. The company established two quality thresholds, one for the Core (3.5) and a second one for the RC (<2).
A waveform was accepted if both thresholds were met.
A total of 83 tuna were sacrificed: 23 by means of HF-PDC,
22 by using LF-PDC, and 28 by using LF-DEAC. The number
471
of tests is not very large owing to their high cost (e.g., the price
of an average-sized tuna can reach $3,000).
For each tested waveform the following statistics were calculated: average weight, weight SD, and mean and SD for the
meat quality score (Core) and spinal injury score (RC). Furthermore, the two most representative indicators of the suitability of
the waveform were obtained as well, these being the percentage
of tuna with meat quality score (Core) greater than 3.5 and the
percentage of tuna with spinal injury score (RC) less or equal
to 2.
Description of the Equipment

The multifunctional and programmable electronic equipment, TEFSYS, for use in fish farming that is described here
has been developed to meet the need for versatility. It can be
used in research and, in a complementary form, in commercial
fish farming for both marine and freshwater species, although
its design has been oriented towards use in seawater. The system allows all the electrical variables used in the process to
be recorded in detail. Furthermore, this equipment is versatile
enough to generate a wide range of waveforms (note that the
traditional use of simple waveforms has been dictated by the
lack of electrofishing equipment with the necessary versatility).
The equipment has two differentiated but interacting parts:
a software section and a hardware section. The software allows
the user to select the waveform that will be applied to the fish
through the electrodes, and this is displayed on the user interface.
Sensors allow the software to identify the physical environment
(air, water, or fish) in which the electrodes are situated, and
activation of the electrical discharge is disabled if the electrodes
are in air or, in cases when they are integrated in a harpoon, if
this harpoon is not piercing a fish.
The user interface is shown in Figure 6. At the top are the
controls that configure the waveform, and the form is drawn in
real time in the graph at the right. At the bottom are two differentiated areas. To the left is a set of indicators that show the state of
the discharge switch of the equipment. Also shown is the physical environment (water, air, or fish) surrounding the harpoon
for seawater electrofishing or the physical environment (water
or air) surrounding the electrodes for freshwater electrofishing.
To the right of the figure, the values of applied voltage and current are displayed during the electrical discharge. These values
are stored for later retrieval and analysis. With the use of this
software and the recording of currents and voltages used during
each capture, it is possible to determine the relationship of each
of the parameters of the electrical wave with the final quality of
the meat.
The electrical hardware is shown in Figure 7. Its main constituent is a full bridge with a LC filter. In addition, an isolation
transformer protects the equipment of the boat. The converter
acts as a power amplifier that reproduces the waveform signal
applied. In the output, the two electrodes that apply electrical
discharges to the fish can be integrated in a harpoon or can be
472
SOTO ET AL.
Waveform
Parameters
100
90
80
70
Waveform 1: High frequency pulsed

direct current (HF-PDC)
60
50
Frequency = 1 kHz
Vpp = 100 V
Duty cycle = 50%
40
30
20
10
0,0
2,0m
4,0m
6,0m
8,0m
10,0m
time
12,0m
14,0m
16,0m
18,0m
20,0m
100
90
80
Waveform 2: Low frequency pulsed

direct current (LF-PDC)
70
60
50
Frequency = 10 Hz
Vpp = 100 V
Duty cycle = 50%
40
30
20
10
0
0,0
20,0m
40,0m
60,0m
80,0m
100,0m
time
120,0m
140,0m
160,0m
180,0m
200,0m
100
80
60
Waveform 3: Low frequency

decreasing exponential alternate
current (LF-DAC)
40
20
0
-20
-40
Frequency = 20 Hz
Vpp = 200 V
-60
-80
-100
0,0
200,0m
400,0m
600,0m
800,0m
1,0 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2,0
time
FIGURE 5. Waveforms used to test programable electrofishing device.
physically independent, depending on the mode of operation

that is required.
If this equipment is used for applications in seawater, such
as for the slaughter of tuna, it will be subjected to abrupt variations of load. The harpoon is fired at the tuna and the discharge
is applied. Sometimes the discharge starts before the electrical
harpoon is inside the tuna, and this can cause changes in the
load impedance of the equipment from about 5 (impedance
of the wires) to 50 (impedance of the wires plus the flesh of
the tuna). Further, the harpoon may subsequently become detached from the tuna, or water may be introduced through the
wound, causing a steep change in the impedance of the equipment. Hence, the converter may be subjected to very high and
unpredictable variations in load of up to 10 fold. Nevertheless,
the output signal must remain as close as possible to the reference value under any situation. Because of this, the control
of the converter must be able to maintain the output waveform
selected, despite wide variations in the load.
The effectiveness of two different types of control for this
kind of converter was compared: one being a feedback control,
and the other being a feedforward control by which the load
current is measured. To perform these tests two control boards
FIGURE 6.
User interface of the new programmable electrofishing device. [Figure available in color online.]
were designed, simulated, and implemented for feedback and

feedforward control, respectively. The main electrical features
of the converter to be controlled are shown in Table 1.
RESULTS
Laboratory Results
After calculation and simulation of both control schemes, it
was necessary to implement and test them first at the laboratory.
To test the equipment, a reference signal was applied and was
replayed by the system analyzed with both control types and
TABLE 1.
Characteristics of the power converter to be controlled.
Parameter
Input voltage
Switch mode
Switching frequency
Output voltage
Dynamic response
473
abrupt load variations. The selected reference waveforms were

a positive offset sinusoidal current and a low-frequency pulsed
DC. The watertuna transition was the object of study for each
of the described controls and waveforms. This is the critical
moment of the electrofishing process, when the signal applied
to the tuna must be exactly equal to the chosen reference value;
otherwise, the good or bad quality of the meat obtained might
not be due to the waveform but might result from a power
surge. The results are shown in Figure 8. With a positive offset
sinusoidal current or a low-frequency pulsed DC and feedback
control (Figure 8a, c) large power surges can be seen when
the load switches from the water to the tuna. With the same
waveforms and feedforward control (Figure 8b, d) large power
surges do not occur when the load switches from water to tuna.
Characteristic
150 V
Unipolar
20 kHz
100 V, + 100 V
[0 Hz, 1.5 kHz]
Field Results
The statistical values for the waveforms tested as described in
the Methods are summarized in Table 2. In the tests performed
with the HF-PDC waveform, the meat quality was good but all
the fish tested suffered serious spinal injuries. With the LF-PDC
waveform the meat quality was optimal for most of the samples.
474
SOTO ET AL.
FIGURE 7.
Electrical hardware circuit of the new programmable electrofishing device.
However, in this case the spinal injuries were lower than those
registered with HF-PDC. Finally, in the tests performed with
LF-DEAC, the meat quality was the best, but the percentage of
spinal injuries in the fish was quite high.
TABLE 2. Results obtained for each tested waveform. HF-PDC = highfrequency pulsed DC, LF-PDC = low-frequency pulsed DC, LF-DEAC =
low-frequency, decreasing exponential AC.
Waveform type
Metric assessed
Number of tuna tested
Average weight (kg)
Weight SD (kg)
Tuna with meat quality
score (Core) > 3.5 (%)
Mean meat quality score
(Core)
Meat quality score (Core)
SD
Tuna with spinal injury
score (RC) 2 (%)
Mean spinal injury score
(RC)
Spinal injury score (RC)
SD
HF- PDC LF-PDC LF-DEAC

23
49.9
11.5
87.0
22
178.3
61.3
86.4
28
223.8
86.8
92.9
4.1
4.5
4.5
0.33
0.62
0.47
0.0
81.8
25.0
3.7
1.8
3.1
0.58
0.89
0.97
DISCUSSION
In the feedback control large power surges can be seen when
the load switches from water to tuna, which is damaging to
the fish. In the harpoon it is very important to have an output
signal as similar as possible to the reference signal, since the
electrical discharge is very often applied before the harpoon has
been driven into the fish. In order to find out the influence of the
waveform on the tuna meat, the applied waveform must have
exactly the selected parameters; otherwise, the results could be
attributable to voltage peaks rather than to the waveform itself.
Even though this control functions to keep the output voltage
similar to the reference whatever the load, it is not able to avoid
power surges when the harpoon switches from water to tuna.
As mentioned previously, this may be an important drawback in
experimental assessments of the influence of waveform on the
quality of tuna meat. By using the feedforward control, we have
managed to remove power surges at the switch from water to
tuna. The output voltage practically equals the reference signal.
With regard to the electroslaughtering field tests, the effects
of each waveform in the Core and RC are summarized in Table 2. The results obtained with HF-PDC and LF-PDC are very
similar to those previously reported with our previous ad hoc
systems (Soto et al. 2006). The HF-PDC waveform produced
spinal injuries above the quality threshold (RC 2) in all specimens with a mean value of meat quality worse than the other
two waveforms (low frequency). For the LF-PDC waveform
spinal injuries were minimal with a good meat quality (Core
475
FIGURE 8. Laboratory results. (a) A positive offset sinusoidal current and control feedback. (b) A positive offset sinusoidal current and control feedforward. (c)
A low-frequency pulsed DC and feedback control. (d) A low-frequency pulsed DC and feedforward control. [Figure available in color online.]
mean = 4.5). If we compare these results with those obtained

for other freshwater species, they are very similar (Lamarque
1990; McMichael 1993; Dalbey 1996; Ainslie 1998; Snyder
2003). In view of these results, we discarded HF-PDC for electroslaughtering fish.
There are no previous studies in freshwater or seawater for
the use of LF-DEAC, which makes it interesting to compare it
with the other low frequency waveform tested, LF-PDC.
Although the use of waveforms LF-PDC and LF-DEAC provided an identical meat quality average value (4.5), the LFDEAC stands out because the number of samples exceeding the
quality threshold (Core > 3.5) is higher with a lower SD value.
Therefore, the meat quality resulting from LF-DEAC is the best
of the three waveforms tested. However, it presents a problem:
high spinal injuries. These are due to the high value of peak-topeak voltage (200 Vpp ) applied at the beginning. This voltage
was set to assure the initial stun of the tuna. In the future we
intend to use the same waveform with a lower initial voltage to
further improve the results.
CONCLUSION
A flexible and programmable equipment for electrofishing in
both seawater and freshwater has been presented. The system
supplies a need that exists in commercial operations, and also
allows research into the effect of the waveform on the quality
of the meat. The equipment is capable of generating any kind
of waveform, identifying the environment the electrodes are,
in and logging the voltage and current values that are applied
to kill the fish. Although the equipment has been designed in
the context of electroslaughtering tuna, it is perfectly compatible
with freshwater fishing; for this, the harpoon used for tuna would
be replaced with electrodes.
476
SOTO ET AL.
The control is the crucial aspect of the design, since it must

achieve an output signal that closely follows the reference value,
even under marked load variations. The feedforward control
proved to be more suitable for this application than the feedback control, since the former responds more rapidly to abrupt
load variations. Furthermore, with the feedforward control large
power surges do not occur, which is one of the most important
aspects of this feature.
Hence, we have achieved a robust and easily operated control
at a low development and manufacturing cost. The control multiplier proposed by (Ryan et al. 1995) has been suppressed simply by replacing the proportional controller by a proportionalintegral one. The equipment has been tested in field conditions
using different waveforms that show, in all cases, a high degree
of reliability and safety for the scuba divers.
Of the tested waveforms, LF-DEAC showed the best results
as far as meat quality is concerned, although spinal injuries were
higher than expected due to the high voltage level. In the future,
we intend to conduct more tests to refine the parameters of the
LF-DEAC waveform and evaluate new signals.
ACKNOWLEDGMENTS
This work was supported by the Council of Science, Technology, Industry and Commerce of the Region of Murcia.
REFERENCES
Ainslie, B. J., J. R. Post, and A. J. Paul. 1998. Effects of pulsed and continuous
DC electrofishing on juvenile rainbow trout. North American Journal of
Fisheries Management 18:905918.
Dalbey, S. R., T. E. McMahon, and W. Fredenberg. 1996. Effect of electrofishing
pulse shape and electrofishing-induced spinal injury on long-term growth
and survival of wild rainbow trout. North American Journal of Fisheries
Management 16:560569.
De la Gandara, F. 2003. Son las tareas propias de un cultivo intensivo de peces
mediterraneos compatibles con su bienestar? [Are the tasks of Mediterranean
fish farming compatible with their welfare?] Pages 17 in I Congreso internacional sobre bienestar animal [Animal welfare congress, 2003 AWC record,
volume 1]. Spanish Department of Agriculture, Murcia.
EFSA (European Food and Safety Authority). 2009. Species-specific welfare
aspects of the main systems of stunning and killing of farmed tuna. EFSA
Journal [online serial] 1072. DOI: 10.2903/j.efsa.2009.1072.
Garca, M. 2002. Procedure for electrostunning and/or electrosacrifice, for industrial application to marine inchthyological species in floating cages. U.S.
patent 6,453,596. September 24, 2002.
Gregory, N. G. 2008. Animal welfare at markets and during transport and
slaughter. Meat Science 80:211.
Kavatic, I., V. Ticina, and V. Franicevic. 2003. Bluefin tuna (Thunnus thynnus
L.) farming on the Croatian coast of the Adriatic Seapresent stage and
future plans, volume 60. Pages 101105 in C. R. Bridges, H. Gordin, and A.
Garcia, editors. Domestication of the bluefin tuna, Thunnus thynnus thynnus.

Cahiers Options Mediterraneennes, Cartagena, Spain.
Kolz, A. L. 1989. A power transfer theory for electrofishing. U.S. Fish and
Wildlife Service Fish and Wildlife Technical Report 22.
Lamarque, P., 1990. Electrophysiology of fish in electrics field. Pages 433 in
I. G. Coux and P. Lamarque, editors. Fishing with electricity: applications in
freshwater fisheries management. Fishing News Books, Oxford, UK.
Lambooij, B., M. A. Gerritzen, H. Reimert, D. Burggraaf, G. Andre, and H. Van
De Vis. 2008. Evaluation of electrical stunning of sea bass (Dicentrarchus
labrax) in seawater and killing by chilling: welfare aspects, product quality
and possibilities for implementation. Aquaculture Research 39:5058.
Lines, J., and S. Kestin. 2004. Electrical stunning of fish: the relationship between the electric field strength and water conductivity. Aquaculture 241:219
234.
Lines, J., and S. Kestin. 2005. Electric stunning of trout: power reduction using
a two-stage stun. Aquacultural Engineering 32:483491.
Lines, J. A., D. H. Robb, S. C. Kestin, S. C. Crook, and T. Benson. 2003. Electric
stunning: a humane slaughter method for trout. Aquacultural Engineering
28:141154.
McMichael, G. A. 1993. Examination of electrofishing injury and short-term
mortality in hatchery rainbow trout. North American Journal of Fisheries
Miyake, P. M., J. M. de la Serna, A. Di Natale, A. Farrugia, I. Katavic, N.
Miyabe, and V. Ticina. 2003. General review of bluefin tuna farming in
the Mediterranean area. Collective Volume of Scientific Papers International
Commission for the Conservation of Atlantic Tunas 55:114124.
Nordgreen, A. H., E. Slinde, D. Mller, and B. Roth. 2008. Effect of various
electric field strengths and current durations on stunning and spinal injuries
of Atlantic herring. Journal of Aquatic Animal Health 20:110115.
Reynolds, J. B. 1996. Electrofishing. Pages 221253 in B. R. Murphy and D. W.
Willis, editors. Fisheries techniques. American Fisheries Society, Bethesda,
Maryland.
Roth, B., D. Moeller, and B. Slinde. 2004. Ability of electric field strength,
frequency, and current duration to stun farmed Atlantic salmon and pollock
and relations to observed injuries using sinusoidal and square wave alternating
current. North American Journal of Aquaculture 66:208216.
Ryan, M. J., and R. D. Lorenz. 1995. A high-performance sine wave inverter controller with capacitor current feedback and back-EMF decoupling. Pages
507513 in Power electronics specialists conference, 1995 PESC 95 record.,
26th annual IEEE, volume 1. Institute of Electrical and Electronics Engineering, Atlanta.
Sharber, N. G., S. W. Carothers, J. P. Sharber, J. C. De Vos, Jr., and D. A.
House. 1994. Reducing electrofishing-induced injury of rainbow trout. North
American Journal of Fisheries Management 14:340346.
Snyder, D. E., 2003. Electrofishing and its harmful effects on fish. U.S. Geological Survey, Information and Technology Report USGS/BRD/ITR20030002, Denver.
Soto, F., J. A. Villarejo, A. Mateo, J. Roca-Dorda, F. De la Gandara, and A.
Garca. 2006. Preliminary experiences in the development of bluefin tuna
Thunnus thynnus (L., 1758) electroslaughtering techniques in rearing cages.
Aquacultural Engineering 34:8391.
Van de Vis, H., S. Kestin, D. Robb, J. Oehlenschlager, B. Lambooij, W. Munkner,
H. Kuhlmann, K. Kloosterboer, M. Tejada, A. Huidobro, H. Ottera, B. Roth,
N. K. Sorensen, L. Akse, H. Byrne, and P. Nesvadba. 2003. Is humane
slaughter of fish possible for industry? Aquaculture Research 34:211220.


A Comparative Analysis of the Economic Risk of Hybrid

Striped Bass Fingerling Production in Ponds and
Indoor Tanks
a
Carole Engle & Pratikshya Sapkota
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University

Drive, Mail Slot 4912, Pine Bluff, Arkansas, 71601, USA
To cite this article: Carole Engle & Pratikshya Sapkota (2012): A Comparative Analysis of the Economic Risk of Hybrid Striped
Bass Fingerling Production in Ponds and Indoor Tanks, North American Journal of Aquaculture, 74:4, 477-484


DOI: 10.1080/15222055.2012.685214
ARTICLE
A Comparative Analysis of the Economic Risk of Hybrid

Striped Bass Fingerling Production in Ponds
and Indoor Tanks
Carole Engle* and Pratikshya Sapkota
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University

Drive, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA
Abstract
A risk analysis was conducted to compare the economic risk of producing hybrid striped bass ( white bass Morone
chrysops striped bass M. saxatilis) fingerlings in ponds and indoor tanks. Cost budgets developed previously
for three pond sizes (0.4, 1.2, and 2.4 ha) and three tank sizes (945, 2,457, and 5,670 L) for each of six scales of
production (50,000, 100,000, 250,000, 500,000, 1,000,000, and 2,000,000 fingerlings/year) were used as the initial
starting point for the development of a spreadsheet-based risk analysis. Probability distributions and cumulative
frequency distributions of break-even prices above total costs were calculated for each scenario. Pond production
was less risky than tank production, and larger pond sizes had lower risk. Survival rates contributed the most to the
economic risk of hybrid striped bass fingerling production. Research is needed to improve survival rates for hybrid
striped bass fingerling production in ponds.
Hybrid striped bass (HSB; white bass Morone chrysops

striped bass M. saxatilis), commonly known as sunshine
bass, fingerling production is characterized by fluctuations in
yield that result primarily from varying survival rates in both
ponds and tanks. Risk has been defined as uncertain consequences (Hardaker et al. 1997), as a situation with more than
one possible outcome (Levy 2006), some of which might be
unfavorable (Kay et al. 2008), and as all the possible outcomes
and the probabilities of their occurring (Olson 2004). Thus, risk
exists when the values of key parameters fluctuate (Engle 2010);
fluctuating yields, survival rates, prices, and costs all result in
risk to the farmer. Improved understanding of how various risk
factors affect the economics of hybrid striped bass fingerling
production may provide guidance for management decisions.
The survival rates of hybrid striped bass fingerlings in commercial ponds have been reported to range from 0% to 70%
(Mike Freeze, Keo Fish Farm, personal communication). In indoor tanks, the survival rates of fry through 15 d posthatch
(dph) have ranged from 3% to 94%, depending upon the types

of feed used (Ludwig 1994a, b; Ludwig et al. 2008; Ludwig
and Lochmann 2000, 2009), rotifer concentration (Ludwig and
Lochmann 2000), tank stocking density (Ludwig and Lochmann
2007), and feeding frequency (Ludwig 2003). The coefficients
of variation for survival ranged from 2% to 104% across these
same studies. However, the majority of studies on hybrid striped
bass fingerling production in tanks have focused only on production through 1421 dph, with little work on culture through
the phase-I size (approximately 1 g [3045 d old]).
Varying survival rates and prices of inputs create economic
risk that is reflected in fluctuating break-even prices. Eklund
et al. (2012) demonstrated that the cost of producing hybrid
striped bass fingerlings would be reduced by 714% in ponds
and 511% in tanks for each 5% improvement in survival. However, the static cost analysis developed by Eklund et al. (2012)
did not allow for explicit evaluation of the effects of the risk of
fluctuations in survival and other parameters.
*Corresponding author: cengle@uaex.edu

Received December 9, 2011; accepted April 9, 2012
477
478
ENGLE AND SAPKOTA
A variety of types of spreadsheet-based risk analysis have

been used in the aquaculture economics literature. These studies have used software add-ons (Crystal Ball, Oracle USA, Inc.,
Redwood City, California; @RISK software, Palisade Corporation, Ithaca, New York) to commercially available spreadsheet
software. Several studies have examined risk levels and the
probability of losses due to alternative management choices.
For example, Valderrama and Engle (2001) used survey data to
measure the levels of financial risk at Honduran shrimp farms,
finding that risk levels were higher on small whiteleg shrimp
Penaeus vannamei farms and for farms using intensive production systems. In Mexico, Martinez and Seijo (2001) measured
the risks associated with various water exchange and aeration
rates in shrimp production, identifying the price of postlarvae,
growth, survival rates, and the harvest-size shrimp price as the
major contributors to risk. Partial aeration was the risk-neutral
choice among three aeration regimes for cage culture of channel catfish Ictalurus punctatus in the United States, although
none of the aeration methods was preferable to the others according to first- or second-degree stochastic dominance criteria
(Nelson et al. 2001). Seijo (2004) showed that alternative harvest scheduling of shrimp can reduce risk. Lower probabilities
of financial losses were found with Nile tilapia Oreochromis
niloticus raised in monoculture with pelleted feeds rather than
rice bran or in the culture of sharptooth catfish Clarias gariepinus (Neira et al. 2009).
There are relatively few studies that have examined risk in
the production of intermediate aquaculture products, such as fingerlings or stocker fish. Engle and Valderrama (2002) showed
that economic risk was lower at catfish fingerling stocking
densities that resulted in higher yields. Pomerleau and Engle
(2003) showed that stocking catfish fingerlings at a medium
density resulted in the lowest risk of the break-even costs of
stocker production being above market prices. The survival rate
of fingerlings was identified as a critical factor in costs and
risk.
Other analyses have examined the feasibility of new species,
investments, production systems, or types of farm organization
from the perspective of risk. For example, Medley et al. (1994)
showed that the economic feasibility of producing Australian
red claw crawfish Cherax quadricarinatus was sensitive to various risk factors, including the cost of the juveniles stocked, the
production of larger crawfish, price, and the length of the growing season. Net present values for shrimp farming investments
in Florida were found to be negative, even with price premiums
(Clark et al. 2010). In Rwanda, fish farming was shown to entail more risk than farming traditional crops such as cabbages
and beans (Hishamunda et al. 1998). The Hishamunda et al.
(1998) study also looked at the form of organization and found
that cooperatives had a higher probability of failure than individually managed farms. Zucker and Anderson (1999) used a
spreadsheet-based risk modeling approach to identify the production and marketing targets most likely to result in economic
success for newly developed technologies in indoor, land-based
summer flounder Paralichthys dentatus systems. Their findings

showed that such facilities should be located near a source of
saltwater and that medium-sized fish products should be sold
mostly to premium market outlets.
Much of the work done on the economics of hybrid striped
bass production has studied costs at the food fish stage. Enterprise budgets have been developed (Dunning and Daniels 2001),
the costs of pond and tank production compared (Gempesaw
et al. 1992b), and the costs of two-and three-phase systems
(DAbramo et al. 2002, 2004) and economies of scale measured
in ponds (Gempesaw et al. 1992a). The costs of various effluent treatment options (Wui and Engle 2004; Sydorovich and
Daniels 2011) and the economic effects of restrictions on access to black carp Mylopharyngodon piceus have been evaluated
(Wui and Engle 2007).
Eklund et al. (2012) developed a comprehensive cost analysis of hybrid striped bass fingerling production in ponds and
indoor tank systems. Three pond sizes (0.4, 1.2, and 2.4 ha) and
three sizes of tanks (945, 2,457, and 5,670 L) were modeled
at six scales of production (50,000, 100,000, 250,000, 500,000,
1,000,000, and 2,000,000). The cost per thousand fingerlings
produced in ponds was estimated to range from US$86 to
$107, compared with $260 to $385 per thousand fingerlings
in tanks. Pond production was the lowest-cost production system. Economies of scale were identified in pond production and,
to some degree, in tank production; costs per thousand fingerlings decreased with increasing numbers of production cycles
per year.
The overall goal of this project was to compare the economic
risk of producing hybrid striped bass fingerlings in various sizes
of ponds and tanks and at various scales of production. The specific objectives were to (1) estimate probability density functions of the break-even prices of hybrid striped bass fingerling
production in various sizes of ponds (0.4, 1.2, and 2.4 ha) and
tanks (945, 2,457, and 5,670 L) for production scales of 50,000,
100,000, 250,000, 500,000, 1,000,000, or 2,000,000 fingerlings
per year; (2) identify the parameters that contribute the most to
the economic risk of hybrid striped bass fingerling production in
ponds and tanks; and (3) compare the economic risks of hybrid
striped bass fingerling production in ponds and tanks.
METHODS
The Eklund et al. (2012) budgets for 15 pond and 17 tank
production scenarios were used as the starting point for development of a risk analysis. Data were compiled for the following
key parameters: (1) feed prices from 1990 to 2009 (Hanson
and Sites 2010); (2) the federal funds rate (Federal Reserve
Board; www.federalreserve.gov) as a measure of the variability in interest rates; (3) wage rates (Bureau of Labor Statistics;
www.bls.gov); (4) electric and gasoline/diesel rates (Department of Energy; www.eia.doe.gov); and (5) the prices of fertilizer, salt, and lime (U.S. Department of Agriculture, Economic
Research Service; www.ers.usda.gov).
479
ECONOMIC RISK OF HYBRID STRIPED BASS FINGERLING PRODUCTION
TABLE 1.
Assumptions for the risk analysis of hybrid striped bass fingerling production in tanks and ponds; n.a. = not applicable.
Variable
Distribution
Survival rate
Triangular
Feeds
$/kg
Normal
Uniform
Uniform
Uniform
Electricity
$/ha
$/h (fry)
$/L
(rotifers)
$/L, $/ha
Gas and diesel
$/ha
Normal
Fertilizer
$/ha
Normal
Lime
Normal
Salt
$/metric
ton
$/kg
normal
Pumping
$/ha
Uniform
Repairs and
maintenance
Oxygen
$/L, $/ha
Triangular
Triangular
Shipping bags
$/tank
rental
$/box
Triangular
Bird control
$/ha
Triangular
Telephone
$/ha
Triangular
Office supplies
$/ha
Triangular
Legal,
accounting
Interest rates
$/ha
Triangular
Normal
Rotifer cysts
$/million
Normal
Rotifer diet
0.5 kg
Normal
Labor
Unit
Uniform
Data set used

Ludwig (2004,
2006)
Hanson and
Sites (2010)
BLSc
BLSc
BLSc
Observation
Department of
Energye
Department of
Energye
None
Department of
Agriculturef
Department of
Agriculturef
Department of
Agriculturef
Department of
Energye
Eklund et al.
(2012)
Eklund et al.
(2012)
Eklund et al.
(2012)
Eklund et al.
(2012)
Eklund et al.
(2012)
Eklund et al.
(2012)
Eklund et al.
(2012)
Federal
Reserve
Boardg
Eklund et al.
(2012)
Eklund et al.
(2012)
Minimum, maximum, and likeliest values used for the triangular distributions.
Mean SD used for the normal distributions.
c
Bureau of Labor Statistics (www.bls.gov).
d
Minimum and maximum used for the uniform distributions.
e
www.eia.doe.gov.
f
www.ers.usda.gov.
g
www.federalreserve.gov
b
Statistical test
used for
normality
Tanks
Ponds
0, 35, 70a
0, 35, 70a
ShapiroWilk
$3.94 0.17
n.a.
6.9, 8.0d
635, 736.1d
n.a.
n.a.
n.a.
0.11, 0.16d
$3.938 0.169b
412.5, 478.20d
n.a.
n.a.
240, 352.63d
Kolmogorov
Smirnov D
test
None
325 73.625b
ShapiroWilk
2,373.8 346.2b
ShapiroWilk
510 2.5b
0.21 0.03b
527.5, 755.1d
None
n.a.
n.a.
n.a.
0.21 0.03b
0.02, 0.06,
0.03a
18.7, 31.2,
25a
45, 75, 60a
125, 485, 242.5a

338, 562, 450a
45, 75, 60a
11.7, 31.2, 15.6a
n.a.
n.a.
37.5, 85, 42.5a

24.7, 27.5, 27.5a
ShapiroWilk
42.3, 70.5,
47.0a
10 5b
42.3, 70.5, 47.0a
380 16.3
n.a.
21.8 0.9
n.a.
10 5b
480
ENGLE AND SAPKOTA
A spreadsheet-based risk analysis was developed in which

distributions estimated from data sets of fluctuating values for
various parameters were substituted for single-point values. Previous risk analyses primarily used triangular distributions for
parameters with fluctuating values. Nelson et al. (2001) used
triangular distributions for all risky parameter values, Neira
et al. (2009) for all but the feed conversion ratio (FCR) and
growth rate, and Valderrama and Engle (2001) for all but yield
and shrimp price. Pomerleau and Engle (2003) used normal distributions for FCR, yield, survival, and fish length but a uniform
distribution for feed price.
In the present study, the data for interest rates and the prices
of feed, lime, and salt were determined to be distributed normally by the ShapiroWilk test (SAS version 9.2), and fuel
and diesel prices were found to be distributed normally by the
KolmogorovSmirnov test using the EasyFit Statistical Program (version 5.4, MathWave Technologies, Dnepropetrovsk,
Ukraine). Minimum wage legislation in the United States establishes a price floor for wages. A uniform distribution was used
that established minimum ($412/ha) and maximum ($478/ha)
rates and assumed that all values within the range were equally
likely. The maximum rate was established based upon the annual average rate of increase in wage rates projected forward
for 5 years. In the same vein, electric rates are not distributed
normally because the price floors are set by public utility commissions. A uniform distribution was established for electric
rates, with the minimum rate being set as current charges and
the maximum rate by projecting forward the annual average
rates of increase for 5 years. Table 1 lists the distributions and
the values used for each parameter.
Monte Carlo simulations with 1,000 iterations were conducted with Crystal Ball to generate probability distributions of
the break-even prices (BEPs) of hybrid striped bass fingerling
production for each of the 32 pond/tank scenarios. The probability of a BEP above total costs being higher than the mean
price in the budget was calculated for each scenario. Cumulative
distribution functions of BEPs above total costs were developed
and presented graphically to show whether first-degree stochastic dominance was present. First-degree stochastic dominance
is present if every point on a cumulative distribution function
is preferable to the corresponding point on another cumulative
distribution function (Levy 2006).
RESULTS
Of the production and cost parameters associated with the
production of hybrid striped bass fingerlings, survival rate was
the most variable, followed by the federal funds rate and fertilizer prices (Figure 1). Gasoline and diesel prices were the
fourth most variable parameters and wage rates the fifth, while
less variation was observed in the prices of lime, electricity, and
salt.
Probability distributions of the break-even prices above total
costs for hybrid striped bass fingerling production were devel-
FIGURE 1. Coefficients of variation for the key parameters in the spreadsheetbased risk analysis.
oped for each of the 32 production scenarios; examples of two

(1 million fingerlings produced in 1.2-ha ponds and in 2,457-L
tanks) are provided in Figure 2, using the convention of expressing frequency of occurrence as probability values. All probability distributions were positively skewed, indicating that there
are proportionately more observations of very high costs of production than of very low costs of production for both pond and
tank scenarios. The tank scenarios, however, had greater probabilities of higher-than-average costs than the pond scenarios
and reflect the overall higher break-even prices for tanks.
FIGURE 2. Probability distributions of break-even prices above total costs for

1 million hybrid striped bass fingerlings in (a) a 1.2-ha pond and (b) a 2,457-L
tank. The values on the y-axis are estimated probabilities of occurrence.
481
TABLE 2. Certainty levels of break-even prices (BEPs; $/1,000 fingerlings)

being higher than the most efficient (lowest-cost) pond sizeproduction scale
combination; n.a. = not applicable.
Pond production
scenario
50,000
100,000
250,000
500,000
1,000,000
2,000,0000
50,000
100,000
250,000
500,000
1,000,000
2,000,0000
FIGURE 3. Cumulative distribution functions of break-even prices above total

costs for 1 million hybrid striped bass fingerlings/year in (a) ponds and (b)
tanks of different sizes. The values on the y-axis are estimated probabilities of
occurrence.
Pond size affected the economic risk of producing hybrid

striped bass fingerlings (Figure 3a). The cumulative distribution functions of BEPs above total costs demonstrate that 1.2-ha
and 2.4-ha ponds exhibit first-order stochastic dominance over
0.4-ha ponds (lines to the left reflect lower production costs for
every given probability), indicating higher economic risk in 0.4ha ponds when holding production constant. Thus, the probabilities of costs being higher than the most efficient (lowest-cost)
pond sizescale combination were higher for the 0.4-ha ponds
(5172%) than for the larger pond sizes (5259% for the 1.2-ha
ponds and 5054% for the 2.4-ha ponds) (Table 2). The cumulative distribution functions for the different sizes of tanks were
more similar to each other, but again the larger tank sizes tended
to dominate when production was held constant (Figure 3b).
The scale of production had little effect on the economic
risk of producing hybrid striped bass fingerlings in ponds (Figure 4a), but some scale effects were found in tank production
(Figure 4b). The cumulative distribution functions were nearly
identical across production scales for ponds, indicating similar
amounts of risk. With tanks, however, the cumulative distribution functions showed first-degree stochastic dominance of
several production scales over others: (1) the highest scales of
production (100,0002 million) dominated the 50,000 scale; (2)
the 100,000, 500,000, 1 million, and 2 million scales dominated
50,000
100,000
250,000
500,000
1,000,000
2,000,0000
BEP
0.4-ha ponds
107
93
89
88
88
86
1.2-ha ponds
n.a.
93
89
88
88
86
2.4-ha ponds
n.a.
n.a.
89
88
88
86
Certainty level
(%)
51
64
68
72
66
68
n.a.
52
55
57
56
59
n.a.
n.a.
54
50
52
54
the 250,000 scale; and (3) the 500,000, 1 million, and 2 million
scales dominated the 100,000 scale (Figure 4b). Thus, larger
scales of production generally entailed less risk for a given tank
size (with the exception of the 100,000 scale dominating the
250,000 scale).
Since tank size and production scale affected the risk of producing hybrid striped bass fingerlings in tanks, the probabilities
of production costs being higher than in the most efficient scenario generally increased with production scale in the small
(945-L) tanks and generally decreased with production scale in
the larger tanks (Table 3). For the lowest production scale, the
probability of higher costs was lower for the smallest tank size.
Tank production of hybrid striped bass fingerlings entailed
higher levels of risk than did pond production (Figure 5). Pond
production dominated tank production by first-degree stochastic
dominance for all scales of production; the probability of tank
production of hybrid striped bass fingerlings costing less than
pond production was 0%.
Survival rate was the greatest source of economic risk for
hybrid striped bass fingerling production in ponds (9496%)
and tanks (9698%) (Tables 4 and 5). The price of lime and
interest rates were the next most important factors contributing
to the risk of raising hybrid striped bass fingerlings in ponds,
whereas electric rates and the price of rotifer cysts were the next
482
ENGLE AND SAPKOTA
FIGURE 5. Comparative economic risks of hybrid striped bass fingerlings

produced in ponds and tanks. The values on the y-axis are estimated probabilities
of occurrence
FIGURE 4. Cumulative distribution functions of break-even prices above total

costs for different scales of production in (a) a 1.2-ha pond and (b) a 2,457-L
tank. The values on the y-axis are estimated probabilities of occurrence.
most important in the case of tanks, although these contributed

substantially less to risk than did the survival rate.
DISCUSSION
Increasing pond size reduced the economic risk of hybrid
striped bass fingerling production. Larger (1.2-ha and 2.4-ha)
ponds entailed less risk than did 0.4-ha ponds at each scale of
production. The economies of larger pond sizes are well documented in pond aquaculture production (Keenum and Waldrop
1988; Engle 2003, 2007, 2010); smaller ponds are associated
with higher per-unit fixed costs of production, which result in
greater financial risk. However, farmers must also consider that

larger ponds can be more difficult to manage because more
hectares are tied up in each management decision.
There have been few studies on the risk of tank aquaculture
production and few that have compared the costs and risk associated with varying sizes of tanks. In this study, tank size had a
small though measurable effect on risk. Recirculating systems
include several component parts that must be sized proportionately (Losordo et al. 1999). Components of sizes that are optimal
for the varying tank sizes are not always available commercially.
When several individual components have excess capacity, the
system operates less efficiently, increasing overall financial risk
(Dunning et al. 1998). This is why the 100,000 production scale
demonstrated slightly less risk than the 250,000 scale. For the
component sizes available on the market, the 100,000 scale used
these components relatively more efficiently. Contracting an engineer to construct all component parts would be prohibitively
expensive for a hybrid striped bass hatchery seeking to add a
recirculating system as just one part of a larger farming operation. Increasing the scale of tank production had less effect on
risk, although some effects were found, particularly with regard
to low scales of production.
Overall, pond production entailed less risk than tank production, primarily due to the increased costs associated with tank
TABLE 3. Certainty levels of break-even prices (BEPs; $/1,000 fingerlings) being higher than the most efficient (lowest-cost) tank sizeproduction scale
combination; n.a. = not applicable.
Certainty level (%) by tank size

Production scale
BEP
945 L
2,457 L
5,670 L
50,000
100,000
250,000
500,000
1,000,000
2,000,000
337
305
275
260
269
269
53
57
59
62
57
61
64
49
66
57
53
55
n.a.
66
51
50
50
52

TABLE 4. Percent contribution of various parameters to the risk of producing
hybrid striped bass fingerlings in different sizes of ponds at a scale of 1 million
fingerlings per cycle.
Parameter
Survival
Lime
Interest rate (real estate)
Legal, accounting
Interest rate (equipment)
Electricity
0.4 ha
1.2 ha
2.4 ha
96
1
1
<1
<1
<1
94
2
<1
<1
<1
<1
96
1
<1
<1
<1
<1
production. This supports findings of Eklund et al. (2012) that

pond production is economically preferable to tank production.
As with the Eklund et al. (2012), study results may differ for
palmetto bass ( striped bass Morone saxatilis white bass
M. chrysops). Tuncer et al. (1990) showed that palmetto bass do
not require the costly rotifer feeding stage. However, sunshine
bass are preferred by industry due to the difficulties of inducing striped bass females to spawn (Hodson et al. 1999) and the
longer time to reach sexual maturity.
Varying survival rates contributed the most to the economic
risk of producing hybrid striped bass fingerlings. Pomerleau and
Engle (2003), in a study of catfish stocker production, also found
that survival-related risks can be critical factors in the overall
cost and risk of producing intermediate aquaculture products.
Engle and Valderrama (2001) found lower economic risks in
catfish fingerling production with management strategies that
resulted in higher yields. Higher yields are affected indirectly
by the survival rate because higher densities are associated with
higher yields. Valderrama and Engle (2001) also found that the
yield risk of shrimp was the primary source of risk. Martinez and
Seijo (2001) explicitly identified survival rate as a partial source
of risk in shrimp production in Mexico. Additional attention
to the effects of survival on production risk is warranted in
the aquaculture economics literature. The hybrid striped bass
literature has few reports of survival through to the phase-I size,
in spite of the number of studies comparing the survival of larval
stages to 1421 dph.
TABLE 5. Percent contribution of various parameters to the risk of producing
hybrid striped bass fingerlings in different sizes of tanks at a scale of 1 million
fingerlings per cycle.
Parameter
Survival
Interest rate (equipment)
Feed (rotifer cyst)
Feed (48% protein)
Salt
Electricity
945 L
2,457 L
5,670 L
97
<1
1
<1
<1
<1
98
<1
<1
<1
<1
<1
96
<1
<1
<1
<1
2
483
CONCLUSIONS
The economic risk of producing hybrid striped bass fingerlings was found to be higher in tanks than in ponds, indicating
that pond production is preferable to tank production. Larger
pond sizes reduced risk, but farmers must also consider the
practical management limits to increasing pond size to produce
hybrid striped bass fingerlings. Survival rate was the greatest
contributing factor to the economic risk of hybrid striped bass
fingerling production. There is great need for research to identify
ways to improve survival through to the phase-I size in ponds.
ACKNOWLEDGMENTS
This study was funded in part by USDAARS Specific Cooperative Agreement 58-6225-8-036. Peter Wui, Anita Kelly,
and Ganesh Karunakaran provided helpful comments on the
manuscript.
REFERENCES
Clark, J. L., R. N. Weldon, C. M. Adams, and F. F. Wirth. 2010. Risk assessment
of a shrimp aquaculture investment in Florida. Aquaculture Economics and
DAbramo, L. R., C. L. Ohs, and T. R. Hanson. 2002. Production and economic
analysis of two-phase and three-phase culture of sunshine bass in earthen
ponds. North American Journal of Aquaculture 64:103112.
DAbramo, L. R., C. L. Ohs, and T. R. Hanson. 2004. Effect of stocking weight
and stocking density on production of hybrid striped bass (sunshine) in earthen
ponds in the second phase of a 2-phase system. Journal of the World Aquaculture Society 35:3345.
Dunning, R., and H. Daniels. 2001. Hybrid striped bass production in ponds:
enterprise budget. Southern Regional Aquaculture Center, Publication 3000,
Stoneville, Mississippi.
Dunning, R. D., R. M. Losordo, and A. O. Hobbs. 1998. The economics of
recirculating tank systems: a spreadsheet for individual analysis. Southern
Regional Aquaculture Center, Publication 456, Stoneville, Mississippi.
Eklund, P., C. Engle, and J. Ludwig. 2012. Comparative cost analysis of hybrid
striped bass fingerling production in ponds and tanks. North American Journal
of Aquaculture 74:3953.
Engle, C. R. 2003. The evolution of farm management, production efficiencies,
and current challenges to catfish production in the United States. Aquaculture
Economics and Management 7:6784.
Engle, C. R. 2007. Species-specific public policy for sustainable development:
the U.S. catfish industry. Pages 313332 in P.-S. Leung, C.-S. Lee, and P.
OBryen, editors. Species and system selection for sustainable aquaculture.
Blackwell Scientific Publications, Oxford, UK.
Engle, C. R. 2010. Aquaculture economics and financing: management and
analysis. Wiley-Blackwell Scientific Publications, Ames, Iowa.
Engle, C. R. and D. Valderrama. 2001. Effect of stocking density on production
characteristics, costs, and risk of producing fingerling channel catfish. North
Engle, C. R. and D. Valderrama. 2002. Production characteristics, costs, and
risk of producing channel catfish, Ictalurus punctatus, fingerlings on farms
with thinning. Journal of Applied Aquaculture 12:5164.
Gempesaw, C. M., J. R. Bacon, and F. F. Wirth. 1992a. Economics of pond size
for hybrid striped bass growout. Journal of the World Aquaculture Society
23:3848.
Gempesaw, C. M., D. Lipton, V. Varma, and J. R. Bacon. 1992b. A comparative
economic analysis of hybrid striped bass production in ponds and tanks. Pages
3141 in N. Stone and V. Pennington, editors. Proceedings of the National
484
ENGLE AND SAPKOTA
Aquaculture Extension Workshop. Arkansas Cooperative Extension Program,

University of Arkansas at Pine Bluff, Pine Bluff.
Hanson, T. R., and D. Sites. 2010. Catfish production data. Mississippi State
University, Stoneville.
Hardaker, J. B., R. B. M. Huirne, and J. R. Anderson. 1997. Coping with risk
in agriculture. CABI Publishing, New York.
Hishamunda, N., C. M. Jolly, and U. Hatch. 1998. Evaluating and managing
risk in small-scale fish farming in a developing economy: an application to
Rwanda. Aquaculture Economics and Management 2:3138.
Hodson, R. J., R. W. Clark, M. S. Hopper, A. S. McGinty, G. M. Weber, and C.
V. Sullivan. 1999. Pages 2332 in T. Tamaru, C. S. Tamaru, J. McVey, and K.
Ikuta, editors. Reproduction of domesticated striped bass: commercial mass
production of fingerlings. University of Hawaii Sea Grant College Program,
United States and Japan Cooperation Program in Natural Resources, Report
28, Honolulu.
Kay, R. D., W. M. Edwards, and P. A. Duffy. 2008. Farm management. McGrawHill, New York.
Keenum, M. E. and J. E. Waldrop. 1988. Economic analysis of farm-raised
catfish production in Mississippi. Mississippi Agriculture and Forestry Experiment Station Technical Bulletin 155.
Levy, H. 2006. Stochastic dominance: investment decision making under uncertainty. Springer, New York.
Losordo, T. M., M. P. Masser, and J. E. Rakocy. 1999. Recirculating aquaculture
tank production systems: a review of component options. Southern Regional
Aquaculture Center, Publication 453, Stoneville, Mississippi.
Ludwig, G. M. 1994a. Tank culture of sunshine bass Morone chrysops
M. saxatilis fry with freshwater rotifers Brachionus calyciflorus and salmon
starter meal as first food sources. Journal of the World Aquaculture Society
25:337341.
Ludwig, G. M. 1994b. Rearing sunshine bass fry on tank-cultured freshwater
rotifers. Aquaculture Magazine (September/October):3945.
Ludwig, G. M. 2003. Tank culture of larval sunshine bass, Morone chrysops
(Rafinesque) M. saxatilis (Walbaum) at three feeding levels. Aquaculture
Research 34:12771285.
Ludwig, G. M., and S. Lochmann. 2000. Culture of sunshine bass Morone
chrysops M. saxatilis, fry in tanks with zooplankton cropped from ponds
with a drum filter. Journal of Applied Aquaculture 10:1126.
Ludwig, G. M., and S. E. Lochmann. 2007. Effect of tank stocking density
on larval sunshine bass growth and survival to the fingerling stage. North
Ludwig, G. M., and S. E. Lochmann. 2009. Tank culture of sunshine bass
fingerlings without using rotifers. North American Journal of Aquaculture
71:224228.
Ludwig, G. M., S. D. Rawles, and S. E. Lochmann. 2008. Effects of rotifer

enrichment on sunshine bass Morone chrysops M. saxatilis larvae growth
and survival and fatty acid composition. Journal of the World Aquaculture
Society 39:158173.
Martinez, J. A., and J. C. Seijo. 2001. Economics of risk and uncertainty of alternative water exchange and aeration rates in semi-intensive
shrimp culture systems. Aquaculture Economics and Management 5:
129145.
Medley, P. B., R. G. Nelson, U. Hatch, D. B. Rouse, and G. F. Pinto. 1994.
Economic feasibility and risk analysis of Australian red claw crayfish Cherax
quadricarinatus aquaculture in the southeastern United States. Journal of the
World Aquaculture Society 25:135146.
Neira, I., C. R. Engle, and C. Nguji. 2009. Economic and risk analysis of tilapia
production in Kenya. Journal of Applied Aquaculture 21:110.
Nelson, R. G., S. A. Duarte, and M. Amasser. 2001. Financial risk analysis of
three aeration regimes in catfish cage culture. Aquaculture Economics and
Olson, K. D. 2004. Farm management: principles and strategies. Iowa State
Press, Ames.
Pomerleau, S., and C. R. Engle. 2003. Production of stocker-size channel catfish:
effect of stocking density on production characteristics, costs, and economic
risk. North American Journal of Aquaculture 65:112119.
Seijo, J. C. 2004. Risk of exceeding bioeconomic limit reference points in
shrimp aquaculture systems. Aquaculture Economics and Management 8:
201212.
Sydorovych, O., and H. Daniels. 2011. Economic analysis of alternative effluent
treatment options for pond production of hybrid striped bass in Aurora, North
Carolina. Aquaculture Economics and Management 15:4670.
Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Acceptance and consumption
of food by striped bass and hybrid larvae. Journal of the World Aquaculture
Society 21:225234.
Valderrama, D., and C. R. Engle. 2001. Risk analysis of shrimp farming in
Honduras. Aquaculture Economics and Management 5:4968.
Wui, Y. S., and C. R. Engle. 2004. A mixed integer programming analysis of effluent treatment options proposed for pond production of hybrid striped bass,
Morone chrysops M. saxatilis. Journal of Applied Aquaculture 15:121
158.
Wui, Y.-S., and C. R. Engle. 2007. The economic impact of restricting use of
black carp for snail control on hybrid striped bass farms. North American
Zucker, D. A., and J. L. Anderson. 1999. A dynamic, stochastic model of a
land-based summer flounder Paralichthys dentatus aquaculture firm. Journal
of the World Aquaculture Society 30:219235.


An Ultrastructure Study of Diet-Related Changes in

Epithelial Tissue of Hybrid Striped Bass Larvae
Margie L. Gallagher
College of Human Ecology, East Carolina University, RW 238 Rivers Building, Greenville,
North Carolina, 27858-4353, USA
To cite this article: Margie L. Gallagher (2012): An Ultrastructure Study of Diet-Related Changes in Epithelial Tissue of Hybrid
Striped Bass Larvae, North American Journal of Aquaculture, 74:4, 489-493


DOI: 10.1080/15222055.2012.676021
COMMUNICATION
An Ultrastructure Study of Diet-Related Changes

in Epithelial Tissue of Hybrid Striped Bass Larvae
Margie L. Gallagher*
North Carolina 27858-4353, USA
Abstract
This study characterized the ultrastructure of normal epidermal and gastrointestinal epithelial cells of larval hybrid striped bass
(white bass Morone chrysops striped bass M. saxatilis) that were
fed live prey (brine shrimp Artemia spp. nauplii) or an artificial dry
diet. Scanning electron microscopy of larvae that were given live
feed revealed normal epidermal cells with microridge structures
and proliferation of neuromasts along the lateral line. Cell junctions had double microridge structures. Transmission electron microscopy of the gut showed keratinized epithelial cells with underlying mucous cells in the foregut, while the midgut was characterized
by columnar epithelial cells with extensive pinocytotic activity. The
hindgut had columnar epithelial cells with centrally located nuclei,
regularly spaced microvilli, and numerous tubular mitochondria
surrounded by rough endoplasmic reticulum. Larvae that received
live feed reached metamorphosis by 27 d posthatch (dph), and
clearly identifiable gastric glands were present. Larvae that were
given the artificial diet did not develop; both epidermal and gastrointestinal cells showed apparent osmotic stress by 10 dph, and
the condition worsened through 24 dph. Stress was indicated by loss
of microridges in epidermal cells and eventual necrosis. Increased
keratinization and reduced pinocytotic activity were observed in
midgut cells, whereas hindgut epithelial cells showed dissociation
of the rough endoplasmic reticulum and reduced numbers of mitochondria. However, tight cell junctions, desmosomes, and double
microridges at cell junctions persisted.
High mortality associated with the period of first feeding to

metamorphosis in striped bass Morone saxatilis and striped bass
hybrids remains problematic for the culture of these fish despite
advances in microencapsulation technologies and culture methods (Langdon and Barrows 2011). Striped bass and hybrids have
a short yolk sac period, with larval stages that possess a very
rudimentary digestive tract (i.e., lacking a functional stomach or
gastric glands) and that must undergo complex metamorphosis
of the digestive system to survive. Ohs et al. (1998) and Ohs
(1995) suggested that the poor survival of striped bass and white
bass Morone chrysops striped bass hybrids (hereafter, hybrid
striped bass) that were offered even sophisticated artificial diets

was due to rapid gut evacuation or a lack of appropriate digestive enzyme activity (or both) in the premetamorphic larvae.
Digestive enzymes are known to be active in larval striped bass
from the time of hatch (Baragi and Lovell 1986); the notable exception is pepsin, which is not produced until the gastric glands
of the stomach become differentiated. Feed technology development efforts have sought to address this issue. The purpose
of the present study was to characterize the normal structure
of the epithelial cells of premetamorphic larval hybrid striped
bass and the changes associated with the use of dry artificial
diets. Epithelial cells of the gastrointestinal (GI) tract mucosa
and epithelial cells of the epidermis (which are continuous with
those of the GI tract) were examined.
METHODS
Hybrid striped bass (age = 4 d posthatch [dph]) were obtained
from Keo Fish Farms (Keo, Arkansas) and immediately placed
into one of nine 40-L cylindrical tanks with rounded bottoms; the
tanks were fitted with center standpipes and were supplied with
gentle aeration via one air stone, similar to the system described
by Denson and Smith (1997). Larvae were given one of three
feed types: (1) newly hatched live brine shrimp Artemia spp.
nauplii, (2) a dry feed consisting of 10% fish silage (prepared
according to Gallagher 1994) and 90% commercially prepared
larval feed, or (3) no feed.
Ten larval fish were sampled from each tank at 4, 10, 20, and
27 dph. As suggested by Perez-Dominguez and Dahm (2011),
larval status was measured based on the striped bass development stages described by Rogers et al. (1982). The stages
are based primarily on yolk sac disappearance; eye, GI, and
fin development; and swimming and feeding behavior. Six larvae were prepared for transmission electron microscopy (TEM)
and scanning electron microscopy (SEM). Whole larvae were
fixed in 1% formalin and 3% glutaraldehyde. Samples were
*E-mail: gallagherm@ecu.edu
Received January 4, 2012; accepted February 17, 2012
489
490
GALLAGHER
washed in sodium cacodylate, postfixed with 1% osmium

tetraoxide, and dehydrated in alcohol. For TEM, specimens
were dissected into foregut, midgut, and hindgut sections and
embedded in Epon 812 resin. Specimen blocks were trimmed,
and silver sections were prepared for examination by TEM using a Phillips CM-10 high-resolution scope. For SEM, whole,
fixed larvae were mounted on aluminum stubs and were sputter
coated with platinum for observation by using an International
Scientific Instruments ISI-40 or a Quanta 200 scanning electron
microscope. To be noted in this study, ultrastructure details had
to occur repeatedly in all larval samples.
Larval fish that were fed live brine shrimp nauplii developed
normally and progressed through the morphological stages ah
described by Rogers et al. (1982). The larvae achieved transformation (stage h; differentiation of rays in dorsal and pectoral fins; initiation of lateral line scales) by 27 dph, exhibiting
clearly identifiable gastric glands with numerous tubular mitochondria surrounded by rough endoplasmic reticulum (RER)
and dense secretion granules (Figure 1) similar to those described by Cataldi et al. (2002) for 12-dph Adriatic sturgeon
Acipenser naccarii. However, larvae that were starved or that
were fed the dry diet did not develop morphologically past stage
d (active pelagic feeding, well-developed mouth, and initial fin
differentiation) despite the fact that food could be seen in the GI
tracts of fish that were given the dry feed. Starved larvae began
to die at 6 dph, and all had died by 19 dph; larvae that received
dry feed did not begin to die until 10 dph, but all had died by
24 dph.
FIGURE 1. Cross section of gastric glands of a hybrid striped bass larva (at
27 d posthatch) that was fed live brine shrimp nauplii; secretion granules (sg)
and microtubule structures (arrows) are apparent. Numerous mitochondria are
also present (scale bar = 1 m). [Figure available in color online.]
Epithelial Cells of the Epidermis

Hybrid striped bass larvae at 4 dph were at stage c (oil globule and yolk nearly absorbed; swimming pelagically) when
they arrived at the laboratory. The epidermis, including the
mouth, exhibited typical microridge structure as described by
Ostrander (2000), and neuromasts were present as noted by
Yufera (2011).
Bereiter-Hahn et al. (1979) described the microridge structures found on the surface of fish epidermis cells over 30 years
ago and proposed a link between microridges and microvilli of
the intestinal mucosa, which is also epithelial tissue that is continuous with the epidermis. The function of microridges remains
unclear; certainly, they provide structural support, as the primary
components are actin and alpha-actinin. However, microridges
also increase surface area and may facilitate the exchange of
gases and ions (Rulifson et al. 1986). Sperry and Wassersug
(2005) proposed that microridges hold protective mucous secretions to the epidermis, thus providing a barrier between the
fish and the surrounding environment.
By 10 dph, larvae that received live feed had progressed to
stage e (yolk sac absorbed; pectoral fins and teeth [covered with
epithelial cells, Figure 2] visible). The epidermis cells were
smaller (<10 m) and continued to exhibit typical microridge
structures (Figure 3A). There was a continuous double microridge structure at the cell junctions; tight junction complexes
and numerous cellular invaginations were present. Larvae that
were given dry feed or that were starved had advanced only to
stage d, and their cells demonstrated apparent osmotic stress
as evidenced by poorly defined microridges and the looseness
of the cell membrane (Figure 3B, C). However, the continuous
double microridge structure persisted in these fish, even
when other microridge structures were scarce or nonexistent
(Figure 3B inset). Starved larvae died by day 19. All fish exhibited the formation of neuromasts (Figure 3) along the lateral line.
FIGURE 2. Epithelial cells of the mouth of a hybrid striped bass larva (at 27
d posthatch) that received a diet of live brine shrimp nauplii; teeth can be seen
emerging from the covering of epithelial cells with microridges (arrows; scale
bar = 10 m).
COMMUNICATION
491
FIGURE 4. Epithelial cells along the lateral line of a hybrid striped bass larva
(at 20 d posthatch) that was fed a dry artificial diet. Note the persistence of
neuromasts (arrow), even when many cells appear to be necrotic (scale bar =
10 m). [Figure available in color online.]
By 20 dph, larvae that were given the live feed had progressed to
stage f or g (differentiation of fins, rudimentary spines, and gas
bladder present) and the neuromasts had proliferated. Larvae
that received dry feed remained at stage d, and numerous epidermal cells exhibited necrosis, although neuromasts persisted
(Figure 4).
FIGURE 3. Epithelial cells along the lateral line, with neuromasts present, in
hybrid striped bass larvae (at 10 d posthatch): (A) a larva that was given live
brine shrimp nauplii as feed, (B) a larva that was fed a dry artificial diet, and (C) a
starved larva. Note the stressed epidermal cells with sparse microridges (arrows;
scale bar = 10 m). Inset in (B) depicts a transmission electron micrograph
(21,000 ) of the epidermis, showing the double microridge structure at the
cellular junction ( ), which persisted even in stressed cells. [Figure available in
color online.]
Epithelial Cells of the Gastrointestinal Tract

At 10 and 20 dph, the foregut epithelial cells of larvae that
were given live feed appeared to be keratinized, with underlying
mucous-secreting (goblet) cells (Figure 5A). As was noted by
Lazo et al. (2011), mucous cells are typical of even undifferentiated esophageal tissue in many fish species and are thought
to serve as a lubricant since fish do not have salivary glands.
Although in some species the goblet cells are often found scattered throughout the intestine at the onset of exogenous feeding
(Lazo et al. 2011), goblet cells were not observed beyond the
foregut in these hybrid striped bass larvae. Instead, the midgut
was characterized by cells with sparse microvillus structures
and many vacuoles that appeared to be pinocytotic in nature
(Figure 5B). The posterior gut was characterized by columnar
epithelial cells, with numerous, regularly spaced microvilli, numerous mitochondria, and extensive RER typical of intestinal
cells (Figure 5B). Mitochondria in all cells contained tubular
rather than lamellar cristae and were similar to mitochondria
that are characteristic of endocrine cells (Bozzola and Russell 1992). The mitochondria were surrounded closely by dense
RER (Figure 5B inset). In the developing larvae before stomach
differentiation, an intestinal-type cell could often be observed
immediately posterior to a pinocytotic-type cell as in Figure 5B.
A few enteroendocrine cells were also observed in this section
of the gut. At 10 dph, larvae that were fed the dry diet had the
same basic features of the gut as did larvae that were given live
feed. In 20-dph larvae that received the dry diet, the midgut cells
GALLAGHER
492
FIGURE 6. Transmission electron micrographs of gastrointestinal cells in hybrid striped bass larvae (at 20 d posthatch) that received dry artificial feed: (A)
pinocytotic cells, showing a marked decline in activity, few or no mitochondria,
numerous free ribosomes ( ), the presence of keratin-like fibrils (black arrow),
the occurrence of tight junctions between cells, and the persistence of desmosomes (white arrow; scale bar = 1 m); and (B) microvillus cells, exhibiting
much fewer mitochondria, rough endoplasmic reticulum that was rounded or
nonexistent, and the presence of free ribosomes ( ; scale bar = 1 m).
showed a high degree of keratinization and pinocytotic activity

had ceased (Figure 6A). Intestinal-type cells did not show normal RER, and mitochondria were sparse. Ribosomes appeared
to be free or associated in spherical clusters (Figure 6B).
FIGURE 5. Epithelial cells in the intestinal tracts of hybrid striped bass larvae
(at 20 d posthatch) that received a diet of live brine shrimp nauplii: (A) foregut,
characterized by highly keratinized (arrow) cells with underlying mucous cells
(mu; scale bar = 10 m); and (B) midgut, characterized by columnar epithelial
cells with centrally located nuclei; the anterior occurrence of pinocytotic cells
with sparsely spaced microvilli, extensive tight junctions (black arrow), and
desmosomes (white arrows); and (immediately posterior to pinocytotic cells)
intestinal absorptive-type cells with slender microvilli (scale bar = 1 m) and
large numbers of tubular mitochondria surrounded by extensive rough endoplasmic reticulum (RER). Inset in (B) shows a transmission electron micrograph
(6,600 ) of tubular mitochondria (mi) and RER.
Conclusions
In the limited number of larval fish that were observed, both
SEM and TEM revealed several aspects of the epithelial cells
of the epidermis and gut during the premetamorphosis period
in larvae that were given a live feed or a dry artificial diet. Microridges and neuromasts were evident by 4 dph. Microridges
of the epidermis were continuous with the gut, as revealed by
scanning electron micrographs of the mouth. Three cell types
were evident in the gut, including highly keratinized cells in
the foregut, pinocytotic cells in the midgut, and intestine-like
cells immediately posterior to the pinocytotic cells. Pinocytotic cells, which presumably develop into gastric glands of the
COMMUNICATION
stomach at metamorphosis, could be seen lying immediately

next to intestine-like cells in the premetamorphic larvae. The
lack of development in larvae that were fed the dry feed may
be due to a lack of digestive enzymes, particularly pepsin, and
a lack of GI development, as has been suggested by others
(Dabrowski and Culver 1991; Ohs 1995; Langdon and Barrows
2011). In addition, I hypothesize that the lack of development in
the hybrid striped bass larvae may have been related to osmotic
stress due to the ingestion of dry feed with a high osmotic load
or (as expressed by Ohs et al. 1998) a high density. In larvae that
received the dry artificial diet, both the epidermal and gut epithelial cells exhibited osmotic stress, as indicated by the loss of
microridge structure, dissociation of the RER, and eventual cell
necrosis. The GI tract is known to be involved in osmoregulation
in fish (Allen et al. 2009); however, more studies are needed to
confirm this idea. I also observed that in larvae receiving the dry
diet, the neuromasts and double microridges at cell junctions
persisted even under stressful conditions, thus demonstrating
the significance of these structures for the integrity of larval
epithelial cells.
REFERENCES
Allen, P. A., J. J. Check Jr., and D. Kultz. 2009. Mechanisms of seawater
acclimations in a primitive, anadromous fish, the green sturgeon. Journal of
Comparative Physiology Part B 179:903920.
Baragi, V., and R. T. Lovell. 1986. Digestive enzyme activities in striped bass
from first feeding through larva development. Transactions of the American
Fisheries Society 15:478484.
Bereiter-Hahn, J., M. Osborn, K. Weber, and M. Voth. 1979. Filament organization and formation of microridges at the surface of fish epidermis. Journal
of Ultrastructure Research 69:316330.
Bozzola, J. J., and L. D. Russell. 1992. Electron microscopy. Jones and Bartlett,
Boston.
Cataldi, E., C. Albano, D. Boglione, L. Dini, G. Monaco, P. Bronzi, and S.
Cataudella. 2002. Acipenser naccarii: fine structure of the alimentary canal
493
with references to its ontogenesis. Journal of Applied Ichthyology 18:329

337.
Dabrowski, K., and D. Culver. 1991. The physiology of larval fish. Digestive
tract and formulation of starter diets. Aquaculture Magazine 17:4961.
Denson, M. R., and T. I. J. Smith. 1997. Tank culture of larval sunshine bass.
Progressive Fish-Culturist 59:5963.
Gallagher, M. L. 1994. Preliminary evaluation of fish silage as a weaning diet
for hybrid striped bass. Bulletin of the Aquaculture Association of Canada
1:2628.
Langdon, C., and R. Barrows. 2011. Microparticulate diets: technology. Pages
335351 in G. J. Holt, editor. Larval fish nutrition. Wiley, West Sussex,
UK.
Lazo, J. P., M. J. Darias, and E. Gisbert. 2011. Ontogeny of the digestive tract.
Pages 546 in G. J. Holt, editor. Larval fish nutrition. Wiley, West Sussex,
UK.
Ohs, C. L. 1995. The development and evaluation of a spray-dried artificial diet
for larval culture of freshwater prawn (Macrobrachium rosenbergii), hybrid
striped bass (Morone saxatilis M. chrysops) and striped bass (M. saxatilis).
Masters thesis. Mississippi State University, Mississippi State.
Ohs, C. L., L. R. DAbramo, R. K. Buddington, H. R. Robinette, J. M. Roethke.
1998. Evaluation of a spray-dried artificial diet for larval culture of freshwater prawn, Macrobrachium rosenbergii, and striped bass, Morone saxatilis.
Aquaculture Nutrition 4:7382.
Ostrander, G. K. 2000. The laboratory fish. Academic Press, London.
Perez-Dominguez, R., and R. Dahm. 2011. Methods for assessing embryonic
and larval growth in fish. Pages 373402 in G. J. Holt, editor. Larval fish
nutrition. Wiley, West Sussex, UK.
Rogers, B. A., D. T. Westin, and S. B. Saila. 1982. Development of techniques
and methodology for the laboratory culture of striped bass, Morone saxatilis.
U.S. Environmental Protection Agency, Report PB-820217795, Washington, D.C.
Rulifson, R. A., J. E. Cooper, and G. Colombo. 1986. Development of feed
and starved striped bass (Morone saxatilis) larvae from the Roanoke River,
North Carolina. North Carolina Department of Natural Resources and Community Development, Albemarle-Pamlico Estuary System Project 9012,
Raleigh.
Sperry, D. G., and R. J. Wassersug. 2005. Proposed function for microridges on
epithelial cells. Anatomical Record 185:253258.
Yufera, M. 2011. Feeding behavior in larval fish. Pages 285305 in G. J. Holt,
editor. Larval fish nutrition. Wiley, West Sussex, UK.


An Ultrastructure Study of Diet-Related Changes in

Epithelial Tissue of Hybrid Striped Bass Larvae
Margie L. Gallagher
North Carolina, 27858-4353, USA
To cite this article: Margie L. Gallagher (2012): An Ultrastructure Study of Diet-Related Changes in Epithelial Tissue of Hybrid
Striped Bass Larvae, North American Journal of Aquaculture, 74:4, 489-493


DOI: 10.1080/15222055.2012.676021
COMMUNICATION
An Ultrastructure Study of Diet-Related Changes

in Epithelial Tissue of Hybrid Striped Bass Larvae
Margie L. Gallagher*
North Carolina 27858-4353, USA
Abstract
This study characterized the ultrastructure of normal epidermal and gastrointestinal epithelial cells of larval hybrid striped bass
(white bass Morone chrysops striped bass M. saxatilis) that were
fed live prey (brine shrimp Artemia spp. nauplii) or an artificial dry
diet. Scanning electron microscopy of larvae that were given live
feed revealed normal epidermal cells with microridge structures
and proliferation of neuromasts along the lateral line. Cell junctions had double microridge structures. Transmission electron microscopy of the gut showed keratinized epithelial cells with underlying mucous cells in the foregut, while the midgut was characterized
by columnar epithelial cells with extensive pinocytotic activity. The
hindgut had columnar epithelial cells with centrally located nuclei,
regularly spaced microvilli, and numerous tubular mitochondria
surrounded by rough endoplasmic reticulum. Larvae that received
live feed reached metamorphosis by 27 d posthatch (dph), and
clearly identifiable gastric glands were present. Larvae that were
given the artificial diet did not develop; both epidermal and gastrointestinal cells showed apparent osmotic stress by 10 dph, and
the condition worsened through 24 dph. Stress was indicated by loss
of microridges in epidermal cells and eventual necrosis. Increased
keratinization and reduced pinocytotic activity were observed in
midgut cells, whereas hindgut epithelial cells showed dissociation
of the rough endoplasmic reticulum and reduced numbers of mitochondria. However, tight cell junctions, desmosomes, and double
microridges at cell junctions persisted.
High mortality associated with the period of first feeding to

metamorphosis in striped bass Morone saxatilis and striped bass
hybrids remains problematic for the culture of these fish despite
advances in microencapsulation technologies and culture methods (Langdon and Barrows 2011). Striped bass and hybrids have
a short yolk sac period, with larval stages that possess a very
rudimentary digestive tract (i.e., lacking a functional stomach or
gastric glands) and that must undergo complex metamorphosis
of the digestive system to survive. Ohs et al. (1998) and Ohs
(1995) suggested that the poor survival of striped bass and white
bass Morone chrysops striped bass hybrids (hereafter, hybrid
striped bass) that were offered even sophisticated artificial diets

was due to rapid gut evacuation or a lack of appropriate digestive enzyme activity (or both) in the premetamorphic larvae.
Digestive enzymes are known to be active in larval striped bass
from the time of hatch (Baragi and Lovell 1986); the notable exception is pepsin, which is not produced until the gastric glands
of the stomach become differentiated. Feed technology development efforts have sought to address this issue. The purpose
of the present study was to characterize the normal structure
of the epithelial cells of premetamorphic larval hybrid striped
bass and the changes associated with the use of dry artificial
diets. Epithelial cells of the gastrointestinal (GI) tract mucosa
and epithelial cells of the epidermis (which are continuous with
those of the GI tract) were examined.
METHODS
Hybrid striped bass (age = 4 d posthatch [dph]) were obtained
from Keo Fish Farms (Keo, Arkansas) and immediately placed
into one of nine 40-L cylindrical tanks with rounded bottoms; the
tanks were fitted with center standpipes and were supplied with
gentle aeration via one air stone, similar to the system described
by Denson and Smith (1997). Larvae were given one of three
feed types: (1) newly hatched live brine shrimp Artemia spp.
nauplii, (2) a dry feed consisting of 10% fish silage (prepared
according to Gallagher 1994) and 90% commercially prepared
larval feed, or (3) no feed.
Ten larval fish were sampled from each tank at 4, 10, 20, and
27 dph. As suggested by Perez-Dominguez and Dahm (2011),
larval status was measured based on the striped bass development stages described by Rogers et al. (1982). The stages
are based primarily on yolk sac disappearance; eye, GI, and
fin development; and swimming and feeding behavior. Six larvae were prepared for transmission electron microscopy (TEM)
and scanning electron microscopy (SEM). Whole larvae were
fixed in 1% formalin and 3% glutaraldehyde. Samples were
*E-mail: gallagherm@ecu.edu
Received January 4, 2012; accepted February 17, 2012
489
490
GALLAGHER
washed in sodium cacodylate, postfixed with 1% osmium

tetraoxide, and dehydrated in alcohol. For TEM, specimens
were dissected into foregut, midgut, and hindgut sections and
embedded in Epon 812 resin. Specimen blocks were trimmed,
and silver sections were prepared for examination by TEM using a Phillips CM-10 high-resolution scope. For SEM, whole,
fixed larvae were mounted on aluminum stubs and were sputter
coated with platinum for observation by using an International
Scientific Instruments ISI-40 or a Quanta 200 scanning electron
microscope. To be noted in this study, ultrastructure details had
to occur repeatedly in all larval samples.
Larval fish that were fed live brine shrimp nauplii developed
normally and progressed through the morphological stages ah
described by Rogers et al. (1982). The larvae achieved transformation (stage h; differentiation of rays in dorsal and pectoral fins; initiation of lateral line scales) by 27 dph, exhibiting
clearly identifiable gastric glands with numerous tubular mitochondria surrounded by rough endoplasmic reticulum (RER)
and dense secretion granules (Figure 1) similar to those described by Cataldi et al. (2002) for 12-dph Adriatic sturgeon
Acipenser naccarii. However, larvae that were starved or that
were fed the dry diet did not develop morphologically past stage
d (active pelagic feeding, well-developed mouth, and initial fin
differentiation) despite the fact that food could be seen in the GI
tracts of fish that were given the dry feed. Starved larvae began
to die at 6 dph, and all had died by 19 dph; larvae that received
dry feed did not begin to die until 10 dph, but all had died by
24 dph.
FIGURE 1. Cross section of gastric glands of a hybrid striped bass larva (at
27 d posthatch) that was fed live brine shrimp nauplii; secretion granules (sg)
and microtubule structures (arrows) are apparent. Numerous mitochondria are
also present (scale bar = 1 m). [Figure available in color online.]
Epithelial Cells of the Epidermis

Hybrid striped bass larvae at 4 dph were at stage c (oil globule and yolk nearly absorbed; swimming pelagically) when
they arrived at the laboratory. The epidermis, including the
mouth, exhibited typical microridge structure as described by
Ostrander (2000), and neuromasts were present as noted by
Yufera (2011).
Bereiter-Hahn et al. (1979) described the microridge structures found on the surface of fish epidermis cells over 30 years
ago and proposed a link between microridges and microvilli of
the intestinal mucosa, which is also epithelial tissue that is continuous with the epidermis. The function of microridges remains
unclear; certainly, they provide structural support, as the primary
components are actin and alpha-actinin. However, microridges
also increase surface area and may facilitate the exchange of
gases and ions (Rulifson et al. 1986). Sperry and Wassersug
(2005) proposed that microridges hold protective mucous secretions to the epidermis, thus providing a barrier between the
fish and the surrounding environment.
By 10 dph, larvae that received live feed had progressed to
stage e (yolk sac absorbed; pectoral fins and teeth [covered with
epithelial cells, Figure 2] visible). The epidermis cells were
smaller (<10 m) and continued to exhibit typical microridge
structures (Figure 3A). There was a continuous double microridge structure at the cell junctions; tight junction complexes
and numerous cellular invaginations were present. Larvae that
were given dry feed or that were starved had advanced only to
stage d, and their cells demonstrated apparent osmotic stress
as evidenced by poorly defined microridges and the looseness
of the cell membrane (Figure 3B, C). However, the continuous
double microridge structure persisted in these fish, even
when other microridge structures were scarce or nonexistent
(Figure 3B inset). Starved larvae died by day 19. All fish exhibited the formation of neuromasts (Figure 3) along the lateral line.
FIGURE 2. Epithelial cells of the mouth of a hybrid striped bass larva (at 27
d posthatch) that received a diet of live brine shrimp nauplii; teeth can be seen
emerging from the covering of epithelial cells with microridges (arrows; scale
bar = 10 m).
COMMUNICATION
491
FIGURE 4. Epithelial cells along the lateral line of a hybrid striped bass larva
(at 20 d posthatch) that was fed a dry artificial diet. Note the persistence of
neuromasts (arrow), even when many cells appear to be necrotic (scale bar =
10 m). [Figure available in color online.]
By 20 dph, larvae that were given the live feed had progressed to
stage f or g (differentiation of fins, rudimentary spines, and gas
bladder present) and the neuromasts had proliferated. Larvae
that received dry feed remained at stage d, and numerous epidermal cells exhibited necrosis, although neuromasts persisted
(Figure 4).
FIGURE 3. Epithelial cells along the lateral line, with neuromasts present, in
hybrid striped bass larvae (at 10 d posthatch): (A) a larva that was given live
brine shrimp nauplii as feed, (B) a larva that was fed a dry artificial diet, and (C) a
starved larva. Note the stressed epidermal cells with sparse microridges (arrows;
scale bar = 10 m). Inset in (B) depicts a transmission electron micrograph
(21,000 ) of the epidermis, showing the double microridge structure at the
cellular junction ( ), which persisted even in stressed cells. [Figure available in
color online.]
Epithelial Cells of the Gastrointestinal Tract

At 10 and 20 dph, the foregut epithelial cells of larvae that
were given live feed appeared to be keratinized, with underlying
mucous-secreting (goblet) cells (Figure 5A). As was noted by
Lazo et al. (2011), mucous cells are typical of even undifferentiated esophageal tissue in many fish species and are thought
to serve as a lubricant since fish do not have salivary glands.
Although in some species the goblet cells are often found scattered throughout the intestine at the onset of exogenous feeding
(Lazo et al. 2011), goblet cells were not observed beyond the
foregut in these hybrid striped bass larvae. Instead, the midgut
was characterized by cells with sparse microvillus structures
and many vacuoles that appeared to be pinocytotic in nature
(Figure 5B). The posterior gut was characterized by columnar
epithelial cells, with numerous, regularly spaced microvilli, numerous mitochondria, and extensive RER typical of intestinal
cells (Figure 5B). Mitochondria in all cells contained tubular
rather than lamellar cristae and were similar to mitochondria
that are characteristic of endocrine cells (Bozzola and Russell 1992). The mitochondria were surrounded closely by dense
RER (Figure 5B inset). In the developing larvae before stomach
differentiation, an intestinal-type cell could often be observed
immediately posterior to a pinocytotic-type cell as in Figure 5B.
A few enteroendocrine cells were also observed in this section
of the gut. At 10 dph, larvae that were fed the dry diet had the
same basic features of the gut as did larvae that were given live
feed. In 20-dph larvae that received the dry diet, the midgut cells
GALLAGHER
492
FIGURE 6. Transmission electron micrographs of gastrointestinal cells in hybrid striped bass larvae (at 20 d posthatch) that received dry artificial feed: (A)
pinocytotic cells, showing a marked decline in activity, few or no mitochondria,
numerous free ribosomes ( ), the presence of keratin-like fibrils (black arrow),
the occurrence of tight junctions between cells, and the persistence of desmosomes (white arrow; scale bar = 1 m); and (B) microvillus cells, exhibiting
much fewer mitochondria, rough endoplasmic reticulum that was rounded or
nonexistent, and the presence of free ribosomes ( ; scale bar = 1 m).
showed a high degree of keratinization and pinocytotic activity

had ceased (Figure 6A). Intestinal-type cells did not show normal RER, and mitochondria were sparse. Ribosomes appeared
to be free or associated in spherical clusters (Figure 6B).
FIGURE 5. Epithelial cells in the intestinal tracts of hybrid striped bass larvae
(at 20 d posthatch) that received a diet of live brine shrimp nauplii: (A) foregut,
characterized by highly keratinized (arrow) cells with underlying mucous cells
(mu; scale bar = 10 m); and (B) midgut, characterized by columnar epithelial
cells with centrally located nuclei; the anterior occurrence of pinocytotic cells
with sparsely spaced microvilli, extensive tight junctions (black arrow), and
desmosomes (white arrows); and (immediately posterior to pinocytotic cells)
intestinal absorptive-type cells with slender microvilli (scale bar = 1 m) and
large numbers of tubular mitochondria surrounded by extensive rough endoplasmic reticulum (RER). Inset in (B) shows a transmission electron micrograph
(6,600 ) of tubular mitochondria (mi) and RER.
Conclusions
In the limited number of larval fish that were observed, both
SEM and TEM revealed several aspects of the epithelial cells
of the epidermis and gut during the premetamorphosis period
in larvae that were given a live feed or a dry artificial diet. Microridges and neuromasts were evident by 4 dph. Microridges
of the epidermis were continuous with the gut, as revealed by
scanning electron micrographs of the mouth. Three cell types
were evident in the gut, including highly keratinized cells in
the foregut, pinocytotic cells in the midgut, and intestine-like
cells immediately posterior to the pinocytotic cells. Pinocytotic cells, which presumably develop into gastric glands of the
COMMUNICATION
stomach at metamorphosis, could be seen lying immediately

next to intestine-like cells in the premetamorphic larvae. The
lack of development in larvae that were fed the dry feed may
be due to a lack of digestive enzymes, particularly pepsin, and
a lack of GI development, as has been suggested by others
(Dabrowski and Culver 1991; Ohs 1995; Langdon and Barrows
2011). In addition, I hypothesize that the lack of development in
the hybrid striped bass larvae may have been related to osmotic
stress due to the ingestion of dry feed with a high osmotic load
or (as expressed by Ohs et al. 1998) a high density. In larvae that
received the dry artificial diet, both the epidermal and gut epithelial cells exhibited osmotic stress, as indicated by the loss of
microridge structure, dissociation of the RER, and eventual cell
necrosis. The GI tract is known to be involved in osmoregulation
in fish (Allen et al. 2009); however, more studies are needed to
confirm this idea. I also observed that in larvae receiving the dry
diet, the neuromasts and double microridges at cell junctions
persisted even under stressful conditions, thus demonstrating
the significance of these structures for the integrity of larval
epithelial cells.
REFERENCES
Allen, P. A., J. J. Check Jr., and D. Kultz. 2009. Mechanisms of seawater
acclimations in a primitive, anadromous fish, the green sturgeon. Journal of
Comparative Physiology Part B 179:903920.
Baragi, V., and R. T. Lovell. 1986. Digestive enzyme activities in striped bass
from first feeding through larva development. Transactions of the American
Bereiter-Hahn, J., M. Osborn, K. Weber, and M. Voth. 1979. Filament organization and formation of microridges at the surface of fish epidermis. Journal
of Ultrastructure Research 69:316330.
Bozzola, J. J., and L. D. Russell. 1992. Electron microscopy. Jones and Bartlett,
Boston.
Cataldi, E., C. Albano, D. Boglione, L. Dini, G. Monaco, P. Bronzi, and S.
Cataudella. 2002. Acipenser naccarii: fine structure of the alimentary canal
493
with references to its ontogenesis. Journal of Applied Ichthyology 18:329

337.
Dabrowski, K., and D. Culver. 1991. The physiology of larval fish. Digestive
tract and formulation of starter diets. Aquaculture Magazine 17:4961.
Denson, M. R., and T. I. J. Smith. 1997. Tank culture of larval sunshine bass.
Gallagher, M. L. 1994. Preliminary evaluation of fish silage as a weaning diet
for hybrid striped bass. Bulletin of the Aquaculture Association of Canada
1:2628.
Langdon, C., and R. Barrows. 2011. Microparticulate diets: technology. Pages
335351 in G. J. Holt, editor. Larval fish nutrition. Wiley, West Sussex,
UK.
Lazo, J. P., M. J. Darias, and E. Gisbert. 2011. Ontogeny of the digestive tract.
Pages 546 in G. J. Holt, editor. Larval fish nutrition. Wiley, West Sussex,
UK.
Ohs, C. L. 1995. The development and evaluation of a spray-dried artificial diet
for larval culture of freshwater prawn (Macrobrachium rosenbergii), hybrid
striped bass (Morone saxatilis M. chrysops) and striped bass (M. saxatilis).
Masters thesis. Mississippi State University, Mississippi State.
Ohs, C. L., L. R. DAbramo, R. K. Buddington, H. R. Robinette, J. M. Roethke.
1998. Evaluation of a spray-dried artificial diet for larval culture of freshwater prawn, Macrobrachium rosenbergii, and striped bass, Morone saxatilis.
Aquaculture Nutrition 4:7382.
Ostrander, G. K. 2000. The laboratory fish. Academic Press, London.
Perez-Dominguez, R., and R. Dahm. 2011. Methods for assessing embryonic
and larval growth in fish. Pages 373402 in G. J. Holt, editor. Larval fish
nutrition. Wiley, West Sussex, UK.
Rogers, B. A., D. T. Westin, and S. B. Saila. 1982. Development of techniques
and methodology for the laboratory culture of striped bass, Morone saxatilis.
U.S. Environmental Protection Agency, Report PB-820217795, Washington, D.C.
Rulifson, R. A., J. E. Cooper, and G. Colombo. 1986. Development of feed
and starved striped bass (Morone saxatilis) larvae from the Roanoke River,
North Carolina. North Carolina Department of Natural Resources and Community Development, Albemarle-Pamlico Estuary System Project 9012,
Raleigh.
Sperry, D. G., and R. J. Wassersug. 2005. Proposed function for microridges on
epithelial cells. Anatomical Record 185:253258.
Yufera, M. 2011. Feeding behavior in larval fish. Pages 285305 in G. J. Holt,
editor. Larval fish nutrition. Wiley, West Sussex, UK.


Effect of Copper Sulfate on Aeromonas hydrophila

Infection in Channel Catfish Fingerlings
a
Julie Bebak , Julio C. Garcia & Ahmed Darwish
U.S. Department of Agriculture, Agriculture Research Service, Aquatic Animal Health

Research Unit, 990 Wire Road, Auburn, Alabama, 36830, USA
b
U.S. Department of Agriculture, Food Safety and Inspection Service, 620 Central Avenue,
Building 2C, Alameda, California, 94501, USA
To cite this article: Julie Bebak, Julio C. Garcia & Ahmed Darwish (2012): Effect of Copper Sulfate on Aeromonas hydrophila
Infection in Channel Catfish Fingerlings, North American Journal of Aquaculture, 74:4, 494-498


American Fisheries Society 2012
DOI: 10.1080/15222055.2012.685212
ARTICLE
Effect of Copper Sulfate on Aeromonas hydrophila Infection

in Channel Catfish Fingerlings
Julie Bebak* and Julio C. Garcia
U.S. Department of Agriculture, Agriculture Research Service, Aquatic Animal Health Research Unit,
990 Wire Road, Auburn, Alabama 36830, USA
Ahmed Darwish
U.S. Department of Agriculture, Food Safety and Inspection Service, 620 Central Avenue, Building 2C,
Alameda, California 94501, USA
Abstract
Motile Aeromonas septicemia results from primary or secondary infection with bacteria from Aeromonas spp.,
including Aeromonas hydrophila. Since 2009, an emerging strain of A. hydrophila has been associated, as a primary
pathogen, with significant morbidity and mortality in the U.S. catfish industry. Two 2 2 factorial experiments
with five replicates were conducted to evaluate whether copper sulfate pentahydrate (CuSO4 ) at a concentration of
1% of total alkalinity (total alkalinity = 98 mg/L as CaCO3 ; total hardness = 60 mg/L as CaCO3 ; pH = 7.4) can
reduce mortality of channel catfish Ictalurus punctatus after their exposure to this emerging strain of A. hydrophila.
In experiment 1, fingerling channel catfish received an 18-h continuous bath exposure to CuSO4 after A. hydrophila
challenge. Survival in the treatments challenged with A. hydrophila, both when exposed or unexposed to CuSO4 , was
significantly lower than survival in sham-exposed controls. Fish exposed to A. hydrophila and treated with copper
sulfate had the lowest percent survival, at 18% (SE, 7.0), and survival was significantly different from the treatment
in which fish were exposed to A. hydrophila but not treated with copper sulfate. In experiment 2, fish received a 4-h
pretreatment with CuSO4 before exposure to A. hydrophila plus a 4-h treatment the next day. In experiment 2, when
fish were exposed to A. hydrophila but not CuSO4 , survival was 80.0% (SE, 5.5). For fish exposed to A. hydrophila and
to CuSO4 , survival was 50.0% (SE, 3.2). The percent mortality in the treatment exposed to A. hydrophila and to CuSO4
was signficantly different from all of the other treatments. This study demonstrated that, under these experimental
conditions, CuSO4 application reduced survival when used as a treatment for infection of fingerling channel catfish
with this strain of A. hydrophila.
Motile Aeromonas septicemia (MAS) affects cultured and

wild warmwater fish worldwide. The MAS results from infection with bacteria from Aeromonas spp., including Aeromonas
hydrophila, which are ubiquitous in the environment. Species
that cause MAS typically act as secondary pathogens and can
cause significant losses in the presence of predisposing stressors.
However, these bacteria can also be primary pathogens (Plumb
1999). Since 2009, A. hydrophila has been responsible for high
rates of mostly acute, but also chronic, morbidity and mortality
in the Alabama catfish industry. Losses of tens of millions
*Corresponding author: jbebakwilliams@gmail.com

Received September 30, 2011; accepted April 9, 2012
494
of pounds of fish have been reported (Hemstreet 2011). This

emerging strain of A. hydrophila can apparently cause significant morbidity and mortality without any predisposing factors.
A hemorrhagic septicemia is associated with this infection that
includes external and internal hemorrhagic lesions in skin and
in viscera (data of L. Khoo and colleagues [paper presented at
the 35th Annual Eastern Fish Health Workshop, 2010]).
In the USA, there are three antibiotics that may be added
to fish feed to treat A. hydrophila infection in catfish Ictalurus
spp. (USFWS AADAP 2011). First, oxytetracycline dihydrate
EFFECT OF COPPER SULFATE ON AEROMONAS HYDROPHILA INFECTION
(Terramycin 200 for Fish, Phibro Animal Health, Ridgefield

Park, New Jersey) is labeled for use under the previous name
for A. hydrophila (Aeromonas liquefaciens). Second, recognizing that U.S. Food and Drug Administration Guidelines (US
FDA 2001) must be followed, a veterinarian may choose to use
Sulfadimethoxine/Ormetoprim (Romet TC, PHARMAQ AS,
Overhalla, Norway) as an extra-label drug. Third, from 2009
through 2011 catfish farmers could participate in the Investigational New Animal Drug Approval process to use florfenicol
(Aquaflor; Schering-Plough Animal Health Corporation, Summit, New Jersey) medicated feed.
A few studies that report the effect of application of copper
sulfate pentahydrate (CuSO4 ) as a treatment for external and
systemic bacterial infections have been published. Griffin
and Mitchell (2007) reported that channel catfish Ictalurus
punctatus fingerlings exposed to copper sulfate for 24 h before
challenge with Edwardsiella ictaluri experienced significantly
reduced mortality. Darwish et al. (2012) found that catfish
fingerlings that received a 4-h CuSO4 treatment given 5.5 h after
exposure to Flavobacterium columnare were significantly more
likely to survive the infection. In contrast, Thomas-Jinu and
Goodwin (2004) found that applying an indefinite bath treatment of CuSO4 starting 20 h after exposure to F. columnare was
not effective at reducing mortality in channel catfish fingerlings.
The purpose of this study was to evaluate whether CuSO4
has an effect on mortality from exposure to this emerging
strain of A. hydrophila. Experiment 1 (Exp 1) was conducted to
determine whether an 18-h continuous bath exposure to CuSO4
results in decreased mortality from A. hydrophila infection.
Experiment 2 (Exp 2) was conducted to determine whether a
4-h pretreatment with CuSO4 before exposure to A. hydrophila
plus a 4-h treatment the next day would reduce mortality from
A. hydrophila infection.
METHODS
Culture conditions and fish.Well water (total alkalinity =
98 mg/L as CaCO3 [Method 8203, Hach Company, Loveland,
Colorado]; total hardness = 60 mg/L as CaCO3 [Method 8213,
Hach Company]; pH = 7.4; total Cu <0.02 mg/L) was used
for both experiments. Mean tank water volume was 42.2 L (SE,
0.4). When water flows were turned on in the tanks, the mean
SE flow rate was 0.11 0.01 L/min. An aquarium heater was
placed in each tank to maintain water temperature at 28.5 C.
Water temperatures were measured in each tank twice daily.
Channel catfish fingerlings (U.S. Department of Agriculture
Agricultural Research Service Industry Pool strain) with a
mean body weight of 10.6 g (SE, 0.4) and 27.3 g (SE, 1.3) were
used for Exp 1 and Exp 2, respectively. Fish were fed a commercial catfish diet (50% protein:16% fat; AquaMax Fingerling
Diet, PMI Nutrition International, Brentwood, Missouri) once
a day to satiation except for the 24-h period preceding A.
hydrophila challenge and when CuSO4 was present in the tanks.
495
Bacterial culture and challenge conditions.An A. hydrophila isolate (ML-1048-K) cultured from the kidney of
a channel catfish that died in 2010 during an outbreak in an
Alabama catfish pond was used for the experimental challenges.
History, clinical signs, and the API 20E Identification System
(Biomerieux, Marcy l Etoile, France) were used to identify this
isolate as a member of this virulent strain of A. hydrophila. The
isolate, which was stored at 80 C until needed, was grown on
sheep blood agar for 24 h at 28 C. Colonies were then used to
inoculate 1 L of tryptic soy broth (TSB), which was incubated,
shaking vigorously on an automated shaker at 130 rpm, for
24 h at 28 C. To enhance virulence after storage, five channel
catfish fingerlings were injected intracoelomically with 100 L
of bacterial broth. The bacterium was reisolated from a dead
fish, transferred into TSB and incubated, shaking vigorously,
for 24 h at 28 C, until it reached an optical density of about 1.0
( = 540 nm; Bio-Rad Smart Spec 3000, Hercules, California).
The A. hydrophila challenge conditions were the same for
Exp 1 and Exp 2. Well water (1 L) was mixed with 100 mL of A.
hydrophila broth (optical density = 1.0). Ten channel catfish fingerlings were challenged for 1.5 h in buckets supplemented with
vigorous aeration. For the sham-exposed controls, fish were immersed in well water with 100 mL of TSB added. At the end of
the challenge, fish were removed from the buckets and returned
to their tanks without transferring excess water. Colony forming
units (CFU)/mL used for the two experiments were determined
from plate counts of 10-fold serial dilutions of the bacteria used
to challenge fish. The challenge concentration was 8.2 107
CFU/mL in Exp 1 and was 1.3 106 CFU/mL in Exp 2.
Copper sulfate. Copper sulfate (CuSO4 5H2 0; SigmaAldrich, St. Louis, Missouri) was used at a treatment
concentration of 1.0% of total alkalinity (Wise et al. 2004), or
0.98 mg/L CuSO4 . Copper (Cu) is 25.4% of the CuSO4 5H2 0
molecule, so this concentration resulted in a targeted Cu concentration of 0.25 mg/L. The CuSO4 was dissolved in distilled
water and added to the tanks. Samples for the measurement
of the total Cu in the water were collected; the 14.5-mL
water sample was preserved with 150 L of concentrated
nitric acid. For dissolved Cu (i.e., cupric ion, Cu + 2), the
14.5 mL of sample was filtered through a 0.20-m filter into
a 15-mL polypropylene tube and 150 L of concentrated
nitric acid was added. Copper concentrations were analyzed
at Arkansas Analytical (Little Rock, Arkansas) with a 730-ES
Simultaneous ICP-AES Analyzer (Agilent Technologies, Santa
Clara, California) following USEPA (1991). The analytical
sensitivity of the method was 0.02 ppm Cu.
Experimental design.Both experiments were set up as the
same 2 2 factorial design, with two levels of exposure to A.
hydrophila and two levels of exposure to CuSO4 . The fish in
treatment 1 (Trt 1) were not exposed to A. hydrophila or CuSO4 ,
the fish in treatment 2 (Trt 2) were exposed to A. hydrophila
but not CuSO4 , the fish in treatment 3 (Trt 3) were not exposed
to A. hydrophila but were exposed to CuSO4 . Treatment 4
(Trt 4) fish were exposed to both A. hydrophila and CuSO4 .
496
BEBAK ET AL.
There were five replicate tanks for each of the four treatments
and a starting density of 10 fish per tank. The fish were
acclimated for at least 96 h after stocking into experimental
tanks. Tank position was randomly assigned. Each experiment
was 14 d in length. Dead fish were removed and counted each
day. The fish in suitable condition were necropsied and cultured
for bacteria.
In Exp 1, at the end of the 1.5-h A. hydrophila challenge,
fish were returned to their tanks. Water flow was turned off 1 h
later and 0.98 mg/L CuSO4 (i.e., 0.25 mg/L Cu) was added to
each tank in Trt 3 and Trt 4. Distilled water was added to each
tank in Trt 1 and Trt 2. About 2 min after CuSO4 was added,
water samples for Cu analysis were taken from three tanks,
then again at 2 h and 18 h later. After 18-h exposure, water was
replaced in all the tanks. Samples for Cu analysis were taken
after replacing the water.
In Exp 2, tanks in Trt 3 and Trt 4 were exposed to a 0.98 mg/L
CuSO4 bath treatment for 4 h before they were exposed to A.
hydrophila. Only water was added to Trt 1 and Trt 2 tanks. Tanks
in all treatments were flushed at the end of the 4-h exposure.
Water samples for Cu analysis were taken 2 min after copper
was added (Time 0), at the end of the 4-h bath treatment, and
then after flushing the tanks. Another 4-h immersion treatment
with CuSO4 or distilled water was carried out 17 h after the first
CuSO4 treatment. Water samples for Cu analysis were taken
2 min after CuSO4 was added and then after flushing the tanks.
Data Analysis.Regression analysis with arcsine squareroot-transformed proportion survival data was used to
determine whether there were statistically significant responses
to treatment for Exp 1 and Exp 2 (StataCorp 2005). Statistical
significance was assigned as P < 0.05.
RESULTS
Experiment 1
The water temperature averaged 29.0 C (SE, 0.06) throughout the experiment. Kidneys from a total of 58 out of 59
(98.3%) dead fish were cultured for bacteria.
Deaths occurred within the first 24 h in the Aeromonasexposed tanks, both in Trt 2 and Trt 4. All fish that died in Trt
2 or Trt 4 had clinical signs of a bacterial septicemia. There
were no deaths in the first 24 h in Trt 3 tanks; the four fish that
died in this treatment were from three tanks and died between
48 and 72 h after the experiment began. During the 14-d
experiment, mortality occurred in all 10 of the tanks in the two
treatments exposed to A. hydrophila. No mortality occurred in
any of the five Trt 1 (i.e., sham-exposed) tanks, with a resulting
mean survival of 100.0% (Table 1). Aeromonas hydrophila was
recovered from every dead fish sampled in Trt 2 and Trt 4 but
not from sham-challenged fish in Trt 3.
Comparison of the treatments in a regression analysis
showed that percent survival was significantly different between treatments (P < 0.001). Treatment 4, which was exposed
to A. hydrophila and treated with Cu, had the lowest percent
TABLE 1. Mean percent survival of channel catfish in treatments for experiment 1 and experiment 2. Treatment 1 (Trt 1) was not exposed to A. hydrophila
or copper sulfate, treatment 2 (Trt 2) was exposed to A. hydrophila but not
copper sulfate, treatment 3 (Trt 3) was not exposed to A. hydrophila but was
exposed to copper sulfate, and treatment 4 (Trt 4) was exposed to A. hydrophila
and copper sulfate.
% Survival (SE)
Treatment
Trt 1
Trt 2
Trt 3
Trt 4
Exposure
Experiment 1
Experiment 2
Sham control
A. hydrophila
Copper sulfate
A. hydrophila +
copper sulfate
100 (0.0)
72.0 (6.0)
92.0 (4.0)
18.0 (7.0)
100 (0.0)
80.0 (5.5)
100 (0.0)
50.0 (3.2)
survival, at 18.0% (SE, 7.0) (Table 1). Survival in this treatment

was significantly different from all other treatments (P < 0.001).
When compared to Trt 1, the sham-challenged tanks, survival
in Trt 4 and Trt 2 were both significantly different, at P < 0.001.
Survival in Trt 3, not exposed to A. hydrophila but treated with
CuSO4 , was not significantly different (P = 0.08) than Trt 1.
After CuSO4 was added to the tanks, mean copper concentrations were 0.264 mg/L (SE, 0.003) and 0.256 mg/L (SE, 0.005)
for total Cu and dissolved Cu, respectively (Table 2). Eighteen
hours later, mean copper concentrations were 0.183 mg/L
(SE, 0.012) and 0.171 mg/L (SE, 0.010) for total Cu and
dissolved Cu, respectively. After replacing the water, dissolved
Cu concentration was <0.02 mg/L.
Experiment 2
The water temperature averaged 28.5 C (SE, 0.06) throughout the experiment. Kidneys from 22 out of 35 (62.9%) dead
fish were cultured for bacteria.
For both treatments (Trt 1 and Trt 3) in which fish were
not exposed to A. hydrophila, there was 100% survival during
the 14-d experiment (Table 1). In the treatment where fish
TABLE 2. Copper concentrations (Cu; mg/L) for experiment 1 and
experiment 2.
Time
Time 0
2h
18 h
Postflush
Trt 1, Time 0
Trt 1, 4 h
Trt 1, Postflush
Trt 2, Time 0
Trt 2, Postflush
Mean total Cu (SE) Mean dissolved Cu (SE)

Experiment 1
0.264 (0.003)
0.183 (0.012)
Experiment 2
0.257 (0.010)
0.249 (0.006)
0.256 (0.005)
0.215 (0.008)
0.171 (0.010)
<0.02 (0.0)
0.238 (0.008)
0.148 (0.006)
<0.02 (0.0)
0.237 (0.007)
0.030 (0.003)
EFFECT OF COPPER SULFATE ON AEROMONAS HYDROPHILA INFECTION
were exposed to A. hydrophila but not CuSO4 , survival was

80.0% (SE, 5.5). In the treatment where fish were exposed to
A. hydrophila and to CuSO4 , survival was 50.0% (SE, 3.2).
The percent survival in these treatments were both significantly
different than the sham-exposed control (Trt 1) at P < 0.001.
The percent mortality in the treatment exposed to A. hydrophila
and to CuSO4 was significantly lower than all of the other
treatments (P < 0.001). Aeromonas hydrophila was isolated
from every dead fish that was cultured.
In Exp 2, just after the CuSO4 treatment was added to the
tank, total Cu was 0.257 mg/L (SE, 0.010) and dissolved Cu
was 0.238 mg/L (SE, 0.008) (Table 2). After tanks were flushed
4 h later, copper concentrations returned to baseline levels of
<0.020 mg/L. The next day, the second treatment resulted
in concentrations of 0.249 mg/L (SE, 0.006) and 0.237 mg/L
(SE, 0.007) for total Cu and dissolved Cu, respectively. After
flushing the tanks, the copper concentration was an average of
0.030 mg/L (SE, 0.003) dissolved Cu.
DISCUSSION
Under the conditions used for this study, CuSO4 application
at 1% of alkalinity was not beneficial when used as a treatment
for infection of fingerling channel catfish with the strain of
A. hydrophila that was first isolated in the Alabama catfish
industry in 2009. When fish infected with A. hydrophila were
treated with an 18-h CuSO4 immersion (Exp 1) mean percent
survival was significantly lower than for those infected with A.
hydrophila but not treated with CuSO4 . A 4-h pretreatment with
CuSO4 before bacterial challenge plus another 4-h immersion
exposure the next day (Exp 2) also was not beneficial.
In Exp 1, the survival in the treatment that was exposed to
CuSO4 but not A. hydrophila was 92%, indicating that there was
some toxic effect of CuSO4 , which was controlled for with the
experimental design that included five replicates per treatment.
The free Cu concentration stayed below the recommended concentration of 1% of alkalinity to avoid toxic effects. Straus and
Tucker (1993) reported a 96-h LC50 of 0.762 mg/L (95% confidence interval, 0.600.95) for water at a temperature of 17.0
0.5 C, a total alkalinity of 76 mg/L CaCO3, and 83 mg/L total
hardness. The total alkalinity of the water in this experiment
was slightly higher, at 90 mg/L CaCO3 , but the total hardness
was lower, at 60 mg/L. Also, the mean water temperature was
29.0 C, 12 C higher than Straus and Tuckers (1993) conditions. These differences in total hardness and water temperature
may have contributed to the slightly increased toxicity seen in
Exp 1.
In Exp 1, the mean survival in the treatment exposed to
both A. hydrophila and CuSO4 was the lowest (18.0%) of the
four treatments, indicating an enhanced toxic effect of CuSO4
application when A. hydrophila is also present. This survival
was significantly lower (P < 0.001) than the mean survival
in the treatment exposed to A. hydrophila but not exposed to
CuSO4 . In Exp 2, the same pattern was observed; the lowest
497
mean survival (50.0%) occurred in fish exposed to both A.

hydrophila and CuSO4 . In addition, mean percent survival
was significantly different in this treatment compared to the
treatment exposed to A. hydrophila only (80.0%) at a P <
0.001. The results in both experiments suggest that CuSO4 may
increase mortality when used as a treatment for A. hydrophila
infection of fingerling channel catfish.
These results contrast with Darwish et al. (in press) who did
find that two applications of CuSO4 at 1% of alkalinity after
exposure to a mixed infection of A. hydrophila and F. columnare
resulted in significantly increased survival of sunshine bass
(female white bass Morone chrysops male striped bass M.
saxatilis). In this case, the A. hydrophila isolate used was from
an outbreak in sunshine bass. The pathogenesis of the bacterial
infection may contribute to the toxicity of the copper. In the
present study, and with this virulent strain of A. hydrophila, the
severe bacterial septicemia that occurs as a result of infection
with this isolate may make the fish more susceptible to the
toxic effects of the copper.
Our results are more consistent with previous reports regarding the detrimental effects of exposure of fish to copper sulfate.
In salmonids, for example, Knittel (1981) and Baker et al.
(1983) also reported that copper may increase susceptibility to
mortality from bacterial infections. Rabago-Castro et al. (2006)
found that at concentrations commonly used in aquaculture,
prophylactic, intermittent exposure of healthy channel catfish
to CuSO4 caused growth suppression, which has also been
shown in salmonids (Kamunde et al. 2002).
This work does not support the use of CuSO4 as a treatment
for this emerging strain of A. hydrophila and indeed suggests
that its use will increase mortality. The diversity of effects that
vary according to fish species and pathogen demonstrate that
much work needs to be done to determine the circumstances, if
any, under which CuSO4 may have useful beneficial effects.
ACKNOWLEDGMENTS
This work was supported by the U.S. Department of Agriculture, Agricultural Research Service, Current Research Information Systems Project Number 6420-32000-024-00D. This
research was conducted in compliance with all relevant federal guidelines and institutional policies. The mention of trade
names or commercial products in this publication is solely for
the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of
Agriculture.
REFERENCES
Baker, R. J., M. D. Knittel, and J. L. Fryer. 1983. Susceptibility of Chinook
salmon, Oncorhynchus tshawytscha (Walbaum), and rainbow trout, Salmo
gairdneri Richardson, to infection with Vibrio anguillarum following sublethal copper exposure. Journal of Fish Diseases 6:267275.
Darwish, A. M., J. Bebak, and K. K. Schrader. In press. Preliminary assessment of Aquaflor, copper sulfate and potassium permanganate for control of
498
BEBAK ET AL.
Aeromonas hydrophila and Flavobacterium columnare infection in sunshine

bass, Morone chrysops female Morone saxatilis male. Journal of Fish
Diseases. DOI: 10.1111/j.1365-2761.2012.01393.x.
Darwish, A. M., D. Mitchell, and D. L. Straus. 2012. Evaluation of a 4-h static
copper sulphate treatment against experimental infection of Flavobacterium
columnare in channel catfish (Ictalurus punctatus). Journal of Aquaculture
Research 43:688695.
Griffin, B. R., and A. J. Mitchell. 2007. Susceptibility of channel catfish,
Ictalurus punctatus (Rafinesque), to Edwardsiella ictaluri challenge following copper sulfate exposure. Journal of Fish Diseases 30:581585.
Hemstreet, B. 2011. Aeromonas status report2010. Fish Farming News
(winter):2.
Kamunde, C., M. Grosell, D. Higgs, and C. M. Wood. 2002. Copper metabolism
in actively growing rainbow trout (Oncorhynchus mykiss): interactions between dietary and waterborne copper uptake. Journal of Experimental Biology
205:279290.
Knittel, M. D. 1981. Susceptibility of steelhead trout Salmo gairdneri Richardson to redmouth infection Yersinia ruckeri following exposure to copper.
Journal of Fish Diseases 4:3340.
Plumb, J. A. 1999. Health maintenance and principal microbial diseases of
cultured fishes. Iowa State University Press, Ames.
Rabago-Castro, J. L., J. G. Sanchez, R. Perez-Castaneda, and A. GonzalezGonzalez. 2006. Effects of the prophylactic use of Romet-30 and copper
sulfate on growth, condition and feeding indices in channel catfish (Ictalurus

puntatus). Aquaculture 253:343349.
StataCorp. 2005. Stata statistical software: release 9. StataCorp, College Station,
Texas.
Straus, D. L., and C. S. Tucker. 1993. Acute toxicity of copper sulfate and
chelated copper to channel catfish Ictalurus punctatus. Journal of the World
Aquaculture Society 24:390395.
Thomas-Jinu, S., and A. E. Goodwin. 2004. Acute columnaris infection in
channel catfish, Ictalurus punctatus (Rafinesque): efficacy of practical treatments for warmwater aquaculture ponds. Journal of Fish Diseases 27:
2328.
US FDA (U.S. Food and Drug Administration). 2001. Compliance policy guide
section 615.115. Extra-label use of medicated feeds for minor species. US
FDA, Washington, D.C.
USEPA (U.S. Environmental Protection Agency). 1991. Methods for chemical analysis of water and wastewater, method 200.7, revision 4.4. USEPA,
Cincinnati, Ohio.
USFWS AADAP (U.S. Fish and Wildlife Service Aquatic Animal Drug
Approval Partnership). 2011. Approved drugs for use in aquaculture, 1st
edition. USFWS AADAP, Bozeman, Montana.
Wise, D. J., A. C. Camus, T. E. Schwedler, and J. S. Terhune. 2004. Health
management. Pages 444502 in C. S. Tucker and J. A. Hargreaves, editors.
Biology and culture of channel catfish. Elsevier, New York.


Embryonic and Early Larval Development in HatcheryReared Common Snook

Carlos Yanes-Roca
Kevan L. Main
a b
, Nicole R. Rhody , Michael Nystrom , Matthew L. Wittenrich &
Institute of Aquaculture, Stirling University, Stirling, Scotland, FK9 4LN, UK
Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida, 34236, USA
Florida Institute of Technology, Department of Biological Sciences, 150 West University

Boulevard, Melbourne, Florida, 32901, USA
To cite this article: Carlos Yanes-Roca, Nicole R. Rhody, Michael Nystrom, Matthew L. Wittenrich & Kevan L. Main (2012):
Embryonic and Early Larval Development in Hatchery-Reared Common Snook, North American Journal of Aquaculture, 74:4,
499-511


DOI: 10.1080/15222055.2012.676013
ARTICLE
Embryonic and Early Larval Development

in Hatchery-Reared Common Snook
Carlos Yanes-Roca*
Institute of Aquaculture, Stirling University, Stirling, Scotland FK9 4LN,
UK; and Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Nicole R. Rhody and Michael Nystrom

Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Matthew L. Wittenrich
Florida Institute of Technology, Department of Biological Sciences, 150 West University Boulevard,
Melbourne, Florida 32901, USA
Kevan L. Main
Mote Marine Laboratory, 1600 Ken Thompson Parkway, Sarasota, Florida 34236, USA
Abstract
To gain an improved understanding of the early life history of common snook Centropomus undecimalis and refine
hatchery production techniques for this species, a combination of digital photography and histological techniques were
used to document the embryonic and early larval development of hatchery-reared individuals. Embryo development
from fertilization to hatching took 15 h at 28 C. Larvae at 2 d posthatch showed fully pigmented eyes, and histological
sections of the digestive tract revealed the presence of cellular structures indicative of a functional gut. This suggests
that common snook larvae have the mechanical ability to detect, capture, and digest prey at 2 d posthatch.
The common snook Centropomus undecimalis is a diadromous, stenothermic, euryhaline, estuarine-dependent species
found in the tropical and subtropical western Atlantic Ocean
and Gulf of Mexico from about 34 N to about 25 S latitude
(Howells et al. 1990). Snook Centropomus spp. are protandric hermaphrodites: some males develop into females between
1 and 7 years of age, having a maximum 20-year lifespan.
The spawning of common snook has been studied for the last
55 years, but despite the importance of common snook as a
popular game fish, the description of its reproductive biology is
incomplete. Common snook in Florida have a daily spawning
cycle in which spawning episodes occur during the late afternoon and the early evening hours during the lunar phases and
during all tidal stages (Taylor et al. 1998).
The identification of critical embryonic and larval stages,

such as eye and gut formation, first feeding, and swim bladder inflation, is essential for a better understanding of any fish species.
This understanding may help to improve common snook larval
rearing techniques and therefore increase larval survival rates.
Although less developed than in adults, the larval digestive
tract is functional when feeding is initiated (Govoni et al. 1986).
Additionally, the digestive tract develops as the larvae grow.
This facilitates changes in the rates of ingestion, digestion, and
the assimilation efficiency influence resulting in improved larval
growth (Sarasquete et al. 1995). The development of the digestive system tract, as well as the possible abnormalities and deficiencies that result from the absence or inadequacy of food, has
been studied in several teleosts (Cousin and Baudin-Laurencin
*Corresponding author: cyanes@mote.org

Received November 10, 2011; accepted February 14, 2012
499
500
YANES-ROCA ET AL.
1985; Avila and Juario 1987; Eckman 1987; Ferrais et al.

1987; Deplano et al. 1991; Boulhic and Gabaudan 1992) and is
defined by the development of key digestive system structures.
Eye development for most fish species is critical for their
survival, especially once exogenous feeding starts (Mani-Ponset
et al. 1996), mainly due to their visual feeding nature (Blaxter
1986; Batty 1987). Prey capture, orientation, schooling, and
eluding predators are other basic activities that rely on vision
(Paul 1983; Blaxter 1986; Porter and Theilacker 1999). The
role of vision in feeding has been investigated using varied
light intensities. Significant differences in the consumption of
prey were found among a variety of fish species including cisco
Coregonus artedi (Jonh and Hasler 1956), sole Solea solea, and
plaice Pleuronectes platessa (Blaxter 1968), with most variations in foraging corresponding to periods of dusk and dawn.
The decrease in rate of feeding corresponds with dusk and dawn
periods. Some species can feed in the dark, e.g., cisco (Jonh
and Hasler 1956), especially when food is present in high concentrations. For example, sole feed in the dark for most of their
larval life and others, like plaice, only at later stages around
metamorphosis (Blaxter 1968). Kawamura (1984), Pankhurst
(1996), and Roo et al. (1999) reported major changes in the visual system in the lecithotrophic phase as preparation for onset
feeding on sparids. At the same time, Roo et al. (1999) studied the relationship between gut and eye development in red
porgy Pagrus pagrus, demonstrating that visual capability was
developed before the onset of exogenous feeding.
No common snook eggs, embryos, or early larvae have been
documented from wild collections. This is a void in information
on the life history and early feeding habits that generally forms
the building blocks of aquaculture protocols.
The objective of this study was to describe the physiological,
embryonic, and larval developmental features of the common
snook.
METHODS
All the samples were collected at the Mote Aquaculture Research Park located in Sarasota, Florida, from May to August
2007. The eggs were stripped from wild females from the field
and fertilized with milt from wild males collected at the same
time using a 92-m seine net deployed from a research vessel.
Once fertilized, the eggs were transported to the main facilities
and stocked in the rearing tanks.
Egg sampling procedures.One hour after fertilization,
common snook embryos were stocked in a 50-L tank at a constant temperature of 28 C and salinity of 35. No aeration was
used, dissolved oxygen was constant at 8 mg/L. The eggs were
kept in the dark, simulating night time light characteristics. A
sample of 10 embryos was collected every hour until hatching
occurred. Once collected, samples were placed under a compound light microscope (Olympus 3500 ) with a dark field.
The egg diameter was measured and embryonic development
was observed and documented using an attached 35 mm camera
(Nikon 500). This sampling was repeated three times during

three separate spawning events over the course of three nights.
Larval sampling procedures.Larvae were sampled from a
3,300-L production tank, where water temperature was maintained at 28 C, salinity at 35 and dissolved oxygen at 10 mg/L.
Five larvae were randomly collected daily from day 0 to day 3
and every two days from day 4 to day 14 after hatching. Once
collected larvae were placed under a light microscope (Olympus
4000), which had a digital camera mounted (Sony 600), larval
pictures were taken under dark field conditions. Standard length
(SL) and myomere height from the specimens were recorded
using a calibrated microscope reticule. This sampling regime
was repeated for seven different spawning events, collecting a
total of 280 larvae, between day 0 and day 14. Another 10 larvae
were collected from day 1 to day 3 after hatching and fixed for
transmission electron microscope (TEM) and scanning electron
microscope (SEM) work.
Electron microscopy specimen fixation and preparation.
The larvae collected for the TEM were fixed in Karnovskys
fixative at 4 C for 2 hours and placed in cacodylate buffer rinse
(pH 7.5) at 4 C until further processing. The tissues were then
transferred to the Electron Microscopy Laboratory, Institute of
Aquaculture, at the University of Stirling, Stirling, Scotland,
where final processing was completed. Once at the processing
unit, the specimens were processed as described in Table 1.
Block sectioning was carried out using a glass knife for
semithin sections (0.5 m) to be observed under the light microscope. For the TEM, block sectioning was carried out with a
diamond knife for ultrathin sections (50 nm). Ultrathin sections
were mounted on copper Formavar carbon-coated grids and
metal stained with uranyl acetate and lead citrate and stained
with a metal stain; this step was carried out following a metal
stain protocol described below. A drop of saturated uranyle acetate was placed on a piece of dental wax and the grid was
floated over the drop for 30 min in the dark. The grid was rinsed
with 50% alcohol and then with distilled water and floated on a
drop of lead citrate for 25 min. The grid was thoroughly rinsed,
dried on filter paper, and immediately removed to avoid dust
accumulation on the grid. Sectioned samples were viewed and
TABLE 1. Mean SD standard length (mm) and myomere height (mm) of
common snook during the first 14 d after hatching (n = 35).
Day
0
2
4
6
8
10
12
14
Standard length
Myomere height
1.785 0.289
2.311 0.434
2.266 0.360
2.520 0.515
2.556 0.454
3.137 0.495
3.580 0.454
4.433 0.780
0.165 0.015
0.211 0.024
0.211 0.059
0.246 0.036
0.277 0.040
0.368 0.047
0.401 0.055
0.549 0.025
EMBRYONIC AND EARLY LARVAL DEVELOPMENT
501
photographed at the Institute of Aquaculture on the TEM (FEI

Tecnai E2 Biotwin).
All the handling during the course of this project followed
the animal welfare rules and regulations from the United States
of America Animal Welfare Institute (Washington, D.C.).
RESULTS
Embryonic Development
The development of common snook embryos at 28 C took
15 h from fertilization to hatching. The fertilized eggs were
spherical, with a homogenous yolk, a smooth chorion, and a single oil droplet. Egg diameters ranged from 0.65 mm to 0.72 mm
(mean = 0.69 mm, n = 300) and the oil droplet diameter ranged
from 0.15 to 0.30 mm (mean = 0.26 mm, n = 300). Fertilization occurred in the field; therefore, no embryonic development
was documented prior to the blastodisc stage (first hour after
fertilization). We estimated that the blastoderm separated from
the yolk approximately 30 min after fertilization. Shortly after
that, the blastoderm was then divided in two blastomeres.
The following is a description of development at approximate
1-h intervals:
One hour and thirteen minutes after fertilization, the second
segmentation or four-cell stage occurred. The second cleavage
furrow developed on two blastomeres at a right angle to the first
cleavage plane. It deepened until each blastomere divided into
two cells of the same size. The oil droplet was larger and gathered toward the vegetal pole (Figure 1, item a). One hour and
thirty-two minutes after fertilization another division occurred.
This time, the blastoderm divided in 16 blastomeres (Figure 1,
item b), also called the 16-cell stage, where the fourth cleavage plane, which parallels the second, divides the two rows of
four blastomeres into four rows of four blastomeres (Figure 1,
item b).
Two hours after fertilization, another division occurred and
the blastoderm divided into 32 blastomeres (Figure 1, item c),
the 32-cell stage, where the fifth cleavage plane divided the
marginal 12 blastomeres meridionally into 24, and the central
four blastomeres horizontally into eight thereby forming two
layers, an outer and an inner layer, in the central region. The
number of marginal cells is 14 (Figure 1, item c). Three hours after fertilization, the blastomeres were still distinct, but the rapid
division was too advanced to observe the number of blastomeres;
this is the late morula stage (Figure 1, item d). At this time, the
planes of the sixth and the later cleavages were difficult to precisely trace. The blastomeres (256512) had different cleavage
planes depending on their positions within the dome-shaped
blastoderm and were arranged in three layers. The peripheral
blastoderms (2124) were flattened in shape. The cells were
arranged in three to four layers but still easily dissociated from
each other (Figure 1, item d).
Four hours and ten minutes after fertilization, the blastomeres
were no longer distinguishable and the blastocoel, or segmentation cavity, began to form. This was the blastula stage (Figure 1,
FIGURE 1. Embryonic development of common snook from fertilization to

hatch (all times are postfertization): a = 1 h, 13 min; b = 1 h, 32 min; c = 2 h;
d = 3 h; e = 4 h, 10 min; f and g = 5 h, 12 min; h = 6 h, 12 min; i = 7 h, 15 min;
j = 8 h, 10 min; k = 9 h, 21 min; l = 10 h, 12 min; m = 11 h, 12 min; n = 12 h,
30 min; o = 13 h, 5 min; p = 14 h, 22 min. [Figure available in color online.]
item e). Projection of the underside of the blastoderm (central

cells) into the yolk sphere was observed. In this stage, some
blastomeres began to cleave asynchronously and to migrate.
Several rows of periblast nuclei were visible around the blastoderm (Figure 1, item e).
Five hours and twelve minutes after fertilization, the blastoderm covered more than half the yolk (Figure 1, item f) and
the blastocoel were completely formed (Figure 1, item f). This
was the mid-gastrula stage, where a streak was visible in the
midline of the embryonic shield projecting into the germ ring
area (Figure 1, item g).
Six hours and twelve minutes post fertilization, the blastoderm covered three-fourths of the yolk sphere (Figure 1, item
h) and the embryonic shield became more clearly visible as a
narrow streak. The enveloping layer extends uniformly over the
yolk sphere through this stage (Figure 1, item h).
Seven hours and fifteen minutes after fertilization, the blastoderm covered more than four-fifths of the yolk surface, leaving
a small area around the vegetal pole exposed (Figure 1, item i).
This was the neurula stage, and the head could be clearly recognized anteriorly in the distinct embryonic body. A beak-like
mass of cells was seen in front of the head, the embryo completes
an arc of 150 . The brain and nerve cord in the arrow-shaped
embryonic body developed as a solid rod of cells. A solid optic
bud appeared on each side of the cephalic end. The beak-like cell
mass was still visible and the blastopore was closed (Figure 1,
item i).
502
YANES-ROCA ET AL.
Eight hours and ten minutes after fertilization, melanophores

were visible on the embryo for first time (Figure 1, item j) and
the optic vesicles were clearly outlined. This is the somite stage,
where a pair of otic (auditory) vesicles appeared at the posterior
region of the head. Depressions began to form at the dorsal
surface of the eye vesicles (Figure 1, item j).
Nine hours and twenty-one minutes after fertilization, the
embryo was strongly pigmented (Figure 1, item k), especially
dorsolaterally. Melanophores on the yolk concentrated on the
ventral surface on each side of the embryo. Some melanophores
were also apparent on the oil droplet (Figure 1, item k).
Ten hours and twelve minutes after fertilization (Figure 1,
item l), the optic vesicles differentiated to form the optic cup
and the lenses began to form. The small otic vesicles appeared,
but they lacked otoliths. The regions of the brain were well
defined, and the neural fold was seen as a median line along the
body (Figure 1, item l).
Eleven hours and twelve minutes after fertilization (Figure 1,
item m), the tubular heart appeared under the head from the
posterior end of the midbrain to the anterior end to the hindbrain.
The body cavity extended further toward the posterior end of the
eye vesicles. The melanophores on the oil droplet were larger
and more distinct; those on the head expanded to outline the
olfactory lobes and optic vesicles. The tail separated from the
yolk (Figure 1, item m).
Twelve hours and thirty minutes after fertilization (Figure 1,
item n), the melanophores were much larger and fewer, forming
aggregations. The anterior portion of the heart, which exhibited
a slow pulsation, extended upward to the anterior end of the
forebrain. Cuverian ducts and the vitello-caudal vein were still
incomplete. The embryonic body encircles nearly three-fourths
of the yolk sphere (Figure 1, item n).
Thirteen hours and five minutes after fertilization (Figure 1,
item o), the blood began to circulate and the heart beat was
faster and more constant. At the same time, the embryo started
moving with quick, short movements mainly produced by the
tail. Melanophores were more compact and they accumulated
in four places along the embryos body (Figure 1, item o).
Fourteen hours and twenty-two minutes after fertilization,
the first larvae hatched (Figure 1, item p). At this time, the
melanophores accumulations could be observed clearly. Five
groups spread along the newly fresh larvae. There were two
on the head region, one over the olfactory region, and another
around the eyes. The other two were situated in the midlateral
and dorsolateral area, and another one in the postanal area (Figure 1, item p). Fifteen hours after fertilization massive hatching
took place; 90% of the eggs hatched at that time.
Larval Development
Common snook larvae hatched 15 h after fertilization at 28 C
(Figure 1, item p), measuring 1.71 mm in mean SL, ranging from
1.38 to 1.84 mm (n = 35). The body of newly hatched yolk
sac larvae was elongated with an oval shape and a mean size of
0.91 mm in length, ranging from 0.85 to 1.02 mm (n = 300), and
with a mean width of 0.58, ranging between 0.41 and 0.72 mm.
A single oil droplet was present, located on the yolk sacs front
side under the head. A transparent voluminous fin fold covered
most of the body (Figure 1, item p). The eyes and mouth were
not formed, and no eye pigmentation was present. The larvae
were concentrated on the surface, floating and moving mainly
in circles due to their limited fin development.
Yolk Sac Stage
On day 0 (24 h after fertilization), common snook larvae
(Figures 2, 3, and 4, item a) measured 2.25 mm (mean SL),
ranging from 1.90 to 2.45 mm (n = 35), and had a mean
myomere height (MH ) of 0.164 mm, ranging from 0.152 to
0.171 mm (n = 35). The yolk sac was reduced in size with a
mean length of 0.51 mm, ranging from 0.31 to 0.63 mm, and
a mean width of 0.35 mm, ranging from 0.29 to 0.45 mm. Eyes
were starting to form, and some pigmentation was observed.
The mouth was not formed, although some definition was
observed. The pectoral fins were starting to develop. In terms
of larval behavior, larvae were mainly at the surface, although
some larvae were distributed in the water column but close to
the surface. Swimming movements were more directional.
Day 1 larvae (Figure 3) had a mean SL of 2.07 mm, ranging from 1.92 to 2.17 mm, and an average MH of 0.202 mm,
ranging from 0.180 to 2.071 mm (n = 35). The yolk sac, although reduced to half the size from the previous day, was still
present with a mean length of 0.16 mm, ranging from 0.12 to
0.20 mm. The eyes were starting to gain definition and the retina
can be now differentiated with eye pigmentation observed but
not fully completed. The optic tectum or primary optic center
of the brain is large (Figure 5A) as in most of fish, which rely
on their sense of vision. Lenses were fully complete and visible
(Figure 5A) and the internal layer organization is well established, although layer thickness is still low. The cornea and the
cartilaginous ring were present, as well as the optic chiasma
right below the infundibulum. The retina layer organization can
be observed (Figure 5B) and the main layers were visible, such
as the outer limiting layer and the outer nuclear layer, where the
columnar nuclear bodies were present although undeveloped.
Also, the outer plexiform layer and inner nuclear layer can be
observed, yet most of layers were not fully developed with some
main organelles missing (Figure 5C), where the outer limiting
layer and the outer nuclear layer were still lacking organelles
definition. Some pigments in the epithelium were present, although still low in levels (Figure 5D). The alimentary canal of
1-d-old larvae shows some differentiation along its length. Cilia,
which help to circulate the contents, can be observed in the lumen (Figure 6BC), and at the same time some irregular small
microvilli appear along the digestive wall. Other organelles observed include the mitochondrion and nucleus of the epithelial
cells, although these organelles were present in low numbers.
Other structures, such as the nonvillous region and the terminal web, were clearly defined although they were lacking in
thickness (Figure 6C).
FIGURE 2.
503
Common snook larval growth during the first 14 d after hatching.
On day 2 (Figures 3 and 4, item b), SL has increased to a

mean 2.31 mm, ranging from 2.12 to 2.71 mm, and with an
average MH of 0.211 mm, ranging from 0.185 to 0.282 mm (n =
35). The yolk sac has been nearly absorbed reducing its size to
a mean length of 0.15 mm, ranging from 0.11 to 0.20 mm. The
oil droplet was still present although reduced in size (Figures 3
and 4, item b). Eyes were formed and pigmented (Figure 5EF)
and the cornea has gained in thickness and was tight against
the lens. The retina layer was more defined and each layer was
thicker. The optical nerve was fully formed and connected to
the main nervous system (Figure 5EJ). For the first time, the
clear layer of pigment epithelium cells was present and the
other layers, such as the outer nuclear layer, were gaining in
complexity (Figure 5FG) with the development of organelles
such as the cones and rods (Figure 5H) nearly completed
(Figure 5I). The mouth was formed and opened (Figures 3 and
4, item b) with the main cartilages, such as the Meckels car-
tilage, the hyposynplectic cartilage, and the basihyal cartilage,

already present (Figure 5E). Also the tongue can be observed
(Figures 4, item b, and 5E). The digestive system was straight
and long, extending past the posterior margin of the yolk sac
and into the ventral fin fold, and although undeveloped, it has
some food inside. The digestive system walls were defined,
no cilia was observed in the lumen, and the microvilli layer
was now well established along the walls (Figure 6D). Also
the microvilli were long and compact (Figure 6EF). The epithelium cell structure was forming and the main organelles
were observed, such as the mitochondrion, the clear and dark
apical cells, and the epithelium nucleus (Figure 6EF); however, organelle numbers were low. Pectoral fins were well developed and functional. Slight caudal fin definition could be
observed. Larvae show a photopositive reaction gathering at
those areas with more light. The distribution of larvae was more
diverse, having larvae distributed in the water column and at
YANES-ROCA ET AL.
504
FIGURE 3. Common snook larval development from day 0 to day 14 after hatching. Length is shown in millimeters.
505
FIGURE 4. Common snook organ development from day 0 to day 14 after hatching: a = day 0, b = day 2, c = day 3, d = day 4, e = day 6, f = day 8, g = day
10, h = day 12, j = day 14. Abbreviations are as follows: A = anus, ADF = anterior dorsal fin, AF = anal fin, AMI = anterior median intestine, C = ceratohyal,
CF = caudal fin, CM = cerebellum, CN = constriction, E = eye, FF = fin fold, H = heart, HS = hyposynplectic cartilage, IN = intestine, L = liver, MC =
Merkels cartilage, MO = medula oblongata, N = notochord, OD = oil droplet, OL = optic lobe of brain, OT = optic tectum, P = pigments, PH = posterior part
of hind, PM = premaxilla, R = rectal area, SB = swim bladder, T = teeth, V = intestine-recto valve, and YS = yolk sac. [Figure available in color online.]
the surface. Burst movements towards prey were commonly

observed.
On day 3 (Figures 3 and 4, item c), snook larvae had a
mean SL of 2.26 mm, ranging from 2.15 to 3.1 mm, and a
mean MH of 0.214 mm, ranging from 0.20 to 0.35 mm (n =
35). The yolk sac was almost completely absorbed and the oil
droplet was still present, although severely reduced. The eyes
had gained pigmentation and were developed. The retina layer
was well structured due to the development of most of the layer
organelles with the cornea fully formed (Figure 5KL). All the
different layers were clearly differentiated, such as the outer
ganglion layer, the inner nuclear layer and the inner ganglion
layer (Figures 5L, 6A). The pigment epithelium cell layer
was fully formed with photoreceptor inclusions (Figure 6A),
the bodies of the photoreceptors were also now well defined

(Figures 5L, 6A). The maxillar jaws were developing and mouth
gap was increasing. The alimentary canal was no longer a long
straight tube; some structure could be observed especially in the
anal region, which, by this stage, was well developed. Structural
epithelium organelles, such as the nucleus, mitochondrion, and
dark vesicles, were all present and in high numbers (Figure 6H
I). The microvilli layer had gained length and was more compact
(Figure 6H). Other important organelles were also present,
such as the Golgi apparatus, rough endoplasmatic reticulum,
desmosome of apical junction and pinocytotic vesicle (Figure 6I). Food was observed in the gut, mainly rotifers and algae.
The swim bladder was formed and showing signs of inflation.
Swimming speed had increased, as well as larvae motility.
YANES-ROCA ET AL.
506
FIGURE 5. Optical SEM and TEM cross sections of common snook larva from day 1 to day 3 posthatch (dph): (A) semithin cross section from the head of a
1-dph common snook larva (scale bar = 100 m; B = buccal cavity, C = cornea, CR = cartilaginous ring, F = infundibulum, INL = inner nuclear layer, L =
lens, OC = optic chiasma, and OT = optic tectum); (B) cross section of the eye of a common snook larva at 1 dph (scale bar = 15 m; CNB = columnar nuclear
bodies [nuclei of photoreceptors], OLM = outer limiting membrane, ONL = outer nuclear layer, and OPL = outer plexiform layer); (C) cross section of the eye of
a common snook larva at 1 dph (scale bar = 10 m); (D) cross section of the eye of a common snook larva at 1 dph (scale bar = 5 m; PE = pigment epithelium);
(E) semithin cross section from the head of a 2-dph common snook larva (scale bar = 90 m; BC = basihyal cartilage, M = Meckels cartilage, ON = optic
nerve, and T = tongue); (F) semithin cross section from the head of a 2-dph common snook larva (scale bar = 40 m); (G) cross section of the a eye of common
snook larva at 2 dph (scale bar = 5 m); (H) electron micrograph of a cross section of the eye of a common snook larva at 2 dph (scale bar = 2 m; C = cones,
PEC = pigment epithelium cell granule, and R = rods); (I) cross section of the eye of a common snook larva at 2 dph (scale bar = 1 m; OD = oil droplet,
PRES = photoreceptor outer segment); (J) cross section of the eye of a common snook larva at 2 dph (scale bar = 10 m); (K) semithin cross section from the
head of a 3-dph common snook larva (scale bar = 100 m); and (L) cross section of the eye of a common snook larva at 3 dph (scale bar = 5 m; B = bodies of
photoreceptors and IGL = inner ganglion layer). [Figure available in color online.]
507
FIGURE 6. Optical SEM and TEM cross sections of the eye and digestive system of common snook larva from 1 to 3 dph (see Figure 5 for abbreviations not
defined here): (A) cross section of the eye of a common snook larva at 3 dph (scale = 10 m); (B) cross section of the antero-medial intestine of a common snook at
1 dph (scale bar = 2 m; C = cilia, L = lumen, M = mitochondrion, MV = microvilli, and N = nucleus); (C) cross section of the rectal area of a common snook
at 1 dph (scale bar = 2 m; NVR = nonvillous region and TW = terminal web); (D) semithin cross section of the antero-medial intestine of a common snook
larva at 2 dph (scale bar = 70 m; CA = clear apical cell in epithelium, DA = dark apical cell in epithelial, DSW = digestive system walls, and N = nucleus of
columnar cell); (E) cross section of the antero-medial intestine of a common snook larva at 2 dph (scale bar = 5 m; LV = light vesicle); (F) cross section of the
rectal area of a common snook larva at 2 dph (scale bar = 1 m); (G) cross section of the antero-medial intestine of a common snook larva at 3 dph (scale bar =
100 m; DD = dark droplet, F = food particles, and N = nucleus of enterocyte); (H) cross section of the antero-medial intestine of a common snook larva at
3 dph (scale bar = 5 m); (I) cross section of the rectal area of a common snook at 3 dph (scale bar = 2 m; D = desmosome of the apical junction, GA = Golgi
apparatus, and RER = rough endoplasmic reticulum); and (J) cross section of the antero-medial intestine of a common snook larva at 3 dph (scale bar = 1 m).
[Figure available in color online.]
508
YANES-ROCA ET AL.
Preflexion Stage
Day 4 larvae (Figures 3 and 4, item d) have a mean SL of
2.35 mm, ranging from 2.22 to 3.23 mm, and a mean MH of
0.211 mm, ranging from 0.20 to 0.28 mm (n = 35). By day
4, the yolk sac has been totally absorbed and the oil droplet
was not present anymore (Figure 3 and 4, item d). The medulla
oblongata can be observed as well as the cerebellum, which was
fully formed. The eyes were fully pigmented, the mouth was
fully functional with more jaw development, and teeth formation
can also be observed. Swim bladder, positioned posterior to the
pectoral fin base and above the stomach, was developed and
inflated. The gut gained in thickness and became partitioned.
Rotifers were observed in the gut.
At this stage larvae were scattered throughout the water column, although no larvae could be seen on the tank bottom.
There were many active swimmers, which spent most of the
time seeking prey.
On day 6 (Figures 3 and 4, item e), the mean SL of snook
larvae was 2.51 mm, ranging from 2.1 to 3.41 mm, with a mean
MH of 0.245 mm, ranging from 0.214 to 0.351 mm (n = 35).
Rows of teeth were present at the same time that mouth gap had
increased and the gas bladder was inflated. The gut was well
partitioned and food was commonly observed inside with 90%
of larvae observed having full stomachs. The fin fold around the
larvae was no longer present and the fins were starting to take
shape, especially the caudal and dorsal fins, and the pectoral fins
were fully functional but still developing.
At day 8, snook larvae (Figures 3 and 4, item f) had a mean
SL of 2.55 mm, ranging from 2.41 to 3.7 mm, with a mean MH
of 0.276 mm, ranging from 0.21 to 0.34 mm (n = 35). At this
stage, the snook larvae have increased body depth, and increased
head size in relation to eye size. The notochord was fully formed
and ends at the caudal fin. There was an increase in volume of
the digestive system, and definition of the different organs was
more apparent. The caudal fin started developing into the shape
of an adult caudal fin, going from a more rounded initial shape
towards a forked fin shape.
Flexion Stage
Ten days after hatching (Figures 3 and 4, item g) common
snook larvae have a mean SL of 3.136 mm, ranging from 3
to 4.2 mm, and the mean MH was 0.381 mm, ranging from
0.291 to 0.41 mm (n = 35). At this stage, notochord flexion had
started, larvae had increased head size, and both maxillar and
premaxillar bones were gaining in thickness and strength. Teeth
were present now in both jaws bones. The digestive system
was well differentiated, with all the organs gaining in volume
and structure. Also, 90% of the observed stomachs were full,
rotifers were observed all along the digestive system. Fin shape
definition continued to develop, especially the dorsal and caudal
fins (Figures 3, 4g).
Common snook larvae at day 12 (Figures 3 and 4, item h) had
a mean SL of 3.57 mm, ranging from 3.41 to 3.84 mm, and an
average MH of 0.401 mm, ranging from 0.266 to 0.467 mm (n =
35). The development of the dorsal and anal fin bases could be
observed. At the same time, notochord flexion was nearly complete. Larval body depth continued to increase and the head size
was still proportionally larger in width than the rest of the body.
By day 14, common snook larvae (Figures 3 and 4, item i) had
a mean SL of 4.43 mm, ranging from 3.57 to 5.71 mm, and their
mean MH was 0.41 mm, ranging from 0.36 to 0.46 mm (n =
35). By this stage, the lower dorsal fin was developing faster
than the upper one and nine rays could be observed. Similar
development happened to the anal fin, although rays were not
as developed and notochord flexion was completed.
During the first 14 d after hatching, development of common
snook was rapid in terms of SL (Figure 2). We observed three
stages: the first stage or the yolk sac stage occurred over a 2-d
period where the SL increased from 1.78 to 2.311 mm. The
next stage was the preflexion stage, which occurred over a 4-d
period, where the increase in SL was less than in the yolk sac
stage, going from an average SL of 2.262.56 mm (Figure 2).
The last stage or the flexion stage occurred over a 6-d period
and the average SL went from 2.56 mm to 4.43 mm, increasing
exponentially each day. The mean SL and MH measurements
during the first 14 d of growth can be seen in Table 1.
DISCUSSION
Common snook Embryonic Development
The developmental features of common snook described in
this paper are typical of most teleost species with planktonic
eggs. No common snook eggs have been described from the
wild, although it is known that, immediately after spawning,
eggs are taken towards the open sea by the outgoing tide and it
is assumed that the following incoming tide brings them back
into the estuarine and mangrove environment. Sampling efforts
to find wild common snook eggs have been carried out, but
the quest has been so far unsuccessful. Many factors may have
resulted in this unsuccessful outcome, such as sampling gear,
location, time, and the inability to recognize common snook
eggs once collected.
The search for buoyant planktonic embryos in the wild will
provide useful information for the development of incubation
protocols in the laboratory. This is important since the influence
of the environmental conditions influences early development
and physiology of the offspring (Blaxter 1992).
The processes of cleavage, formation of layers, and morphogenesis of teleost eggs during incubation have been described in a number of standard textbooks (such as Rudnick
1955, Waddington 1956, and Smith 1957), with Oppenheimer
(1947) and Devillers (1961) stressing structural changes from
the viewpoint of experimental embryology. Most of those descriptions were done for freshwater species, although extensive
work has been done for some marine species including silver
warehou Seriolella punctata (Grimes and Robertson 1981), gilthead bream Sparus aurata, European flounder Platichthys flesus, dab Limanda limanda, Atlantic herring Clupea harengus,
Atlantic cod Gadus morhua, plaice, and Atlantic salmon Salmo

salar.
Common snook eggs are similar to those of many other fish
species in their round shape, although in the anchovy Engraulis
spp. and bitterling Rhodeus sericeus the eggs are ovoid and in
certain gobies (family Gobiidae) the eggs are pear shaped. At the
same time, the common snook embryo has only one oil globule,
whereas more than one can be found in other marine species
(Simpson 1956) and in most fish species that have telolecithal
eggs with the yolk more concentrated at the vegetative pole.
As with hagfish (family Myxinidae) and elasmobranches, common snook exhibits meroblastic cleavage, even though snook
are teleosts. They do not have the holoblastic cleavage that
characterizes lampreys (family Petromyzontidae). Other groups,
such as bowfin Amia spp., gar Lepisosteus spp., and sturgeon
Acipenser spp., have intermediate features. All the above embryo characteristics, plus the melanophores pattern, egg size,
and oil droplet size described in the results, will make the recognition of common snook embryos more accurate and provide a
tool for egg quality evaluation in common snook culture.
Common snook Morphological Development
Common snook larval development was described in 1982
by Lau and Shafland. They primarily focused on larvae from
14 d old and beyond, mainly looking at osteological, cephalic,
and fin development. Also, Wittenrich et al. (2009) described the
osteological development of the feeding apparatus with feeding
performance, but other than this no reported common snook
development studies have been carried out.
Eye Development
At hatching, common snook larvae have an unpigmented
and nonfunctional visual system; many other species are similar
in this respect (e.g., sole, Atlantic mackerel Scomber scombrus, whiting Merlangius merlangus, European pilchard Sardina pilchardus, and Pacific sardine Sardinops sagax caeruleus),
while some others have pigmented eyes (e.g., plaice, Atlantic
cod, Atlantic herring, and salmonids) (Blaxter and Staines
1970). Blaxter (1986) described the important role that vision
has on fish larvae orientation, as a consequence of them being
visual feeders. Fish species with a relatively small yolk sac need
to develop fast in order to survive. In common snook larvae, just
like in sparids such as madai Pagrus major (Kawamura 1984)
and New Zealand snapper Pagrus auratus (Pankhurst 1996),
the most important changes in the eye structure occur in the
lecitotrophic larvae as a preparation for prey capture.
One-day-old common snook larvae had all the basic structural elements necessary for visual function, but most of them
were incomplete. This was an indication that the eye was about
to become functional. Kawamura (1984) found that the visual
system of madai is functional at 36 h posthatch when visual cells
and pigments are present and nerve optic fibers connect with the
optic tectum. In common snook, the visual system could not be
functional at day 1 posthatch, principally because the pigmenta-
509
tion pattern, responsible for photon absorption, was very sparse

at this stage. Retinas of most fish larvae mainly have greensensitive single cones (Evans and Browman 2004). This is the
case of 2-d-old common snook larvae: although the retina was
not as fully developed as in the adult stage, all the retina structural layers were complete but not fully functional. Histological
observations support the partial functionality of the eye; by this
time the pigment cell layer was present and the optic nerves
were connected to the optic tectum. Thus, the visual system was
completely ready for prey capture.
The pure cone retina has been found in the earlier stages
of many teleost larvae such as Pacific salmon Oncorhynchus
spp. (Ali 1959), Atlantic herring (Blaxter and Jones 1967), and
plaice (Blaxter 1968). However, at first feeding common snook
are only equipped with simple cones, as with madai (Kawamura
1984) and New Zealand snapper (Pankhurst 1996), and rods and
twin cones appear at metamorphosis (Blaxter and Staines 1970).
At day 3 the common snook larval retina has well-developed presumptive cone receptors, the pigmentation layer is fully complete, the structural layers are fully functional and differentiated,
and the formation of the rod precursors can be spotted, although
they are scarce.
Overall, by day 2 posthatching the visual system of common
snook larvae is developed sufficiently to locate and capture prey,
although due to the underdeveloped stage of the rods (which
provide better vision in low light [OConnell 1981; Kawamura
1984; Pankhurst 1996]) adequate light conditions are necessary
to optimize their ability to capture prey (Huse 1993). Taking the
rod development into consideration, light intensity during larval
development should be altered accordingly, and to investigate
this matter more samples of older larvae should be examined
using histology.
Digestive System Development
At hatching, common snook larvae had a simple undifferentiated straight gut linked to an unstructured mouth and anus, as described in other teleost species (Stroband and Dabrowski 1979;
Govoni 1980; Cousin and Baudin-Laurencin 1985; Govoni et al.
1986; Boulhic and Gabaudan 1992; Bisbal and Bengston 1995;
Roo et al. 1999; Pena et al. 2004). It is generally assumed that
lipid absorption takes place in the anterior intestine, and based
on this assumption and the importance that lipids have over the
larval development, all the histological work done in this study
was based on the anterior intestine development.
During the first 2 d, the common snook larval digestive system undergoes major changes. The anus and the mouth open,
gut cells undergo significant growth, development of organelles
is increased, and the intestinal valve is formed. On day 1, the alimentary canal differentiation is starting to appear, the mouth is
starting to be formed, but jaw cartilages are still developing. At
the cell level, organelles can be observed but their underdeveloped stage and low numbers render the digestive system unable
to function. The alimentary canal epithelium at some parts of
the luminal surface showed the presence of microvilli, although
510
YANES-ROCA ET AL.
in low numbers and underdeveloped, at the same time ciliated

cells were present in the lumen. Such ciliated cells were not
found after day 1, and similar findings were observed in other
species (Govoni et al. 1986; Loewe and Eckman 1988; Calzada
et al. 1998). The presence of these ciliated cells in larvae (which
did not have any peristaltic movements) was reported by Iwai
(1964) and contributes to the circulation of the intestinal contents (mainly yolk). Common snook larvae by day 2 posthatch
had absorbed the yolk sac, which was also observed in goldlined seabream Rhabosargus sarba (Ibrahim 2004) and in sea
bass Lates calcifer (Walford and Lam 1993). Other species,
such as Atlantic cod and sheephead seabream Archosargus probatocephalus (John and Tucker 1987; Kjrsvik et al. 1991),
exhausted their yolk sac at 4 d after hatching. The mouth was
open and the main jaw cartilages such as the Meckels cartilage
or the hyomandibular cartilage were present. These results were
also seen in other species, such as rainbow darter Etheostoma
caeruleum, white sucker Catostomus commersonii, logperch
Percina caprodes and Atlantic cod (McElman and Balon 1981;
Paine and Balon 1985; Kjrsvik et al. 1991). The dermal bones,
such as the maxillary and the premaxillary are formed later on;
however, food was observed in the gut.
Morphological and histological observations suggest that 2
d after hatching common snook larvae possess digestive organs
enabling digestion, absorption, and metabolization of endogenous food. At the onset of day 2 posthatch, enterocytes are morphologically developed, yet as with cod larvae (Kjrsvik et al.
1991), the digestive mechanisms are immature and their functionality relies on lipid absorption and possibly temporary lipid
storage in the anterior part of the gut (Tanaka 1972a; Stroband
and Dabrowski 1979). Although no histology of the rectal area
was done, the epithelial cells are probably responsible for food
protein ingestion and intracellular digestion (Iwai and Tanaka
1968; Tanaka 1972b; Watanabe 1982a, b).
Day 3 common snook larvae had continued to develop.
Mouth jaw cartilages were gaining in definition and speeding
their functionality. The alimentary canal is now more structured
with a clear differentiation between the different gut parts. Organelles, such as mitochondrion, rough endoplasmatic reticule,
and the golghi apparatus, increased in numbers. The microvilli
layer is increased in length and consistency.
In conclusion, the alimentary canal of common snook larvae
develops from a undifferentiated tube at hatching to a complex
tract before the onset of day 2 after hatching (when the yolk sac
is exhausted). This fast development is parallel to that of the
visual system; synchronization of the formation of these two
systems is important for prey capture and predator avoidance.
Together with the digestive and visual system development,
common snook larvae have a partially developed fin structure
that allows them to move in the water column and approach
prey.
Like their feeding habits and behavior, the location of common snook larvae in the planktonic column is unknown. This
paper has described the common snook larval development dur-
ing the first 14 d in laboratory conditions in order to provide

information for future studies on wild common snook larvae
and for aquaculture purposes. Although there are many theories
regarding the diet of common snook larvae, there are no reported
studies on this topic; common snook larval diets are one of the
major bottlenecks in snook aquaculture. In addition, issues such
as prey type and size during the first 57 d are still poorly understood. Therefore, the collection of wild larvae will provide
useful information to identify the optimal prey, improve the
rearing protocol, and increase knowledge of the wild common
snook larvae ecology.
ACKNOWLEDGMENTS
This work was supported by grants from the Institute of
Aquaculture at Stirling University, the Florida Fish and Wildlife
Conservation Commission, the National Oceanic and Atmospheric Administration funded research consortium, the Science and the Consortium for Ocean Replenishment, and the
Mote Scientific Foundation.
REFERENCES
Ali, M. A. 1959. The ocular structure, retimotor and photobehavioral responses
of juvenile Pacific salmon. Canadian journal of Zoology 39:511526.
Avila, E. M., and J. M. Juario. 1987. Yolk and oil-globule utilization and development morphology of the digestive tract epithelium in larval rabbitfish,
Sinatus guttatus. Aquaculture 65:319331.
Batty, R. S. 1987. Effect of light intensity on activity and food-searching of larval
herring (Clupea harengus): a laboratory study. Marine Biology 94:323327.
Blaxter, J. H. S. 1992. The effect of temperature on larval fishes. Netherlands
Journal of Zoology 42:336357.
Blaxter, J. H. S. 1986. Development of sense organs and behaviour of teleost
larvae with special reference to feeding and predator avoidance. Transactions
of the American Fisheries Society 115:98114.
Blaxter, J. H. S. 1968. Light intensity, vision and feeding in young place. Journal
of Experimental Marine Biology and Ecology 48:3953.
Blaxter, J. H. S., and M. P. Jones. 1967. The development of the retina and
retinomotor responses in the herring. Journal of Marine Biology 47:677697.
Blaxter, J. H. S., and M. Staines. 1970. Pure cone retina and retinomotor responses in larval teleosts. Journal of the Marine Biology Association 50:449
460.
Bisbal, G. A., and D. A. Bengston. 1995. Development of the digestive tract in
larval summer flounder. Journal of Fish Biology 47:277291.
Boulhic, M., and J. Gabaudan. 1992. Histological study of the organogenesis
of the digestive system and swim bladder of the Dover sole, (Solea solea)
(Linnaeus 1758). Aquaculture 102:373396.
Calzada, A., A. Medina, and M. L. Gonzalez de Canales. 1998. Fine structure
of the intestine development in cultured sea bream larvae. Journal of Fish
Biology 53:340365.
Cousin, J. C. B., and F. Baudin-Laurencin. 1985. Morphogenèse de lappareil
digestif et de la vessie gazeuse du turbot, Scophthalmus maximus. [Morphogenesis of the digestive system and swim bladder of the turbot Scophthalmus
maximus.] Aquaculture 47:305319.
Deplano, M., J. P. Diaz, M. Connes, R. Kentouri-Divanach, and F. Cavalier.
1991. Appearance of lipid-absorption capacities in larvae of the sea bass, Dicentrarchus labrax during transition to the exotrophic phase. Marine Biology
108:361371.
Devillers, C. 1961. Structural and dynamic aspects of the development of the
teleostean egg. Advance Morphogenesis 1:379428.

Eckman, R. 1987. Pathological changes in the midgut epithelium on grayling,
Thymallus thymallus L., and larvae reared on different kinds of food, and
their relation to mortality and growth. Journal of Fish Diseases 10:9199.
Evans, B. I., and H. Browman. 2004. Variation in the development of fish retina.
Pages 145166 in J. J. Govoni, editor. The development of form and function
in fishes and the question of larval adaptation. American Fisheries Society,
Symposium 40, Bethesda, Maryland.
Ferrais, R. P., J. D. Tan, and M. C. de laCruz. 1987. Development of the digestive
tract of milkfish, Chanos chanos (Forskal): histology and histochemistry.
Govoni, J. J. 1980. Morphological, histological, functional aspects of alimentary
canal and associated organ development in larval (Leiostomus xanthurus).
Reveu Canadiene 39:6980.
Govoni, J. J., G. W. Boenhlert, and Y. Watanabe. 1986. The physiology of
digestion in fish larvae. Environmental Biology of Fish 16:5977.
Grimes, P. J., and D. A. Robertson. 1981. Egg and larva development of the
silver warehou Seriolella punctata (Pisces: Centrolophidae). New Zealand
Journal of Marine and Freshwater Research 15:261266.
Howells, R. G., A. J. Sonski, P. L. Shafland, and B. D. Hilton. 1990. Lower
temperature tolerance of snook, Centropomus undecimalis. Northeast Gulf
Science 11:155158.
Huse, I. 1993. First feeding larval sensory perception and behavioural programming: implications for systems and procedures in culture. Pages 146152 in
B. T. Walther and H. J. Fyhn, editors. Physiology and biochemical aspects of
fish development. University of Bergen, Bergen, Norway.
Ibrahim, F. S. 2004. Reproductive biology of wild gold lined seabream, Rhabdosargus sarba captive breeding and larval development in the Sultanate of
Oman. Doctoral dissertation. Stirling University, Stirling, Scotland.
Iwai, T. 1964. Feeding and ciliary conveying mechanisms in larvae of salmonid
fish Plecoglossus altivelis. Physiological Ecology 12:3844.
Iwai, T., and M. Tanaka. 1968. The comparative study of the digestive tract of
the teleost larvae-III. Epithelial cells in the posterior gut of helfbreak larvae.
Bulletin of the Japanese Society of Scientific Fisheries 34:4448.
John, W., and J. R. Tucker. 1987. Sheephead culture and preliminary evaluation
for farming. The progressive fish culturist 49:224228.
Jonh, K. R., and A. D. Hasler. 1956. Observations on some factors affecting the
hatching of eggs and the survival of young shallow-water cisco Leucichthys
artedi Le Suer, in Lake Mendota, Wisconsin. Limnology and Oceanography
1:176194.
Kawamura, G. 1984. The visual cell morphology of Pagrus pagrus and its
adaptive changes with shift from pelagic and benthic habitats. Bulletin of
Japanese Society of Scientific Fisheries 50:19751980.
Kjrsvik, E., T. Van derMeer, J. Kryvi, S. Arnfinnson, and P. G. Kvenseth.
1991. Early development of the digestive tract of cod larvae, Gadus morhua
L., during start feeding and starvation. Journal of fish Biology 38:115.
Lau, S. R., and P. L. Shafland. 1982. Larval development of snook Centropomus
undecimalis (Pisces: Centropomidae). Copeia 1982:618627.
Loewe, H., and R. Eckman 1988. The ontogeny of the alimentary tract
of coregonid larvae: normal development. Journal of Fish Biology 33:
841850.
Mani-Ponset, L., E. Guyot, J. P. Diaz, and R. Connes. 1996. Utilization of yolk
reserves during postembryonic development in three teleostean species: the
sea bream, (Sparus aurata), the sea bass, (Dicentrarchus labrax) and the
pike-perch (Stizostedion lucioperca). Marine Biology 126:539547.
McElman, J. F., and E. K. Balon. 1981. Early ontogeny of the white sucker,
Catostomus comersoni, with steps of saltatory development. Journal of Fish
Biology 5:191224.
OConnell, C. P. 1981. Development of organ systems in the northern anchovy,
Engraulis mordax, and other teleosts. American Zoology 21:429446.
511
Oppenheimer, J. M. 1947. Organization of the teleost blastoderm. Quarterly

review of Biology 22:105118.
Paine, M. D., and E. K. Balon. 1985. Early development of the northern log
perch, Percina caprodes semifsciata, according to the theory of the saltatory
ontogeny. Environmental Biology of Fishes 11:173190.
Pankhurst, N. W. 1996. Changes in visual morphology throughout life history
stages of the New Zealand snapper, Pagrus auratus. New Zealand Journal of
Marine Freshwater Research 30:7990.
Paul, A. J. 1983. Light, temperature, nauplii concentrations, and prey capture
by first feeding pollock larvae (Theragra chalcogramma). Marine Ecology
Progress Series 13:175179.
Pena, R., S. Dumas, R. Saldivar-Lucio, G. Garcia, A. Trasvin, and D. HernandezCeballos. 2004. The effects of light intensity on first feeding of the spotted
sand bass Paralabrax maculatofasciatus Steindachner larvae. Aquaculture
Research 35:345349.
Porter, S. M., and G. H. Theilacker. 1999. The development of the digestive
tract and eye in larval walleye pollock, Theragra chalcograma. U.S. National
Marine Fisheries Service Fishery Bulletin 97:722729.
Roo, F. J., J. Socorro, M. S. Izquierdo, M. J. Caballero, C. M. Hernandez-Cruz,
A. Fernandez, and H. Fernadez-Palacios. 1999. Development of red porgy
(Pagrus pagrus) visual system in relation with changes in the digestive tract
and larval feeding habits. Aquaculture 179:499511.
Rudnick, D. 1955. Teleost and birds. Pages 297314 in B. H. Willier, P. A. Weiss,
and V. Hamburger, editors. Analysis of development. Saunders, Philadelphia.
Sarasquete, M. C., A. Polo, and M. Yufera. 1995. Histology and histochemistry
of the development of the digestive system of larval gilthead seabream, Sparus
aurata L. Aquaculture 130:7992.
Simpson, A. C. 1956. The pelagic phase. Pages 207250 in M. Graham, editor.
Sea fisheries: their investigation in the United Kingdom. Arnold, London.
Smith, S. 1957. Early development hatching. Pages 323359 in M. E. Brown,
editor. The physiology of fishes. Academic Press, New York.
Stroband, H. W. J., and K. R. Dabrowski. 1979. Morphological and physiological aspects of the digestive system and feeding in freshwater fish larvae.
Pages 355376 in M. Fontaine, editor. La nutrition des poissons. [Fish nutri
tion.] Centre National dEtudes
et de Recommandations sur la Nutrition et
lAlimentation, Paris.
Tanaka, M. 1972a. Studies on the structure and function of the digestive system
in teleost larvae -IV changes in the epithelium related to fat absorption in
the anterio-medium part of the intestine after feeding. Japanese Journal of
Ichthyology 19:1525.
Tanaka, M. 1972b. Studies on the structure and function of the digestive system in teleost larvae-V. Epithelial changes in the posterior gut and protein
ingestion. Japanese Journal of Ichthyology 19:172180.
Taylor, R. G., H. J. Grier, and J. A. Whittington. 1998. Spawning rhythms of
common snook in Florida. Journal of Fish Biology 53:502520.
Waddington, C. H. 1956. Principles of development. Allen and Unwin, London.
Walford, J., and T. J. Lam. 1993. Development of digestive tract and proteolytic
enzyme activity in sea bass (Lates calcifer) larvae and juveniles. Aquaculture
109:187205.
Watanabe, Y. 1982a. Intercellular digestion of horseradish peroxidase by the
intestinal cells of teleost larvae and juveniles. Bulletin of the Japanese Society
of Scientific Fisheries 48:3742.
Watanabe, Y. 1982b. Ultraestructure of epithelial cells of the antero-median
intestine and the rectum in larval and juvenile teleost. Bulletin of the Faculty
of Fisheries 33:217228.
Wittenrich, M. L., N. R. Rhody, R. G. Turigan, and K. L. Main. 2009. Coupling
osteological development of the feeding apparatus with feeding performance
in common snook, Centropomus undecimalis, larvae: identifying morphological constrains to feeding. Aquaculture 294:221227.


An Injectable, Slow-Release Implantation Method for

Exposing Fish to Chemicals over a Period of Weeks
a
Gerald E. Zaroogian , Ruth E. Gutjahr-Gobell , Doranne Borsay Horowitz , Saro

a
Jayaraman , Mark Cantwell , Clinton O. Chichester & Lesley J. Mills
U.S. Environmental Protection Agency, Office of Research and Development, National

Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, 27 Tarzwell
Drive, Narragansett, Rhode Island, 02882, USA
b
Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island,

Kingston, Rhode Island, 02881, USA
To cite this article: Gerald E. Zaroogian, Ruth E. Gutjahr-Gobell, Doranne Borsay Horowitz, Saro Jayaraman, Mark Cantwell,
Clinton O. Chichester & Lesley J. Mills (2012): An Injectable, Slow-Release Implantation Method for Exposing Fish to Chemicals
over a Period of Weeks, North American Journal of Aquaculture, 74:4, 512-521


DOI: 10.1080/15222055.2012.697097
ARTICLE
An Injectable, Slow-Release Implantation Method for

Exposing Fish to Chemicals over a Period of Weeks
Gerald E. Zaroogian, Ruth E. Gutjahr-Gobell, Doranne Borsay Horowitz,
Saro Jayaraman, and Mark Cantwell
U.S. Environmental Protection Agency, Office of Research and Development,

National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division,
27 Tarzwell Drive, Narragansett, Rhode Island 02882, USA
Clinton O. Chichester
Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island, Kingston,
Rhode Island 02881, USA
Lesley J. Mills*
U.S. Environmental Protection Agency, Office of Research and Development,
National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division,
27 Tarzwell Drive, Narragansett, Rhode Island 02882, USA
Abstract
A slow-release, injectable implant method was developed for administering test chemicals to cunners Tautogolabrus
adspersus. The implant is composed of a matrix of a test chemical homogenized in a mixture of Ethocel (Dow Chemical)
and coconut oil. The effectiveness of a subcutaneous implant of this matrix in vivo was determined by tracing plasma
concentrations of three separate chemicals (estradiol, ethynylestradiol, and atrazine) over time in treated male
cunners. Release from the implant was determined based on the percentage of the implanted concentration of test
chemical (plus metabolites) that was detected in fish plasma over a 12-week period after implantation. Circulating
estrogen concentrations measured in plasma from two different cunners that received the estradiol implant were
almost identical, indicating that there is a reasonably even distribution of test chemical within the Ethocelcoconut oil
preparation and that individual variability may be minimal for release of test chemical from the implant. Metabolites
of estradiol and atrazine were a major portion of the circulating concentration of these chemicals. Estradiol and
atrazine demonstrated metabolic and clearance profiles that were very different from those of the xenoestrogen
ethynylestradiol. A follow-up in vitro study was conducted to further characterize the release of estradiol from the
implant matrix. Results showed a rapid release of estradiol from the matrix bolus during the first 24 h, followed by a
more gradual release over subsequent days. The in vitro tests indicated that measuring in vivo plasma concentrations
may not accurately reflect the release rate of a chemical from the implant matrix, in part because metabolism and
clearance affect the circulating concentrations in vivo.
For years, aquaculturists and aquatic toxicologists have had

an interest in finding effective and efficient ways to administer
hormones or other chemicals to fish in a sustained, controlled
manner. Administration of hormones to fish is used for multiple
*Corresponding author: mills.lesley@epa.gov

Received January 12, 2012; accepted May 19, 2012
512
purposes in aquaculture, including the manipulation of reproduction, the induction of rapid growth, and the production of
monosex populations (Pandian and Sheela 1995; Shelton and
Mims 2003). In the past, exogenous hormones and chemicals
SLOW-RELEASE METHOD FOR EXPOSING FISH TO CHEMICALS
have been administered to fish through a variety of methods,

including diet (e.g., Gutjahr-Gobell et al. 1999; Patyna et al.
1999; Bayley et al. 2002; Madsen et al. 2003; Keen et al. 2005),
water column exposure (e.g., Hunter and Donaldson 1983;
Gimeno et al. 1998; Kramer et al. 1998; Metcalfe et al. 2001;
Segner et al. 2003; Arslan and Phelps 2004; Lahnsteiner et al.
2006), and oral gavage (Sundararaj and Goswami 1968; Pizza
and OConnor 1983). Although exposure through such methods may be desirable when trying to simulate environmental
routes of exposure, the administration of hormones or chemicals through injection or surgical implantation may be more
practical in circumstances where a direct route of exposure with
minimal effort is needed.
Incorporation of test chemicals into a vegetable oil vehicle
(e.g., olive oil, corn oil, peanut oil, sesame oil, coconut oil, or
cocoa butter), cholesterol, a cholesterolcocoa butter matrix,
or a cholesterolanimal lard matrix has been used by various
researchers for intraperitoneal implantation of chemicals into
fish (Leatherland 1985; Pankhurst et al. 1986; Carolsfeld et al.
1988; Crim et al. 1988; Cyr and Eales 1989; Garcia 1989;
Bisbal and Specker 1991; Donohoe and Curtis 1996; Black
et al. 1998; Mandiki et al. 2004; Sangiao-Alvarellos et al.
2005; Wang et al. 2005). A consideration with intraperitoneal
implantation is that the test chemical is placed adjacent to key
abdominal organs (e.g., gonads or liver), which could result
in an increased, and perhaps undesirable, local exposure of
these organs. However, cholesterol and some oils appear to
provide a prolonged release. For example, Yamada et al. (1997)
reported that after rainbow trout Oncorhynchus mykiss received
intraperitoneal implantation with a steroid in coconut oil, an
initial increase in plasma steroid concentration was observed 1 d
after implantation, followed by a gradual decrease over the subsequent 15 d. They suggested that coconut oil, which solidifies
when implanted, was responsible for the slow release of steroid
into the peritoneal cavity. Wang at al. (2005) reported similar
results when using cocoa butter as a carrier for the injection
of cortisol into grass carp Ctenopharyngodon idella. Sherwood
et al. (1988) found that cholesterol-based pellets yielded
sustained releases of gonadotropin-releasing hormone analog
(GnRHa) in vitro over a period of 28 d. Crim et al. (1988) found
that both intraperitoneal and intramuscular pellet implants
composed of plasma gonadotropin and cholesterol (with or
without cocoa butter) yielded sustained releases of luteinizing
hormone releasing hormone analog in juvenile rainbow trout.
Implantation of crystalline exogenous compounds enclosed
in silicone or silastic tubing has been used with some success,
resulting in biological effects lasting several weeks to months in
some cases (Pickering and Duston 1983; Pankhurst et al. 1986;
Trudeau et al. 1991; Joakim Larsson et al. 2002; Yamaguchi et al.
2004; Goncalves et al. 2007). However, this method requires that
minor surgery be conducted to implant the capsule into each
fish. In addition, chemicals that are hydrophilic do not penetrate
easily through the hydrophobic matrix, thus decreasing implant
effectiveness (Mylonas and Zohar 2000).
513
A less-invasive method that has been used with some success

is the incorporation of chemicals into coconut oil or into a cocoa
buttercholesterol mixture. The mixture may then be injected
into fish intramuscularly or subcutaneously (Lee et al. 1986;
Scott et al. 1999). Coconut oil has been used successfully in our
laboratory (U.S. Environmental Protection Agency [USEPA],
National Health and Environmental Effects Research Laboratory [NHEERL]) to implant exogenous estrogens and other
chemicals subcutaneously into summer flounder Paralichthys
dentatus (Mills et al. 2001).
Our research into the biological effects of endocrinedisrupting chemicals on fish necessitated the development of
an inexpensive delivery method that could be used on our test
species, the cunner Tautogolabrus adspersus. Since laboratory
water exposure of cunners was impractical due to the size and
spawning behavior of this species, the goal of the present study
was to develop and test the utility of a minimally invasive,
injectable implant that would continuously release exogenous
chemicals into the circulation of treated fish over time. We report
here on the suitability and effectiveness of our implant method
by tracing plasma concentrations of three separate chemicals
(estradiol, ethynylestradiol, and atrazine) that were implanted
by subcutaneous injection just below the dorsal fin of laboratoryheld cunners. Plasma concentrations of test chemical and major
metabolites were measured over time. To better characterize the
release of chemical from our implant without the in vivo effects of metabolism and clearance, we also examined release of
radio-labeled estradiol from boluses of implant matrix in vitro.
METHODS
Animals
Cunners were collected from the East Passage of
Narragansett Bay, Rhode Island, off a large stone pier at
the southeastern end of Jamestown (Conanicut Island) during
the summer in 19992005. Details of collection methods are
provided by Gutjahr-Gobell et al. (2002). Briefly, cunners were
captured by using modified Gee minnow traps (Memphis Net
and Twine, Memphis, Tennessee) and were transported to the
laboratory in coolers filled with ambient seawater. Fish were
held in large, aerated rectangular (4,400 L) or round (1,000
L) holding tanks that received flow-through Narragansett Bay
seawater. Cunners were fed an ad libitum ration of thawed and
chopped krill, squid, and mussels; the fish were held over winter
in the laboratory at the ambient temperatures and photoperiod
of Narragansett Bay.
Chemicals
Ethocel Standard FP Premium 10 (a formulation of ethyl
cellulose) was obtained from Dow Chemical (Midland, Michigan). Refined coconut oil (100% pure), distributed by Spectrum
Essentials (Petaluma, California), was purchased locally.
17-estradiol (1,3,5-[10]-estratrien-3,17-diol; Chemical
Abstracts Service [CAS] Number 50-28-2) and estrone
514
ZAROOGIAN ET AL.
(1,3,5[10]-estratrien-3-ol-17-one; CAS Number 53-16-7) were

obtained from Steraloids (Wilton, New Hampshire). 17ethynylestradiol
(17-ethynyl-1,3,5[10]-estratriene-3,17diol; CAS Number 57-63-6), estriol (1,3,5[10]-estratriene3,16,17-triol; CAS Number 50-27-1), and pyridine (highperformance liquid chromatography grade; CAS Number 11086-1) were purchased from Sigma-Aldrich (St. Louis, Missouri).
The derivatizing agent bis(trimethylsilyl)trifluoroacetamide
(BSTFA) was obtained from Supelco (Bellefonte, Pennsylvania). Atrazine (6-chloro-N-ethyl-N -[1-methylethyl]1,3,5-triazine-2,4-diamine; CAS Number 1912-24-9) and
its metabolites deethylatrazine (DEA; 2-chloro-4-amino6-isopropylamino-s-triazine), deisopropylatrazine (DIA; 2chloro-4-ethylamino-6-amino-s-triazine), and diaminochlorotriazine (DACT; 2-chloro-4,6-diamino-1,3,5-triazine) were
provided to the Reproductive Toxicology Division (USEPA,
NHEERL) by Syngenta Crop Protection (Greensboro, North
Carolina). The deuterated estradiol (17-estradiol-d4 ) and
atrazine (atrazine-ethylamine-d5 ) used as internal standards
for analytical chemistry procedures were purchased from
Cambridge Isotope Laboratories (Andover, Massachusetts).
Preparation of the Injectable Implant Matrix
Stock solutions of the test chemicals were prepared in different solvents, depending upon their solubilities, as follows: estradiol was dissolved in a mixture of 0.5 mL of ethanol and 0.5 mL
of acetone to make a 100-mg/mL stock solution; ethynylestradiol was dissolved in acetone to prepare a 100-mg/mL stock
solution; and atrazine was dissolved in a mixture of 0.5 mL of
chloroform and 0.5 mL of ethanol to prepare a 75-mg/mL stock
solution.
To prepare the slow-release matrix, Ethocel was dissolved in
methylene chloride. The solution was then reduced in volume to
approximately 1 mL. The desired amount of test chemical stock
solution was added to the Ethocelmethylene chloride mixture
and was mixed well. The ratio of Ethocel to chemical in the
ethynylestradiol and atrazine preparations was approximately
40:1. The 40:1 ratio was chosen based upon prior experimentation with the solubilities of these compounds in the final implant matrix. Our objective was to maximize the amount of test
chemical that was incorporated into the matrix without having
precipitation occur during later steps of the implant preparation
process. In the preparation of the estradiol implant, the ratio of
Ethocel to estradiol was limited to a maximum of approximately
17:1 because any concentration of Ethocel exceeding 50 mg/mL
was prone to precipitation during subsequent steps.
Slow-release matrix was made in 5-mL batches. Liquid coconut oil (2 mL), used as a carrier for the slow-release matrix, was added to the Ethocelmethylene chloridetest chemical mixture. The mixture was homogenized thoroughly until
emulsified. An additional 2-mL quantity of coconut oil was
added, and the mixture was homogenized again. Since preliminary experiments showed that residual solvent alone caused
irritation and necrosis at the injection site in fish that received
the implant matrix containing no test chemical, the mixture was

placed in a light-tight vial and stirred under vacuum on a magnetic stirrer until the solvent had completely dissipated. The
implant preparation was then brought up to a final volume of
5 mL by adding coconut oil and mixing thoroughly. Control
implants were prepared with the same amounts of solvent and
Ethocel as the highest test chemical concentration but with no
chemical added. Just prior to injection, the implant matrix was
liquified by gently warming to approximately 35 C. Each fish
was lightly anesthetized with tricaine methanesulfonate (MS222). By use of a 1-mL glass syringe with an 18-gauge needle,
the appropriate liquified matrix (at 2 L/g of fish wet weight)
was implanted subcutaneously into each fish just below the dorsal fin. Gentle finger pressure was applied over the injection site
for a few seconds until the coconut oil in the bolus solidified,
thus preventing any leakage of the matrix.
Experimental Design
All fish tanks were aerated and received 1820 C flowthrough seawater at a flow rate of 1 L/min. To provide submerged cover for the fish, each tank also contained a 20-cm
length of 10-cm-diameter polyvinyl chloride pipe. Fish were allowed to acclimate to experimental conditions for 35 d before
an experiment started. All fish were fed fresh or thawed mussels
ad libitum every day. Further details of the experimental system
are described by Gutjahr-Gobell et al. (2002).
Each plasma sample was obtained from blood that was drawn
from a single fish; individual samples were not pooled for analysis. Percentages of each chemical in the fish plasma samples
taken over time were calculated by using the following formula:
%Fp =
Cp
100,
Wf D
where %Fp is the percentage of the nominal concentration of

implanted chemical that was detected in the fish plasma, Cp is the
amount of chemical (g) measured in 1 mL of fish plasma, Wf
is the weight of the fish (g), and D is the nominal concentration
of chemical (g/g) administered to the fish.
Chemical analysis of boluses was not possible in either the in
vivo or in vitro studies because of the high bolus lipid content.
Estradiol treatment.A repeat sampling design was used to
obtain blood samples during this experiment, meaning the same
fish was sampled at six time points. This design required the use
of very large cunners that could tolerate repeated blood draws.
Because the availability of sufficiently large fish was limited,
only two fish were used. Each of two large male fish received a
single concentration of estradiol (6 g/g of fish wet weight) in
the slow-release implant matrix. Fish wet weight was approximately 180 g, and fish length (measured as total length) was
approximately 22 cm. Each individual was kept in its own tank
an 80-cm-tall, 114-L-capacity, high-density polyethylene barrel
(47-cm diameter) with a clear Plexiglas cover. Just before implantation, 0.4 mL of blood was drawn from a caudal vein of each
fish by using a 1-mL tuberculin syringe with a 22-gauge needle rinsed with a heparin sodium salt solution (1,000 units/mL;
United States Biochemical). These pretreatment samples were
used as controls. Subsequent blood samples (0.4 mL) were taken
at 3 h after implantation and at 1, 2, 3, 4, and 7 d after implantation. All blood samples were processed as described by GutjahrGobell et al. (2002). Plasma was stored at 80 C until assayed.
Ethynylestradiol treatment.Five smaller male cunners
were used for the ethynylestradiol study. Wet weight of fish
ranged from 23 to 36 g and averaged 30 g (SD = 5). Length was
between 12.5 and 14.0 cm and averaged 13.2 cm (SD = 0.6).
Four males received implanted ethynylestradiol at 1.2 g/g of
fish weight, and one male received a control matrix containing
no ethynylestradiol. Fish were housed individually in 38-L
(10-gal) glass aquaria with aeration and flow-through seawater.
Each fish was sampled twice during the experiment by using
methods previously described (Gutjahr-Gobell et al. 2002).
Blood samples were drawn from one fish at 1 and 8 d after
implantation, from a second fish at 3 and 10 d postimplantation,
from a third fish at 5 and 12 d postimplantation, and from
a fourth fish at 7 and 14 d postimplantation. Blood from the
control fish was sampled at 7 and 14 d after control matrix
implantation. All blood samples were processed as described
by Gutjahr-Gobell et al. (2002). Plasma samples were stored at
80 C until assayed.
Atrazine treatment.Six male cunners were housed individually in 38-L (10-gal) glass aquaria with aeration and flowthrough seawater. Wet weight of fish ranged from 32 to 47 g,
with a mean of 41 g (SD = 6). Length was between 13.1 and
14.6 cm, with a mean of 14.0 cm (SD = 0.6). Four fish were
given atrazine in slow-release matrix at 2.4 g/g of fish weight
on day 0 and were sampled at 1, 3, 5, and 7 d postimplantation.
A different fish was bled on each sampling day, and as much
blood as possible was drawn from each fish. The process of
sampling and separating plasma was as described by GutjahrGobell et al. (2002). Slow-release matrix containing no atrazine
was implanted into two control fish; one of the control fish was
sampled on day 1, and the other control fish was sampled on
day 7. Plasma samples were stored at 80 C until assayed.
In vitro study.To better characterize the release of chemical from our implant without the influence of metabolism or
clearance, we conducted an in vitro study to quantify the release
of estradiol from the implant matrix into Leibovitzs L-15 culture medium (used as a surrogate for fish plasma). The average
wet weight of cunners used in our in vivo laboratory exposure
experiments was approximately 35 g, which meant that a fish
receiving 2 L of matrix per gram of wet weight would receive a
bolus of 0.07 mL. We therefore chose a bolus volume of 0.07 mL
for the in vitro study based on the 2-L/g dose received by a
typical 35-g fish.
Slow-release matrix was prepared in the same way as
the estradiol implantation matrix described earlier for the in
vivo experiment, except that 50 L of tritiated estradiol (0.99
curie/mmol in ethanol) were added to the Ethocel. Only a very
515
small amount of the total estradiol in each implant matrix was

tritiated estradiol. After the solvents had dissipated, the slowrelease matrix was brought up to a final volume of 5 mL with
coconut oil and was homogenized thoroughly. A tuberculin syringe with the tip cut off was used to draw up 0.07 mL of
slow-release matrix for each bolus. The syringe and matrix were
chilled at 20 C for 48 h to solidify the matrix into a compact
bolus. After 48 h, each chilled bolus was expelled from the syringe into a scintillation vial containing 2.0 mL of L-15 medium
at 18 C. Scintillation vials were then incubated at 18 C and
gently shaken on an orbit shaker (Lab-Line Instruments) until
sampled.
We prepared matrix at two dosage levels. The first preparation contained 210 g of estradiol and 3.08 mg of Ethocel per
bolus, resulting in an Ethocel: estradiol ratio of approximately
15:1. If implanted into a 35-g fish, this bolus would deliver an
estradiol concentration of 6 g/g of fish wet weight. The second
preparation contained 42 g of estradiol and 0.7 mg of Ethocel
per bolus, resulting in an Ethocel : estradiol ratio of approximately 17:1. If implanted, this latter bolus would deliver an
estradiol concentration of 1.2 g/g of fish wet weight.
Tritiated estradiol released from boluses was measured at
time 0; at 1, 3, and 6 h; and at 1, 2, 3, 4, 7, 9, and 11 d. Three replicates at each dosage concentration were sampled at each time
period (n = 66). For sampling, a bolus was removed and 3 mL of
Hionic-Fluor scintillation cocktail (PerkinElmer) were added to
the remaining L-15 medium. Radioactivity in the medium was
counted in disintegrations per minute (dpm) by using a Packard
Tri-Carb Liquid Scintillation Analyzer (Model 2500TR;
PerkinElmer). A new bolus was used for each sampling.
The cumulative percentage of estradiol measured in the L-15
medium at each sampling time was calculated as 100 {[mean
dpm of the three 3H-estradiol-spiked replicate samples at time t
(d)]/[mean dpm at time 0]}.
Analytical Chemistry
Estrogens.Ethynylestradiol, estradiol, estrone, and estriol
were extracted from the plasma samples before being analyzed
by gas chromatography (GC)mass spectrometry (MS). The
method used to extract estrogens from fish plasma was adapted
from Lopez de Alda and Barcelo (2001) and Jeannot et al.
(2002), with some modifications. Fish plasma samples were
ground to a fine powder with anhydrous sodium sulfate. Three
milliliters of diethyl ether and known amounts of the internal
standard, 17-estradiol-d4 , were added to each sample. Samples were then vortexed (30 s), sonicated (10 min), and centrifuged (1,200 revolutions/min; 10 min). By use of disposable pipettes, the supernatants were carefully transferred into
2-dram screw-top vials. This extraction procedure was repeated
two more times. The combined extracts were then volume reduced, solvent exchanged to methanol, and adjusted to a final
volume of 2 mL with a 30:70 solution of water: methanol.
The extract mixture was passed through a C-18 cartridge
(Sep-Pak Environmental tC18 cartridge WAT036800; Waters,
516
ZAROOGIAN ET AL.
Milford, Massachusetts) that was previously conditioned with

water (5 mL) and methanol (5 mL). After the cartridge was
dried under full vacuum for 1 min, estrogens were eluted with
8 mL of acetonitrile. The acetonitrile was then volume reduced
and taken to dryness under a gentle stream of nitrogen. For derivitization, 25 L each of BSTFA and pyridine were added to
each sample, which was then heated at 65 C for 25 min. The
sample was allowed to cool and was vortexed; 2 L were then
immediately injected into an Agilent 6890 gas chromatograph
(Agilent Technologies, Santa Clara, California) that was fitted
with a mass spectrometer (5973N; Agilent) equipped with a
60-m DB-5ms capillary column (Agilent J&W Scientific).
Further details of the analytical method are as described by
Hernando et al. (2004).
A standard curve was made by the addition of known amounts
of ethynylestradiol, estradiol, and estrone that were brought to
1 mL with hexane. Method detection limits were 4.96, 8.50, and
11.4 ng/mL for estradiol, estrone, and ethynylestradiol, respectively. Only a single point calibration was performed for estriol;
no method detection limit was determined.
Atrazine.Plasma samples were extracted and analyzed for
atrazine and the metabolites DACT, DIA, and DEA based on
methods described by Brzezicki et al. (2003). Final extract vol-
ume was 0.5 mL. Two microliters of extract were injected into an
Agilent 6890 gas chromatograph fitted with an Agilent 5973N
mass spectrometer that was equipped with a DB-17ms analytical column (Agilent J&W Scientific). Method detection limits
were 0.53, 0.49, 0.23, and 0.16 ng/mL for atrazine, DACT, DIA,
and DEA, respectively.
RESULTS
Estradiol
Prior to treatment, no measurable estradiol, estrone, or estriol was found in the plasma of the two male cunners used
for the estradiol experiment. Concentrations of estradiol and
estrone detected in the plasma were summed to determine the
total percentage of implanted estrogen that was in circulation at
any given time. The concentration of estrogen measured in the
cunner plasma after implantation with estradiol at 6 g/g of fish
wet weight is shown in Table 1. Chemical analysis of plasma for
estrogens revealed that in estradiol-treated fish, estrone made up
1927% of the total circulating estrogen by 3 h after estradiol
implantation, and the estrone percentage increased to between
35% and 56% at subsequent sampling times. No estriol was detected in any plasma sample. The total concentration of estrogen
TABLE 1. Concentrations (ng/mL) of the test chemicals and their metabolites detected over time in plasma from male cunners that received the implanted slowrelease matrix. Percentage of the total is indicated in parentheses ( = indicates no sample; ND = no chemical was detected; DACT = diaminochlorotriazine;
DIA = deisopropylatrazine; DEA = deethylatrazine).
Time after implantation

Chemical or
metabolite
Fish Aa
Estradiol
Estrone
Estriol
Total
Fish B
Estradiol
Estrone
Estriol
Total
3h
2d
3d
4d
5d
7d
8d
10 d
12 d
14 d
Estradiol Treatment (6 g/g of fish wet weight)

4,428 (73)
1,672 (27)
ND
6,100
765 (56)
597 (44)
ND
1,362
451 (53)
396 (47)
ND
847
255 (53)
222 (47)
ND
477
136 (44)
170 (56)
ND
306
44.7 (46)
52.8 (54)
ND
97.5
ND
ND
ND
ND
4,903 (81)
1,115 (19)
ND
6,018
780 (59)
553 (41)
ND
1,333
648 (52)
608 (48)
ND
1,256
396 (65)
212 (35)
ND
608
77.8 (64)
43.7 (36)
ND
121.5
ND
ND
ND
ND
ND
ND
ND
ND
176
ND
224
ND
172
ND
77.3
ND
Ethynylestradiol
Estrone
Atrazine
DACT
DIA
DEA
Total
1d
Fish A was found dead on day 11.
Ethynylestradiol Treatment (1.2 g/g of fish wet weight)

710
1,455
664
493
99
172
ND
ND
Atrazine Treatment (2.4 g/g fish wet weight)
17.9 (18)
2.4 (8)
ND
ND
15.1 (15)
18.1 (62)
6.0 (100) 11.9 (82)

34.6 (35)
4.8 (16)
ND
ND
30.9 (31)
3.8 (13)
ND
2.6 (18)
98.5
29.1
6.0
14.5
FIGURE 1. Percentage of estradiol plus estrone measured in 1 mL of fish

plasma over time relative to the total amount of estradiol implanted in two male
cunners.
measured in the plasma of the two males was similar at every

sampling time (Table 1).
The percentage of estrogen (estradiol and estrone) in plasma
at any given time was very low in comparison with the concentration of implanted estradiol; the maximum estrogen percentage (i.e.,%Fp ) of approximately 0.56% was observed at 3 h
postimplantation (Figure 1). At 7 d postimplantation, the %Fp
had dropped to an average of 0.01%, representing a plasma
estrogen concentration of about 110 ng/mL. Plasma estrogen
was not detectable by day 10. Since the method detection limit
for estradiol was 4.96 ng/mL using GCMS, estrogen may have
been present at concentrations below this detection limit in the
day-10 samples.
Ethynylestradiol
No ethynylestradiol was detected in the plasma of cunners with control matrix implants. In fish that received the
ethynylestradiol implant, a small concentration of estrone was
detected on days 1 and 3 (Table 1); the commercial stock of
ethynylestradiol that was used to prepare the implants was
analyzed, and estrone was found to be a minor contaminant.
The maximum concentration of ethynylestradiol (1,455 ng/mL
of plasma) was observed at 3 d postimplantation and represented about 4% of the implanted concentration. On days 1,
5, and 7, circulating ethynylestradiol (%Fp ) was 1.7, 1.8, and
1.4%, respectively, of the implanted concentration (Figure 2).
Ethynylestradiol was still detectable in plasma from treated cunners at 14 d after implantation (Table 1).
Atrazine
No atrazine was detected in the two males that received the
control matrix. Atrazine metabolites made up a large percentage
(82100%) of the total chemical in circulation (Table 1). At 5
and 7 d after atrazine implantation, only metabolites were still
detectable in the plasma of treated fish. By 7 d postimplantation,
517
FIGURE 2. Percentage of ethynylestradiol plus estrone measured in 1 mL of

fish plasma over time relative to the total amount of ethynylestradiol implanted
in male cunners.
82% of the total chemical detected was DACT and the remaining
18% was DEA, whereas no atrazine or DIA was detected.
The percentage of implanted atrazine that was in circulation
at any given time (%Fp ; Figure 3) was calculated based on
the total detected concentration of atrazine plus its three major
metabolites (DACT, DIA, and DEA). The maximum %Fp
was 0.09%, which was measured in plasma sampled at 1 d
postimplantation. By day 3, the percentage had dropped to
0.03%; on days 5 and 7,%Fp was further reduced to 0.006%
and 0.017%, respectively (Figure 3).
In Vitro Study
Within the first 24 h of incubation, there was a rapid release of
estradiol from the boluses, averaging 9.2% of the estradiol in the
FIGURE 3. Percentage of atrazine plus its metabolites (diaminochlorotriazine,

deisopropylatrazine, and deethylatrazine) measured in 1 mL of fish plasma over
time relative to the total amount of atrazine implanted in male cunners.
518
ZAROOGIAN ET AL.
TABLE 2. Mean (SD in parentheses) cumulative percentage of estradiol measured in L-15 medium from slow-release matrix boluses over time during the in
vitro study.
Time
Percentage of
6-g/g bolus
Percentage of
1.2-g/g bolus
1h
3h
6h
1d
2d
3d
5d
7d
9d
11 d
2.5 (0.4)
3.9 (0.3)
5.9 (1.7)
9.2 (2.8)
11.2 (0.6)
10.6 (0.4)
13.3 (0.5)
13.7 (0.9)
15.1 (2.4)
15.2 (0.9)
3.7 (0.3)
8.9 (0.8)
11.9 (1.6)
17.5 (2.5)
21.6 (0.7)
24.6 (0.4)
32.4 (5.4)
35.0 (4.5)
31.6 (6.8)
30.0 (1.5)
6.0-g/g boluses and 17.5% of the estradiol in the 1.2-g/g boluses (Table 2). This rate of release slowed appreciably after the
first day to a more-gradual average release (over the next 10 d) of
0.6% per day from the 6.0-g/g boluses and 1.3% per day from
the 1.2-g/g boluses. Release was approximately twice as rapid
for the 1.2-g/g boluses, in which the Ethocel amount was only
about one-fifth of that added to the 6.0-g/g boluses (Table 2).
A visual comparison of the data points and logarithmic trend
lines in Figures 4 and 5 shows that the patterns of estradiol release from the boluses into fish plasma and L-15 medium were
quite similar, with a sharp increase occurring during the first
24 h, followed by a more gradual release for the rest of the observation period. However, much greater concentrations of total
estrogen were retrieved from the L-15 mediumapproximately
FIGURE 4. Cumulative percentage of estradiol released (based on the radioactivity, in disintegrations per minute [dpm], of tritiated estradiol) as measured in
1 mL of L-15 medium relative to the total amount of estradiol in the 6-g/g
and 1.2-g/g boluses during the in vitro study. Logarithmic trend lines for each
dosage are plotted for reference.
FIGURE 5. Cumulative percentage of total estrogen (estradiol plus estrone)

in the plasma of two male cunners that received the estradiol treatment (6-g/g
dosage) during the in vivo study. A logarithmic trend line is plotted for reference.
an order of magnitude greater than the concentrations in the cunner plasma.

DISCUSSION
A number of implant systems (reviewed by Mylonas and
Zohar 2000) have been developed for delivering a sustained
dose of hormone to large fish in captivity. These systems may
be adaptable to administering other chemicals of interest as
well. A currently available example for hormone administration
is the commercial product Ovaplant (Western Chemical), which
is implanted by use of the commercially available Ralogun
pellet injector (Schering-Plough Animal Health). However, the
manufacture of products like these requires technical expertise
and specialized equipment (Mylonas and Zohar 2000). We
were interested in developing an injectable implant that would
be inexpensive yet practical to prepare and administer in the
laboratory. Ideally, the implant would release a small amount
of test chemical into a fishs circulation over a period of weeks
and could be used for a wide variety of chemicals. In addition,
we wanted to accurately deliver the test chemical based on fish
weight while causing only minimal trauma to the fish; thus, we
surmised that a liquid injectable delivery method was the best
choice. Our earlier experiments with juvenile summer flounder
indicated that coconut oil was a good matrix carrier because it
could be injected as a lukewarm liquid that would quickly solidify (Mills et al. 2001). To ensure that the release of the selected
test chemical was slow enough, we opted to mix an ethyl cellulose polymer, Ethocel, into our coconut oil implant. Ethocel is
used in pharmaceutical products that require controlled-release
dosage formulations (Dow Chemical 2005). The resulting
implant preparation of Ethocel and test chemical in a coconut oil
carrier was relatively simple to prepare and easy to inject, and it
left a bolus that was still discernable when fish were necropsied
at 2 weeks postimplantation. Although we acknowledge that

our in vivo experiment results could have been enhanced by
increasing the number of study fish, the results presented
here show that estradiol, ethynylestradiol, atrazine, and the
associated metabolites were detectable in the plasma of treated
fish for 714 d after initial implantation. This indicates that our
implant is an effective method for the administration of these
chemicals for up to 2 weeks. Further in vivo experimentation
would be necessary to determine whether, with modification,
the implant can deliver chemical for a longer period of time.
In Vivo Experiments
The similar concentrations of circulating estrogen measured
in the two individual cunners that received estradiol at 6 g/g
indicate that (1) there is a reasonably even distribution of test
chemical within the Ethocelcoconut oil preparation and (2) individual variability is apparently minimal for the release of test
chemical from the implant. Because estradiol was implanted at
the highest total test concentration (6 g/g) among the three
chemicals, we expected that plasma concentrations would be
higher for estradiol than for the other chemicals. This was
borne out: the average plasma concentration of chemical in the
two estradiol-implanted fish was approximately 1,350-ng/mL
estrogen at 1 d after implantation compared with 710 ng/mL
for the ethynylestradiol-implanted fish and 105 ng/mL for the
atrazine-implanted fish (Table 1). Because the ratio of Ethocel
to test chemical was lower in the estradiol implant (17:1) than
in the ethynylestradiol and atrazine implants (both 40:1), we
also expected that the percentage of estradiol detected in circulation might be higher than the detected levels of the other
chemicals. However, the percentage of estrogen (estradiol plus
its metabolite, estrone) measured in plasma from both of the
estradiol-implanted fish on day 1 after implantation was just
above 0.1%approximately equal to the atrazine percentage
(0.1%) and lower than the ethynylestradiol percentage (1.7%).
Rapid metabolism and clearance of estradiol from the plasma
of the fish provide a possible explanation. Our data show that at
3 h after estradiol implantation, 1927% of the total circulating
estrogen in estradiol-treated male cunners was the metabolite
estrone, and the estrone percentage increased to 3556% of the
circulating estrogen in plasma at subsequent sampling times. On
day 7, estrogen concentrations were still elevated in comparison
with normal physiological concentrations in male fish, which
are in the picograms per milliliter range (Miura et al. 1999;
Amer et al. 2001; Geraudie et al. 2010).
Specker and Chandlee (2003) reported that within 1 d, larval
and juvenile summer flounder depurated estradiol taken up from
water; Zohar (1982) demonstrated that estradiol had a half-life
of less than 30 min in the bloodstream of rainbow trout. We
might expect the release of estradiol from the Ethocelcoconut
oil bolus to be initially rapid as the reservoir of steroid in zones
nearest to the surface of the bolus diffuse into the bloodstream.
Remaining steroid must diffuse further to reach the surface of
the bolus, so that the release rate slows over time until it becomes
519
lower than the clearance rate. A decreasing rate of release from

the bolus, in combination with metabolism and clearance by the
fish, may partially explain why estrogen was undetectable in the
plasma of treated cunners by day 10 postimplantation.
In contrast to the apparently rapid rate of estradiol
metabolism and clearance, the xenoestrogen ethynylestradiol
appeared to be neither metabolized nor rapidly excreted.
Ethynylestradiol reached its peak concentration in cunner
plasma (i.e., %Fp = 4%) on day 3 after implantation, indicating that enterohepatic recirculation of this xenoestrogen, as
documented in rainbow trout (Schultz et al. 2001), could also be
occurring in cunners. Schultz et al. (2001) found that after injection, ethynylestradiol was extensively conjugated and secreted
into the bile of treated rainbow trout. When the gall bladder
emptied, the stored ethynylestradiol was released into the gut,
where most was deconjugated and reabsorbed, effectively redosing the rainbow trout with ethynylestradiol. Our data suggest
that a similar phenomenon took place in cunners. The concentration of ethynylestradiol measured in the circulation of cunners
more than doubled from day 1 to day 3 postimplantation. In
contrast, circulating levels of estradiol and atrazine from implants were characterized by an initial burst within the first day
after implantation, followed by a decline thereafter. This initial
burst of chemical is also characteristic of other implant delivery systems (Mylonas and Zohar 2000). Interestingly, estrone, a
contaminant in our ethynylestradiol stock, was detectable only
in plasma samples taken at 1 and 3 d after implantation, thus
lending further credence to the premise that natural estrogens
are cleared from the circulation of cunners much faster than the
xenoestrogen ethynylestradiol.
Metabolism appears to be a major consideration in the case
of atrazine implantation. On day 1 postimplantation, approximately 81% of the circulating chemical from the implant was in
the form of three atrazine metabolites: DACT, DIA, and DEA. If
we had measured only the concentration of atrazine (not metabolites) in the plasma of cunners, none would have been detected
after day 3. By including the major metabolites in our chemical analysis, we were able to detect the atrazine metabolites
in plasma samples up to day 7 after implantation. Detection of
metabolites and of the parent compound is a major advantage
of using GCMS to quantify the release of chemical from an
implant.
In Vitro Study
In vitro, maximum concentrations of estrogen measured in
the L-15 medium surrounding the boluses were much higher
than those measured in fish plasma during the in vivo exposures
(9.2% in vitro versus 0.56% in vivo). These results indicate that
measurement of in vivo plasma estrogen concentrations does not
capture the actual release rate from an implanted bolus because
uptake, metabolism, and clearance rates all contribute to the
plasma concentrations observed in vivo.
Comparing the release rates of estrogen between in vitro
preparations, a higher maximum release rate was observed in
520
ZAROOGIAN ET AL.
the 1.2-g/g boluses (17.5%) than in the 6-g/g boluses (9.2%).

The higher release rate is likely related to the smaller amount of
Ethocel added to the matrix in the 1.2-g/g preparation (0.7 mg
of Ethocel/bolus) relative to the 6-g/g preparation (3.08 mg of
Ethocel/bolus). Because Ethocel forms a film (Dow Chemical
2005) that reduces diffusion, increasing the amount of Ethocel
in the matrix would be expected to result in a slower release rate
of test chemical, just as we observed. The behavior of Ethocel
in our implants is in direct contrast to that reported for cellulose.
Sherwood et al. (1988) observed higher release rates of GnRHa
with increased amounts of cellulose in cellulosecholesterol
pellets. Sherwood et al. (1988) hypothesized that the cellulose
expanded and became soft upon hydration, providing diffusion
channels and promoting degradation of the matrix. However,
water-insoluble Ethocel forms a film that reduces diffusion, and
our results indicate that diffusion rates can be manipulated by
increasing or decreasing the amount of Ethocel in a matrix
preparation.
Conclusions
We have developed a long-lasting, easily prepared, injectable
implant consisting of a test chemical in an Ethocelcoconut oil
matrix. This matrix can be used in fish to deliver chemicals
for release over a period of 1 week or more. Our implants
eliminate the need for a pellet press, pellet injector implant gun,
or surgical incisions. Implants are deposited subcutaneously,
far away from sensitive organs in the peritoneal cavity; they are
soft and pliable, not brittle or crumbly, and are easily injected by
using a syringe. We used the implant to conduct fish reproductive
studies with endocrine-disrupting chemicals, but this injectable
implant may also provide aquaculturists with an inexpensive,
simple alternative to current procedures used for administering
hormones or other treatments to fish.
In cunners, estradiol and atrazine demonstrated much different metabolism and clearance profiles than ethynylestradiol.
This information suggests that fish plasma concentrations do not
always represent the true release rate of a test chemical from the
implant matrix due to metabolism, clearance, or recirculation of
the chemical in the intact animal.
ACKNOWLEDGMENTS
We thank Martha Simoneau for revising, editing, proofreading, and formatting this manuscript. We also thank Richard J.
Pruell for supplying technical advice during the research; Alycia
Collins for providing technical assistance; and Roxanne Johnson, Janet Nye, Joseph LiVolsi, and Bryan Taplin for providing
valuable comments on an earlier version of the manuscript. This
research would not have been possible without the support and
encouragement given by Suzanne Ayvazian. The manuscript
has been reviewed and approved for publication by the USEPA,
Office of Research and Development, NHEERL, Atlantic Ecology Division as Contribution Number AED-11-089. Approval
does not signify that the contents of this report necessarily
reflect the views and policies of the USEPA. Mention of trade
names or commercial products does not constitute endorsement

or recommendation for use by the USEPA.
REFERENCES
Amer, M. A., T. Miura, C. Miura, and K. Yamauchi. 2001. Involvement of sex
steroid hormones in the early stages of spermatogenesis in Japanese huchen
(Hucho perryi). Biology of Reproduction 65:10571066.
Arslan, T., and R. P. Phelps. 2004. Production of monosex male black crappie, Pomoxis nigromaculatus, populations by multiple androgen immersion.
Bayley, M., M. Junge, and E. Baatrup. 2002. Exposure of juvenile guppies to
three antiandrogens causes demasculinization and a reduced sperm count in
adult males. Aquatic Toxicology 56:227239.
Bisbal, G. A., and J. L. Specker. 1991. Cortisol stimulates hypo-osmoregulatory
ability in Atlantic salmon, Salmo salar L. Journal of Fish Biology 39:
421432.
Black, D. E., R. Gutjahr-Gobell, R. J. Pruell, B. Bergen, and A. E. McElroy.
1998. Effects of a mixture of non-ortho- and mono-ortho-polychlorinated
biphenyls on reproduction in Fundulus heteroclitus (Linnaeus). Environmental Toxicology and Chemistry 17:13961404.
Brzezicki, J. M., M. E. Andersen, B. K. Cranmer, and J. D. Tessari. 2003. Quantitative identification of atrazine and its chlorinated metabolites in plasma.
Journal of Analytical Toxicology 27:569573.
Carolsfeld, J., N. M. Sherwood, H. Kreiberg, and S. A. Sower. 1988. Induced
sexual maturation of herring using GnRH quick-release cholesterol pellets.
Crim, L. W., N. M. Sherwood, and C. E. Wilson. 1988. Sustained hormone release: II. effectiveness of LHRH analog (LHRHa) administration by
either single time injection or cholesterol pellet implantation on plasma gonadotropin levels in a bioassay model fish, the juvenile rainbow trout. Aquaculture 74:8795.
Cyr, D. G., and J. G. Eales. 1989. Effects of short-term 17-estradiol treatment
on the properties of T3 -binding proteins in the plasma of immature rainbow
trout, Salmo gairdneri. Journal of Experimental Zoology 252:245251.
Donohoe, R. M., and L. R. Curtis. 1996. Estrogenic activity of chlordecone,
o,p -DDT and o,p -DDE in juvenile rainbow trout: induction of vitellogenesis
and interaction with hepatic estrogen binding sites. Aquatic Toxicology 36:
3152.
Dow Chemical. 2005. Dow cellulosics: ETHOCELTM ethylcellulose polymers technical handbook. Dow Chemical, Midland, Michigan. Available:
www.dow.com/dowwolff/en/pdfs/192-00818.pdf. (January 2012).
Garcia, L. M. B. 1989. Dose-dependent spawning response of mature female
sea bass, Lates calcarifer (Bloch), to pelleted luteinizing hormone-releasing
hormone analogue (LHRHa). Aquaculture 77:8596.
Geraudie, P., M. Gerbron, and C. Minier. 2010. Seasonal variations and alterations of sex steroid levels during the reproductive cycle of male roach
(Rutilis rutilis). Marine Environmental Research 69(Supplement 1):5355.
Gimeno, S., H. Komen, S. Jobling, J. Sumpter, and T. Bowmer. 1998. Demasculinisation of sexually mature male common carp, Cyprinus carpio,
exposed to 4-tert-pentylphenol during spermatogenesis. Aquatic Toxicology
43:93109.
Goncalves, D., J. Alpedrinha, M. Teles, and R. F. Oliveira. 2007. Endocrine
control of sexual behavior in sneaker males of the peacock blenny Salaria
pavo: effects of castration, aromatase inhibition, testosterone and estradiol.
Hormones and Behavior 51:534541.
Gutjahr-Gobell, R. E., D. E. Black, L. J. Mills, R. J. Pruell, B. K. Taplin,
and S. Jayaraman. 1999. Feeding the mummichog (Fundulus heteroclitus)
a diet spiked with non-ortho- and mono-ortho-substituted polychlorinated
biphenyls: accumulation and effects. Environmental Toxicology and Chemistry 18:699707.
Gutjahr-Gobell, R. E., M. Huber, D. J. Borsay Horowitz, G. E. Zaroogian,
and L. J. Mills. 2002. A temperate reef fish, Tautogolabrus adspersus
(Walbaum), as a potential model species for laboratory studies evaluating

reproductive effects of chemical exposure. Environmental Toxicology and
Chemistry 21:380389.
Hernando, M. D., M. J. Mezcua, M. J. Gomez, O. Malato, A. Aguera, and
A. R. Fernandez-Alba. 2004. Comparative study of analytical methods involving gas chromatographymass spectrometry after derivatization and gas
chromatographytandem mass spectrometry for the determination of selected
endocrine disrupting compounds in wastewaters. Journal of Chromatography
A 1047:129135.
Hunter, G. A., and E. M. Donaldson. 1983. Hormonal sex control and its
application to fish culture. Pages 223303 in W. S. Hoar, D. J. Randall,
and E. M. Donaldson, editors. Fish physiology, volume 9B. Academic Press,
New York.
Jeannot, R., H. Sabik, E. Sauvard, T. Dagnac, and K. Dohrendorf. 2002. Determination of endocrine-disrupting compounds in environmental samples
using gas and liquid chromatography with mass spectrometry. Journal of
Chromatography A 974:143159.
Joakim Larsson, D. G., T. S. Sperry, and P. Thomas. 2002. Regulation of androgen receptors in Atlantic croaker brains by testosterone and estradiol. General
and Comparative Endocrinology 128:224230.
Keen, P. L., D. A. Higgs, K. J. Hall, and M. Ikonomou. 2005. Effects of dietary exposure of 4-nonylphenol on growth and smoltification of juvenile
coho salmon (Oncorhynchus kisutch). Science of the Total Environment 349:
8194.
Kramer, V. J., S. Miles-Richardson, S. L. Pierens, and J. P. Giesy. 1998. Reproductive impairment and induction of alkaline-labile phosphate, a biomarker
of estrogen exposure, in fathead minnows (Pimephales promelas) exposed to
waterborne 17-estradiol. Aquatic Toxicology 40:335360.
Lahnsteiner, F., B. Berger, M. Kletzl, and T. Weismann. 2006. Effect of 17estradiol on gamete quality and maturation in two salmonid species. Aquatic
Toxicology 79:124131.
Leatherland, J. F. 1985. Effects of 17-estradiol and methyl testosterone on the
activity of the thyroid gland in rainbow trout, Salmo gairdneri Richardson.
General and Comparative Endocrinology 60:343352.
Lee, C. S., C. S. Tamaru, and C. D. Kelley. 1986. Technique for making chronicrelease LHRH-a and 17-methyltestosterone pellets for intramuscular implantation in fishes. Aquaculture 59:161168.
Lopez de Alda, M. J., and D. Barcelo. 2001. Use of solid-phase extraction in
various of its modalities for sample preparation in the determination of estrogens and progestogens in sediment and water. Journal of Chromatography A
938:145153.
Madsen, L. L., B. Korsgaard, and P. Bjerregaard. 2003. Estrogenic effects in
flounder Platichthys flesus orally exposed to 4-tert-octylphenol. Aquatic Toxicology 64:393405.
Mandiki, S. N. M., M. B. Houbart, I. Babiak, E. Vandeloise, J. N. Gardeur, and P.
Kestmont. 2004. Are sex steroids involved in the sexual growth dimorphism
in Eurasian perch juveniles? Physiology and Behavior 80:603609.
Metcalfe, C. D., T. L. Metcalfe, Y. Kiparissis, B. G. Koenig, C. Khan, R. J.
Hughes, T. R. Croley, R. E. March, and T. Potter. 2001. Estrogenic potency
of chemicals detected in sewage treatment plant effluents as determined by
in vivo assays with Japanese medaka (Oryzias latipes). Environmental Toxicology and Chemistry 20:297308.
Mills, L. J., R. E. Gutjahr-Gobell, R. A. Haebler, D. J. Borsay Horowitz, S.
Jayaraman, R. J. Pruell, R. A. McKinney, G. R. Gardner, and G. E. Zaroogian.
2001. Effects of estrogenic (o,p-DDT; octylphenol) and anti-androgenic
(p,p-DDE) chemicals on indicators of endocrine status in juvenile male
summer flounder (Paralichthys dentatus). Aquatic Toxicology 52:157176.
Miura, T., C. Miura, T. Ohta, M. R. Nader, T. Todo, and K. Yamauchi. 1999.
Estradiol-17 stimulates the renewal of spermatogonial stem cells in males.
Biochemical and Biophysical Research Communications 264:230234.
Mylonas, C. C., and Y. Zohar. 2000. Use of GnRHa-delivery systems for the
control of reproduction in fish. Reviews in Fish Biology and Fisheries 10:
463491.
521
Pandian, T. J., and S. G. Sheela. 1995. Hormonal induction of sex reversal in

fish. Aquaculture 138:122.
Pankhurst, N. W., N. E. Stacey, and R. E. Peter. 1986. An evaluation of
techniques for the administration of 17-estradiol to teleosts. Aquaculture
52:145155.
Patyna, P. J., R. A. Davi, T. F. Parkerton, R. P. Brown, and K. R. Cooper. 1999.
A proposed multigeneration protocol for Japanese medaka (Oryzias latipes)
to evaluate effects of endocrine disruptors. Science of the Total Environment
233:211220.
Pickering, A. D., and J. Duston. 1983. Administration of cortisol to brown trout,
Salmo trutta L., and its effects on the susceptibility to Saprolegnia infection
and furunculosis. Journal of Fish Biology 23:163175.
Pizza, J. C., and J. M. OConnor. 1983. PCB dynamics in Hudson River striped
bass: II. accumulation from dietary sources. Aquatic Toxicology 3:313327.
Sangiao-Alvarellos, S., J. M. Mguez, and J. L. Soengas. 2005. Actions of
growth hormone on carbohydrate metabolism and osmoregulation of rainbow
trout (Oncorhynchus mykiss). General and Comparative Endocrinology 141:
214225.
Schultz, I. R., G. Orner, J. L. Merdink, and A. Skillman. 2001. Doseresponse
relationships and pharmacokinetics of vitellogenin in rainbow trout after
intravascular administration of 17-ethynylestradiol. Aquatic Toxicology
51:305318.
Scott, A. P., P. R. Witthames, E. L. M. Vermeirssen, and J. Carolsfeld. 1999.
Prolonged-release gonadotrophin-releasing hormone analogue implants enhance oocyte final maturation and ovulation, and increase plasma concentrations of sulphated C21 steroids in North Sea plaice. Journal of Fish Biology
55:316328.
Segner, H., K. Caroll, M. Fenske, C. R. Janssen, G. Maack, D. Pascoe, C.
Schafers, G. F. Vandenbergh, M. Watts, and A. Wenzel. 2003. Identification
of endocrine-disrupting effects in aquatic vertebrates and invertebrates: report
from the European IDEA project. Ecotoxicology and Environmental Safety
54:302314.
Shelton, W. L., and S. D. Mims. 2003. Fabrication of silastic implants for in vivo
steroid delivery in fish. North American Journal of Aquaculture 65:158161.
Sherwood, N. M., L. W. Crim, J. Carolsfeld, and S. M. Walters. 1988. Sustained hormone release: I. characteristics of in vitro release of gonadotropinreleasing hormone analogue (GnRH-A) from pellets. Aquaculture 74:
7586.
Specker, J. L., and M. K. Chandlee. 2003. Methodology for estradiol treatment
in marine larval and juvenile fish: uptake and clearance in summer flounder.
Sundararaj, B. I., and S. V. Goswami. 1968. Effects of estrogen, progesterone,
and testosterone on the pituitary and ovary of catfish, Heteropneustes fossilis
(Bloch). Journal of Experimental Zoology 169:211227.
Trudeau, V. L., R. E. Peter, and B. D. Sloley. 1991. Testosterone and estradiol potentiate the serum gonadotropin response to gonadotropin-releasing
hormone in goldfish. Biology of Reproduction 44:951960.
Wang, W. B., A. H. Li, T. Z. Cai, and J. G. Wang. 2005. Effects of intraperitoneal
injection of cortisol on non-specific immune functions of Ctenopharyngodon
idella. Journal of Fish Biology 67:779793.
Yamada, H., R. Satoh, T. Yamashita, A. Kambegawa, and M. Iwata. 1997. Development of a time-resolved fluoroimmunoassay (TR-FIA) for testosterone:
measurement of serum testosterone concentrations after testosterone treatment in the rainbow trout (Oncorhynchus mykiss). General and Comparative
Endocrinology 106:181188.
Yamaguchi, S., H. Kagawa, K. Gen, K. Okuzawa, and M. Matsuyama. 2004.
Silicone implants for delivery of estradiol-17and 11-ketotestosterone to red
seabream Pagrus major. Aquaculture 239:485496.
Zohar, Y. 1982. Levolution de la pulsatilite et des cycles nycthemerau de
la secretion gonadotrope chez la truite arc-en-ciel femelle en relation avec
le cycle sexuel annuel et par rapport a lactivite steroidogène de lovaire.
Doctoral dissertation. Universite Pierre et Marie Curie, Paris.


Characterization of Greater Amberjack Microsatellite

Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco
Jacks, and Banded Rudderfish
a
Mark A. Renshaw , Alejandro Buentello & John R. Gold
Center for Biosystematics and Biodiversity, Texas A&M University, College Station, Texas,
77843-2258, USA
b
Schillinger Genetics, 4401 Westown Parkway, Suite 225, West Des Moines, Iowa, 50266, USA
To cite this article: Mark A. Renshaw, Alejandro Buentello & John R. Gold (2012): Characterization of Greater Amberjack
Microsatellite Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco Jacks, and Banded Rudderfish, North American Journal
of Aquaculture, 74:4, 522-529


DOI: 10.1080/15222055.2012.686959
TECHNICAL NOTE
Characterization of Greater Amberjack Microsatellite

Markers in Lesser Amberjacks, Yellowtail Jacks, Almaco
Jacks, and Banded Rudderfish
Mark A. Renshaw*
Center for Biosystematics and Biodiversity, Texas A&M University, College Station,
Texas 77843-2258, USA
Alejandro Buentello
Schillinger Genetics, 4401 Westown Parkway, Suite 225, West Des Moines, Iowa 50266, USA
John R. Gold
Center for Biosystematics and Biodiversity, Texas A&M University, College Station,
Texas 77843-2258, USA
Abstract
Thirty-one microsatellite markers that were previously isolated
from and characterized in greater amberjacks Seriola dumerili
were assayed for cross-species amplification in four other members
of the carangid genus Seriola: the lesser amberjack S. fasciata,
yellowtail jack S. lalandi, almaco jack S. rivoliana, and banded
rudderfish S. zonata. The number of markers that consistently
amplified and were polymorphic ranged from 16 in yellowtail jacks
to 25 in lesser amberjacks. The microsatellites characterized in this
study will be useful for a variety of applications, including stock
structure assessments of wild fish and parentage assignments of
farmed fish.
Five species of the carangid genus Seriola support commercial and recreational fisheries along the coast of the
mainland United States: the greater amberjack S. dumerili,
lesser amberjack S. fasciata, yellowtail jack S. lalandi, almaco
jack S. rivoliana, and banded rudderfish S. zonata. Three of the
species (greater amberjack, yellowtail jack, and almaco jack)
also are important components of commercial aquaculture
production. The commercial aquaculture of greater amberjacks
and yellowtail jacks is currently being researched or is already
in progress in Japan, Australia, New Zealand, and a number
of Mediterranean countries, including Italy, Spain, Malta, and
Greece (Repulles-Albelda et al. 2008; Hamasaki et al. 2009;
*Corresponding author: mrenshaw@nd.edu
Received February 7, 2012; accepted April 16, 2012
522
Stuart and Drawbridge, in press). In the United States, almaco

jacks are produced commercially in Hawaii by using offshore
aquaculture cages (Simpson 2011), and the production of
yellowtail jacks is currently under research in California by
utilizing broodfish that are collected locally from the wild
(Stuart and Drawbridge, in press).
Nuclear-encoded microsatellites are useful markers for elucidating stock structure of wild fish (Miller et al. 2011) and
can contribute to fisheries management decisions (Reiss et al.
2009). Microsatellites also are valuable for use in aquaculture
programs, as they can serve as tools for parentage assignments
(Gold et al. 2010), species authentication (Iguchi et al. 2012),
and marker-assisted selection in farmed fish (Liu and Cordes
2004). In this paper, we evaluate the amplification of 31 microsatellites developed from the genomic DNA of greater amberjacks for cross-amplification in lesser amberjacks, yellowtail
jacks, almaco jacks, and banded rudderfish.
METHODS
Ninety-six fish were assayed in this study: 19 lesser
amberjacks (all from Johns Island, South Carolina), 29
yellowtail jacks (all from San Diego, California), 30 almaco
jacks (all from Johns Island), and 18 banded rudderfish (7
from Johns Island; 11 from Panama City, Florida). Fin clips
were taken and placed in a 95% solution of ethanol (Florida
TECHNICAL NOTE
samples), sarkosylurea (South Carolina samples), or a 20%

solution of DMSO (California samples). Genomic DNA was
extracted using a standard phenylchloroform protocol. The
PCR primer sequences followed those given by Renshaw
et al. (2006, 2007) for microsatellites isolated from the DNA
of greater amberjacks. Unlabeled and fluorescently labeled
primers (6-FAM and HEX) were obtained from Integrated
DNA Technologies; primers that were fluorescently labeled
with NED were obtained from Applied Biosystems, Inc. (ABI).
The fluorescent label was attached to one of the primers from
each microsatellite marker pair (Table 1).
The PCR amplifications were performed in 10-L reactions
by using 1 L of DNA, 1 Colorless GoTaq Flexi Buffer
(Promega), 2-mM MgCl2 , 200 m of each dNTP, 5 pmol of each
primer (forward and reverse), and 0.5 unit of GoTaq Flexi DNA
Polymerase (Promega). Cycling conditions employed (1) an initial denaturation step at 95 C for 3 min; (2) seven cycles with
denaturation at 95 C for 30 s, first annealing temperature (defined below) for 45 s, and extension at 72 C for 1 min; (3) seven
cycles with denaturation at 95 C for 30 s, second annealing temperature for 45 s, and extension at 72 C for 1 min; (4) 38 cycles
with denaturation at 95 C for 30 s, third annealing temperature
for 45 s, and extension at 72 C for 1 min; and (5) a final extension
step at 72 C for 10 min. The first annealing temperature was the
one listed by Renshaw et al. (2006, 2007), the second annealing
temperature was 2 C lower than the first annealing temperature,
and the third annealing temperature was 4 C lower than the first
annealing temperature (Table 1). Amplified PCR products were
run on an ABI 377 Automated Sequencer. Allele sizes were estimated with the GeneScan 400HD ROX Size Standard (ABI);
allele sizes were determined using GeneScan version 3.1.2 and
Genotyper version 2.5 (ABI). Genetic variability of each microsatellite marker was measured by the number of alleles,
gene diversity (expected heterozygosity HE ), and observed heterozygosity (HO ) as calculated in Genetic Data Analysis (GDA)
software (Lewis and Zaykin 2001). Fishers exact tests as implemented in GDA were used to test for significant departures from
HardyWeinberg expectations at individual microsatellites and
for departures from genotypic equilibrium at pairs of microsatellites; Bonferroni corrections for multiple tests (Rice 1989) were
applied manually to the P-value outputs from GDA. MicroChecker software (Van Oosterhout et al. 2004) was used to test
for evidence of null alleles, scoring errors due to stuttering, and
scoring errors due to large-allele dropout at each microsatellite
marker.

Of the 31 microsatellites that were assayed, only Sdu23 did
not consistently amplify for at least one of the four species.
Summary data for the remaining 30 microsatellites are presented
in Table 1; previously published data for greater amberjacks
523
(Renshaw et al. 2006, 2007) are also included in Table 1 to

facilitate comparisons across species. In total, 25 markers were
polymorphic for lesser amberjacks, and the number of alleles
ranged from 2 to 16; HE ranged from 0.053 to 0.930, while
HO ranged from 0.053 to 0.947. After Bonferroni correction for
multiple tests (Rice 1989), genotypes at all markers conformed
to HardyWeinberg expectations and all pairs of microsatellites
were in genotypic equilibrium; analysis with Micro-Checker
indicated no evident issues.
Sixteen markers were polymorphic for yellowtail jacks, with
the number of alleles ranging from 2 to 14. The HE ranged from
0.100 to 0.859, and HO ranged from 0.103 to 1.000. After Bonferroni correction, genotypes at two markers (Sdu29 and Sdu46)
deviated significantly from HardyWeinberg expectations and
17 microsatellite pairs deviated significantly from genotypic
equilibrium (Sdu29Sdu1, Sdu29Sdu2, Sdu29Sdu4, Sdu29
Sdu10, Sdu29Sdu19, Sdu29Sdu21, Sdu29Sdu31, Sdu29
Sdu4, and Sdu46Sdu19). Analysis with Micro-Checker indicated the possibility of null alleles at Sdu29.
Twenty-three markers were polymorphic for almaco jacks,
and the number of alleles ranged from 2 to 22; HE ranged from
0.033 to 0.951, while HO ranged from 0.033 to 0.933. After
Bonferroni correction, genotypes at all markers conformed to
HardyWeinberg expectations and all pairs of microsatellites
were in genotypic equilibrium. Analysis with Micro-Checker
indicated the possibility of null alleles at Sdu10.
Eighteen markers were polymorphic for banded rudderfish.
The number of alleles ranged from 2 to 17, HE ranged from
0.203 to 0.954, and HO ranged from 0.111 to 1.000. After Bonferroni correction, genotypes at two markers (Sdu12 and Sdu31)
deviated significantly from HardyWeinberg expectations and
one microsatellite pair (Sdu31Sdu43) deviated significantly
from genotypic equilibrium. Analysis with Micro-Checker
indicated the possibility of null alleles at both Sdu12 and
Sdu31.
For fisheries managers, the 30 microsatellites characterized
in the present study can be used as tools for assessing stock
structure and will provide valuable population genetic indices
within each species. The markers will also facilitate the improvement of aquaculture programs through a variety of applications.
For aquaculture facilities producing larvae that are subsequently
grown for future sale, parentage assignments (Gold et al. 2010)
can identify broodstock contribution and can be used to maximize mating design and output. Differences in microsatellite
allele sizes between species can be used to authenticate the
labeling of aquaculture products as they go to market (Iguchi
et al. 2012). As part of a larger set of markers, these microsatellites can be employed for the mapping of quantitative trait loci
and the marker-assisted selection of future broodfish (Liu and
Cordes 2004).
524
(GACA)20
(GAA)8
(CAA)8
(GACA)8 (GGCA)4
(GACA)6
(TAA)3 CAA(TAA)10
CAA(TAA)5
(GATA)12
CTATATTCACTCTGTTGCC ++
GTGTAGGAGAGACTGTAAG
CGGTGTATTGTTACTGTGAC +
TCGTCTCTGATTGGTTAG
GGAAATAGTTTGGATCACGCTGG +++
GGATGCTCAGTGAAGTTGTGC
GTAAGGATTTGTCATGTAGCC ++
GGAGACGAGTTCTCTTTGC
CCAAAGCAGGTGAAAGTGA +
GGTCCATACAACAACTCAG
Sdu2
Sdu3
Sdu4
Sdu5
Sdu6
Repeat
sequence*
CGTTTCCATCGCACTTTT +++
GCTAACACTCACTGGTG
Primer sequence
(5 3 )a*
Sdu1
Microsatellite
50
48
46
56
54
52
60
58
56
53
51
49
50
48
46
53
51
49
T Ab
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Spc
19, 4
29, 7
30, 8
N/A
28, 11
N/A
29, 3
30, 5
18, 3
29, 5
19, 2
N/A
30, 4
N/A
29, 5
N/A
29, 3
30, 5
N/A
29, 10
19, 13
N/A
30, 1
18, 13
23, 5
19, 10
N/A
30, 11
N/A
29, 12
N, N A d
289309
352420
306362
N/A
305385
N/A
135147
129144
141150
145160
208211
N/A
214223
N/A
212227
N/A
284300
281287xx
N/A
320360
250295
N/A
202
203257
206236
227267
N/A
215259
N/A
231279
Size
rangee
0.605, 0.789
0.808, 0.897
0.403, 0.333
N/A
0.751, 0.714
N/A
0.327, 0.345
0.532, 0.433
0.252, 0.278
0.716, 0.670
0.444, 0.526
N/A
0.715, 0.800
N/A
0.602, 0.690
N/A
0.493, 0.448
0.275, 0.267
N/A
0.808, 0.897
0.899, 0.947
N/A
0.000, 0.000
0.921, 0.889
0.560, 0.478
0.892, 0.842
N/A
0.860, 0.767
N/A
0.844, 0.759
HE , H O f
MicroCheckerh
0.0363
0.0103
0.1334
N/A
N/A
0.252
N/A
N/A
0.4566
0.1278
1.0000
0.918
0.5997
N/A
N/A
0.5003
N/A
N/A
0.717
N/A
N/A
0.6056
0.5084
N/A
N/A
0.972
0.8891
N/A
N/A
1.0000
0.3231
0.390
0.0309
N/A
N/A
0.1031
N/A
N/A
0.043
(Continued on next page)
PHW g
TABLE 1. Summary data for 30 greater amberjack microsatellite markers that were characterized in lesser amberjacks, yellowtail jacks, almaco jacks, and banded rudderfish. The fluorescently labeled
primer is in bold text, with the appropriate label signified by the number of plus signs (+ = 6-FAM; ++ = HEX; +++ = NED). Asterisks indicate information that was taken from earlier descriptions
of these microsatellites (Renshaw et al. 2006 for Sdu1Sdu27; Renshaw et al. 2007 for Sdu29Sdu46); all information included for greater amberjacks was published previously (Renshaw et al. 2006,
2007). N/A indicates that the marker failed to amplify consistently.
525
Continued.
(CAA)8
(GAA)9
(GA)4 (GAA)9
(GAA)18
(CAA)7
(GACA)5 GGCA
(GACA)5 GGCA
(GACA)7
CCAGTCTATGAAACACAACC +
CCTGAAGCGATGAAGCGT
CTGTTGTCCTTCCAGAC +++
CCACATCGTCTGAATAGC
CCAAGTCCTCCTGCTACTACCAT +
CCTTGTGGATGACCTGTTTG
GCTCTCGTGTGTTACTCAAG +
GCAACTGTCAGATCCTCCA
CCACAAGTTATCACAAGCCACC ++
GCTTTGTCCCCTGTGTGCTG
Sdu8
Sdu9
Sdu10
Sdu11
Sdu12
Repeat
sequence*
CACTTTCAACTGGAACACC ++
GGTTCTGCTGGCTCATTG
Primer sequence
(5 3 )a*
Sdu7
Microsatellite
TABLE 1.
60
48
46
56
54
52
56
54
52
53
51
49
56
54
52
56
54
52
T Ab
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Spc
19, 1
N/A
30, 3
N/A
26, 3
19, 3
29, 1
30, 2
18, 1
29, 3
19, 8
29, 1
30, 1
N/A
29, 2
19, 2
29, 4
30, 12
18, 9
29, 15
19, 1
29, 1
30, 2
18, 1
29, 2
N/A
N/A
30, 7
18, 9
29, 10
N, N A d
337
N/A
346352
N/A
343364
101107
95
101107
95
105111
233249
237
219
N/A
230238
273276
261282
285321
287314
295346
168
168
168171
168
169172
N/A
N/A
232260
229277
237313
Size
rangee
0.000, 0.000
N/A
0.501, 0.367
N/A
0.361, 0.269
0.152, 0.158
0.000, 0.000
0.033, 0.033
0.000, 0.000
0.068, 0.069
0.528, 0.579
0.000, 0.000
0.000, 0.000
N/A
0.131, 0.138
0.462, 0.474
0.577, 0.552
0.793, 0.633
0.887, 0.944
0.902, 0.966
0.000, 0.000
0.000, 0.000
0.066, 0.067
0.000, 0.000
0.160, 0.172
N/A
N/A
0.807, 0.767
0.792, 0.500
0.776, 0.690
HE , H O f
1.0000
N/A
0.1678
N/A
0.102
1.0000
1.0000
1.0000
1.0000
1.000
0.8750
1.0000
1.0000
N/A
1.000
1.0000
0.8619
0.0722
0.9606
0.412
1.0000
1.0000
1.0000
1.0000
1.000
N/A
N/A
0.0759
0.0013
0.550
PHW g
N/A
N/A
N/A
N/A
N/A
MicroCheckerh
526
Continued.
(CAA)11
(CAGA)16
(GATA)25
(GAA)21
(GATA)13
(GA)14
GCATTCTGGCATTAGCAT +++
GGTACTCTAGTTAGCCCTAC
CTCAGGACAATGTTGGTAG +
GCTAACAAGTTCACGACAT
CATTCTCCAAGTATGTGACCTC ++
GCTCTATGCGAATACCTCCA
CCTTCTGTCTTGACTCTGC +++
CGATTCATCCAGCTTTAGG
CCTTGCCATACCGATGCCAG +
GACTGCTCTGCCTGCTTGTTG
Sdu19
Sdu21
Sdu22
Sdu27
Sdu29
Repeat
sequence*
GAGTTGTACTGTGGTAAAC +
GGACATTAGAGTCTGTGG
Primer sequence
(5 3 )a*
Sdu16
Microsatellite
TABLE 1.
60
58
56
56
54
52
56
54
52
56
54
52
56
54
52
50
48
46
T Ab
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Spc
19, 3
29, 1
30, 2
18, 3
29, 4
19, 7
29, 2
30, 7
18, 2
29, 10
19, 16
29, 10
30, 22
18, 17
27, 20
19, 6
29, 1
30, 1
18, 3
29, 11
19, 3
29, 1
30, 7
18, 8
29, 8
19, 2
29, 8
30, 4
18, 7
29, 11
N, N A d
108120
102
105108
96105
114126
212236
212216
232256
212216
236272
277369
297389
289385xx
301389
264380
300318
292
295
295304
311341
313317xx
303
290322
324352
266298
299301
319365
311321
309325
311377
Size
rangee
0.198, 0.211
0.000, 0.000
0.033, 0.033
0.294, 0.167
0.462, 0.552
0.805, 0.895
0.100, 0.103
0.848, 0.833
0.203, 0.111
0.823, 0.724
0.930, 0.947
0.789, 0.897
0.951, 0.867
0.954, 1.000
0.950, 0.926
0.762, 0.632
0.000, 0.000
0.000, 0.000
0.586, 0.611
0.832, 0.862
0.284, 0.263
0.000, 0.000
0.605, 0.567
0.857, 0.778
0.783, 0.630
0.235, 0.263
0.808, 0.345
0.571, 0.667
0.765, 0.833
0.847, 0.793
HE , H O f
MicroCheckerh
1.0000
1.0000
1.0000
0.0397
0.174
0.0441
1.0000
0.0181
0.1691
0.178
0.2709
0.9997
0.0863
1.0000
0.697
0.7447
1.0000
1.0000
0.2134
0.694
0.3553
1.0000
0.8688
0.4006
0.060
1.0000
0.0000
N
0.4981
0.3300
0.487
(Continued on next page)
PHW g
527
Continued.
(CA)14
(CA)17
(GA)13
(GA)20
(GA)23
(CA)16
CCTGTGAGAGCATTTGGTAT ++
GTGCTTGTCTCTTCTGTCAT
CCTCTAACAGCCACAATCA ++
GCTCTTCACCTTCCTCATA
CCTTGTGTTGTATCTGCTGTAA +++
GGAATAAACCTCGTCTGTCA
CTGTTATGAAGCAGTGAAGAGG +
GGACCATCCTGCTCTGACA
CCTCTAATGGACTTCAGCG +++
GGTTATTTTGAGAGCCGTC
Sdu32
Sdu33
Sdu34
Sdu36
Sdu37
Repeat
sequence*
CACATTTGGACGGATTCTTC +
GCTGTTATCCTCCAGTGCT
Primer sequence
(5 3 )a*
Sdu31
Microsatellite
TABLE 1.
53
51
49
56
54
52
56
54
52
56
54
52
53
51
49
56
54
52
T Ab
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Spc
19, 4
29, 9
30, 20
18, 7
29, 7
19, 10
29, 9
30, 8
18, 12
29, 21
19, 7
29, 2
30, 5
18, 7
29, 3
19, 4
29, 1
30, 3
N/A
29, 8
19, 3
29, 1
30, 16
18, 4
29, 9
19, 8
29, 14
N/A
18, 12
29, 25
N, N A d
8496
106140
90164
86104
8498
93125
97131
99165
103133
99177
194212
210212
194202
188216
202208
93103
194
9197
N/A
84118
194198
186
192250
198210
200226
164180
173281
N/A
167219
160278
Size
rangee
0.745, 0.684
0.781, 0.759
0.941, 0.933
0.578, 0.333
0.738, 0.759
0.851, 0.789
0.653, 0.862
0.811, 0.700
0.921, 0.944
0.948, 0.862
0.802, 0.789
0.290, 0.345
0.671, 0.600
0.838, 0.833
0.296, 0.276
0.681, 0.737
0.000, 0.000
0.532, 0.567
N/A
0.778, 0.517
0.522, 0.474
0.000, 0.000
0.912, 0.933
0.487, 0.500
0.815, 0.690
0.858, 0.842
0.859, 1.000
N/A
0.887, 0.833
0.889, 0.828
HE , H O f
0.4228
0.2719
0.2394
0.0028
0.650
0.5653
0.8653
0.1963
0.8969
0.231
0.8488
0.5616
0.4063
0.4069
0.598
0.8497
1.0000
0.5847
N/A
0.028
0.4838
1.0000
0.4456
0.3928
0.132
0.3169
0.2088
N/A
0.5094
0.448
PHW g
N/A
N/A
MicroCheckerh
528
Continued.
(CA)17 7bp (CA)5
(CA)18
(CA)28
(GA)12
(GA)30
CGATGCTTTCAACTCCGACACAC +++
CCATCCTTCATCAGCAACAACATCC
AGCGTGGACAGTTTATGG ++
GTCTGTTTACTGGTCGCA
GGAACATTTGGAGCCATAAGAC
CAGAAGAAGAGCGTGGTGGAGAG +++
GGTAATGGGAGGTGTGAGTGT ++
CCTTCTCCTGTTAATCCATCTCC
GCAGTGTGAGCCATACATTAC +++
CTACAGGACAAAAGCCATT
Sdu40
Sdu41
Sdu43
Sdu44
Sdu46
53
51
49
56
54
52
60
58
56
53
51
49
64
62
60
56
54
52
T Ab
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Sfa
Sla
Sri
Szo
Sdu
Spc
19, 4
29, 2
30, 4
N/A
29, 6
19, 7
29, 3
30, 12
18, 3
28, 9
19, 6
N/A
N/A
N/A
29, 10
19, 5
29, 10
30, 22
18, 14
29, 11
19, 3
29, 2
30, 1
18, 10
29, 3
19, 2
29, 5
N/A
18, 1
29, 13
N, N A d
145153
137139
137147
N/A
154180
187201
181185
196232
189193
197235
96112
N/A
N/A
N/A
96130
259271
283315
272340
265305
276298
118124
117119
114
116136
114124
238252
232244
N/A
211
217259
Size
rangee
0.642, 0.684
0.242, 0.276
0.653, 0.533
N/A
0.338, 0.345
0.805, 0.737
0.617, 0.655
0.838, 0.900
0.541, 0.778
0.825, 0.893
0.718, 0.842
N/A
N/A
N/A
0.636, 0.724
0.518, 0.579
0.845, 0.931
0.941, 0.900
0.913, 0.833
0.835, 0.828
0.656, 0.579
0.100, 0.103
0.000, 0.000
0.813, 0.722
0.402, 0.379
0.053, 0.053
0.749, 1.000
N/A
0.000, 0.000
0.817, 0.897
HE , H O f
1.0000
1.0000
0.0491
N/A
0.688
0.1863
0.7009
0.3556
0.0975
0.884
0.3794
N/A
N/A
N/A
0.548
0.8453
0.2253
0.6472
0.1897
0.654
0.5681
1.0000
1.0000
0.6769
0.574
1.0000
0.0003
N/A
1.0000
0.628
PHW g
N/A
N/A
N/A
N/A
N/A
MicroCheckerh
Primer sequences are forward (top) and reverse (bottom).

TA is the annealing temperature ( C) used for PCR amplification (see Methods text).
c
Species (Sp) characterized are the lesser amberjack (Sfa), yellowtail jack (Sla), almaco jack (Sri), banded rudderfish (Szo), and greater amberjack (Sdu).
d
N is the number of individuals assayed; NA is the number of alleles detected.
e
Size range (bp) refers to alleles that have been discovered thus far; the superscript xx indicates that alleles were not spaced as anticipated (Sdu4, Sdu21, and Sdu27).
f
HE is expected heterozygosity; HO is observed heterozygosity.
g
PHW is the probability of deviation from HardyWeinberg expectations; deviations that were significant after Bonferroni correction (Rice 1989) are shown in bold italics.
h
Possible issues with loci as indicated by Micro-Checker software (Van Oosterhout et al. 2004): N = evidence for null alleles; = no evident issues. Micro-Checker was not utilized with the previously published data for
greater amberjacks.
(CA)16
Repeat
sequence*
AGTGGCTTCTGCTGCTGT ++
CGTGTGCGTGCTTGTAAA
Primer sequence
(5 3 )a*
Sdu39
Microsatellite
TABLE 1.
TECHNICAL NOTE
ACKNOWLEDGMENTS
We thank the following for fish sampling: D. Player,
E. Muhammed, and B. White (South Carolina Department
of Natural Resources); K. Gruenthal and M. Drawbridge
(HubbsSeaWorld Research Institute); T. Morris (Rancheros del
March); and R. Allman, D. DeVries, and B. Walling (National
Marine Fisheries Service). Funding was provided by Texas
AgriLife Research (Project H-6703) and Texas A&M
UniversityNational Council of Science and Technology, Mexico (Project 2010-004). This paper is Number 92 in the series
Genetic Studies in Marine Fishes and Contribution Number
205 of the Center for Biosystematics and Biodiversity at Texas
A&M University.
REFERENCES
Gold, J. R., M. A. Renshaw, E. Saillant, and R. R. Vega. 2010. Spawning frequency of brood dams and sires in a marine fish stock-enhancement hatchery.
Journal of Fish Biology 77:10301040.
Hamasaki, K., K. Tsuruoka, K. Teruya, H. Hashimoto, K. Hamada,
T. Hotta, and K. Mushiake. 2009. Feeding habits of hatchery-reared
larvae of greater amberjack, Seriola dumerili. Aquaculture 288:216
225.
Iguchi, J., Y. Takashima, A. Namikoshi, and M. Yamashita. 2012. Species
identification method for marine products of Seriola and related species.
Fisheries Science 78:197206.
529
Lewis, P. O., and D. Zaykin. 2001. Genetic Data Analysis: computer program for the analysis of allelic data, version 1.0 (d16c). Available:
http://hydrodictyon.eeb.uconn.edu/people/plewis/software.php. (July 2012).
Liu, Z. J., and J. F. Cordes. 2004. DNA marker technologies and their applications in aquaculture genetics. Aquaculture 238:137.
Miller, P. A., A. J. Fitch, M. Gardner, K. S. Hutson, and G. Mair. 2011. Genetic
population structure of yellowtail kingfish (Seriola lalandi) in temperate
Australian waters inferred from microsatellite markers and mitochondrial
DNA. Aquaculture 319:328336.
Reiss, H., G. Hoarau, M. Dickey-Collas, and W. J. Wolff. 2009. Genetic population structure of marine fish: mismatch between biological and fisheries
management units. Fish and Fisheries 10:361395.
Renshaw, M. A., J. C. Patton, C. E. Rexroad III, and J. R. Gold. 2006. PCR
primers for trinucleotide and tetranucleotide microsatellites in greater amberjack, Seriola dumerili. Molecular Ecology Notes 6:11621164.
Renshaw, M. A., J. C. Patton, C. E. Rexroad III, and J. R. Gold. 2007. Isolation
and characterization of dinucleotide microsatellites in greater amberjack,
Seriola dumerili. Conservation Genetics 8:10091011.
Repulles-Albelda, A., F. E. Montero, A. S. Holzer, K. Ogawa, K. S. Hutson, and
J. A. Raga. 2008. Speciation of the Paradeontacylix spp. (Sanguinicolidae)
of Seriola dumerili. Two new species of the genus Paradeontacylix from the
Mediterranean. Parasitology International 57:405414.
Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223225.
Simpson, S. 2011. The blue food revolution. Scientific American 304:5461.
Stuart, K. R., and M. A. Drawbridge. In press. Captive spawning and larval
rearing of California yellowtail (Seriola lalandi). Aquaculture Research. DOI:
10.1111/j.1365-2109.2011.03077.x.
Van Oosterhout, C., W. F. Hutchinson, and P. Shipley. 2004. Micro-Checker:
software for identifying and correcting genotyping errors in microsatellite
data. Molecular Ecology Notes 4:535538.


Dietary Lipid Levels Affect Growth and Fatty Acid

Profiles of Malaysian Mahseer Tor tambroides
a
Ehsan Ramezani-Fard , Mohd Salleh Kamarudin , Che Roos Saad , Sharr Azni Harmin &
Goh Yong Meng
Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, Selangor,

Serdang, 43400, Malaysia
b
Department of Veterinary Preclinical Sciences, Faculty of Veterinary Medicine, Universiti

Putra Malaysia, Selangor, Serdang, 43400, Malaysia
To cite this article: Ehsan Ramezani-Fard, Mohd Salleh Kamarudin, Che Roos Saad, Sharr Azni Harmin & Goh Yong Meng
(2012): Dietary Lipid Levels Affect Growth and Fatty Acid Profiles of Malaysian Mahseer Tor tambroides , North American
Journal of Aquaculture, 74:4, 530-536


DOI: 10.1080/15222055.2012.690829
COMMUNICATION
Dietary Lipid Levels Affect Growth and Fatty Acid Profiles

of Malaysian Mahseer Tor tambroides
Ehsan Ramezani-Fard, Mohd Salleh Kamarudin,* Che Roos Saad,
and Sharr Azni Harmin
Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 Serdang,
Selangor, Malaysia
Goh Yong Meng

Department of Veterinary Preclinical Sciences, Faculty of Veterinary Medicine,
Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia
Abstract
After protein, the second major essential macronutrient in fish
diet is lipid. This study was conducted to determine the optimum
level of dietary lipid for the best growth performance of juvenile
Malaysian mahseer Tor tambroides. Four isonitrogenous diets containing 40% crude protein were formulated to contain different
levels of lipid (5, 10, 15, or 20% on an as-fed basis). Cod liver oil
was incorporated into the feed as the main dietary lipid source used
to formulate the diets while residual oil coming from other ingredients contributed about 5% of dietary lipid. The experimental diets
were labeled as L5, L10, L15, or L20 to denote the levels of dietary
lipid. Fish were fed the experimental diets in triplicate groups for
63 d. Growth performance, survival rate, and daily feed intake
by Malaysian mahseer significantly decreased when fed diets in
which levels of dietary lipid increased from 5% to 20%. However,
the growth performance did not vary significantly between fish fed
the L5 and L10 diets. The increase in dietary lipid significantly
increased the hepatosomatic index but did not influence the viscerosomatic index. Increasing dietary lipid levels also decreased
the lipid content in the whole body composition of fish. The monounsaturated fatty acid (MUFA) and n-3 polyunsaturated fatty
acid (PUFA) contents of fish liver significantly increased with the
increase of dietary lipid. The results of this study suggested that
5% dietary lipid is sufficient for the best survival rate and growth
performance of juvenile Malaysian mahseer.
The Malaysian mahseer Tor tambroides (also known as

Thai mahseer) is one of the most important and highly valued
cyprinids throughout the Himalayan region and southeast Asia
(Ambak et al. 2007). Since the decline in natural populations of
this species owing to overexploitation and destruction of natural
*Corresponding author: msalleh@agri.upm.edu.my
Received January 21, 2012; accepted April 29, 2012
530
habitats (Ingram et al. 2007), there has been great interest in the
conservation and culture of Malaysian mahseer (Ramezani-Fard
et al. 2011a). Efforts have been made to replenish its natural
stock through several conservation programs and to culture the
fish to meet the high demand for this species (Kamarudin et al.,
in press). The first success in induced spawning of pond-reared
Malaysian mahseer broodstock through hormonal treatment
was reported by Ingram et al. (2005). After this breakthrough,
development of an optimized diet that meets the nutritional
requirements of this fish is crucial to ensure the success of
Malaysian mahseer conservation and aquaculture goals. Earlier
works showed that juveniles of this species required 4050%
dietary protein for the optimal growth performance (Ng et al.
2008; Misieng et al. 2011).
Lipids, such as essential fatty acids (EFAs), phospholipids,
and fat-soluble vitamins, are major essential macronutrients in
fish diets and act as a source of energy (Sargent et al. 2002).
However, incorporation of traditional marine-derived lipids in
aquafeeds has become increasingly costly owing to the recent
scarcity and high price of fish oil. Bazaz and Keshavanath (1993)
reported a positive effect of increasing dietary lipid on the
growth performance of Deccan mahseer T. khudree. An excessive amount of dietary lipid can reduce fish growth by reducing
feed consumption and can increase the amount of lipid deposited
in the carcass (Wang et al. 2005). Therefore, the optimal dietary
lipid level plays a critical role in the increase of growth performance and decrease of feed costs. The present study was carried
out to investigate the effects of different levels of dietary lipid on
growth performance, survival rate, body composition, and fatty
531
COMMUNICATION
acid (FA) profiles of muscle and liver of juvenile Malaysian

mahseer.
METHODS
Diet preparation.Four isonitrogenous diets with 40%
crude protein and 5, 10, 15, or 20% crude fat (on an as-fed
basis) were formulated, and experimental diets were labeled as
L5, L10, L15, or L20 (Tables 1, 2). Cod liver oil was incorporated into the feed as the main dietary lipid source used to
formulate the diets while residual oil coming from other ingredients (fish meal, soy meal, corn meal) contributed about 5%
of dietary lipid. Dry ingredients were thoroughly mixed using a
dough mixer. This was followed by adding dietary oils and distilled water to produce a homogenous dough. The moist dough
was then screw-pressed through a 2-mm die and the feed pellets
formed were oven dried and stored at 20 C.
Rearing and sampling.Wild caught juvenile Malaysian
mahseer were provided by a local supplier and their authenticity was verified using taxonomical and morphological features
(Rainboth 1996; Kottelat 1998). The fish were transferred to the
Aquaculture Experimental Station, Universiti Putra Malaysia,
TABLE 1. Feed ingredients and proximate compositions of the test diets
fed to juvenile Malaysian mahseer. Four isonitrogenous diets with 40% crude
protein and 5, 10, 15, or 20% crude fat (on an as-fed basis) were formulated,
and experimental diets were labeled as L5, L10, L15, or L20.
TABLE 2. Fatty acid composition (% of total fatty acid) of the experimental

diets fed to juvenile Malaysian mahseer. See Table 1 for explanation of diet
labels.
Diets
Fatty acid
14:0
16:0
16:1n-7
18:0
18:1n-9
18:2n-6
18:3n-3
20:0
20:1n-9
20:2n-6
20:4n-6
20:5n-3
22:0
22:1n-11
22:5n-3
22:6n-3
SFA
MUFA
n-3 PUFA
n-3 : n-6 ratio
L5
L10
L15
L20
1.10
25.76
3.50
6.38
31.87
19.70
0.88
0.70
0.34
0
0
4.43
0.28
0
1.05
4.00
34.22
35.72
10.37
0.53
3.12
22.81
4.44
5.85
26.34
12.03
0.76
0.10
3.85
1.19
0.23
6.29
0.23
3.79
2.25
6.74
32.11
38.42
16.03
1.19
3.49
21.23
5.49
5.19
23.26
6.34
0.63
0.23
6.30
0.94
0.51
7.89
0.21
6.87
2.43
9.01
30.35
41.92
19.95
2.56
4.35
18.35
6.95
4.38
21.40
3.83
0.51
0.31
8.31
0.74
0.64
8.21
0.15
8.74
2.89
10.23
27.54
45.40
21.85
4.19
Diets
Diet parameter
L5
L10
L15
Ingredient (g/100 g on as-fed basis)

38.0
38.0
38.0
Fishmeala
Soy meal
13.0
13.0
13.0
14.8
14.8
14.8
Caseinb
Corn meal
32.2
28.4
23.1
3.8
9.1
Fish oilc
1.0
1.0
1.0
Vitamin premixd
1.0
1.0
1.0
Mineral premixe
Proximate analysis (% on as-fed basis)
Crude protein
40.4
40.1
39.7
Crude lipid
4.7
9.4
14.8
Ash
10.4
10.8
10.6
32.1
28.6
25.0
Carbohydratesf
Gross energy (kJ/g)
17.3
18.0
19.2
Dry matter
87.6
88.9
90.1
a
L20
38.0
13.0
14.8
17.6
14.6
1.0
1.0
39.6
19.6
10.9
19.1
20.1
89.2
Malaysian fish meal (63% crude protein).

Casein from bovine milk (Sigma-Aldrich).
Cod liver oil (Seven Seas).
d
Vitamin premix (g/kg premix): ascorbic acid, 45; myo-inositol, 5; choline chloride,
75; niacin, 4.5; riboflavin, 1; pyridoxine, 1; thiamin mononitrate, 0.92; Ca-pantothenate,
3; retinyl acetate, 0.6; cholecalciferol, 0.083; vitamin K menadione, 1.67; -tocopheryl
acetate (500 IU/g), 8; biotin, 0.02; folic acid, 0.09; vitamin B12 , 0.001; cellulose, 845.11.
e
Mineral premix (g/kg premix): KCL, 90; KI, 0.04; CaHPO42H2 O, 500; NaCl, 40;
CuSO4 5H2 O, 3; ZnSO4 7H2 O, 4; CoSO4 , 0.02; FeSO4 7H2 O, 20; MnSO4 H2 O, 3; CaCo3 ,
215; MgOH, 124; Na2 SeO3 , 0.03; NaF, 1.
f
Carbohydrates = dry matter (protein + lipid + ash).
b
c
and acclimated to laboratory conditions for 2 weeks in 1,000L fiberglass tanks equipped with a recirculating system that
continuously purified water through a series of mechanical and
biofilter systems. During the acclimation period, fish were on a
diet containing 40% crude protein and 5% crude fat and were
fed twice per day. After this period, 12 rectangular-shaped glass
aquaria (65 L) were each stocked with 10 fish of similar size.
The rearing conditions and water quality of this laboratory setting were fully explained by Ramezani-Fard et al. (2011b). In
summary, each aquarium was equipped with a recirculating system. Water temperature was between 27.5 C and 29 C, while
pH was between 8 and 8.8. Fish were fed twice per day (0900
and 1600 hours) close to apparent satiation. After a 10-d conditioning period, the feeding trial was commenced using fish with
a mean initial weight of 2.6 g (SD, 0.2). All treatments were
conducted in a controlled rearing condition, each of which was
triplicated. Fish in each aquarium were batch-weighed at the
start and end of the experiment, as well as every 3 weeks, and
the feeding trial was conducted for 9 weeks.
Six fish of the initial batch as well as six fish from each
treatment (two fish per tank) at the end of the experiment were
sacrificed, weighed individually, and kept frozen at 80 C for
subsequent whole-body proximate analysis. Similarly, an additional six fish were individually weighed, sacrificed, and dissected. Their liver and viscera were extracted and weighed. The
532
RAMEZANI-FARD ET AL.
dissected fish were subsequently dressed, and muscle from the

area between the lateral and dorsal line was removed. The collected liver and muscle were immediately stored at 80 C
for further FA analyses. The following growth performance
and feed utilization efficiency parameters were also calculated:
weight gain (WG), specific growth rate (SGR), feed conversion
ratio (FCR), daily feed intake (DFI), and protein efficiency ratio
(PER).
Biochemical analysis.Prior to chemical analysis, the whole
body and fillets of fish were lyophilized in triplicate groups per
sample (all fish from a treatment group were combined and triplicate samples were taken from a group) for 48 h, and the lost
moisture was calculated. Proximate composition and FA analysis of samples were described in Ramezani-Fard et al. (2012).
In summary, crude protein and crude lipid were determined
by the Kjeldahl method and Soxhlet extraction, respectively.
The ash content was determined by incinerating the dry sample at 600 C for 4 h, and the gross energy was measured by
direct combustion in an adiabatic bomb calorimeter. Lipid from
feed, liver, and lyophilized fillet was extracted with a chloroform : methanol (2:1, v:v) mixture. Fatty acid methyl esters
(FAMEs) were then separated and quantified on a fused silica
capillary column (Supelco SP-2330; 30 m 0.25 mm; film
thickness, 0.20 mm) in a gas chromatograph (Agilent 7890N,
Agilent Technologies, Santa Clara, California). Fatty acids were
identified by comparing the relative retention time with the reference standards (Supelco, 37 components FAME Mix; Supelco,
Bellefonte, Pennsylvania) and menhaden oil and expressed as
the area percentage of FAMEs.
Statistical analysis.All experimental data were analyzed
by ANOVA. The mean differences were evaluated using Dun-
cans multiple-range test. Homogeneity of variances was tested

using Levenes test, and data identified as nonhomogeneous
were subjected to arcsine transformation before statistical analysis. All analysis was carried out with SPSS 15 for Windows
(SPSS, Chicago, Illinois) and the difference was considered significant at P < 0.05.
RESULTS
Growth performance, survival rate, feed efficiency, and body
indices of Malaysian mahseer fed different diets are presented
in Table 3. Growth of fish significantly decreased as dietary lipid
increased from 5% to 20%. However, the growth performances
of fish fed the L5 and L10 diets were not significantly different.
No mortality was observed in fish fed either the L5 or L10 diet.
The survival rate was significantly lower in fish fed the L15 or
L20 diet compared with those fed the other diets. The fish fed
the L5 or L10 diet had the highest DFI while fish fed the L20
diet had the lowest intake of feed. The best and worst FCR were
observed in fish fed the L20 and L10 diets, respectively. The
hepatosomatic index (HSI) was significantly lower in fish fed
the L5 or L10 diet than in those fed the L15 or L20 diet. The
viscerosomatic indices (VSIs) were not significantly influenced
by the diets.
The whole-body protein, lipid, and moisture contents of
Malaysian mahseer were significantly influenced by the experimental diets (Table 4). Fish fed the L5 and L10 diets had significantly higher whole-body lipid than those fed the L15 and
L20 diets. Increased moisture content and decreased protein
content were also observed in the whole-body composition of
fish when fish were fed the L20 diet. There were no significant
TABLE 3. Growth performance, survival rate, feed utilization efficiency, and body indices of juvenile Malaysian mahseer fed the experimental diet for 63 d.
See Table 1 for explanation of diet labels. Mean SE (n = 3 except for HSI and VSI where n = 6). Values within the same row with different lowercase letters
are significantly different at P < 0.05.
Diets
Parameters
Initial body weight (g)
Final body weight (g)
WGa (g)
SGRb (%/d)
Survival (%)
FCRc
DFId (%/d)
PERe
HSIf
VSIg
a
L5
L10
L15
L20
2.70 0.09
9.19 0.57 z
6.50 0.52 z
1.94 0.08 z
100.0 z
2.08 0.08 y
3.58 0.03 z
1.12 0.04
1.30 0.04 y
6.67 0.34
2.70 0.08
8.29 0.13 zy
5.59 0.06 z
1.78 0.02 z
100.0 z
2.40 0.05 z
3.83 0.06 z
1.08 0.05
1.62 0.06 zy
7.14 0.33
2.86 0.05
7.32 0.14 y
4.46 0.18 y
1.49 0.06 y
77.8 2.8 y
1.68 0.01 x
2.49 0.17 y
1.27 0.14
2.46 0.33 z
6.68 0.56
3.0 0.15
7.20 0.30 y
4.21 0.15 y
1.39 0.01 y
75.0 4.8 y
1.42 0.04 w
1.85 0.04 x
1.34 0.13
2.43 0.58 z
6.23 0.35
0.19
0.01
0.00
0.00
0.00
0.00
0.00
0.27
0.02
0.39
Final weight gain.

Specific growth rate: [(loge final mean weight loge initial mean weight) / experimental days] 100.
c
Feed conversion ratio: dry feed intake (g) / wet weight gain (g).
d
Daily feed intake: 100 dry feed intake (g) / [(total final body weight + total initial body weight) / 2] experimental days.
e
Protein efficiency ratio: wet weight gain (g) / protein intake (g).
f
Hepatosomatic index: 100 liver weight (g) / body weight (g).
g
Viscerosomatic index: 100 visceral weight (g) / body weight (g).
b
533
COMMUNICATION
TABLE 4. Whole-body proximate composition, liver lipid, and muscle lipid content (% of wet weight) of juvenile Malaysian mahseer fed the experimental diets
for 63 d. See Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letters are significantly different at
P < 0.05.
Diets
Parameter
Moisture
Protein
Fat
Ash
Carbohydratea
Liver fat
Muscle fat
Initial
L5
L10
L15
L20
68.4 0.1
17.3 0.1
9.5 0.1
3.5 0.0
0.6 0.1
19.6 1.0
5.5 0.1
68.2 0.1 y
17.5 0.2 z
9.8 0.0 z
3.6 0.1
0.8 0.1 z
20.1 0.6
5.1 0.1
67.6 0.2 y
17.7 0.1 z
9.7 0.1 z
3.6 0.0
1.3 0.2 zy
19.4 0.6
5.1 0.3
68.1 0.3 y
17.4 0.1 z
8.1 0.0 y
3.5 0.1
2.6 0.3 y
19.1 1.0
5.2 0.1
70.7 0.3 z
16.0 0.1 y
7.9 0.0 x
3.6 0.1
2.7 0.4 y
21.0 1.3
5.1 0.0
0.00
0.02
0.00
0.42
0.03
0.75
0.64
Carbohydrates = dry matter (protein + lipid + ash).
differences between total lipid contents of liver and muscle of

fish fed different diets (Table 4).
Palmitic acid (16:0) and oleic acid (18:1n-9) were the most
abundant FAs in the muscle of Malaysian mahseer. In addition to these FAs, stearic acid (18:0), linoleic acid (18:2n-6),
and docosahexaenoic acid (DHA) (22:6n-3) were the other
FAs contributing most to the total FA content of fish muscle
(Table 5). The increase of fish oil inclusion in the diet led to
a significant decrease of saturated fatty acid (SFA) as well as
an increase of monounsaturated fatty acid (MUFA) in fish mus-
cle. The concentration of eicosapentaenoic acid (EPA; 20:5n-3)

significantly increased with the increase of dietary lipid level.
However, the muscle contents of most of long-chain polyunsaturated fatty acids (LC-PUFAs) including 22:6n-3, 20:5n-3,
and 20:4n-6 decreased at the end of the experiment compared
with their initial amounts. There were no significant differences
between the concentrations of 20:4n-6, 22:5n-3, and 22:6n-3 in
the muscle of fish fed different diets.
Similar to the muscle, the concentrations of SFAs, MUFAs,
and n-3 PUFAs in the fish liver were significantly affected by the
TABLE 5. Fatty acid composition (% of total fatty acid) of muscle tissue of juvenile Malaysian mahseer at the beginning and end of the experimental period.
See Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letter are significantly different at P < 0.05.
Diets
Fatty acid
Initial
L5
L10
L15
L20
14:0
14:1
16:0
16:1n-7
18:0
18:1n-9
18:2n-6
18:3n-3
20:0
20:1n-9
20:3n-6
20:4n-6
20:5n-3
22:1n-11
22:5n-3
22:6n-3
SFA
MUFA
n-3 PUFA
n-3 : n-6 ratio
3.6
0.21
25.0
5.7
10.2
24.2
7.9
0.18
2.60
1.3
0.54
6.1
3.6
0.21
1.3
7.2
41.4
31.7
12.3
0.85
3.5 0.0 z
0.28 0.08
28.4 0.2 z
3.7 0.1 x
8.5 0.1
32.4 0.6
8.2 0.2 z
0.20 0.02
0.80 0.02
1.6 0.1 x
0.74 0.10 z
3.1 0.4
1.4 0.1 y
3.6 0.0 z
0.22 0.03
28.3 0.1 z
6.3 0.1 zy
8.4 0.4
30.6 0.6
7.0 0.0 y
0.22 0.03
0.61 0.02
2.3 0.0 x
0.48 0.05 y
2.6 0.2
1.5 0.0 zy
0.64 0.02 x
1.7 0.1
5.6 0.6
41.0 0.5 z
40.0 0.6 x
8.9 0.7
0.89 0.09 zy
3.3 0.0 y
0.2 0.02
25.2 0.3 y
6.0 0.2 y
8.2 0.4
31.9 0.2
6.6 0.2 y
0.35 0.03
0.77 0.07
4.1 0.1 y
0.42 0.03 y
2.5 0.3
1.8 0.1 z
1.72 0.37 y
1.7 0.2
5.6 0.2
37.4 0.4 y
43.8 0.5 y
9.4 0.2
0.99 0.03 z
3.3 0.1 y
0.13 0.03
21.4 0.3 x
6.6 0.2 z
8.3 0.3
32.0 0.5
6.6 0.1 y
0.32 0.06
0.75 0.13
5.5 0.4 z
0.39 0.04 y
2.5 0.1
1.8 0.2 z
2.84 0.05 z
1.7 0.2
5.8 0.5
33.8 0.2 x
47.1 0.6 z
9.6 0.6
1.01 0.05 z
0.00
0.22
0.00
0.00
0.89
0.12
0.00
0.05
0.33
0.00
0.01
0.33
0.03
0.00
0.99
0.94
0.00
0.00
0.57
0.01
1.7 0.1
5.4 0.2
41.3 0.3 z
38.0 0.4 w
8.7 0.3
0.72 0.01 y
534
TABLE 6. Fatty acid composition (% of total fatty acid) of liver tissue of juvenile Malaysian mahseer at the beginning and end of the experimental period. See
Table 1 for explanation of diet labels. Mean SE, n = 3; values within the same row with different lowercase letters are significantly different at P < 0.05.
Diets
Fatty acid
Initial
L5
L10
L15
L20
14:0
16:0
16:1n-7
18:0
18:1n-9
18:2n-6
18:3n-3
20:0
20:1n-9
20:2n-6
20:3n-6
20:4n-6
20:5n-3
22:1n-11
22:5n-3
22:6n-3
SFA
MUFA
n-3 PUFA
n-3 : n-6 ratio
4.1
26.0
5.9
9.1
35.8
6.2
0.17
3.5
0.46
0.74
4.6
0.2
0.68
0.24
2.2
42.8
42.4
3.3
0.29
4.4 0.1
28.6 0.3 z
6.5 0.2
8.5 0.1 z
39.9 0.5 z
4.9 0.2 y
0.07 0.01 w
0.56 0.03
1.4 0.1 w
0.12 0.01 y
0.61 0.02
1.1 0.1
0.66 0.05 x
0.21 0.03 x
2.4 0.1 x
42.0 0.5 z
47.9 0.6 zy
3.4 0.1 w
0.50 0.02 x
4.6 0.0
28.1 0.2 z
6.4 0.2
7.4 0.2 y
35.9 0.3 y
6.4 0.1 z
0.23 0.01 x
0.60 0.09
2.8 0.0 x
0.27 0.01 z
0.77 0.07
1.4 0.2
0.82 0.11 yx
1.0 0.1 x
0.51 0.02 y
2.8 0.1 x
40.6 0.3 zy
46.2 0.4 y
4.4 0.2 x
0.50 0.01 x
4.8 0.6
26.7 0.7 y
6.6 0.3
7.4 0.2 y
32.3 0.9 x
5.8 0.2 z
0.35 0.01 y
0.83 0.07
5.0 0.1 y
0.13 0.07 y
0.68 0.16
1.4 0.3
1.04 0.06 zy
2.3 0.2 y
0.52 0.05 y
4.2 0.2 y
39.7 1.0 y
46.2 1.1 y
6.1 0.3 y
0.76 0.03 y
4.3 0.1
23.4 0.1 x
7.6 0.4
7.1 0.5 y
30.8 0.2 x
4.1 0.3 x
0.48 0.07 z
0.62 0.07
7.5 0.2 z
0.25 0.01 z
0.37 0.03
1.3 0.0
1.14 0.07 z
4.3 0.2 z
1.12 0.06 z
5.6 0.3 z
35.4 0.5 x
50.2 0.7 z
8.4 0.4 z
1.39 0.08 z
0.68
0.00
0.06
0.03
0.00
0.00
0.00
0.09
0.00
0.03
0.06
0.68
0.00
0.00
0.00
0.00
0.00
0.01
0.00
0.00
diets (Table 6). The increase of dietary lipid level significantly

decreased the concentration of SFA and 18:1n-9 in the liver.
Increased total n-3 PUFA content was observed in the liver of
fish, increasing from 3.38 to 8.37 (% of total FA) when fish were
fed increasing amounts of dietary lipid. A notable decrease of
20:4n-6 content was observed in the liver of fish fed all the
experimental diets compared with the initial amount. However,
liver content of 20:4n-6 did not differ significantly among fish
fed diets with increasing lipid levels.
DISCUSSION
Earlier studies showed that increasing dietary lipid levels are
likely to improve or be ineffective in improving the growth performance of many marine carnivorous species such as white
seabass Atractoscion nobilis and Atlantic halibut Hippoglossus
hippoglossus (Bright et al. 2005; Martins et al. 2007; Lopez
et al. 2009). However, a high dietary lipid level can inhibit the
growth of some freshwater species such as hybrid tilapia Oreochromis niloticus Oreochromis aureus, silver barb Puntius
gonionotus, and grass carp Ctenopharyngodon idella (Chou and
Shiau 1996; Du et al. 2005; Mohanta et al. 2008). The reduction
of growth performance of Malaysian mahseer with increasing
dietary lipid was in agreement with the latter group of farmed
freshwater species. As shown in earlier studies (Satpathy et al.
2003; Hamre et al. 2004; Alam et al. 2009a), this study also
revealed that FCR decreased with increasing dietary lipid. It is
well established that fish adjust their feed intake to meet their
digestible energy requirements (Du et al. 2005). Although the
feed intake of some carnivorous species such as Atlantic halibut
and southern flounder Paralichthys lethostigma is not extremely
affected by dietary lipid level (Martins et al. 2007; Alam et al.
2009b), the feed intake of most omnivorous species reduces with
increasing dietary lipid level (Pei et al. 2004; Du et al. 2005;
Wang et al. 2005). In the present study, DFI decreased dramatically with the increase of dietary lipid level. Considering that
the test diets were not isocaloric, fish on high lipid diets may
reduce their feed intake for energy regulation purposes. This
reduction can lead to reduced growth because of insufficient
protein intake (Lee and Kim 2001).
An early work by Ng et al. (2008) showed that optimal growth
of Malaysian mahseer fingerlings is achieved when they are fed
a semipurified diet containing 10% lipid and 1819 kJ/g gross
energy. According to their study, the increase of dietary lipid up
to 15% does not improve nor inhibit the growth performance of
fish. It should be noted that corn oil supplied more than 50%
of lipid in their study, while fish oil was the primary source
of lipid in the present study. High amounts of dietary fish oil
have been hypothesized to reduce growth performance of some
warmwater fishes such as tilapia and some carps (Alava 1998;
Du et al. 2008).
COMMUNICATION
Proximate analysis showed that whole-body lipid content

of Malaysian mahseer decreased when their dietary lipid was
increased from 10% to 15%. De Silva et al. (2001) found similar
results in freshwater shortfin eel Anguilla australis that were fed
excess amounts of dietary lipid in which the muscle lipid content
was reduced after a 10-week feeding trial. The present study
also showed that the HSI significantly increased as dietary lipid
levels increased; however, such an increase did not influence
the VSI. The HSI in Malaysian mahseer fed either the 15%
or 20% lipid diets were higher compared with that in fish fed
the 5% lipid diets. The increased HSI, along with the lack of
lipid accumulation in the liver, could be associated with the
accumulation of carbohydrate molecules.
Several authors have noted that tissue FA profiles reflect the
dietary FA intake composition (dos Santos et al. 1993; AliyuPaiko et al. 2010). This is in agreement with increasing levels of
MUFAs and n-3 PUFAs in both muscle and liver of Malaysian
mahseer fed increasing levels of dietary lipid. The increase of
long-chain MUFAs (20:1n-9 and 22:1n-11), which were prevalent in fish oil, was notable in Malaysian mahseer tissues in response to the increase of cod liver oil in the diets. The amounts of
n-3 PUFAs, particularly 22:6n-3, increased in the liver of fish fed
diets with 15% or 20% lipid compared with those fed the other
diets. A notable reduction of 20:4n-6 levels in the tissues of fish
fed all the experimental diets compared with its initial amount
may be associated with deficiency of 18:2n-6 and 18:3n-3, the
parental FAs of 20:4n-6, in the diets. This reduction can be prevented by partial replacement of dietary fish oil with vegetable
oils containing high amounts of C18 PUFA. The findings of this
study suggest that a high level of n-3 LC-PUFA is not required
in the diet of Malaysian mahseer. In agreement with this finding,
some freshwater fish species may have a higher requirement for
n-6 PUFA over n-3 PUFA in the diet, particularly those species
capable of synthesizing LC-PUFAs de novo from 18-carbon
precursors such as 18:2n-6 (Mishra and Samantaray 2004;
Zuraini et al. 2006). In conclusion, this study demonstrated that
a diet with 5% lipid is sufficient to provide the best survival rate
and growth performance in juvenile Malaysian mahseer. A dietary lipid level beyond 10% containing more than 3.8% fish oil
reduces both survival rate and growth performance of this fish.
ACKNOWLEDGMENTS
This study was supported through a Malaysian government
E-Science grant no. 05-01-04-SF0209. The authors thank Mahdi
Ebrahimi for technical assistance.
REFERENCES
Alam, M. S., W. O. Watanabe, P. M. Carroll, and T. Rezek. 2009a. Effects of
dietary protein and lipid levels on growth performance and body composition
of black sea bass Centropristis striata (Linnaeus 1758) during grow-out in a
pilot-scale marine recirculating system. Aquaculture Research 40:442449.
Alam, M. S., W. O. Watanabe, and H. V. Daniels. 2009b. Effect of different dietary protein and lipid levels on growth performance and body composition of
535
juvenile southern flounder, Paralichthys lethostigma, reared in a recirculating

aquaculture system. Journal of the World Aquaculture Society 40:513521.
Alava, V. R. 1998. Effect of salinity, dietary lipid source and level on growth of
milkfish (Chanos chanos) fry. Aquaculture 167:229236.
Aliyu-Paiko, M., R. Hashim, A. S. Chong, L. Yogarajah, and A. M. El-Sayed.
2010. Influence of different sources and levels of dietary protein and lipid
on the growth, feed efficiency, muscle composition and fatty acid profile of
snakehead Channa striatus (Bloch, 1793) fingerling. Aquaculture Research
41:13651376.
Ambak, M. A., A. H. Ashraf, and S. Budin. 2007. Conservation of the Malaysian
mahseer in Nenggiri basin through community action. Pages 217228 in S.
S. Siraj, A. Christianus, C. K. Ng, and S. S. De Silva, editors. Proceedings
of the international symposium on the mahseer. Malaysian Fisheries Society,
Kuala Lumpur.
Bazaz, M. M., and P. Keshavanath. 1993. Effect of feeding different levels of
sardine oil on growth, muscle composition and digestive enzyme activities of
mahseer, Tor khudree. Aquaculture 115:111119.
Bright, L. A., S. D. Coyle, and J. H. Tidwell. 2005. Effect of dietary lipid level
and protein energy ratio on growth and body composition of largemouth bass
Micropterus salmoides. Journal of the World Aquaculture Society 36:129
134.
Chou, B. S., and S. Y. Shiau. 1996. Optimal dietary lipid level for growth
of juvenile hybrid tilapia, Oreochromis niloticus Oreochromis aureus.
De Silva, S. S., R. M. Gunasekera, G. Gooley, and B. A. Ingram. 2001. Growth
of Australian shortfin eel (Anguilla australis) elvers given different dietary
protein and lipid levels. Aquaculture Nutrition 7:5357.
dos Santos, J., I. C. Burkow, and M. Jobling. 1993. Patterns of growth and lipid
deposition in cod (Gadus morhua L.) fed natural prey and fish-based feeds.
Du, Z. Y., P. Clouet, L. M. Huang, P. Degrace, W. H. Zheng, J. G. He, L. X. Tian,
and Y. J. Liu. 2008. Utilization of different dietary lipid sources at high level
in herbivorous grass carp (Ctenopharyngodon idella): mechanism related to
hepatic fatty acid oxidation. Aquaculture Nutrition 14:7792.
Du, Z. Y., Y. J. Liu, L. X. Tian, J. T. Wang, Y. Wang, and G. Y. Liang. 2005.
Effect of dietary lipid level on growth, feed utilization and body composition
by juvenile grass carp (Ctenopharyngodon idella). Aquaculture Nutrition
11:139146.
Hamre, K., R. Christiansen, R. Waagb, A. Maage, B. E. Torstensen, B. Lygren,
. Lie, E. Wathne, and S. Albrektsen. 2004. Antioxidant vitamins, minerals
and lipid levels in diets for Atlantic salmon (Salmo salar, L.): effects on
growth performance and fillet quality. Aquaculture Nutrition 10:113123.
Ingram, B. A., S. Sungan, G. J. Gooley, S. Y. Sim, D. Tinggi, and S. S. De Silva.
2005. Induced spawning, larval development and rearing of two indigenous
Malaysian mahseer, Tor tambroides and T. douronensis. Aquaculture Research 36:983995.
Ingram, B. A., S. Sungan, D. Tinggi, S. Y. Sim, G. J. Gooley, and S. S. De Silva.
2007. Development in the spawning of Tor tambroides and T. douronensis in
captivity. Pages 123126 in S. S. Siraj, A. Christianus, C. K. Ng, and S. S. De
Silva, editors. Proceedings of the international symposium on the mahseer.
Malaysian Fisheries Society, Kuala Lumpur.
Kamarudin, M. S., E. Ramezani-Fard, C. R. Saad, and S. A. Harmin. In
press. Effects of dietary fish oil replacement by various vegetable oils
on growth performance, body composition and fatty acid profile of juvenile Malaysian mahseer, Tor tambroides. Aquaculture Nutrition. DOI:
10.1111/j.1365-2095.2011.00907.x.
Kottelat, M. 1998. Fishes of the Nam Theun and Xe Bangfai basins, Laos,
with diagnoses of twenty-two new species (Teleostei: Cyprinidae, Balitoridae, Cobitidae, Coiidae and Odontobutidae). Ichthyological Exploration of
Freshwaters 9:1128.
Lee, S. M., and K. D. Kim. 2001. Effects of dietary protein and energy levels
on the growth, protein utilization and body composition of juvenile masu
salmon (Oncorhynchus masou Brevoort). Aquaculture Research 32(Supplement 1):3945.
536
Lopez, L. M., E. Durazo, M. T. Viana, M. Drawbridge, and D. P. Bureau.

2009. Effect of dietary lipid levels on performance, body composition and
fatty acid profile of juvenile white seabass, Atractoscion nobilis. Aquaculture
289:101105.
Martins, D. A., L. M. P. Valente, and S. P. Lall. 2007. Effects of dietary lipid level
on growth and lipid utilization by juvenile Atlantic halibut (Hippoglossus
hippoglossus, L.). Aquaculture 263:150158.
Mishra, K., and K. Samantaray. 2004. Interacting effects of dietary lipid level
and temperature on growth, body composition and fatty acid profile of rohu,
Labeo rohita (Hamilton). Aquaculture Nutrition 10:359369.
Misieng, J. D., M. S. Kamarudin, and M. Musa. 2011. Optimum dietary protein requirement of Malaysian mahseer (Tor tambroides) fingerling. Pakistan
Journal of Biological Sciences 14:232235.
Mohanta, K. N., S. N. Mohanty, J. K. Jena, and N. P. Sahu. 2008. Optimal
dietary lipid level of silver barb, Puntius gonionotus fingerlings in relation to
growth, nutrient retention and digestibility, muscle nucleic acid content and
digestive enzyme activity. Aquaculture Nutrition 14:350359.
Ng, W. K., N. Abdullah, and S. S. De Silva. 2008. The dietary protein requirement of the Malaysian mahseer, Tor tambroides (Bleeker), and the lack of
protein-sparing action by dietary lipid. Aquaculture 284:201206.
Pei, Z., S. Xie, W. Lei, X. Zhu, and Y. Yang. 2004. Comparative study on
the effect of dietary lipid level on growth and feed utilization for gibel carp
(Carassius auratus gibelio) and Chinese longsnout catfish (Leiocassis longirostris Gunther). Aquaculture Nutrition 10:209216.
Rainboth, W. J. 1996. Fishes of the Cambodian Mekong. Food and Agriculture
Organization of the United Nations, Rome.
Ramezani-Fard, E., M. S. Kamarudin, S. A. Harmin, and C. R. Saad. 2012.

Dietary saturated and omega-3 fatty acids affect growth and fatty acid profiles
of Malaysian mahseer. European Journal of Lipid Science and Technology
114:185193.
Ramezani-Fard, E., M. S. Kamarudin, S. A. Harmin, C. R. Saad, M. K. Abd
Satar, and S. K. Daud. 2011a. Ontogenic development of the mouth and
digestive tract in larval Malaysian mahseer, Tor tambroides Bleeker. Journal
of Applied Ichthyology 27:920927.
Ramezani-Fard, E., M. S. Kamarudin, C. R. Saad, and S. A. Harmin. 2011b.
Changes over time in muscle fatty acid composition of Malaysian mahseer,
Tor tambroides, fed different dietary lipid percentage. African Journal of
Biotechnology 10:1825618265.
Sargent, J. R., D. R. Tocher, and J. G. Bell. 2002. The lipids. Pages 181257 in
J. E. Halver and R. W. Hardy, editors. Fish nutrition, third edition. Academic
Press, San Diego, California.
Satpathy, B. B., D. Mukherjee, and A. K. Ray. 2003. Effects of dietary protein
and lipid levels on growth, feed conversion and body composition in rohu,
Labeo rohita (Hamilton), fingerlings. Aquaculture Nutrition 9:1724.
Wang, J. T., Y. J. Liu, L. X. Tian, K. S. Mai, Z. Y. Du, Y. Wang, and H. J. Yang.
2005. Effect of dietary lipid level on growth performance, lipid deposition,
hepatic lipogenesis in juvenile cobia (Rachycentron canadum). Aquaculture
249:439447.
Zuraini, A., M. N. Somchit, M. H. Solihah, Y. M. Goh, A. K. Arifah, M. S.
Zakaria, N. Somchit, M. A. Rajion, Z. A. Zakaria, and A. M. Mat Jais. 2006.
Fatty acid and amino acid composition of three local Malaysian Channa spp.
fish. Food Chemistry 97:674678.


Evidence of Female Heterogamety in Largemouth Bass,

Based on Sex Ratio of Gynogenetic Progeny
a
Robert P. Glennon , Boris Gomelsky , Kyle J. Schneider , Anita M. Kelly & Alf Haukenes
c
a
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas, 72086, USA
Aquaculture Research Center, Kentucky State University, 103 Athletic Drive, Frankfort,
c
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine
Bluff, Arkansas, 71601, USA
To cite this article: Robert P. Glennon, Boris Gomelsky, Kyle J. Schneider, Anita M. Kelly & Alf Haukenes (2012): Evidence
of Female Heterogamety in Largemouth Bass, Based on Sex Ratio of Gynogenetic Progeny, North American Journal of
Aquaculture, 74:4, 537-540


DOI: 10.1080/15222055.2012.700906
COMMUNICATION

Robert P. Glennon
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas 72086, USA

Kentucky 40601, USA
Anita M. Kelly and Alf Haukenes

Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff,
Arkansas 71601, USA
Abstract
Meiotic gynogenetic progeny in largemouth bass Micropterus
salmoides have been obtained by inseminating largemouth bass
eggs with UV-irradiated sperm from white bass Morone chrysops
or striped bass Morone saxatilis and suppressing the second meiotic division by hydrostatic pressure. The sex composition of gynogenetic progeny was determined by dissection or ultrasound investigation of 1-year-old fish. Among the 21 fish analyzed, 7 fish
(33.3%) were male and 14 fish (66.7%) were female. The presence
of males in meiotic gynogenetic progeny suggests the existence of
female heterogamety (WZ females, ZZ males) in largemouth bass.
The largemouth bass Micropterus salmoides is the most popular game fish in the United States. Females of this species
exhibit faster growth and attain larger size than do the males
(Padfield 1951). Manipulating the sex ratio of largemouth bass
populations is of great interest because of its potential to produce higher ratios of females or monosex female populations
for the purposes of creating larger and faster growing fish for
aquaculture and trophy sport fishing. Several studies have evaluated hormonal sex reversal with the goal of shifting sex ratios under influence of androgens or estrogens in largemouth
bass (Garrett 1989; Porter 1996; Al-Ablani and Phelps 2001;
Arslan et al. 2009). Induced triploidy has also been tested in
this species for reproduction control and obtaining potentially
larger fish (Garrett et al. 1992; Fries et al. 2002; Neal et al.
2004).
The optimal method for production of monosex fish populations involves crossing previously obtained sex-reversed fish
with normal breeding fish (Purdom 1993; Donaldson 1996). The
scheme of crosses directed to production of monosex progenies
depends on the type of heterogamety in given fish species. In
fishes, both male (XY males, XX females) and female (WZ
females, ZZ males) heterogamety are described; sometimes different types of heterogamety are revealed in closely related
species (Dabrowski et al. 2000; Devlin and Nagahama 2002).
Until now, there were no data on sex determination mechanism
in largemouth bass.
One of the basic methods for determining the type of heterogamety in fish is based on sex composition of gynogenetic
progenies. In the case of male heterogamety (XY/XX), gynogenetic progenies usually consist of females only; the presence of both females and males indicates female heterogamety
(WZ/ZZ; Thorgaard 1983; Van Eenennaam et al. 1999; Devlin
and Nagahama 2002). This article presents data on sex composition of a gynogenetic progeny in largemouth bass that suggests
the existence of female heterogamety in this species.
METHODS
Experiments on production of gynogenetic progeny in largemouth bass were conducted at the facilities of J. M. Malone and
Son, Inc. (Lonoke, Arkansas) in April 2009. To induce gynogenetic development, largemouth bass eggs were inseminated

Received January 19, 2012; accepted June 4, 2012
537
538
GLENNON ET AL.
with UV-irradiated sperm from white bass Morone chrysops or

striped bass Morone saxatilis. For production of gynogenetic
diploids, the second meiotic division in eggs was suppressed by
application of hydrostatic pressure.
Ovulation was induced in mature largemouth bass females
by a single intramuscular injection of human chorionic gonadotropin (HCG, 4,000 IU/kg), after which they were placed
into separate 246-L fiberglass tanks. The time between hormonal injection and ovation was 3640 h. White bass and striped
bass males were each given one intramuscular HCG injection
(1,000 IU/kg). The sperm obtained was irradiated based on techniques previously used for white bass sperm irradiation in experiments on induced gynogenesis in black crappie Pomoxis
nigromaculatus (Gomelsky et al. 2000). Specifically, sperm
was irradiated at a dose of 1,000 J/m2 with a FisherBiotech
UV microprocessor-controlled Crosslinker (FB-UVXL-1000;
Fisher Scientific). For irradiation, 2 mL of sperm diluted 10fold with 0.85% NaCl solution was placed in 6-cm-diameter
glass Petri dishes. Parameters of pressure shock for suppression
of the second meiotic division in largemouth bass were chosen
based on published data from experiments on induced triploidy
(Garrett et al. 1992; Neal et al. 2004); pressure shock of 8,000
psi for 1 min was initiated 5 min after the eggs were inseminated
with irradiated sperm. The fertilized eggs were placed into a 1L hydrostatic pressure chamber and the desired pressure was
achieved using a 20 ton hydraulic jack.
In total, ovulated eggs were produced from 12 females. Nine
trials on insemination of eggs with irradiated sperm and application of hydrostatic pressure were performed; in some trials
eggs from several females were mixed. After pressure treatment,
eggs were put into separate 246-L fiberglass tanks for embryo
incubation and hatching of larvae. The resulting swim-up gynogenetic larvae were reared in the same tanks in which embryos
had been incubated and fed with live zooplankton supplied from
filtered pond water for 30 days after hatching.
Two hundred and sixty gynogens survived to the fry stage.
Sixty gynogenetic fish were randomly selected and reared in
1 m3 tanks at a separate facility on live foods (minnow and
bluegill) for 1 year. The remaining 200 gynogenic fish were
moved to a different facility for hormonal sex reversal by androgen treatment; the results of that study will be reported in a
separate publication.
Twenty-one gynogenetic fish from the untreated group survived to the age of one year. On April 19, 2010, 15 fish were
lost due to equipment failure; they were dissected for sex determination by macroscopic observation of gonads and their total
length was recorded. The six surviving gynogenetic fish from
this group were sexed 3 days later by ultrasound investigation
using Tela-Vet 1000 equipped with a 58 MHz linear transducer
(Classic Medical, Tequesta, Florida). Ultrasound is a noninvasive method that has been shown to accurately determine the
sex in several species of fish (Martin et al. 1983; Reimers et al.
1987; Shields et al. 1993; Blythe et al. 1994; Colombo et al.
2004; Masoudifard et al. 2011). The fish were anesthetized with
quinaldine sulfate and placed in a pan of water. Each fish was

held submerged in the water against a styrofoam pad to prevent
double imaging while the ultrasound was performed. The probe
was placed near the abdomen to determine the sex of each fish.
Female fish were identified by the presence of eggs in the ovarian sacs; males had no eggs but showed the presence of testes
structures.
RESULTS
From 15 dissected fish, 7 fish were male and 8 fish were female. Males and females had well-developed maturing testes
and ovaries, respectively, with normal gonad morphological
structure and color for this species. Mean lengths (SD) of
males and females were 27.09 ( 1.17) cm and 25.44 ( 0.95)
cm, respectively.
The six remaining 1-year-old fish were identified as females
by ultrasound investigation. In the summer of 2011, at 2 years
of age, further observations of these fish confirmed their identification as females; when pressed on the abdomen, no sperm
was expressed and other traits typical for females were observed
(e.g., swollen abdomen).
In total, of the 21 gynogenetic fish analyzed, 7 (33.3%) were
male and 14 (66.7%) were female.
DISCUSSION
The use of heterologous sperm, instead of sperm from the
same species, is a very effective method for experiments on
diploid gynogenesis in fish. If hybrids between two species
are nonviable, survivors are exclusively of gynogenetic origin
(Chourrout 1987; Mims et al. 1997; Dabrowski et al. 2000;
Gomelsky et al. 2000). Preliminary experiments conducted at
the Aquaculture Research Center of Kentucky State University
in 2008 have shown that white bass spermatozoa were capable to
fertilizing largemouth bass eggs, but the hybrids obtained were
nonviable and perished before or soon after hatching. Since the
present study used only irradiated sperm of Morone species for
induction of gynogenetic development in largemouth bass eggs,
all obtained larvae could be only of gynogenetic origin. In an
earlier study, irradiated sperm of white bass was used for induced
gynogenesis in another species of the family Centrarchidae,
black crappie (Gomelsky et al. 2000).
Usually the presence of males in meiotic gynogenetic progenies suggests the existence of female heterogamety (WZ/ZZ)
in given species. Other possible factors such as autosomal and
environmental influences on sex determination can also result in
appearance of gynogenetic males (Devlin and Nagahama 2002;
Flynn et al. 2006). Males in meiotic gynogenetic progenies
were described in plaice Pleuronectes platessa (Purdom and
Lincoln 1973), blue tilapia Oreochromis aureus (Penman et al.
1987), common barbel Barbus barbus (Castelli 1994), muskellunge Esox masquinongy (Dabrowski et al. 2000), white sturgeon Acipenser transmontanus (Van Eenennaam et al. 1999),
shortnose sturgeon A. brevirostrum (Flynn et al. 2006), and
COMMUNICATION
some other fish. The frequency of heterozygotes in a meiotic gynogenetic progeny obtained from a female that is heterozygous for some gene depends on the frequency of crossing over between the gene and centromere during prophase
of the first meiotic division; the proportion of heterozygotes
increases with increasing crossing over frequency (Thompson
1983; Thorgaard et al. 1983). Similarly, when meiotic gynogenetic progeny is obtained from heterogametic female (WZ),
the resulting sex ratio depends on the frequency of crossing
over between a sex-determining gene and centromere. If WW
fish are viable, then in the absence of crossing over, the meiotic gynogenetic progeny obtained from WZ female will consist of WW females (so called super-females) and ZZ males
with ratio 1:1. Crossing over between the sex-determining gene
and the centromere will result in the appearance of WZ females and, with increasing recombination rates, the phenotypic sex ratio will be shifted towards prevalence of females.
If all three categories are viable, the proportion of ZZ males
should be approximately equal to the proportion of WW superfemales (Van Eenennaam et al. 1999; Devlin and Nagahama
2002).
One third (33%) of the meiotic gynogenetic progeny of largemouth bass were males. This is similar to the proportion of males
in meiotic gynogenetic progenies observed in plaice (37%;
Purdom and Lincoln 1973), muskellunge (40%; Dabrowski et al.
2000), and shortnose sturgeon (35%; Flynn et al. 2006). If
WW super-females in largemouth bass are viable, their proportion in gynogenetic progeny should be approximately 33%,
the same as proportions of ZZ males and WZ females. The
WW super-females are of special interest with regard to possible genetic sex regulation, since crossing them with normal
ZZ males should produce all-female progeny (WZ). Crossing
of WW super-females with phenotypic WW males obtained by
sex reversal will allow for production of WW super-females in
mass quantities.
The existence of female heterogamety in largemouth bass
should be confirmed later by identification of WW fish in test
crosses.
ACKNOWLEDGMENTS
REFERENCES
Al-Ablani, S. A., and R. P. Phelps. 2001. Induction of feminization in largemouth
bass (Micropterus salmoides) by oral administration of estradiol-17 beta or
diethylstilbestrol and associated pathological effects. Journal of Aquaculture
in the Tropics 16:185195.
Arslan, T., R. P. Phelps, and J. A. Osborne. 2009. Effects of oestradiol17 or 17-methyltestosterone administration on gonadal differentiation of
largemouth bass Micropterus salmoides (Lacepède). Aquaculture Research
40:18131822.
539
Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Determination of sex and maturational status of striped bass (Morone saxatilis)
using ultrasonic imaging. Aquaculture 125:175184.
Castelli, M. 1994. Study on sex determination in the common barbel (Barbus
barbus L.) (Pisces, Cyprinidae) using gynogenesis. Pages 509519 in A. R.
Beaumont, editor. Genetics and evolution of aquatic organisms. Chapman
and Hall, London.
Chourrout, D. 1987. Genetic manipulations in fish: review of methods. Pages
111126 in K. Tiews, editor. Selection, hybridization, and genetic engineering
in aquaculture, volume 2. Heenemann Verlags, Berlin.
Colombo, R. E., P. S. Wills, and J. E. Garvey. 2004. Use of ultrasound imaging
to determine sex of shovelnose sturgeon. North American Journal of Fisheries
Dabrowski, K., J. Rinchard, F. Lin, M. A. Garcia-Abiado, and D. Schmidt.
2000. Induction of gynogenesis in muskellunge with irradiated sperm of
yellow perch proves diploid muskellunge male homogamety. Journal of Experimental Zoology 287:96105.
Devlin, R. H., and Y. Nagahama. 2002. Sex determination and sex differentiation
in fish: an overview of genetic, physiological, and environmental influences.
Donaldson, E. M. 1996. Manipulation of reproduction in farmed fish. Animal
Reproduction Science 42:381392.
Flynn, S. R., M. Matsuoka, M. Reith, D. J. Martin-Robichaud, and T. J. Benfey.
2006. Gynogenesis and sex determination in shortnose sturgeon, Acipencer
brevirostrum Lesuere. Aquaculture 253:721727.
Fries, L. T., G. L. Kurten, J. Isaac Jr., T. Engelhardt, D. Lyon, and D. G.
Smith. 2002. Mass production of polyploid Florida largemouth bass for
stocking public waters in Texas. Pages 393399 in D. P. Philipp and M. S.
Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society, Symposium 31, Bethesda, Maryland.
Garrett, G. P. 1989. Hormonal sex control of largemouth bass. Progressive
Fish-Culturist 51:146148.
Garrett, G. P., M. C. F. Birkner, and J. R. Gold. 1992. Triploidy induction in
largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture
1(3):2734.
Gomelsky, B., S. D. Mims, R. J. Onders, W. L. Shelton, K. Dabrowski, and
M. A. Garcia-Abiado. 2000. Induced gynogenesis in black crappie. North
Martin, R. W., J. Myers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983.
Ultrasonic imaging, a potential tool for sex determination of live fish. North
Masoudifard, M., A. R. Vajhi, M. Moghim, R. M. Nazari, A. R. Naghavi,
and M. Sohrabnejad. 2011. High validity sex determination of three years
old cultured beluga sturgeon (Huso huso) using ultrasonography. Journal of
Applied Ichthyology 27:643647.
Mims, S. D., W. L. Shelton, O. Linhart, and C. Wang. 1997. Induced meiotic
gynogenesis of paddlefish Polyodon spathula. Journal of the World Aquaculture Society 28:334343.
Neal, J. W., D. M. Neal, R. L. Noble, and M. V. McGee. 2004. Artificial propagation and induction of triploidy in largemouth bass Micropterus salmoides
and ploidy discrimination using erythrocyte length. Journal of the World
Padfield, J. H., Jr. 1951. Age and growth differentiation between the sexes of
the largemouth black bass, Micropterous salmoides (Lacepede). Journal of
the Tennessee Academy of Science 26:4254.
Penman, D. J., M. S. Shah, J. A. Beardmore, and D. O. F. Skibinski. 1987. Sex
ratios of gynogenetic and triploid tilapia. Pages 267276 in K. Tiews, editor.
Selection, hybridization, and genetic engineering in aquaculture, volume 2.
Heenemann Verlags, Berlin.
Porter, M. D. 1996. Effects of methyltestosterone on largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 6(4):3945.
Purdom, C. E. 1993. Genetics and fish breeding. Chapman and Hall, London.
Purdom, C. E., and R. F. Lincoln. 1973. Chromosome manipulation in fish.
Pages 8389 in J. H. Schroder, editor. Genetics and mutagenesis of fish.
Springer-Verlag, New York.
540
GLENNON ET AL.
Reimers, E., P. Landmark, T. Sorsdal, E. Bohmer, and T. Solum. 1987.

Determination of salmonids sex, maturation and size: an ultrasound
and photocell approach. Aquaculture Magazine 13(November/December):
4144.
Shields, R. J., J. Davenport, C. Young, and P. L. Smith. 1993. Oocyte maturation and ovulation in the Atlantic halibut, Hippoglossus hippoglossus (L.),
examined using ultrasonography. Aquaculture Research 24:181186.
Thompson, D. 1983. The efficiency of induced diploid gynogenesis in inbreeding. Aquaculture 33:237244.
Thorgaard, G. H. 1983. Chromosome set manipulation and sex control in fish.

Pages 405434 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, editors.
Fish physiology, volume 9B. Academic Press, New York.
Thorgaard, G. H., F. W. Allendorf, and K. L. Knudsen. 1983. Gene-centromere
mapping in rainbow trout: high interference over long map distances. Genetics
103:771783.
Van Eenennaam, A. L., J. P. Van Eenennaam, J. F. Medrano, and S. I. Doroshov.
1999. Evidence of female heterogametic genetic sex determination in white
sturgeon. Journal of Heredity 90:231233.



a
Robert P. Glennon , Boris Gomelsky , Kyle J. Schneider , Anita M. Kelly & Alf Haukenes
c
a
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas, 72086, USA
c
Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine
Bluff, Arkansas, 71601, USA
To cite this article: Robert P. Glennon, Boris Gomelsky, Kyle J. Schneider, Anita M. Kelly & Alf Haukenes (2012): Evidence
of Female Heterogamety in Largemouth Bass, Based on Sex Ratio of Gynogenetic Progeny, North American Journal of


DOI: 10.1080/15222055.2012.700906
COMMUNICATION

Robert P. Glennon
J. M. Malone and Son, Incorporated, 1156 Malone Lake Road, Lonoke, Arkansas 72086, USA

Kentucky 40601, USA
Anita M. Kelly and Alf Haukenes

Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, Mail Slot 4912, Pine Bluff,
Arkansas 71601, USA
Abstract
Meiotic gynogenetic progeny in largemouth bass Micropterus
salmoides have been obtained by inseminating largemouth bass
eggs with UV-irradiated sperm from white bass Morone chrysops
or striped bass Morone saxatilis and suppressing the second meiotic division by hydrostatic pressure. The sex composition of gynogenetic progeny was determined by dissection or ultrasound investigation of 1-year-old fish. Among the 21 fish analyzed, 7 fish
(33.3%) were male and 14 fish (66.7%) were female. The presence
of males in meiotic gynogenetic progeny suggests the existence of
female heterogamety (WZ females, ZZ males) in largemouth bass.
The largemouth bass Micropterus salmoides is the most popular game fish in the United States. Females of this species
exhibit faster growth and attain larger size than do the males
(Padfield 1951). Manipulating the sex ratio of largemouth bass
populations is of great interest because of its potential to produce higher ratios of females or monosex female populations
for the purposes of creating larger and faster growing fish for
aquaculture and trophy sport fishing. Several studies have evaluated hormonal sex reversal with the goal of shifting sex ratios under influence of androgens or estrogens in largemouth
bass (Garrett 1989; Porter 1996; Al-Ablani and Phelps 2001;
Arslan et al. 2009). Induced triploidy has also been tested in
this species for reproduction control and obtaining potentially
larger fish (Garrett et al. 1992; Fries et al. 2002; Neal et al.
2004).
The optimal method for production of monosex fish populations involves crossing previously obtained sex-reversed fish
with normal breeding fish (Purdom 1993; Donaldson 1996). The
scheme of crosses directed to production of monosex progenies
depends on the type of heterogamety in given fish species. In
fishes, both male (XY males, XX females) and female (WZ
females, ZZ males) heterogamety are described; sometimes different types of heterogamety are revealed in closely related
species (Dabrowski et al. 2000; Devlin and Nagahama 2002).
Until now, there were no data on sex determination mechanism
in largemouth bass.
One of the basic methods for determining the type of heterogamety in fish is based on sex composition of gynogenetic
progenies. In the case of male heterogamety (XY/XX), gynogenetic progenies usually consist of females only; the presence of both females and males indicates female heterogamety
(WZ/ZZ; Thorgaard 1983; Van Eenennaam et al. 1999; Devlin
and Nagahama 2002). This article presents data on sex composition of a gynogenetic progeny in largemouth bass that suggests
the existence of female heterogamety in this species.
METHODS
Experiments on production of gynogenetic progeny in largemouth bass were conducted at the facilities of J. M. Malone and
Son, Inc. (Lonoke, Arkansas) in April 2009. To induce gynogenetic development, largemouth bass eggs were inseminated

Received January 19, 2012; accepted June 4, 2012
537
538
GLENNON ET AL.
with UV-irradiated sperm from white bass Morone chrysops or

striped bass Morone saxatilis. For production of gynogenetic
diploids, the second meiotic division in eggs was suppressed by
application of hydrostatic pressure.
Ovulation was induced in mature largemouth bass females
by a single intramuscular injection of human chorionic gonadotropin (HCG, 4,000 IU/kg), after which they were placed
into separate 246-L fiberglass tanks. The time between hormonal injection and ovation was 3640 h. White bass and striped
bass males were each given one intramuscular HCG injection
(1,000 IU/kg). The sperm obtained was irradiated based on techniques previously used for white bass sperm irradiation in experiments on induced gynogenesis in black crappie Pomoxis
nigromaculatus (Gomelsky et al. 2000). Specifically, sperm
was irradiated at a dose of 1,000 J/m2 with a FisherBiotech
UV microprocessor-controlled Crosslinker (FB-UVXL-1000;
Fisher Scientific). For irradiation, 2 mL of sperm diluted 10fold with 0.85% NaCl solution was placed in 6-cm-diameter
glass Petri dishes. Parameters of pressure shock for suppression
of the second meiotic division in largemouth bass were chosen
based on published data from experiments on induced triploidy
(Garrett et al. 1992; Neal et al. 2004); pressure shock of 8,000
psi for 1 min was initiated 5 min after the eggs were inseminated
with irradiated sperm. The fertilized eggs were placed into a 1L hydrostatic pressure chamber and the desired pressure was
achieved using a 20 ton hydraulic jack.
In total, ovulated eggs were produced from 12 females. Nine
trials on insemination of eggs with irradiated sperm and application of hydrostatic pressure were performed; in some trials
eggs from several females were mixed. After pressure treatment,
eggs were put into separate 246-L fiberglass tanks for embryo
incubation and hatching of larvae. The resulting swim-up gynogenetic larvae were reared in the same tanks in which embryos
had been incubated and fed with live zooplankton supplied from
filtered pond water for 30 days after hatching.
Two hundred and sixty gynogens survived to the fry stage.
Sixty gynogenetic fish were randomly selected and reared in
1 m3 tanks at a separate facility on live foods (minnow and
bluegill) for 1 year. The remaining 200 gynogenic fish were
moved to a different facility for hormonal sex reversal by androgen treatment; the results of that study will be reported in a
separate publication.
Twenty-one gynogenetic fish from the untreated group survived to the age of one year. On April 19, 2010, 15 fish were
lost due to equipment failure; they were dissected for sex determination by macroscopic observation of gonads and their total
length was recorded. The six surviving gynogenetic fish from
this group were sexed 3 days later by ultrasound investigation
using Tela-Vet 1000 equipped with a 58 MHz linear transducer
(Classic Medical, Tequesta, Florida). Ultrasound is a noninvasive method that has been shown to accurately determine the
sex in several species of fish (Martin et al. 1983; Reimers et al.
1987; Shields et al. 1993; Blythe et al. 1994; Colombo et al.
2004; Masoudifard et al. 2011). The fish were anesthetized with
quinaldine sulfate and placed in a pan of water. Each fish was

held submerged in the water against a styrofoam pad to prevent
double imaging while the ultrasound was performed. The probe
was placed near the abdomen to determine the sex of each fish.
Female fish were identified by the presence of eggs in the ovarian sacs; males had no eggs but showed the presence of testes
structures.
RESULTS
From 15 dissected fish, 7 fish were male and 8 fish were female. Males and females had well-developed maturing testes
and ovaries, respectively, with normal gonad morphological
structure and color for this species. Mean lengths (SD) of
males and females were 27.09 ( 1.17) cm and 25.44 ( 0.95)
cm, respectively.
The six remaining 1-year-old fish were identified as females
by ultrasound investigation. In the summer of 2011, at 2 years
of age, further observations of these fish confirmed their identification as females; when pressed on the abdomen, no sperm
was expressed and other traits typical for females were observed
(e.g., swollen abdomen).
In total, of the 21 gynogenetic fish analyzed, 7 (33.3%) were
male and 14 (66.7%) were female.
DISCUSSION
The use of heterologous sperm, instead of sperm from the
same species, is a very effective method for experiments on
diploid gynogenesis in fish. If hybrids between two species
are nonviable, survivors are exclusively of gynogenetic origin
(Chourrout 1987; Mims et al. 1997; Dabrowski et al. 2000;
Gomelsky et al. 2000). Preliminary experiments conducted at
the Aquaculture Research Center of Kentucky State University
in 2008 have shown that white bass spermatozoa were capable to
fertilizing largemouth bass eggs, but the hybrids obtained were
nonviable and perished before or soon after hatching. Since the
present study used only irradiated sperm of Morone species for
induction of gynogenetic development in largemouth bass eggs,
all obtained larvae could be only of gynogenetic origin. In an
earlier study, irradiated sperm of white bass was used for induced
gynogenesis in another species of the family Centrarchidae,
black crappie (Gomelsky et al. 2000).
Usually the presence of males in meiotic gynogenetic progenies suggests the existence of female heterogamety (WZ/ZZ)
in given species. Other possible factors such as autosomal and
environmental influences on sex determination can also result in
appearance of gynogenetic males (Devlin and Nagahama 2002;
Flynn et al. 2006). Males in meiotic gynogenetic progenies
were described in plaice Pleuronectes platessa (Purdom and
Lincoln 1973), blue tilapia Oreochromis aureus (Penman et al.
1987), common barbel Barbus barbus (Castelli 1994), muskellunge Esox masquinongy (Dabrowski et al. 2000), white sturgeon Acipenser transmontanus (Van Eenennaam et al. 1999),
shortnose sturgeon A. brevirostrum (Flynn et al. 2006), and
COMMUNICATION
some other fish. The frequency of heterozygotes in a meiotic gynogenetic progeny obtained from a female that is heterozygous for some gene depends on the frequency of crossing over between the gene and centromere during prophase
of the first meiotic division; the proportion of heterozygotes
increases with increasing crossing over frequency (Thompson
1983; Thorgaard et al. 1983). Similarly, when meiotic gynogenetic progeny is obtained from heterogametic female (WZ),
the resulting sex ratio depends on the frequency of crossing
over between a sex-determining gene and centromere. If WW
fish are viable, then in the absence of crossing over, the meiotic gynogenetic progeny obtained from WZ female will consist of WW females (so called super-females) and ZZ males
with ratio 1:1. Crossing over between the sex-determining gene
and the centromere will result in the appearance of WZ females and, with increasing recombination rates, the phenotypic sex ratio will be shifted towards prevalence of females.
If all three categories are viable, the proportion of ZZ males
should be approximately equal to the proportion of WW superfemales (Van Eenennaam et al. 1999; Devlin and Nagahama
2002).
One third (33%) of the meiotic gynogenetic progeny of largemouth bass were males. This is similar to the proportion of males
in meiotic gynogenetic progenies observed in plaice (37%;
Purdom and Lincoln 1973), muskellunge (40%; Dabrowski et al.
2000), and shortnose sturgeon (35%; Flynn et al. 2006). If
WW super-females in largemouth bass are viable, their proportion in gynogenetic progeny should be approximately 33%,
the same as proportions of ZZ males and WZ females. The
WW super-females are of special interest with regard to possible genetic sex regulation, since crossing them with normal
ZZ males should produce all-female progeny (WZ). Crossing
of WW super-females with phenotypic WW males obtained by
sex reversal will allow for production of WW super-females in
mass quantities.
The existence of female heterogamety in largemouth bass
should be confirmed later by identification of WW fish in test
crosses.
ACKNOWLEDGMENTS
REFERENCES
Al-Ablani, S. A., and R. P. Phelps. 2001. Induction of feminization in largemouth
bass (Micropterus salmoides) by oral administration of estradiol-17 beta or
diethylstilbestrol and associated pathological effects. Journal of Aquaculture
in the Tropics 16:185195.
Arslan, T., R. P. Phelps, and J. A. Osborne. 2009. Effects of oestradiol17 or 17-methyltestosterone administration on gonadal differentiation of
largemouth bass Micropterus salmoides (Lacepède). Aquaculture Research
40:18131822.
539
Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Determination of sex and maturational status of striped bass (Morone saxatilis)
using ultrasonic imaging. Aquaculture 125:175184.
Castelli, M. 1994. Study on sex determination in the common barbel (Barbus
barbus L.) (Pisces, Cyprinidae) using gynogenesis. Pages 509519 in A. R.
Beaumont, editor. Genetics and evolution of aquatic organisms. Chapman
and Hall, London.
Chourrout, D. 1987. Genetic manipulations in fish: review of methods. Pages
111126 in K. Tiews, editor. Selection, hybridization, and genetic engineering
in aquaculture, volume 2. Heenemann Verlags, Berlin.
Colombo, R. E., P. S. Wills, and J. E. Garvey. 2004. Use of ultrasound imaging
to determine sex of shovelnose sturgeon. North American Journal of Fisheries
Dabrowski, K., J. Rinchard, F. Lin, M. A. Garcia-Abiado, and D. Schmidt.
2000. Induction of gynogenesis in muskellunge with irradiated sperm of
yellow perch proves diploid muskellunge male homogamety. Journal of Experimental Zoology 287:96105.
Devlin, R. H., and Y. Nagahama. 2002. Sex determination and sex differentiation
in fish: an overview of genetic, physiological, and environmental influences.
Donaldson, E. M. 1996. Manipulation of reproduction in farmed fish. Animal
Reproduction Science 42:381392.
Flynn, S. R., M. Matsuoka, M. Reith, D. J. Martin-Robichaud, and T. J. Benfey.
2006. Gynogenesis and sex determination in shortnose sturgeon, Acipencer
brevirostrum Lesuere. Aquaculture 253:721727.
Fries, L. T., G. L. Kurten, J. Isaac Jr., T. Engelhardt, D. Lyon, and D. G.
Smith. 2002. Mass production of polyploid Florida largemouth bass for
stocking public waters in Texas. Pages 393399 in D. P. Philipp and M. S.
Ridgway, editors. Black bass: ecology, conservation, and management. American Fisheries Society, Symposium 31, Bethesda, Maryland.
Garrett, G. P. 1989. Hormonal sex control of largemouth bass. Progressive
Garrett, G. P., M. C. F. Birkner, and J. R. Gold. 1992. Triploidy induction in
largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture
1(3):2734.
Gomelsky, B., S. D. Mims, R. J. Onders, W. L. Shelton, K. Dabrowski, and
M. A. Garcia-Abiado. 2000. Induced gynogenesis in black crappie. North
Martin, R. W., J. Myers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983.
Ultrasonic imaging, a potential tool for sex determination of live fish. North
Masoudifard, M., A. R. Vajhi, M. Moghim, R. M. Nazari, A. R. Naghavi,
and M. Sohrabnejad. 2011. High validity sex determination of three years
old cultured beluga sturgeon (Huso huso) using ultrasonography. Journal of
Applied Ichthyology 27:643647.
Mims, S. D., W. L. Shelton, O. Linhart, and C. Wang. 1997. Induced meiotic
gynogenesis of paddlefish Polyodon spathula. Journal of the World Aquaculture Society 28:334343.
Neal, J. W., D. M. Neal, R. L. Noble, and M. V. McGee. 2004. Artificial propagation and induction of triploidy in largemouth bass Micropterus salmoides
and ploidy discrimination using erythrocyte length. Journal of the World
Padfield, J. H., Jr. 1951. Age and growth differentiation between the sexes of
the largemouth black bass, Micropterous salmoides (Lacepede). Journal of
the Tennessee Academy of Science 26:4254.
Penman, D. J., M. S. Shah, J. A. Beardmore, and D. O. F. Skibinski. 1987. Sex
ratios of gynogenetic and triploid tilapia. Pages 267276 in K. Tiews, editor.
Selection, hybridization, and genetic engineering in aquaculture, volume 2.
Heenemann Verlags, Berlin.
Porter, M. D. 1996. Effects of methyltestosterone on largemouth bass, Micropterous salmoides. Journal of Applied Aquaculture 6(4):3945.
Purdom, C. E. 1993. Genetics and fish breeding. Chapman and Hall, London.
Purdom, C. E., and R. F. Lincoln. 1973. Chromosome manipulation in fish.
Pages 8389 in J. H. Schroder, editor. Genetics and mutagenesis of fish.
Springer-Verlag, New York.
540
GLENNON ET AL.
Reimers, E., P. Landmark, T. Sorsdal, E. Bohmer, and T. Solum. 1987.

Determination of salmonids sex, maturation and size: an ultrasound
and photocell approach. Aquaculture Magazine 13(November/December):
4144.
Shields, R. J., J. Davenport, C. Young, and P. L. Smith. 1993. Oocyte maturation and ovulation in the Atlantic halibut, Hippoglossus hippoglossus (L.),
examined using ultrasonography. Aquaculture Research 24:181186.
Thompson, D. 1983. The efficiency of induced diploid gynogenesis in inbreeding. Aquaculture 33:237244.
Thorgaard, G. H. 1983. Chromosome set manipulation and sex control in fish.

Pages 405434 in W. S. Hoar, D. J. Randall, and E. M. Donaldson, editors.
Fish physiology, volume 9B. Academic Press, New York.
Thorgaard, G. H., F. W. Allendorf, and K. L. Knudsen. 1983. Gene-centromere
mapping in rainbow trout: high interference over long map distances. Genetics
103:771783.
Van Eenennaam, A. L., J. P. Van Eenennaam, J. F. Medrano, and S. I. Doroshov.
1999. Evidence of female heterogametic genetic sex determination in white
sturgeon. Journal of Heredity 90:231233.


Effects of Two Anesthetics on Survival of Juvenile

Culter mongolicus during a Simulated Transport
Experiment
Mingli Lin
a
a b
, Qidong Wang
a
a b
, Yuguo Xia
, Tanglin Zhang & Shaowen Ye
a b
, Brian R. Murphy , Zhongjie Li , Jiashou Liu
State Key Laboratory of Freshwater Ecology and Biotechnology, Institute of Hydrobiology,

Chinese Academy of Sciences, Wuhan, 430072, China
b
Graduate University of Chinese Academy of Sciences, Beijing, 100039, China
Department of Fish and Wildlife Conservation and Conservation Management Institute,

Virginia Polytechnic Institute and State University, Blacksburg, Virginia, 24061, USA
To cite this article: Mingli Lin, Qidong Wang, Yuguo Xia, Brian R. Murphy, Zhongjie Li, Jiashou Liu, Tanglin Zhang & Shaowen
Ye (2012): Effects of Two Anesthetics on Survival of Juvenile Culter mongolicus during a Simulated Transport Experiment,
North American Journal of Aquaculture, 74:4, 541-546


DOI: 10.1080/15222055.2012.700905
ARTICLE
Effects of Two Anesthetics on Survival of Juvenile Culter

mongolicus during a Simulated Transport Experiment
Mingli Lin, Qidong Wang, and Yuguo Xia
Chinese Academy of Sciences, Wuhan 430072, China;
and Graduate University of Chinese Academy of Sciences, Beijing 100039, China
Brian R. Murphy
Department of Fish and Wildlife Conservation and Conservation Management Institute,
Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061, USA
Zhongjie Li, Jiashou Liu, Tanglin Zhang, and Shaowen Ye*

Chinese Academy of Sciences, Wuhan 430072, China
Abstract
Cultivation of the redtail culter Culter mongolicus has been increasing substantially over the last decade along the
Yangtze River basin; such increases in production lead to increased juvenile transportation. However, redtail culter
juveniles have high transport mortality rates due to a strong stress response that is exacerbated by the accumulation
of toxic metabolic waste. Through a 24-h simulated transport experiment (sampling every 6 h), we assessed effects of
tricaine methanesulfonate (MS-222) at 10 mg/L of water, 20 mg/L, and 40 mg/L on redtail culter survival and water
quality parameters, and similarly we assessed clove oil at 2 mg/L of water, 5 mg/L, and 10 mg/L. None of the anesthetics
significantly improved water quality during the initial 612 h of the experiment. However, MS-222 treatments at the
first 1224 h of the experiment had significantly higher dissolved oxygen (DO), ammonia, and pH than the control
but failed to decrease un-ionized ammonia content. In contrast, the clove oil treatment significantly reduced the
un-ionized ammonia but failed to improve DO and pH at 1224 h. The improvements in water quality were reflected
in cumulative mortality, MS-222 and clove oil anesthetic treatments having significantly lower cumulative mortality
than the control at 1224 h. The MS-222 and clove oil slowed water quality deterioration, ensured a better transport
environment, and improved juvenile survival during transportation. We recommend 5 mg/L clove oil be used when
transporting juvenile redtail culters because that concentration improves fish survival while keeping cost low.
Transport of juvenile fishes is an important procedure in

aquaculture operations and fisheries management. The transportation of juveniles in closed polyethylene bags charged with
oxygen is considered a simple and effective method, particularly for air travel, and this practice is widespread throughout
the world (Berka 1986; Lim et al. 2003). However, water quality
generally deteriorates in this closed system due to fish respiration and excretion, pH and dissolved oxygen decreasing and
ammonia and un-ionized ammonia increasing (Guo et al. 1995;

Pramod et al. 2010). Deterioration of water quality stresses
the nervous, immunological, and hormonal systems of juvenile
fishes, and it could cause mortality if it deteriorates beyond normal tolerance limits of fish (Urbinati et al. 2004; Harmon 2009).
Additionally, juveniles may be physically damaged during handling and transport procedures (Murai et al. 1979). Therefore,
juvenile fish transport in polyethylene bags usually has high
*Corresponding author: yeshw@ihb.ac.cn

Received December 21, 2011; accepted June 4, 2012
541
542
LIN ET AL.
mortality, especially during long-distance or high-density transportation (Teo et al. 1989; Gomes et al. 2003).
Anesthetics are widely used before and during transport
to slow metabolism and reduce stress of fish (Harmon 2009;
Pramod et al. 2010). Anesthetics reduce the contractile force
of the ventricular myocardium and alter gill haemodynamics
(Hill et al. 2002) by depressing neuronal activity (Spath and
Schweickert 1977; Arnolds et al. 2002) and preventing plasma
cortisol elevation (Iversen et al. 2003). These neural and physiological variations reduce fish activity and metabolic rate, thus
decreasing oxygen uptake and carbon dioxide and ammonia production (Zhuang et al. 2009). Therefore, water quality does not
decline as rapidly during fish transport when some anesthetics
are added (Berka 1986; Park et al. 2009).
When anesthetics are used at proper doses, physical damage may also be reduced because fish are calmer and less active. Consequently, anesthetics generally reduce fish mortality during transportation (Estudillo and Duray 2003; Pramod
et al. 2010). Somewhat paradoxically, anesthetics have also been
shown to cause an acute stress response immediately after sedation (Trushenski et al. 2012; Zahl et al. 2012). During this
acute response, the stress hormone cortisol may increase, leading to increased levels of glucose and lactate in the blood after
sedation. Compared with the levels of these chemicals that occur after a fish is stressed by handling, the amounts released in
response to anesthesia are low (Zahl et al. 2012). Levels of cortisol, glucose, and lactate in the blood return to normal within
6 h after sedation (Trushenski et al. 2012). A number of anesthetics, including tricaine methanesulfonate (MS-222), benzocaine, 2-phenoxyethanol, quinaldine sulphate, metomidate, and
lidocaine have been used in juvenile fish transportation (Guo
et al. 1995; Park et al. 2009; Pramod et al. 2010). Among the
broad spectrum of anesthetics, MS-222 appears to be used most
frequently (Marking and Meyer 1985; Berka 1986). Clove oil
(major constituent eugenol: 2-methoxy-4-[2-propenyl] phenol),
is another common fish anesthetic due to its efficacy, affordability, and short withdrawal period (Harper 2003). Although
many studies have assessed the effects of anesthetics on ornamental fish transported in polyethylene bags (Teo et al. 1989;
Pramod et al. 2010) and on low-density transportation of cultivated juvenile fish (Guo et al. 1995), the effects of anesthetics on
juvenile survival and water quality during high-density transport
in polyethylene bags are not well studied.
Redtail culter (or Mongolian culter) Culter mongolicus is an
important piscivorous fish in China, widely distributed in lakes,
rivers, reservoirs, and other freshwater bodies (CAS 1976). The
species is characterized by high price and high market potential
(Zhang 2008). However, its population has declined seriously
during recent decades, mainly due to overfishing, habitat destruction, and water pollution (Zhang 2005; Ye 2007). To rebuild
the population, Chinas Ministry of Agriculture initiated stock
enhancement programs in some lakes of the Yangtze River basin.
Furthermore, pond and cage farming of redtail culters has developed rapidly, increasing the need for juvenile transportation.
However, high mortality has been observed during transport

for redtail culters (Xu et al. 2009). Our objective in this study
was to evaluate the effects of MS-222 and clove oil on survival
of redtail culter juveniles during transportation in polyethylene
bags. Furthermore, we monitored water quality parameters to
help define important mechanisms of anesthetic action.
METHODS
Experimental fish.We obtained juvenile redtail culters
from Niushan Lake Fish Breeding Center, Wuhan, Hubei
Province, China. Juveniles were stocked in a cultivation pond
for 1 month prior to the experiment. They were fed a commercial
pelleted diet in the pond, but feeding was terminated 1 d before
the experiment. Juveniles were 41 d old when the experiment
began (mean SE: total length = 46.09 0.91 mm, body
weight = 0.75 0.04 g).
Anesthetics.Two commercial anesthetics, MS-222 (Sigma
Chemicals, St. Louis, Missouri) and chemically pure clove oil
(Shanghai Feixiang Chemical Factory, Shanghai, China), were
used to lightly sedate the fish. Each anesthetic was tested at
three concentrations based on previous experience, MS-222 at
10, 20, and 40 mg/L of water and clove oil at 2, 5, and 10 mg/L.
Sodium bicarbonate was used in MS-222 solutions to adjust pH
to 7.0.
Experimental setup.The experiment was conducted on 31
July 2010. Juveniles scoop-netted after being seined from the
pond were batch weighted and transferred to bags. Eighty-four
bags used in this experiment (7 treatments 4 time samples
3 replicates). The clear plastic bags were 20 L (40 cm wide,
63 cm high) and were sealed to be airtight after adding 5 L of
water,15 L oxygen, the fish, and their anesthetic treatment dose.
Transport density was 50 g of fish/bag (average, 67 fish/bag) or
10 g of fish per 1 L of water (13 fish/L of water). This density was
based on actual transport practices and a previous experiment
to measure oxygen consumption of juveniles.
All the bags were put in a vehicle and transported for approximately 1 h to the laboratory of the Institute of Hydrobiology,
Chinese Academy of Sciences. Natural light was maintained
in the laboratory, and an air conditioner was used to maintain
the air temperature at 25 C. The experiment lasted for 24 h,
and three bags from each treatment were examined at each 6-h
interval to assess water quality.
Transport water came from the cultivation pond: temperature = 29 C, dissolved oxygen (DO) = 6 mg/L, pH = 7.8,
conductivity = 325 S/cm, ammonia = 0.47 mg/L). During the
experiment, DO (YSI Model 85 instrument, YSI Inc., Beijing,
China) and pH (YSI pH100) were measured immediately after
the bags were opened, and total ammonia was titrated within
6 h after the bags were opened. Total ammonia content was
determined by the Nessler reagent spectrophotometric method
(Huang et al. 1999). Concentration of un-ionized ammonia was
calculated by multiplying the total ammonia by the appropriate
conversion factor according to the measured water temperature
543
EFFECTS OF ANESTHETICS IN TRANSPORT
7.8
Control
MS-10
MS-20
MS-40
CO-2
CO-5
CO-10
7.6
pH value
and pH (Emerson et al. 1975). Dead fish were counted and

weighed at each sampling period (6, 12, 18, and 24 h).
Statistical analysis.Cumulative mortality (CM) was calculated as CMi = (WDi /WTi )100%, where CMi was cumulative mortality, WDi was weight of dead individuals, and
WTi was total weight of juveniles, and i was the ith hour.
To determine significant differences between treatments, water
quality variables and fish mortality were analyzed using oneway ANOVA, followed by least-significant difference (LSD)
multiple-comparisons analysis. Statistical Package for the Social Sciences (SPSS) Version 15.0 was used as statistical software, and the significance level for statistical analyses was set
at = 0.05.
7.4
a
a
ab
7.2
7.0
6.8
6
12
18
24
Time(hours)
FIGURE 2. Changes of pH values in water used in simulated transportation

of redtail culter juveniles. The water was treated with one of three MS-222
treatments or one of three clove oil treatments (the treatment abbreviations are
explained in Figure 1). At each time point, values accompanied by different
letters are significantly different.
(P < 0.05). However, the results for total ammonia differed from
the results for un-ionized ammonia. The control group had lower
un-ionized ammonia than the two anesthetic treatments at 6 h
and 12 h (Figure 4), but un-ionized ammonia in the control group
increased quickly after 12 h, exceeding the un-ionized ammonia
levels for all treatment groups at 24 h. Both MS-222 and clove
oil treatments displayed no significant differences in un-ionized
ammonia concentration during the first 12 h (P > 0.05), but
significantly lower concentrations of un-ionized ammonia were
observed in clove oil treatments at 18 h and 24 h (P < 0.05)
than in the MS-222 treatments.
20
14
a
a
ad
ab
15
bcd
bc
c
a
10
Control
CO-2
CO-5
CO-10
MS-10
MS-20
MS-40
Control
CO-2
CO-5
CO-10
MS-10
MS-20
MS-40
12
ab
ab
abc
bc
bc
Ammonia(NH -N, mg/L)
Dissolved oxygen (mg/L)
b
bc
bc
c
RESULTS
The DO level remained high in all treatments through the
first 12 h of the experiment (Figure 1), but DO at 18 h and 24 h
was significantly higher in MS-222 treatments than in both the
control and clove oil treatments (P < 0.05). The pH declined
sharply during the first 6 h of the experiment, and no significant differences were observed between control and anesthetic
treatments. However, pH of MS-222 treatments declined significantly more slowly than in the control and clove oil treatments
after 12 h (P < 0.05; Figure 2).
Total ammonia increased rapidly during the experiment. In
the control group, total ammonia increased from 0.47 mg/L at
the beginning of experiment to 11.01 mg/L (SE, 0.48) at the
end (Figure 3). There were no significant differences in total
ammonia between control and anesthetic treatments before the
initial 12 h, but concentration of ammonia in the control was
significantly higher than in anesthetic treatments at 18 h and 24 h
10
ab
ab
ab
ab
b
bc
b
b
c
a
2
c
0
0
12
18
24
Time(hours)
0
0
12
18
24
Time(hours)
FIGURE 1. Changes in dissolved oxygen concentration in water used in simulated transportation of redtail culter juveniles. The water was treated with one
of three MS-222 treatments (10 mg/L [MS-10], 20 mg/L [MS-20], or 40 mg/L
[MS-40]) or one of three clove oil treatments (2 mg/L [CO-2], 5 mg/L [CO-5], or
10 mg/L [CO-10]). At each time point, values accompanied by different letters
are significantly different.
FIGURE 3. Changes in total ammonia in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three MS-222
treatments or one of three clove oil treatments (the treatment abbreviations are
explained in Figure 1). At each time point, values accompanied by different
letters are significantly different.
544
LIN ET AL.
Un-ionized Ammonia(NH , mg/L)
.06
Control
MS-10
MS-20
MS-40
CO-2
CO-5
CO-10
.05
.04
a
a
a
ab
a
a
ab
bc
bc
bc
c
ab
ab
.03
d
d
b
.02
b
b
.01
0
12
18
24
Time(hours)
the control group at 45.5% (SE, 10.8) was significantly higher

than mortality in all anesthetics treatments (P < 0.05) except
clove oil at 10 mg/L.
FIGURE 4. Changes in un-ionized ammonia in water used in simulated transportation of redtail culter juveniles. The water was treated with one of three
MS-222 treatments or one of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by
different letters are significantly different.
Mortality of juvenile redtail culters occurred throughout the

experiment, and the highest mortality rate was observed during
the first 6 h (Figure 5). At 12 h, mortality did not increase significantly, and no significant differences were observed among
treatments (P > 0.05). At 18 h, cumulative mortality of the control group was significantly higher than mortality in all anesthetic treatments (P < 0.05). At 24 h, cumulative mortality in
DISCUSSION
Our research showed that total ammonia and un-ionized
ammonia concentrations increased during simulated transport
and pH and DO decreased. Similar results have been observed
in other species, including the southern platyfish Xiphophorus
maculatus, and Indian tiger barb Puntius filamentosus (Amend
et al. 1982; Guo et al. 1995; Pramod et al. 2010). As fish respire,
they consume oxygen consume and accumulate waste products.
Fish excrete carbon dioxide and consume DO through respiration, which decreases water acidity and DO (Harmon 2009). The
major pathways for ammonia production are transamination and
deamination of adenylates in fish muscle and gill tissue (Randall
and Wright 1987). Un-ionized ammonia and ionized ammonia
are at a dynamic equilibrium in water. Un-ionized ammonia
content increases when pH, temperature, and ionized ammonia
concentration increase (Emerson et al. 1975). Water temperature
was constant and pH decreased during the experiment. Therefore, the increase in un-ionized ammonia we observed in the
control group was due to sharp increases of total ammonia.
None of the anesthetics significantly improved water quality parameters during the 612-h period. However, all levels of
anesthetic treatments benefited water quality during the 1224h period, though MS-222 and clove oil had different mechanisms and different impacts on water variables. The MS-222
FIGURE 5. Cumulative mortality of redtail culter juveniles during simulated transportation. The water was treated with one of three MS-222 treatments or one
of three clove oil treatments (the treatment abbreviations are explained in Figure 1). At each time point, values accompanied by different letters are significantly
different.
EFFECTS OF ANESTHETICS IN TRANSPORT
treatments had significantly higher DO, ammonia, and pH than

the controls during 1224 h, but failed to decrease un-ionized
ammonia content. In contrast, the clove oil treatment significantly reduced the un-ionized ammonia but failed to improve
DO and pH during 1224 h. These results have also been observed in southern platyfish, where 2-phenoxyethanol, quinaldine sulphate, metomedate, and MS-222 had different effects
on water quality (Guo et al. 1995).
Juveniles become agitated when transported at high density,
which leads to more oxygen consumption and metabolic waste
products (Berka 1986). This stress response may explain different effects on water quality variables during different phases.
However, the anesthetic mechanisms that affect water metrics
and the differences between anesthetics, still are not clearly
understood.
Generally, mortality increases with time during juvenile
transport due to the deterioration of water quality (Pramod et al.
2010). Among the factors affecting fish mortality, DO is considered the most important (Berka 1986). Moreover, high concentration of ammonia in water causes high ammonia levels and
pH in fish blood, which damage the red blood cells and gills,
affect osmoregulation, and increase the oxygen demand of fish
(Lawson 1995). Compared with ionized ammonia, un-ionized
ammonia is considered more toxic to fish. Though the maximum safe concentration of un-ionized ammonia is unknown,
0.0125 mg/L is commonly accepted by fish culturists (Meade
1985). Sublethal un-ionized ammonia concentrations are known
to cause behavioral, physiological, and histologic changes in
fish; high concentrations directly result in mortalities (Evans
et al. 2006). Furthermore, water variables can act together; the
ability of fish to use oxygen depends on their tolerance to stress,
water temperature, pH, and concentrations of carbon dioxide
and metabolic products such as ammonia (Berka 1986). Furthermore, un-ionized ammonia toxicity increases when DO is
low (Merkens and Downing 1957).
Compared with the control group, both MS-222 and clove oil
significantly improved survival of juvenile redtail culters at 6 h
and 18 h. This result was similar to that of Pramod et al. (2010),
who found that juvenile survival during transportation improved
when anesthetics were added. Higher survival of anesthetized
fish at and beyond 18 h can be attributed to the better water
quality maintained by the actions of the anesthetics.
Additionally, we found that mortality was higher in the control group than in anesthetic treatments even at 6 h, despite the
fact that no significant differences in water quality were observed between them. We speculate that this was the result of
physical damage avoidance due to stress suppression from anesthetics because clove oil and MS-222 have been shown to reduce
fish stress (Iversen et al. 2003; Inoue et al. 2005). Regardless of
the mechanisms, our results demonstrate that MS-222 and clove
oil are useful in short-term and long-term transport of juvenile
redtail culters.
In conclusion, our study showed that MS- 222 and clove
oil reduced deterioration of water quality, thus improving juve-
545
nile redtail culter survival during transport. Although MS-222

and clove oil were equally effective, we recommend the use of
5 mg/L clove oil because that anesthetic and dose provide a
cost-effective means of improving fish survival during transport
in China. However, other countries may restrict the use of clove
oil for food fish, so further research into eugenol and other clove
oil derivatives is needed.
ACKNOWLEDGMENTS
We express our thanks to technician Xinnian Chen for
assistance in tagging operations. We also thank M. D. Klopfer
and two anonymous reviewers for their critical reviews of
this manuscript. This research was financially supported by
the National Natural Science Foundation of China (grants
30830025 and 30900182) and National Science and Technology
Supporting Program (grant 2012BAD25B08). Participation of
B. R. Murphy was supported by the AcornAlcinda Foundation,
Lewes, Delaware.
REFERENCES
Amend, D. F., T. R. Croy, B. A. Goven, K. A. Johnson, and D. H. McCarthy.
1982. Transportation of fish in closed systems: methods to control ammonia,
carbon dioxide, pH, and bacterial growth. Transactions of the American
Arnolds, D. E. W., S. J. Zottoli, C. E. Adams, S. M. Dineen, S. Fevrier,
Y. Guo, and A. J. Pascal. 2002. Physiological effects of tricaine on the
supramedullary/dorsal neurons of the cunner, Tautogolabrus adspersus. Biological Bulletin 203:188189.
Berka, R. 1986. The transport of live fish: a review. EIFAC (European Inland
Fisheries Advisory Commission) Technical Paper 48.
CAS (Chinese Academy of Sciences). 1976. Fishes of the Yangtze River. CAS,
Department of Ichthyology, Institute of Hydrobiology, Science Press, Beijing.
Emerson, K., R. C. Russo, R. E. Lund, and R. V. Thurston. 1975. Aqueous
ammonia equilibrium calculations: effect of pH and temperature. Journal of
the Fisheries Research Board of Canada 32:23792383.
Estudillo, C. B., and M. N. Duray. 2003. Transport of hatchery-reared and wild
grouper larvae, Epinephelus sp. Aquaculture 219:279290.
Evans, J. J., D. J. Pasnik, G. C. Brill, and P. H. Klesius. 2006. Un-ionized
ammonia exposure in Nile tilapia: toxicity, stress response, and susceptibility
to Streptococcus agalactiae. North American Journal of Aquaculture 68:23
33.
Gomes, L. C., R. Roubach, C. A. R. M. Araujo-Lima, A. R. Chippari-Gomes,
N. P. Lopes, and E. C. Urbinati. 2003. Effect of fish density during transportation on stress and mortality of juvenile tambaqui Colossoma macropomum.
Journal of the World Aquaculture Society 34:7684.
Guo, F. C., L. H. Teo, and T. W. Chen. 1995. Effects of anaesthetics on the water
parameters in a simulated transport experiment of platyfish, Xiphophorus
maculatus (Gunther). Aquaculture Research 26:265271.
Harmon, T. S. 2009. Methods for reducing stressors and maintaining water
quality associated with live fish transport in tanks: a review of the basics.
Reviews in Aquaculture 1:5866.
Harper, C. 2003. Status of clove oil and eugenol for anesthesia of fish. Aquaculture Magazine 29(6):4142.
Hill, J. V., W. Davison, and M. E. Forster. 2002. The effects of fish anaesthetics
(MS222, metomidate and AQUI-S) on heart ventricle, the cardiac vagus and
branchial vessels from Chinook salmon (Oncorhynchus tshawytscha). Fish
Physiology and Biochemistry 27:1928.
Huang, X., W. Chen, and Q. Cai. 1999. Standard methods for observation
and analysis in Chinese ecosystem research network-survey: observation and
analysis of lake ecology. Science Press, Beijing.
546
LIN ET AL.
Inoue, L. A. K. A., L. O. B. Afonso, G. K. Iwama, and G. Moraes. 2005. Effects

of clove oil on the stress response of matrinxa (Brycon cephalus) subjected
to transport. Acta Amazonica 35:289295.
Iversen, M., B. Finstad, R. S. McKinley, and R. A. Eliassen. 2003. The efficacy
R
of metomidate, clove oil, Aqui-STM and Benzoak
as anaesthetics in Atlantic
salmon (Salmo salar L.) smolts, and their potential stress-reducing capacity.
Lawson, T. B. 1995. Fundamentals of aquacultural engineering. Chapman and
Hall, New York.
Lim, L. C., P. Dhert, and P. Sorgeloos. 2003. Recent developments and improvements in ornamental fish packaging systems for air transport. Aquaculture
Research 34:923935.
Marking, L. L., and F. P. Meyer. 1985. Are better anesthetics needed in fisheries?
Fisheries 10(6):25.
Meade, J. W. 1985. Allowable ammonia for fish culture. Progressive FishCulturist 47:135145.
Merkens, J. C., and K. M. Downing. 1957. The effect of tension of dissolved
oxygen on the toxicity of un-ionized ammonia to several species of fish.
Annals of Applied Biology 45:521527.
Murai, T., J. W. Andrews, and J. W. Muller. 1979. Fingerling American shad:
effect of Valium, MS-222, and sodium chloride on handling mortality. Progressive Fish-Culturist 41:2729.
Park, I. S., M. O. Park, J. W. Hur, D. S. Kim, Y. J. Chang, Y. J. Kim, J. Y.
Park, and S. C. Johnson. 2009. Anesthetic effects of lidocaine-hydrochloride
on water parameters in simulated transport experiment of juvenile winter
flounder, Pleuronectes americanus. Aquaculture 294:7679.
Pramod, P. K., T. P. Sajeevan, A. Ramachandran, S. Thampy, and S. S. Pai. 2010.
Effects of two anesthetics on water quality during simulated transport of a
tropical ornamental fish, the Indian tiger barb Puntius filamentosus. North
Randall, D. J., and P. A. Wright. 1987. Ammonia distribution and excretion in
fish. Fish Physiology and Biochemistry 3:107120.
Spath, M., and W. Schweickert. 1977. The effect of metacaine (MS-222) on the
activity of the efferent and afferent nerves in the teleost lateral-line system.
Naunyn-Schmiedebergs Archives of Pharmacology 297:916.
Teo, L. H., T. W. Chen, and B. H. Lee. 1989. Packaging of the guppy, Poecilia reticulata, for air transport in a closed system. Aquaculture 78:321
332.
Trushenski, J. T., J. D. Bowker, B. R. Gause, and B. L. Mulligan. 2012. Chemical
and electrical approaches to sedation of hybrid striped bass: induction, recovery, and physiological responses to sedation. Transactions of the American
Urbinati, E. C., J. Sampaio de Abreu, A. C. da Silva Camargo, and M. A.
Landinez Parra. 2004. Loading and transport stress of juvenile matrinxa
(Brycon cephalus, Characidae) at various densities. Aquaculture 229:389
400.
Xu, W., L. W. Geng, Z. H. Jia, C. T. Li, J. S. Yi, and C. B. Zhang. 2009.
Preliminary studies on artificial propagation of Culter mongolicus mongolicus
in Jingpo Lake. Freshwater Fisheries 4:6366.
Ye, S. 2007. Studies on fish communities and trophic network model of shallow
lakes along the medium reach of the Yangtze River. Doctoral dissertation.
Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan.
Zahl, I. H., O. Samuelsen, and A. Kiessling. 2012. Anaesthesia of farmed fish:
implications for welfare. Fish Physiology and Biochemistry 38:201218.
Zhang, T. 2005. Life-history strategies, trophic patterns and community structure in the fishes of Lake Biandantang. Doctoral dissertation. Institute of
Hydrobiology, Chinese Academy of Sciences, Wuhan.
Zhang, X., Z. Ruan, and B. Xiong. 2008. Age and growth characteristics of Culter mongolicus in Poyand Lake. Transactions of Oceanology and Limnology
3:137143.
Zhuang, P., B. Xu, L. Z. Zhang, and G. P. Feng, and L. H. Chen. 2009. Effects
of MS-222 and clove oil on oxygen consumption rate and ammonia excretion
rate of juvenile Chinese sturgeon, Acipenser sinensis. Journal of Fishery
Sciences of China 4:612618.


Effect of Photoperiod on Growth and Feed Efficiency of

Fingerling Walleye
a
Thomas M. Harder , Gordon G. Gotsch & Robert C. Summerfelt

a
McGraw Wildlife Foundation, Post Office Box 9, Dundee, Illinois, 60118, USA
Department of Natural Resource Ecology and Management, Iowa State University, Ames,
Iowa, 50011-3221, USA
To cite this article: Thomas M. Harder, Gordon G. Gotsch & Robert C. Summerfelt (2012): Effect of Photoperiod on Growth
and Feed Efficiency of Fingerling Walleye, North American Journal of Aquaculture, 74:4, 547-552


DOI: 10.1080/15222055.2012.700907
ARTICLE
Effect of Photoperiod on Growth and Feed Efficiency

of Fingerling Walleye
Thomas M. Harder* and Gordon G. Gotsch
McGraw Wildlife Foundation, Post Office Box 9, Dundee, Illinois 60118, USA
Robert C. Summerfelt
Department of Natural Resource Ecology and Management, Iowa State University, Ames,
Iowa 50011-3221, USA
Abstract
We examined the effect of photoperiod on growth and feed efficiency of fingerling walleyes Sander vitreus cultured
in a water reuse aquaculture system in northern Illinois from fall to late winter. We evaluated three photoperiod
regimens over 24-h days: continuous light (24 h), 18-h light, and 12-h light. Twelve, 0.55-m3 tanks (four/treatment)
were each stocked with 100 feed-trained walleyes (10.9 kg/m3; 196203 mm total length, 6166 g in weight). Survival
over the 131-d culture interval was greater than 98.8% in all photoperiod treatments. All measures of growth were
significantly slower and feeding efficiency was lower for fish in the 12 h of light treatment than for the 24-h and
18-h light treatments. Differences in performance between the 24-h and 18-h treatments were not significant, but all
measures of growth and feed efficiency were higher for fish in continuous light than in the 18-h group. The findings
support use of continuous, in-tank lighting for intensive culture of fingerlings walleye.
The response of walleyes Sander vitreus to light intensity is

of importance in both extensive and intensive culture. Walleyes
change from positive phototaxis to negative when they are 32
40 mm. They are attracted to high light intensity (7,800 lx) up
to 8 weeks after hatching, but older fish preferred intensities
of 24 lx (Bulkowski and Meade 1983). Attraction of fry and
small walleye fingerlings to light has been used in pond culture
to concentrate fish around in-pond feeders and to aggregate fish
for nighttime pond-harvesting operations (Harder and Gotsch
2007). Walleye larviculture has been done with overhead light
of 100 lx (Moore 1996) to 680 lx (Colesante 1996).
Scherer (1976) found a consistent inverse relationship between overhead light intensity (200, 20, and 2 lx) and vertical
position of fingerling walleyes. Tank color and light have been
evaluated to determine effects on growth and survival during habituation of pond-cultured juvenile walleyes to dry diets. Tests
have included black compared with blue tank walls (Harder and
Summerfelt 1996), uncovered tanks with overhead light of low

intensity (<16 lx; Kuipers and Summerfelt 1994), tanks in a
lighted room with turbid water (Johnson and Rudacille 2010),
tanks in dimly lighted rooms with in-tank lights (Siegwarth and
Summerfelt 1992), covered tanks with in-tank lighting (Nagel
1996), and tanks in a dark room with in-tank light (Johnson
and Rudacille 2010). Although the visual system of juvenile
and adult walleyes is biologically adapted to be sensitive to
low light intensity, Clayton et al. (2009) found that very dim
overhead light (about 10 lx) with high turbidity (111 NTU) reduced survival, suggesting that although walleyes prefer low
light intensity, there is a light threshold below which feeding
and survival is negatively affected.
Nickum (1986) stated that there has not been a comparison
between a scheduled photoperiod and continuous lighting for
the culture of walleyes. Longer day length is said to stimulate
growth of fish (Boeuf and Le Bail 1999). Continuous light rather
*Corresponding author: tharder@mcgrawwildlife.org

Received February 6, 2012; accepted June 1, 2012
547
548
HARDER ET AL.
than a 12-h light : 12-h dark cycle resulted in higher survival

of larval walleye. This was attributed to the beneficial effect of
continuous light in keeping fish dispersed in the water column
(Loadman et al. 1986). He recommended a photoperiod of 16 h
light: 8 h dark (16 L: 8 D) to simulate summer solstice at midlatitudes in the northern hemisphere (15.23 h light : 8.77 h dark
on June 21 in northern Illinois where our study was conducted).
The study by Huh et al. (1976) is the only study we found
that related photoperiod and growth of juvenile walleyes. Their
data showed only a small difference between growth of walleye
fingerlings exposed to a photoperiod with either an 8-h or 16-h
light interval.
The expanding use of indoor, intensive water reuse aquaculture systems (WRAS) for culture of many species of fish,
including walleyes (Summerfelt and Penne 2007), provides the
opportunity to control environmental conditions to optimize fish
growth. These past studies demonstrate that photoperiod has the
potential for increasing growth rate and enhancing utilization of
feed, but the topic has not been thoroughly studied. The objective of this study was to evaluate growth and measures of
feed conversion ratio (FCR) and feed efficiency ratio (FER) of
juvenile walleyes cultured in a WRAS under three photoperiod
treatments over 24-h days: continuous (i.e., 24 h) light (24 L),
18 L, and 12 L (this treatment being slightly longer than the
natural photoperiod of 1011 h light for this season at this
location).
METHODS
Culture system.Fish were cultured in a WRAS at the
McGraw Wildlife Foundation, Dundee, Illinois (88 17 W, 42
02 N), at an elevation of 247 m (810 ft) msl. The design features
of this WRAS were similar to that described by Summerfelt
(1996), except that the system at McGraw site included use
of liquid oxygen infused into an oxygen cone to supersaturate
the water supply to the culture tanks. The culture tanks were
maintained at volume of 0.55 m3 and a water depth of 52 cm.
The mean water inflow (14 L/m) provided an exchange rate of
1.5 exchanges/h.
Lighting.Each tank was randomly assigned to a photoperiod treatment, and there were four replicates for each of the
three photoperiod treatments. Tanks were covered and located
in a dark room, and each tank was equipped with a single submerged light similar in design to that illustrated by Siegwarth
and Summerfelt (1992). The submerged lights were constructed
with 5-cm, schedule-40 polyvinyl chloride (PVC) pipe with a
clear acrylic cap on one end that was positioned 5 cm below the
water surface. Light was emitted through the clear cap from a
single 150-mA, 6.3-V DC lamp. These small lamps require only
0.945 W of energy per bulb. Power was supplied to the lights
from a single battery charger, which had an output of 12 V DC at
2 amp. Except for a small opening under the feeder and another
for the water inlet, the tanks were covered with foam board, and
overhead fluorescent lights were turned off. We were unable
to measure light intensity underwater, but light intensity at the

water surface was 36 lx measured 30 cm below the light. Light
duration was controlled for each treatment by a 24-h timer.
Feeding schedule.Total daily feed amounts for each tank
were dispensed at 5-min intervals within each treatments feeding
schedule:
24 L: from 1000 hours one day to 0800 hours the next
day (22 h/d);
18 L: from 1000 hours one day to 0200 hours the next
(16 h/d);
12 L: from 1000 hours to 2000 hours in the same day
(10 h/d).
The lights came on at 0800 hours in the 12 L and 18 L
treatments and were on continuously in the 24 L treatment. Feed
was dispensed with auger feeders controlled by electric timer.
Feeding was not done between 0800 and 1000 hours when tank
cleaning and water samples were collected. Fish showed a weak
startle response to the sudden light-on.
Water quality.Water samples were collected between 0800
and 1000 hours from each tank. Daily measurements were taken
of temperature and dissolved oxygen (YSI model 550, YSI Inc.,
Yellow Springs, Ohio). The pH (YSI model 63), nitrite, alkalinity (ALK), and total ammonia nitrogen (TAN; Hach FF-2, Hach
Company, Loveland, Colorado) were measured once weekly.
Values for carbon dioxide (CO2 ) were calculated: [CO2 ] =
ALK 10(6.3-pH) (Summerfelt, S.T. 1996); un-ionized ammonia was calculated as (TAN) (% un-ionized NH3 )/100, where
NH3 was at pH and temperature (Piper et al. 1982).
Measures of performance. On day 1 of the experiment, 100
fish were stocked into each of 12 tanks, from which a sample
of 25 fish were measured to determine initial mean lengths,
weights, tank biomass, and feeding rates. Tank means of initial
fish total length and weight ranged from 196 to 203 mm and
6166 g. Initial stocked biomass was 6 kg/tank and density
was10.9 kg/m3. Thereafter, a sample of 25 fish were weighed
and measured at 2-week intervals to obtain interval growth rates
and adjust feed amounts to maintain the same 2% of body weight
per day. At the conclusion of the 131-d culture interval (March
8, 2011), all fish from each tank were counted, weighed, and
measured individually.
Growth was expressed as relative growth (weight gain as a
percentage of initial weight). Absolute growth rate (mm/d and
g/d) was calculated, as was specific (logarithmic) growth rate
(SGR as percent per day): SGR = 100 [loge final weight
(loge initial weight/t)], where t represents the number of days
between initial and final weight.
Feed conversion ratio (FCR) and feed efficiency ratio (FER)
were calculated by standard formulas (all weights being grams):
FCR = total feed fed/total live weight gain and FER = (live
weight gain/total feed fed)100.
An ANOVA was used to test differences in performance on
the tank means of the growth and food conversion variables
among the treatments with = 0.05. When the P-value of
549
EFFECT OF PHOTOPERIOD ON FINGERLING WALLEYE
TABLE 1. Water quality means during 131-d culture interval of walleyes in photoperiod testing, including dissolved oxygen (DO), total ammonia nitrogen
(TAN), un-ioinized ammonia (UIA), alkalinity (ALK), and CO2 .
Photoperiod
(hours of light/d)
24
18
12
Concentration (mg/L)
pH
DO
NO2
TAN
UIA
ALK
CO2
21.9
21.9
21.8
7.7
7.6
7.7
8.9
8.2
8.5
0.6
0.5
0.5
0.38
0.33
0.30
0.007
0.006
0.006
202
200
201
10.1
9.9
9.2
an ANOVA was significant, Fishers protected least significant

difference multiple range test was used to determine differences
among treatments.
Fish and feed.The walleyes used in this study were cultured from gametes obtained from Wolf Lake, Wisconsin, in
2010. Eggs were water hardened on site then disinfected with
povidone-iodine (100 mg/L of water for 10 min) before transport
to the McGraw Foundation for incubation. They were hatched
in May, stocked in a fertilized pond, and pond reared for about
45 d. After harvest from the pond, they were transferred to indoor tanks where they were habituated to a commercial dry
diet (Otohime C2, Reed Mariculture, San Jose, California) and
subsequently weaned to 1-mm WG 9206 (Nelsons Silver Cup,
Tooele, Utah), which was increased to 4-mm WG as fish grew
(phase II). At the onset of the present study (phase III), fish
were still on 4-mm WG but were converted to the 4-mm, slow
sinking, extruded, production steelhead diet (Silver Cup Fish
Feed, Tooele, Utah) in December 2010; they remained on this
diet until final inventory on March 8, 2011. The manufactures
steelhead diet label indicates typical composition: 46% protein,
16% fat, 1% fiber, and 4,325 kcal/kg digestible energy content.

The feeding rate was 2% of body weight/d, recalculated every
2 weeks through the 131-d of the study.
RESULTS
Water Quality
Because of the use of supersaturated water, mean dissolved
oxygen for all tanks averaged 97% saturation (Table 1). The
water supply had an average temperature of 6.9 C (range, 3.8
10.4 C), but it was possible to maintain mean temperature in the
culture tanks at 21.821.9 C without heating the water because
the tanks and the components of the WRAS were located in
a heated room. All other water quality measures were similar
between treatments.
Survival
Survival was approximately 99% in all treatments (Table 2),
which is greater than typically reported for intensive culture
of advanced fingerlings. At Rathbun Hatchery in southcentral
TABLE 2. Comparative performance of juvenile walleyes cultured in tanks for 131 d with light durations of 24, 18, or 12 h per 24-h day. Within a row, means
with the same letters are not significantly different (P > 0.05).
Photoperiod treatment groups (means SE)

Variable
Total length (mm)
Total weight (g)
Relative weighta
Length growth (mm/d)
Weight growth (g/d)
SGRb
Growth (% weight gain)
FCRc
FERd
Biomass density (kg/m3)
Survival (%)
24 h/d
282.4
218.1
99.5
0.65
1.18
0.96
250.0
1.72
58.3
39.2
99.2
1.2 z
3.2 z
1.2 z
0.01 z
0.03 z
0.02 z
8.1 z
0.07 z
2.4 z
0.6 z
0.5 z
18 h/d
280.8
207.6
96.2
0.63
1.10
0.91
230.8
1.82
54.7
37.2
98.8
2.7 zy
2.6 z
1.8 zy
0.01 z
0.02 z
0.01 z
3.9 z
0.06 z
1.8 z
0.6 z
0.9 z
12 h/d
ANOVA
P-value
0.04
<0.01
<0.01
0.02
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
0.59
276.2
189.4
92.2
0.59
0.96
0.84
196.5
2.13
47.0
34.7
99.8
2.3 y
4.7 y
0.2 y
0.02 y
0.03 y
0.02 y
6.9 y
0.03 y
0.6 y
0.7 y
0.5 z
a
Relative weight (Wr ) was calculated from the standard weight (Ws ) formula of Anderson and Neumann (1996) as log10 Ws = 5.453 + (3.180 log10 TL), where TL is total
length (mm).
b
SGR is specific growth rate (percent/d) = [ln(final weight) ln(initial weight)/d]100.
c
FCR is feed conversion ratio = total feed fed/total live weight gain, where weights are grams.
d
FER is feed efficiency ratio = (live weight gain/total feed fed)100, where weights are grams.
550
HARDER ET AL.
Iowa survival of top-graded phase III walleyes fed in outside

tanks and exposed to a natural photoperiod over 90 d averaged
88.9% (range, 8397%) over a 10-year interval (Summerfelt
et al. 2011).
Growth, Feeding Efficiency, and Biomass
Relative growth, absolute growth (length and weight), and
SGR were not significantly (P > 0.05) different for fish in the
24 L or 18 L treatments, but fish with continuous light had higher
mean values for all measures of growth. All growth measures
for both the 24 L and 18 L were significantly greater than those
of the 12 L fish (Table 2).
Feed conversion ratio was significantly lower and feeding
efficiency ratio significantly higher (P < 0.01) for the 24 L fish
than for the 12 L fish (Table 2). Again, there was no difference
between fish in the 24 L and 18 L treatments.
Final densities in the three treatments in our study ranged
from 34.739.2 kg/m3. The ending tank biomass was highest
(39.2 kg/m3, SE = 0.06) for the 24 L treatment.
DISCUSSION
Temperature treatment means in our study were slightly
lower than physiological optimum temperature of 25.3 C at
45 lx, based on measures of metabolic rate (Cai and Summerfelt
1992). Nitrite nitrogen (range, 0.50.6 mg/L) was higher than
recommended levels (<0.1 mg/L) for optimal health of coldwater and warmwater fish in intensive culture (Wedemeyer 1996),
but nitrite (NO2 ) is in a pH-dependent equilibrium with nitrous
acid (HNO2 ), which is the more toxic form. At the pH in our
study tanks (7.67.7), all of the total nitrite would be present
in the ionized form that is not readily diffusible by the gills
(Colt and Tomasso 2001). Water quality measures were similar
between treatments and were within the suggested range for
optimal fish health (Wedemeyer 1996).
Published growth rates for walleye fingerlings vary with fish
size; however, smaller fish typically grow faster than larger fish,
and temperatures closer to optimum result in faster growth.
Growth rates (mm/d) in our study ranged from 0.59 to 0.65 mm/d
for fish that were intially196203 mm total length and ending
at 276282 mm. Siegwarth and Summerfelt (1992) reported
that walleye fingerlings (176216 mm) at 25 C grew at a rate
of 0.55 mm/d. Walleyes of larger size (285324 mm) and cultured at lower temperature (20.7 C) had much slower growth
(0.31 mm/d; Siegwarth and Summerfelt 1993). Summerfelt
(1996) summarized walleye fingerling growth data from several studies and calculated a mean of means growth rate of
0.61 mm/d. Summerfelt et al. (2011) reported phase-III growth
of walleye fingerlings from 229 to 250 mm of 1.41.5 mm/d and
1.11.2 g/d. Final densities in the three treatments in our study
(34.739.2 kg/m3) were more than twice that of the 16.8 kg/m3
reported for the phase-III culture at the Rathbun Hatchery, Morovia, Iowa, one of the few state hatcheries in the country that use
an intensive culture systems for raising phase-III walleyes (Summerfelt et al. 2011). Phase-III to food-size walleyes in a WRAS
were grown to a density of 72.1 kg/m3 (Summerfelt 1996), which
suggests that given good water quality, walleyes can be raised
to densities comparable to that of many other cultured fish.
In regard to research on photoperiod affect, an important
point is to be certain that light affects fish growth through better food conversion efficiency and not just through stimulated
food intake (Boeuf and Le Bail 1999). The present research
demonstrates an improvement in growth and feed efficiency as
photoperiod increases. Fish in our three photoperiod treatments
were fed the same daily rate (2%) of the total biomass at 5-min
intervals when the light was on (except for 2 h for tank cleaning
and water quality measurements). So, there was a substantial
difference in the total number of feedings per day among the
three treatments ([hours of light 2]60 min/5 min): 264 daily
feedings for 24 L, 192 for 18 L, and 120 12 L. If feeding frequency was the treatment rather than photoperiod, the analysis
would show the same statistical differences because feeding
frequency per day and photoperiod were the same. A previous
study, (Phillips et al. 1998) found that growth rates and food
conversion did not differ significantly between feeding frequencies of 9 or 90/d and 3 or 30/d for fingerling walleyes raised in
intensive culture with light : dark photoperiod of 16: 8. Thus,
the photoperiod effect is the total hours of light for feeding.
The significant difference in FCR among the three treatments,
with FCR increasing as photoperiod shortened, substantiates
this hypothesis.
Natural photoperiod during the interval of our study was even
less than that of the shortest photoperiod treatment (12 L); thus,
natural light or an artificial photoperiod that mimics natural day
length during this interval is insufficient to produce the best
growth. The findings and most literature support a hypothesis
that continuous light encourages continuous feeding and maximizes growth and feeding efficiency. Thus, to obtain maximum
fallspring growth benefit for culture of juvenile walleyes in a
WRAS, it is important to have at least 18-h to 24-h of artificial
light of low intensity.
The basic biological mechanism for enhanced growth with
longer photoperiod is not known, but it may result from stimulation of the hypothalamicpituitary axis (Jobling 2010). However, the benefit of extended photoperiod has been reported
for many unrelated, freshwater and marine fishes. First-feeding
Arctic char Salvelinus alpinus subjected to 24-h light and continuous feed availability had lower mortality and higher mean
final weights with less size variation within treatments than fish
under restricted feeding and ambient photoperiod (Burke et al.
2005). Red seabream Pagrus major reared with 24-h continuous
light showed the highest total weight gain and specific growth
rate of four different photoperiod regimes tested. Food intake
and feeding efficiency was also highest for fish in continuous
light than for fish in a balanced light : dark treatment of 12:12 h
(Biswas et al. 2008).
EFFECT OF PHOTOPERIOD ON FINGERLING WALLEYE
Constant photoperiod promotes overall growth regardless of

temperature, but constant temperature and photoperiod together
produces the best overall growth performance in both male
and female yellow perch Perca flavescens, and manipulation
of both can also inhibit maturation (Shewmon 2005). Petit et al.
(2003) reported faster growth of largemouth bass Micropterus
salmoides cultured with continuous light than fish held under
12:12 h light : dark cycle. They attributed the faster growth to
greater feed consumption of fish exposed to continuous light,
even when both groups of fish were fed the same daily ratio.
With respect to the cost of energy consumed by using the
continuous (24 L) lighting system we used, four tanks with intank lighting would have an expenditure of only $0.0073/d at
a charge rate of $0.08/kilowatt-hour. By comparison, if 20-W
bulbs were used in overhead lighting of four tanks, the energy
cost would be $0.12/d at the same charge rate. Because a 24-h
photoperiod is more beneficial to walleye growth and feeding
efficiency than a 12-h photoperiod and the cost of this type of
lighting is minimal, a continuous lighting regime should be used
in walleye culture.
The price of feed varies with the specific attributes of the feed,
the quantity, shipping costs, and other variables, but whatever
the price, it is generally the single most expensive variable cost
item for fish culture. Therefore, it is imperative for the culture of
any species to implement cultural technology that will provide
for the lowest FCR. In the our study, the FCR was 1.72 for the
24 L fish and 1. 82 for 18 L, both being significantly lower than
for the 2.13 for the 12 L fish (Table 2). This tells us that the
quantity of feed per unit weight gain will be 25% greater for
fish cultured with light for only 12 h/d than for fish cultured at
24 h/d. Thus, the daily photoperiod becomes an important part
of cultural technology for this species that will have substantial
impact on production costs.
This study further demonstrates the utility of a WRAS for
culture of fingerling walleyes for food fish or enhancement
stocking. Continuing culture of fall to spring fingerlings under
optimum conditions of temperature and light would provide fish
at stocking that are far less vulnerable to predators, as well as
shorten the time between stocking and when the fish would reach
a minimum harvestable length. Larscheid (1995) and Santucci
and Wahl (1993) reported that stocking larger walleye fingerlings provides a better economic return than does stocking fry or
summer fingerlings. A larger fish is less vulnerable to poststocking mortality from predation. Growing walleyes overwinter in
a WRAS for stocking in spring or early summer should provide
the stocked fish with enhanced abundance of prey that would
increase their growth and shorten the time between stocking and
harvest.
ACKNOWLEDGMENTS
We thank Steve Newman of the Wisconsin Department of
Natural Resources, Escanaba Research Station, for supplying a
551
boat and nets to collect spawning walleyes. We appreciate the

assistance of our summer intern Lucas Brown.
REFERENCES
Anderson, R. O., and R. M. Neumann. 1996. Length, weight, and associated
structural indices. Pages 447482 in B. R. Murphy and D. W. Willis, editors.
Fisheries techniques, 2nd edition. American Fisheries Society, Bethesda,
Maryland.
Biswas, A. K., M. Seoka, Y. Tanaka, K. Takii, and H. Kumai. 2008. Use of
photoperiod manipulation to stimulate the growth performance of juvenile
red sea bream (Pagrus major). World Aquaculture 39(2):1215.
Boeuf, G., and P. Y. Le Bail. 1999. Does light have an influence on fish growth?
Bulkowski, L., and J. W. Meade. 1983. Changes in phototaxis during early
development of walleye. Transactions of the American Fisheries Society
112:445447.
Burke, M. G., M. R. Kirk, N. A. MacBeth, D. J. Bevan, and R. D. Moccia.
2005. Influence of photoperiod and feed delivery on growth and survival
of first-feeding Arctic char. North American Journal of Aquaculture 67:
344350.
Cai, Y., and R. C. Summerfelt. 1992. Effects of temperature and size on oxygen consumption and ammonia excretion by walleye. Aquaculture 104:127
138.
Clayton, R. D., J. E. Morris, and R. C. Summerfelt. 2009. Effect of turbidity
duration on culture of walleye larvae. North American Journal of Aquaculture
71:174177.
Colesante, R. T. 1996. Intensive culture of walleye using brine shrimp and
formulated diets. Pages 191194 in R. C. Summerfelt, editor. Walleye culture
manual. North Central Regional Aquaculture Center Publications, Iowa State
University, Ames.
Colt, J. E., and J. R. Tomasso. 2001. Hatchery water supply and treatment. Pages
91186 in G. A. Wedemeyer, editor. Fish hatchery management, 2nd edition.
American Fisheries Society, Bethesda, Maryland.
Harder, T. M., and G. G. Gotsch. 2007. Nighttime lighting and feeding in ponds
enhance survival of fingerling walleyes during habituation to manufactured
feed. North American Journal of Aquaculture 69:250256.
Harder, T. M., and R. C. Summerfelt. 1996. Effects of tank color and size on
the success of training walleye fingerlings to formulated feed. Pages 631
636 in G. S. Libey and M. B. Timmons, editors. Successes and failures
in commercial recirculating aquaculture: aquacultural engineering society
proceedings, volume 2. Northeast Regional Agricultural Engineering Service,
NRAES-98, Cornell University, Ithaca, New York.
Huh, H. T., H. E. Calbert, and D. A. Stuiber. 1976. Effects of temperature
and light on growth of yellow perch and walleye using formulated feed.
Transactions of the American Fisheries Society 105:254258.
Jobling, M. 2010. Fish culture: the rearing environment. Pages 3360 in N. R.
Le Francois, M. Jobling, C. Carter, P. U. Blier, and A. Savoie, editors. Finfish
aquaculture diversification. CAB International, Oxfordshire, UK.
Johnson, J. A., and J. B. Rudacille. 2010. Intensive culture of walleye converted
to a dry diet as small fingerlings. Iowa Department of Natural Resources,
Federal Aid in Sport Fish Restoration, Project F-160-R, Completion Report,
Des Moines.
Kuipers, K. L., and R. C. Summerfelt. 1994. Converting pond-reared walleye
fingerlings to formulated feeds: effects of diet, temperature, and stocking
density. Journal of Applied Aquaculture 4(2):3157.
Larscheid, J. G. 1995. Development of an optimal stocking regime for walleyes
in East Okoboji Lake, Iowa. Pages 472483 in H. L. Schramm Jr. and
R. G. Piper, editors. Uses and effects of cultured fishes in aquatic ecosystems.
American Fisheries Society, Symposium 15, Bethesda, Maryland.
Loadman, N. L., G. E. E. Moodie, and J. A. Mathias. 1986. Significance of
cannibalism in larval walleye (Stizostedion vitreum). Canadian Journal of
Fisheries and Aquatic Sciences 43:613618.
552
HARDER ET AL.
Moore, A. A. 1996. Intensive culture of walleye fry on formulated feed.

Pages 195197 in R. C. Summerfelt, editor. Walleye culture manual. North
Central Regional Aquaculture Center Publications, Iowa State University,
Ames.
Nagel, T. 1996. Intensive culture of fingerling walleye on formulated feeds.
Pages 205207 in R. C. Summerfelt, editor. Walleye culture manual. North
Central Regional Aquaculture Center Publications, Iowa State University,
Ames.
Nickum, J. G. 1986. Walleye. Pages 115126 in R. R. Stickney, editor. Culture
of nonsalmonid freshwater fishes. CRC Press, Boca Raton, Florida.
Petit, G., M. Beauchaud, J. Attia, and B. Buisson. 2003. Food intake and growth
of largemouth bass (Micropterus salmoides) held under alternated light/dark
cycle (12L:12D) or exposed to continuous light. Aquaculture 228:397
401.
Phillips, T. A., R. C. Summerfelt, and R. D. Clayton. 1998. Feeding frequency
effects on water quality and growth of walleye fingerlings in intensive culture.
Piper, R. G., I. B. McElwain, L. E. Orme, J. P. McCraren, L. G. Fowler, and J. R.
Leonard. 1982. Fish hatchery management. U.S. Fish and Wildlife Service,
Washington, D.C.
Santucci, V. J., Jr., and D. H. Wahl. 1993. Factors influencing survival and
growth of stocked walleye (Stizostedion vitreum) in a centrarchid-dominated
impoundment. Canadian Journal of Fisheries and Aquatic Sciences 50:1548
1558.
Scherer, E. 1976. Overhead-light intensity and vertical positioning of the walleye, Stizostedion vitreum vitreum. Journal of the Fisheries Research Board of
Canada 33:289292.
Shewmon, L. N. 2005. Culture methods for growth enhancement and off-season
production of yellow perch Perca flavescens. Masters thesis. North Carolina
State University, Raleigh.
Siegwarth, G. L., and R. C. Summerfelt. 1992. Light and temperature effects
on performance of walleye and hybrid walleye fingerlings reared intensively.
Siegwarth, G. L., and R. C. Summerfelt. 1993. Performance comparison and
growth models for walleyes and walleye sauger hybrids reared for two
years in intensive culture. Progressive Fish-Culturist 55:229235.
Summerfelt, R. C., J. A. Johnson, and C. P. Clouse. 2011. Culture of walleye,
sauger, and hybrid walleye. Pages 451570 in B. A. Barton, editor. Biology,
management, and culture of walleye and sauger. American Fisheries Society,
Bethesda, Maryland.
Summerfelt, R. C., and C. R. Penne. 2007. Nutrient retention by fish in a
multispecies recirculating aquaculture facility. International Journal of Recirculating Aquaculture 8:4364.
Summerfelt, S. T. 1996. Aquaculture of walleye as a food fish. Pages 215230
in R. C. Summerfelt, editor. Walleye culture manual. North Central Regional
Aquaculture Center Publications Office, Iowa State University, Ames.
Wedemeyer, G. A. 1996. Physiology of fish in intensive culture systems.
Chapman and Hall, New York.


Effects of Winter Feeding on Growth, Body

Composition, and Processing Traits of Co-Cultured Blue
Catfish, Channel Catfish, and Channel Catfish Blue
Catfish Hybrids
Brian G. Bosworth
U.S. Department of Agriculture, Agricultural Research Service, Catfish Genetics Research

Unit, Post Office Box 38, Stoneville, Mississippi, 38776, USA
To cite this article: Brian G. Bosworth (2012): Effects of Winter Feeding on Growth, Body Composition, and Processing
Traits of Co-Cultured Blue Catfish, Channel Catfish, and Channel Catfish Blue Catfish Hybrids, North American Journal of


American Fisheries Society 2012
DOI: 10.1080/15222055.2012.686958
ARTICLE
Effects of Winter Feeding on Growth, Body Composition,

and Processing Traits of Co-Cultured Blue Catfish, Channel
Catfish, and Channel Catfish Blue Catfish Hybrids
Brian G. Bosworth*
U.S. Department of Agriculture, Agricultural Research Service, Catfish Genetics Research Unit,
Post Office Box 38, Stoneville, Mississippi 38776, USA
Abstract
The effects of winter feeding on growth, body composition, and processing yield were compared for co-cultured
blue catfish Ictalurus furcatus, channel catfish I. punctatus, and channel catfish blue catfish hybrids. Fish (0.4
0.7 kg) from each genetic group were stocked communally at 5,625 fish/ha in ten 0.04-ha ponds during mid-November.
Fish in five ponds were fed at 2% of initial body weight twice per week; fish in the other five ponds were not fed.
The study was terminated after 14 weeks, and fish were weighed and processed. Analyzed traits were weight change
(%), survival, and yields (%) of carcass, shank fillet, nugget, head, viscera, skin, intraperitoneal fat, liver, and ovary.
Among fed fish, hybrids gained the most weight, channel catfish had intermediate weight gain, and blue catfish
gained the least weight. Unfed blue catfish lost more weight than unfed channel catfish and hybrids. Survival was
not different among genetic groups or between feeding regimes. Interactions among main effects for processing and
body composition traits made generalizations difficult, but carcass yield was consistently higher for blue catfish and
hybrids than for channel catfish, higher for females than for males, and higher for fed fish than for unfed fish. Shank
fillet yield was higher for hybrids than for blue catfish and channel catfish, higher for females than for males, and
higher for fed fish than for unfed fish. Nugget yield was higher for blue catfish than for channel catfish and hybrids
and was higher for fed fish than for unfed fish. Blue catfish and hybrids had higher intraperitoneal fat yield and
lower liver and ovary yield than channel catfish. Fed fish had higher intraperitoneal fat and liver yield than unfed
fish. Winter feeding improved growth and fillet yield in all groups, but the benefits of winter feeding were lower for
blue catfish than for channel catfish and hybrid catfish.
Production of channel catfish Ictalurus punctatus is the

largest aquaculture enterprise in the United States, representing
more than half of aquaculture production by weight. The majority of feeding and growth occurs between April and October,
when water temperatures are conducive to catfish feeding activity (Tucker and Hargreaves 2004). During the winter months,
many catfish producers withhold feed or provide very little feed
because cold water temperatures reduce feeding activity and
because increased rainfall often limits the access of feeding
equipment to areas around pond levees. In addition, the weight
that is lost due to the withholding of feed in the winter months
is typically regained fairly quickly through compensatory
growth when feeding resumes in the spring (Robinson et al.
2001). However, studies of channel catfish indicate that winter feeding results in weight gain (Reagan and Robinette 1979;
Robinette et al. 1985; Kim and Lovell 1995) or reduced weight
loss (Heidinger 1975; Nanninga et al. 2011) in comparison with
fish that are unfed. Effects of winter feeding on processing yield
(carcass and fillet yield) are inconsistent; some reports indicate
that winter feeding improves processing yield (Kim and Lovell
1995), while other studies show no effect (Nanninga et al. 2011).
Less is known about the effects of winter feeding on blue
catfish I. furcatus. Grant and Robinette (1992) reported that
winter-fed, market-weight channel catfish gained more weight
than blue catfish. Tidwell et al. (1995) reported that winter-fed
fingerling channel catfish and blue catfish lost weight at similar
*E-mail: brian.bosworth@ars.usda.gov
Received November 29, 2011; accepted April 16, 2012
553
554
BOSWORTH
rates, although those authors did not compare weight loss in the
fed groups with that in unfed control groups.
Production of channel catfish blue catfish hybrids has increased substantially in the last 5 years, but information on the
effects of winter feeding on hybrid catfish growth and processing yield is not available. Anecdotal reports indicate that hybrid
catfish feed more aggressively than channel catfish at cold water
temperatures.
The objectives of this study were to determine effects of
winter feeding on growth, body composition, and processing
traits of co-cultured channel catfish, blue catfish, and channel
catfish blue catfish hybrids. Producers and processors can use
this information to make decisions about the benefits of winter
feeding for a particular genetic group of catfish.
METHODS
Channel catfish of the U.S. Department of Agriculture
(USDA) 103 strain, blue catfish of the D&B strain, and hybrids
produced by crossing USDA 103 females with D&B males were
used in this study. All fish were approximately 17 months old
when the study was initiated, and all were treated similarly (i.e.,
in terms of stocking density, feeding regime, feed, etc.) prior to
the study. Market-weight fish (0.40.7 kg) from each genetic
group were stocked communally in ten 0.04-ha ponds during
mid-November 2006; each pond was stocked with 75 fish of
each group. Fish were stocked communally due to limited pond
availability. Previous trials at the Catfish Genetics Research
Unit (USDA Agricultural Research Service) have indicated that
5 replicate ponds/treatment are required to obtain adequate statistical power for feeding trials. Therefore, separate culture of
the genetic groups would have required 30 pondsmore than
were available. From each pond, 15 fish of each genetic group
were randomly selected for tagging with a PIT tag (Allflex USA,
Inc., Dallas, Texas) to allow tracking of individual fish growth.
Tagged fish were weighed individually, and gender was recorded
at stocking. The additional, untagged fish (60 fish/group for each
pond) were counted, weighed as a group, and stocked into each
pond to increase the fish density to 5,625 fish/ha. A sample
of 50 fish from each genetic group was processed at the time
of stocking to provide initial values for processing and body
composition traits. Fish in five ponds were fed a 32% protein,
slow-sink pellet (Delta Western, Indianola, Mississippi) at 2%
of initial body weight twice per week (Monday and Thursday)
regardless of water temperature. Fish in the other five ponds
were not fed.
Feeding was terminated after 14 weeks. Feed was withheld
for 4 d to allow clearance of feed from the gastrointestinal tracts
of the fish. Ponds were then seined, and the tagged and untagged
fish were sorted into genetic groups based on visual inspection.
Gender and individual weight were recorded for tagged fish. Untagged fish in each pond were sorted by genetic group, counted,
and weighed as a group. From each pond, 1015 fish/genetic
group were selected for processing; these fish were stunned by
electricity, mechanically deheaded (Baader 166; Baader, Inc.,

Lubeck, Germany), mechanically skinned (Collum Jet 470-SS;
Collum Tool, Greenville, Mississippi), and eviscerated by hand.
Carcasses were placed individually in plastic bags and then
placed in coolers on ice. The next day, carcasses were filleted
by an experienced employee from a catfish processing plant.
Gender was recorded, and weights of the gutted carcass, gutted
deheaded carcass, gutteddeheadedskinned carcass, shank
fillet, nugget (rib meat), intraperitoneal fat, and liver were measured from individual fish. Ovary weight was recorded for females. Intraperitoneal fat, liver, and ovary were weighed to the
nearest 0.1 g; carcass traits were weighed to the nearest 0.5 g.
Weights of head, skin, and viscera were estimated by subtraction of the appropriate carcass weights (for example, weight of
viscera = whole weight gutted carcass weight). Survival was
determined for each genetic group in fed and unfed treatments.
The food conversion ratio (FCR) was estimated for fed fish as
follows: (weight of feed fed)/(total weight of fish harvested
total weight of fish stocked).
Water temperatures for the study period were obtained from a
database maintained by the Mississippi State University Extension Service. At the beginning of the study, chlorides in ponds
were adjusted to approximately 100 mg/L by the addition of
NaCl. Other water quality variables were not measured.
Analyzed traits included percent change in weight of tagged
individuals, survival, and yields (relative to whole weight) of
head, viscera, gutteddeheadedskinned carcass, skin, shank
fillet, nugget, intraperitoneal fat, liver, and ovary. Traits were
analyzed by ANOVA. The model for percent weight change,
processing yield traits, and body composition traits included
fixed effects of gender, feeding regime, and genetic group and
relevant interactions; the random effects of pond within feeding regime and pond genetic group gender within feeding
regime were also included. Fish weight was used as a covariate for fillet yield and visceral fat. Survival was analyzed with
a model that included the fixed effects of feeding regime and
genetic group and the random effects of pond within feeding
regime and pond genetic group within feeding regime. The
FCR is reported here but was not analyzed since FCR was undefined for the unfed treatment. Weight gain of untagged fish
was not analyzed because gender data were not recorded for untagged fish and because means for percent weight change were
similar to those for individually weighed fish. Statistical analyses were conducted with the MIXED procedure in the Statistical
Analysis System version 9.1 (SAS Institute, Inc., Cary, North
Carolina), and differences among fixed effects were declared
significant at P < 0.05.
The choice to use a co-culture experimental design was primarily due to the lack of adequate replicate ponds for rearing
each genetic group separately. Therefore, the differences observed among genetic groups in response to winter feeding may
555
WINTER FEEDING EFFECTS ON CATFISH AND CATFISH HYBRIDS
be due, at least in part, to the co-culture design used. However,

consistencies between the data observed with the co-culture design and data observed during other studies in which fish were
cultured as separate groups suggest that the impact of co-culture
was not substantial. In addition, the overall effects of the fed
versus unfed treatments should be relevant, and comparisons
of fed and unfed fish within each genetic group should provide useful information despite the necessity of the co-culture
design.
Mean water temperature for the study period was 10.8 C
(range = 5.120.5 C) and was similar to the average water temperature for the same period as recorded during the previous
7 years (20002006) at this location (10.5 C). Therefore, temperatures for the study period were typical for a winter in the
Mississippi Delta, and results of the study were not influenced
by an abnormally warm or cold winter.
Survival was not different among genetic groups (survival =
95.1% for blue catfish, 93.4% for channel catfish, and 96.2%
for hybrids) and was not affected by feeding regime (survival =
94.5% for fed fish; 95.4% for unfed fish). Grant and Robinette
(1992) reported similar high survival rates for market-sized blue
catfish and channel catfish (98% and 99%, respectively) that
were fed over the winter. Kim and Lovell (1995) and Nanninga
et al. (2011) also reported high survival (>90%) and no difference in survival between winter-fed and unfed channel catfish.
Fish were visually sorted to genetic group by experienced personnel; misidentification of genetic group is rare for fish of the
size used in this study.
Although commercial producers report catfish mortalities associated with winter kill, visceral toxicosis of catfish (VTC), and
other diseases during winter months, the results of this study and
others suggest that there are no direct benefits of winter feeding
on catfish survival. However, there has been speculation that
winter feeding may indirectly reduce mortalities due to VTC,
which is believed to be caused by the ingestion of botulism
toxin that is present in fish carcasses decaying on pond bottoms
(Gaunt et al. 2007). Winter feeding could reduce the likelihood
that catfish will feed on decaying carcasses and may thereby reduce the incidence of VTC (Nanninga et al. 2011). In addition,
there is evidence that winter feeding improves survival during
the subsequent spring (Kim and Lovell 1995).
Initial values for trait means of genetic groups and sexes
are presented in Table 1; final values for trait means of genetic
groups, sexes, and feeding regimes are presented in Table 2.
Initial values are primarily presented as a reference to reflect
changes in traits over the course of the study. The main focus
of the discussion is related to the data that were collected at the
termination of the trial.
Hybrid catfish were larger at stocking (672 g) than blue
catfish (534 g) and channel catfish (520 g); genders did not
differ in initial weight. Percent weight change was affected
by genetic group and feeding regime; differential responses of
genetic groups to feeding regime resulted in a significant feeding regime genetic group interaction (Table 2). Gender did
not affect percent weight change. Among fed fish, blue catfish
had the lowest percent weight gain, channel catfish exhibited
TABLE 1. Least-squares means of initial values for viscera, head, carcass, skin, shank fillet, nugget, liver, intraperitoneal (IP) fat, and ovary yields (all values
are percentages relative to whole weight) from winter-fed and unfed blue catfish (BC), channel catfish (CC), and channel catfish blue catfish hybrids (HC). For
a given effect (genetic group, gender, or interaction) and a given yield characteristic, values with different letters are significantly different (P < 0.05). Significant
effects are summarized (G = genetic group; S = gender [sex]; G S = genetic group gender interaction).
Effect, group, or statistic

Genetic group
BC (n = 50)
CC (n = 50)
HC (n = 50)
SE
Gender
Female (n = 55)
Male (n = 95)
SE
Genetic group gender
BC (n = 18)
BC (n = 33)
CC (n = 17)
CC (n = 33)
HC (n = 20)
HC (n = 29)
SE
Significant effects
Viscera
Head
Carcass
Skin
10.3 z
9.7 y
9.9 zy
0.2
19.5 z
24.1 y
20.4 x
0.3
63.5
60.1
63.3
0.3
6.6
6.2
6.3
0.2
10.1
9.8
0.2
20.5 z
22.2 y
0.2
63.1 z
61.5 y
0.3
6.3
6.5
0.2
19.4 z
19.6 z
22.5 y
25.7 x
19.6 z
21.2 z
0.4
G, S, G S
63.9 z
63.2 z
61.6 x
58.5 w
63.8 z
62.9 z
0.5
G, S, G S
6.6
6.7
5.9
6.4
6.3
6.3
0.3
10.2 z
10.5 z
10.0 zy
9.4 y
10.3 z
9.6 y
0.3
G, G S
Shank fillet
Nugget
Liver
IP fat
Ovary
37.3 z
36.9 z
39.2 y
0.3
10.9 z
9.1 y
9.2 y
0.2
1.03 z
1.22 y
1.08 z
0.05
3.79 z
2.81 y
4.58 x
0.16
0.20 z
1.20 y
0.51 z
0.19
38.4 z
37.2 y
0.3
9.7
9.8
0.1
1.12
1.11
0.04
3.70
3.76
0.13
10.8
11.0
9.1
9.1
9.1
9.3
0.2
G
1.02
1.05
1.28
1.17
1.05
1.12
0.07
G
3.70
3.88
2.73
2.89
4.67
4.50
0.22
G
37.5 zy
37.2 z
38.3 y
35.5 x
39.5 w
38.8 yw
0.5
G, S, G S
556
BOSWORTH
TABLE 2. Least-squares means of final values for weight change (%) and for viscera, head, carcass, skin, shank fillet, nugget, liver, intraperitoneal (IP) fat, and
ovary yields (all yield values are percentages relative to whole weight) from winter-fed and unfed blue catfish (BC), channel catfish (CC), and channel catfish blue
catfish hybrids (HC). For a given effect (genetic group, gender, feeding regime, or interaction) and a given yield characteristic, values with different letters are
significantly different (P < 0.05). Significant effects are summarized (G = genetic group; F = feeding regime; S = gender [sex]; F G = feeding regime genetic
group interaction; G S = genetic group gender interaction; F G S = feeding regime genetic group gender interaction).
Effect, group,
or statistic
Weight
change
Viscera
Genetic group
BC (n = 126)
1.3 z
9.7 z
CC (n = 118)
5.0 y
9.3 y
HC (n = 122)
8.1 x
9.1 y
SE
0.7
0.1
Gender
Female (n = 149)
4.0
9.8 z
Male (n = 217)
4.0
8.9 y
SE
0.5
0.1
Feeding regime
Fed (n = 191)
11.4 z
9.8 z
Unfed (n = 175)
3.4 y
8.9 y
SE
0.7
0.1
Feeding regime genetic group gender
Fed BC (n = 24)
5.3 z
10.1 z
Fed BC (n = 40)
5.6 z
9.7 zx
Fed CC (n = 28)
11.3 y
11.0 y
Fed CC (n = 33)
13.2 y
8.6 x
Fed HC (n = 28)
17.5 x
10.2 z
Fed HC (n = 38)
15.6 yx
9.3 x
Unfed BC (n = 31)
7.8 w
9.7 zx
Unfed BC (n = 43) 7.6 w
9.1 x
1.9 v
9.0 x
Unfed CC (n = 25)
Unfed CC (n = 32) 2.5 v
8.6 x
Unfed HC (n = 25) 0.8 v
8.9 x
Unfed HC (n = 31)
0.00 v
7.9 w
SE
1.8
0.3
Significant effects
G, F,
F, G, S,
FG
G S,
FGS
Head
Carcass
Skin
Shank
fillet
Nugget
Liver
IP fat
Ovary
0.23 z
1.83 y
0.65 z
0.22
19.5 z
23.8 y
20.2 x
0.2
63.3 z
59.9 y
64.2 x
0.2
7.6 z
7.0 y
6.5 x
0.1
36.2 z
35.6 y
38.3 x
0.2
10.6 z
8.7 y
9.2 x
0.1
0.96 z
1.20 y
1.06 z
0.04
3.2 z
1.5 y
3.1 z
0.1
20.4 z
21.9 y
0.2
62.8 z
62.1 y
0.2
6.9
7.1
0.1
37.2 z
36.3 y
0.2
9.3 z
9.6 y
0.1
1.12 z
1.02 y
0.03
2.7
2.6
0.1
20.5 z
21.8 y
0.2
62.9 z
62.0 y
0.2
6.8 z
7.3 y
0.1
37.4 z
36.1 y
0.2
9.7 z
9.2 y
0.1
1.15 z
0.99 y
0.04
2.9 z
2.3 y
0.1
18.5 z
19.4 zy
22.0 x
24.3 v
18.8 z
20.0 zy
19.8 zy
20.2 zy
23.2 w
25.6 u
20.2 zy
21.7 x
0.4
F, G, S,
GS
64.2 z
7.2 z
63.7 z
7.3 z
60.5 y
6.5 y
60.1 y
6.9 zy
64.7 z
6.3 yx
64.5 z
6.3 yx
62.4 x
8.0 w
62.8 x
7.8 w
60.8 y
6.9 zy
58.1 w
7.6 w
64.2 z
6.6 y
63.5 z
6.8 zy
0.4
0.2
F, G, S, F, G, S,
G S,
GS
FGS
intermediate weight gain, and hybrid catfish had the greatest

weight gain (Table 2). Grant and Robinette (1992) also reported
a similar pattern of better growth in winter-fed channel catfish
(18% increase in body weight) than in winter-fed blue catfish
(9% increase in body weight). In the present study, fed hybrid
catfish had the highest percent increase in weight, supporting
anecdotal observations that hybrids feed more aggressively in
cold water than blue catfish or channel catfish. However, the
difference between hybrids and channel catfish was not large
(4.4%).
The differences in growth among genetic groups in response
to winter feeding may be associated with inherent biological or
behavioral differences among the groups. Feeding activity and
37.3 z
37.0 z
36.9 z
35.5 y
39.3 x
38.3 w
35.3 y
35.3 y
36.2 zy
33.9 v
37.8 zw
37.8 zw
0.4
F, G, S,
GS
11.0 z
1.00
3.4
10.8 z
1.06
3.3
8.4 y
1.44
2.0
9.3 x
1.14
1.7
9.1 x
1.18
3.5
9.7 w
1.07
3.8
10.1 wv 0.92
3.4
10.3 v
0.86
2.8
8.5 y
1.12
1.2
8.6 y
1.08
1.2
8.7 y
1.06
2.7
9.2 x
0.92
2.6
0.2
0.08
0.2
F, G,
F, G, S F, G,
GS
FG
0.93
0.88
0.19
0.02 z
2.27 y
0.50 zx
0.44 zx
1.38 w
0.82 x
0.30
G, F
G
appetite at cold temperatures may simply be reduced in blue

catfish relative to channel catfish and hybrid catfish. Another
possibility is that because blue catfish tend to school more tightly
and to stay higher in the water column than channel catfish, they
are less likely to scavenge for sinking feed from the pond bottom.
Competition for food in a co-culture situation is another potential cause of growth differences among genetic groups of
winter-fed fish. Blue catfish may be less-aggressive feeders than
channel catfish or hybrids and may have been outcompeted for
the feed. However, Grant and Robinette (1992) also reported
slower growth for winter-fed blue catfish than for channel catfish when the two groups were reared separately. In addition,
the feeding rate used in the present study (2% of body weight/d
twice per week) is considerably higher than that suggested for

winter feeding of catfish (Robinson et al. 2001). Therefore, it
seems unlikely that competition for food among co-cultured
genetic groups resulted in the differential growth response to
winter feeding.
Another possible factor resulting in the winter growth differences among genetic groups was competition related to fish size.
Hybrids grew the fastest and were also larger at stocking, and
large fish typically dominate feeding activity. However, channel catfish and blue catfish were similar in size at stocking, but
the fed blue catfish had substantially lower growth than the fed
channel catfish. In addition, there was no effect of initial weight
on percent gain for fed fish within each genetic group; this result indicates that, at least within genetic groups, there were
no substantial growth differences related to size-based competition. Hybrid catfish typically grow faster than channel catfish
and blue catfish when the three groups are grown separately at
warm water temperatures. Taken as a whole, it seems that the
differences among groups were primarily attributable to genetic
differences rather than to size-based competition, although the
effects of stocking weight on growth cannot be determined given
the co-culture study design.
Unfed blue catfish lost more weight than unfed channel
catfish or unfed hybrids. The fairly low weight losses for unfed channel catfish and hybrids were unexpected, as overwinter weight losses in unfed channel catfish are generally reported to be in the 510% range (Lovell and Sirikul 1975;
Kim and Lovell 1995). One possible explanation for the relatively low weight loss in unfed hybrids and channel catfish
was the presence of significant numbers of small crayfish observed in the ponds at harvest. Ponds were drained and dried
prior to stocking and no crayfish were noticed, but crayfish
typically hatch and grow during the winter; therefore, it is possible that catfish in both unfed and fed ponds were consuming crayfish. However, because crayfish occurred in all ponds,
their presence was unlikely to have caused the large differences in growth among genetic groups and between feeding
regimes.
Although the reported magnitude of benefit from winter feeding (i.e., in terms of catfish weight gain) is inconsistent among
studies, all studies report greater weight loss in unfed fish than
in fed fish. The differences in the magnitude of response to
winter feeding may be related to the frequency and duration
of feeding. In studies reporting a weight increase in response to
winter feeding, fish are typically fed two or more times per week
(Reagan and Robinette 1979; Grant and Robinette 1992;
Burtle and Newton 1993; Kim and Lovell 1995; present study).
In studies that have reported weight loss, fish were fed based
on a schedule that required higher water temperatures to permit
feeding, thereby reducing the number of times the fish were fed
(Tidwell et al. 1995; Nanninga et al. 2011). In several studies
that have demonstrated large weight gains in catfish (Reagan
and Robinette 1979; Burtle and Newton 1993; Kim and Lovell
1995), the time frame defined as winter included fairly warm
557
months (October, March, and/or April), which may have allowed

for greater feeding activity and higher growth.
The FCR of fed fish in this study was 6.2, which was within
the range of FCRs (312) that have been reported for other winter feeding trials with channel catfish and blue catfish (Reagan
and Robinette 1979; Robinette et al. 1985; Grant and Robinette
1992; Kim and Lovell 1995). The FCRs reported for winter
feeding are considerably higher than those typical for the normal feeding season (2.02.5). High FCR (and thus high feed
costs relative to fish weight gain) is one of the reasons why
many catfish producers do not provide feed during the winter.
At current catfish feed prices (US$425.00 per ton) and an FCR
of 6.2, the cost of feed to produce 1 kg of fish is approximately
$2.90, nearly identical to the current pond-bank catfish price per
kilogram. Development of new feeds for winter feeding or the
incorporation of winter feeding into new, high-fish-density production systems (e.g., partitioned aquaculture systems; Brune
et al. 2004) may result in improved FCRs and better economics
for winter feeding. High-density systems may improve winter
FCRs because the fish are confined to a small area and may feed
more efficiently.
Groups were co-cultured, and therefore it was not possible
to determine the effects of genetic group or gender on FCR.
However, the poor growth of winter-fed blue catfish suggests
that those fish would have had poorer FCRs than fed channel
catfish or hybrids. Grant and Robinette (1992) reported FCRs
of 11.59 and 5.89 for winter-fed blue catfish and channel catfish, respectively, supporting the idea that blue catfish will have
poorer FCRs than channel catfish during winter feeding. Therefore, I recommend either (1) withholding feed from blue catfish
during winter or (2) applying a lower winter feeding rate for
blue catfish than for channel catfish or hybrid catfish.
Most of the initial and final values for body composition and
processing yield traits were affected by genetic group, gender,
and feeding regime; significant interactions between genetic
group and gender and among feeding regime, genetic group,
and gender were common (Tables 1, 2). Head yields at the start
and end of the study were lowest for blue catfish, intermediate for hybrids, and highest for channel catfish. Male channel
catfish had larger heads than females, whereas the difference
between male and female hybrids was intermediate and the
difference between male and female blue catfish was not significant; thus, a gender genetic group interaction was observed
at the start and end of the study. Fed fish had lower head yield
than unfed fish. Head size is unlikely to actually increase, but
during feed restriction the size of the head (which is primarily
bone) remains relatively constant while other body components
(fat, muscle, etc.) are mobilized and utilized for energy needs;
thus, the head weight expressed as a proportion of whole-body
weight increases in unfed fish. The same pattern of gender, genetic group, and gender genetic group interaction effects on
head yield has also been reported previously for channel catfish,
blue catfish, and hybrid catfish (Bosworth et al. 2004; Jiang
et al. 2008). Bosworth and Wolters (2005) observed that after a
558
BOSWORTH
period of feed restriction in the fall, unfed channel catfish had

a higher head yield than fed fish, similar to the results of the
present study.
At the start and end of the study, percent yield of viscera
was higher for blue catfish than for channel catfish and hybrids.
There was a consistent gender genetic group effect across
time due to channel catfish and hybrid females having significantly more viscera than males, whereas the effects of gender
on yield of viscera were minimal in blue catfish. Winter feeding resulted in consistently higher viscera yield across genetic
groups. Blue catfish and hybrids had more intraperitoneal fat
than channel catfish at the start and end of the study, and fed fish
had more intraperitoneal fat than unfed fish. Gender did not affect intraperitoneal fat level. Intraperitoneal fat levels increased
as body weight increased in blue catfish and hybrids. Initial
and final liver yields were higher for channel catfish than for
blue catfish and hybrids. Females had larger livers than males
throughout the study, and fed fish had larger livers than unfed
fish. Channel catfish females had larger ovaries than hybrid or
blue catfish females at the start and end of the study. A feeding regime genetic group interaction was observed due to the
higher ovary yield in fed channel catfish females relative to unfed females, while feeding regime had no effect on ovary yield
of hybrid or blue catfish females. The observed differences in
ovary yield among blue catfish, channel catfish, and hybrids are
similar to those reported previously (Grant and Robinette 1992;
Bosworth et al. 2004) and are likely the result of blue catfish
maturing at a later age than channel catfish (Graham 1999).
Bosworth and Wolters (2005) reported that fall feed restriction
of channel catfish resulted in effects on viscera yield similar
to those observed in the current study (i.e., higher viscera, intraperitoneal fat, liver, and ovary yields in fed fish than in unfed
fish).
At the start of the study, the yield of deheadedgutted
skinned carcasses (marketed as whole fish by catfish processors) was higher for blue catfish and hybrid catfish than for
channel catfish; at the end of the study, the carcass yield was
highest for hybrids, intermediate for blue catfish, and lowest
for channel catfish. Fed fish had a higher carcass yield than
unfed fish, but the genetic group feeding regime interaction
was significant due to the greater effect of feeding regime on
carcass yield for blue catfish than for hybrids or channel catfish.
Females had higher carcass yields than males, but a significant
genetic group gender interaction was present at both the start
and end of the study because the effect of gender on carcass
yield was greater for channel catfish than for blue catfish or
hybrid catfish. Higher carcass yields for hybrids and blue catfish in comparison with channel catfish have been reported in
other studies (Grant and Robinette 1992; Argue et al. 2003;
Bosworth et al. 2004; Li et al. 2004; Jiang et al. 2008). The
smaller head size in hybrids and blue catfish relative to channel catfish appears to be the main factor influencing the higher
carcass yields for hybrids and blue catfish in comparison with
channel catfish. The observed higher carcass yield for females
than for males in the channel catfish and hybrid groups was
also reported by Bosworth et al. (2004). Effects of feed restriction on carcass yield have been inconsistent, with some reports
showing a response of lower carcass yield in feed-restricted catfish (i.e., as was observed in the present study; Kim and Lovell
1995; Li et al. 2004, 2006) and others showing no effect of
feed restriction on carcass yield (Bosworth and Wolters 2005;
Nanninga et al. 2011).
Initial yield of shank fillets, the highest value and highest
volume product sold by catfish processors, was greater for hybrid catfish than for blue catfish and channel catfish, which had
similar fillet yields. At the end of the study, hybrids had the
highest shank fillet yield, blue catfish had an intermediate fillet
yield, and channel catfish had the lowest fillet yield. Females
had a higher fillet yield than males at the start and end of the
study. Gender genetic group interactions for fillet yield were
consistent across time and resulted from gender-based differences being large for channel catfish, intermediate for hybrids,
and nonsignificant for blue catfish. Fed fish had a higher fillet
yield than unfed fish when averaged across genetic groups and
genders. Differences in fillet yield among channel catfish, blue
catfish, and hybrids were similar to those reported previously by
Bosworth et al. (2004) and Jiang et al. (2008) and indicate that
hybrid catfish generally exhibit a fillet yield that is superior to
the yield obtained from blue catfish or channel catfish. Similar
to the results of this study, previous studies have demonstrated
that restricting feeding typically reduces fillet yield in catfish
(Li et al. 2004, 2006; Bosworth and Wolters 2005). Higher fillet
yields for female channel catfish and female hybrids relative to
males have also been reported (Bosworth et al. 2004; Bosworth
and Wolters 2005).
Nugget yield was higher for blue catfish than for channel
catfish and hybrids at the start and end of the study. Fed fish had
a higher nugget yield than unfed fish. Males had a higher nugget
yield than females at the end of the study, but nugget yield
did not differ between genders at the start of the study. Male
channel catfish and male hybrids had higher nugget yields than
females, while gender did not affect nugget yield in blue catfish,
thus resulting in a significant gender genetic group effect.
Bosworth et al. (2004) and Jiang et al. (2008) also reported
a higher nugget yield in blue catfish relative to other genetic
groups.
In summary, winter feeding had positive effects on growth
and processing yield of channel catfish, blue catfish, and hybrid catfish. However, the benefits of winter feeding on growth
and processing yield were greater in channel catfish and hybrids than in blue catfish. Hybrid catfish had the best combined response (i.e., growth and processing yield) to winter
feeding. Producers and processors would realize the most benefit from winter-fed hybrid catfish relative to winter-fed channel catfish or blue catfish. However, actual economic benefits
will depend on the price of fingerlings (hybrid fingerlings cost
more) and the development of feeding methods that will reduce FCRs for winter feeding. The issue that remains to be
addressed is how to efficiently and economically feed catfish

during winter, when feeding activity is sporadic and not easily
observed.
REFERENCES
Argue B. J., Z. Liu, and R. A. Dunham. 2003. Dressout and fillet yields of
channel catfish, Ictalurus punctatus, blue catfish, Ictalurus furcatus, and their
F1, F2 and backcross hybrids. Aquaculture 228:8190.
Bosworth, B. G., and W. R. Wolters. 2005. Effects of short-term feed restriction
on production, processing, and body shape traits in market-weight channel
catfish, Ictalurus punctatus. Aquaculture Research 36:344351.
Bosworth, B. G., W. R. Wolters, J. L. Silva, R. S. Chamul, and S. Park. 2004.
Comparison of production, meat yield, and meat quality traits of NWAC 103
line channel catfish (Ictalurus punctatus), Norris line channel catfish, and
channel catfish female blue catfish male (I. furcatus) F1 hybrids. North
Brune, D. E., G. Schwartz, A. G. Eversole, J. A. Collier, and T. E. Schwedler.
2004. Partitioned Aquaculture Systems. Southern Regional Aquaculture Center, Publication 4500, Stoneville, Mississippi.
Burtle, G. J., and L. G. Newton. 1993. Winter feeding frequency for channel
catfish in cages. Progressive Fish-Culturist 55:137139.
Gaunt, P. S., S. R. Kalb, and J. R. Barr. 2007. Detection of botulinum type E toxin
in channel catfish with visceral toxicosis syndrome using catfish bioassay and
endopep mass spectrometry. Journal of Veterinary Diagnostic Investigation
19:349354.
Graham, K. 1999. The review of the biology and management of blue catfish.
Pages 3749 in E. R. Irwin, W. A. Hubert, C. F. Rabeni, H. L. Schramm
Jr., and T. Coon, editors. Catfish 2000: proceedings of the international ictalurid symposium. American Fisheries Society, Symposium 24, Bethesda,
Maryland.
Grant, J. C., and H. R. Robinette. 1992. Commercially important traits of blue
and channel catfish as related to second summer, winter, and third summer
growth. Aquaculture 105:3745.
Heidinger, R. C. 1975. Growth of hybrid sunfishes and channel catfish at
low temperatures. Transactions of the American Fisheries Society 104:333
334.
559
Jiang M., W. H. Daniels, H. J. Pine, and J. A. Chappell. 2008. Production

and processing trait comparisons of channel catfish, blue catfish, and their
hybrids grown in earthen ponds. Journal of the World Aquaculture Society
39:736745.
Kim, M. K., and R. T. Lovell. 1995. Effect of overwinter feeding regime on body
weight, body composition and resistance to Edwardsiella ictaluri in channel
catfish, Ictalurus punctatus. Aquaculture 134:237246.
Li, M. H., E. H. Robinson, B. B. Manning, and B. G. Bosworth. 2004. Effect
of dietary protein concentration on production characteristics of pond-raised
channel catfish Ictalurus punctatus fed once daily or once every other day to
satiation. North American Journal of Aquaculture 66:184190.
Li, M. H., E. H. Robinson, D. F. Oberle, and B. G. Bosworth. 2006. Effects of dietary protein concentration and feeding regime on channel catfish,
Ictalurus punctatus, production. Journal of the World Aquaculture Society 37:
370377.
Lovell, R. T., and B. Sirikul. 1975. Winter feeding of channel catfish. Proceedings of the Annual Conference Southeastern Association of Game and Fish
Commissioners 28(1974):208216.
Nanninga, A. S., C. R. Engle, N. Stone, and A. E. Goodwin. 2011. Effects of
winter feeding of channel catfish on production in multibatch systems. North
American Journal of Aquaculture 73:1, 6067.
Reagan, R. E., and R. H. Robinette. 1979. Feeding of channel catfish fingerlings
in mild and severe winters in Mississippi. Proceedings of the Annual Conference Southeastern Association of Fish and Wildlife Agencies 32(1978):426
428.
Robinette, R. H., R. L. Busch, S. H. Newton, C. J. Haskins, S. Davis, and R. R.
Stickney.1985. Winter feeding of channel catfish in Mississippi, Arkansas,
and Texas. Proceedings of the Annual Conference Southeastern Association
of Fish and Wildlife Agencies 36(1982):162171.
Robinson, E. H., M. H. Li, and B. B. Manning. 2001. A practical guide to
nutrition, feeds, and feeding (second revision). Mississippi Agricultural and
Forestry Experiment Station Technical Bulletin 1113.
Tidwell, J. H., C. D. Webster, and S. D. Mims. 1995. Effects of two dietary protein levels on body weight and composition of juvenile blue
and channel catfish during the winter. Journal of Applied Aquaculture 5:
4554.
Tucker, C. S., and J. A. Hargreaves. 2004. Biology and culture of channel catfish.
Elsevier, Amsterdam.


Efficacy and Physiological Responses of Grass Carp to

Different Sedation Techniques: I. Effects of Various
Chemicals on Sedation and Blood Chemistry
a
Brian R. Gause , Jesse T. Trushenski , John C. Bowzer & James D. Bowker
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125
Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois, 62901-6511, USA
b
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050
Bridger Canyon Road, Bozeman, Montana, 59715, USA
To cite this article: Brian R. Gause, Jesse T. Trushenski, John C. Bowzer & James D. Bowker (2012): Efficacy and Physiological
Responses of Grass Carp to Different Sedation Techniques: I. Effects of Various Chemicals on Sedation and Blood Chemistry,
North American Journal of Aquaculture, 74:4, 560-566


DOI: 10.1080/15222055.2012.691013
TECHNICAL NOTE
Efficacy and Physiological Responses of Grass Carp

to Different Sedation Techniques: I. Effects of Various
Chemicals on Sedation and Blood Chemistry
Brian R. Gause, Jesse T. Trushenski,* and John C. Bowzer
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive,
Life Science II, Room 173, Carbondale, Illinois 62901-6511, USA
James D. Bowker
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program,
4050 Bridger Canyon Road, Bozeman, Montana 59715, USA
Abstract
Grass carp Ctenopharyngodon idella are commonly used as a
low cost, biological control for aquatic vegetation in aquaculture
ponds and other private and public waters. In order to minimize
the risk of establishing self-sustaining populations in U.S. waters,
many states now require grass carp be certified as triploid prior to
sale and stocking. To facilitate ploidy testing, grass carp are typically sedated before collecting blood samples. Chemical sedatives
such as tricaine methanesulfonate (MS-222) and carbon dioxide
(CO2 ) are most commonly used to sedate fish, but there is increasing interest in other chemical sedatives such as benzocaine
and eugenol. We evaluated time to induction to Stage IV sedation
and recovery, survival, and postsedation blood chemistry of grass
carp (301 8 g, mean SE) sedated with MS-222 (150 mg/L),
benzocaine (150 mg/L), eugenol (60 mg/L), or CO2 (400 mg/L).
Induction times for all sedatives excluding CO2 (14.9 min) were
less than 2.4 min (range, 1.52.4 min). Average recovery time after
induction was 5.8 min (range, 2.88.3 min) excluding benzocaine,
which had a recovery time of 15.4 min. Survival was high and
unaffected by sedative option. Plasma cortisol and lactate levels
peaked between 0.5 and 1 h postinduction before returning to resting levels at 6 h postinduction. No obvious changes were observed
in blood glucose or hematocrit. Each of the sedatives was effective
in sedating grass carp, and though changes in blood chemistry indicated that an acute stress response occurred, the response was
transient. Although each of the evaluated sedatives would facilitate ploidy testing, some strategies may be more appropriate than
others based on FDA approval status and access to the sedative
compound, handling time, withdrawal period, and on-site conditions and resources.
Grass carp Ctenopharyngodon idella are commonly used as a

low cost, biological control for aquatic vegetation in aquaculture
*Corresponding author: saluski@siu.edu
Received October 24, 2011; accepted April 13, 2012
560
ponds and other private and public waters (Masser 2002). Concerns regarding the establishment of self-sustaining populations
of this nonnative species have led to bans on stocking fertile,
diploid grass carp in many states (Kelly et al. 2011). Triploid
fish, rendered functionally sterile through the interfering effects of ploidy manipulation on gametogenesis (Benfey 1999;
Zajicek et al. 2011), may be legally stocked in some states, but
triploidy must be verified prior to sale and stocking in those
states allowing such fish (Zajicek et al. 2011). A rapid triploidy
verification test developed by Wattendorf (1986) requires only
a small blood sample for analysis to verify ploidy state. Fish
are typically sedated to facilitate blood sampling, but there are
a limited number of drugs or chemical sedatives currently available for this purpose that are approved by the U.S. Food and
Drug Administration (FDA) or are otherwise made available by
the FDA for use.
The only drug currently approved by the FDA for the temporary immobilization of fish is tricaine methanesulfonate, most
commonly referred to as MS-222. The use of MS-222 is limited
to ictalurids, salmonids, esocids, percids, or other laboratory and
hatchery fishes at water temperatures greater than 10 C. Users
of MS-222 must adhere to a 21-d withdrawal period prior to fish
being released or slaughtered for consumption. Fish must be
fed and kept healthy during this holding period, and in the case
of grass carp, must also be maintained in separate holding systems to maintain validity of the ploidy verification tests. Owing
to limitations of suitable holding tanks, it is often impractical
to hold segregated fish for an extended period of time and doing so probably contributes to additional costs for producers and
TECHNICAL NOTE
customers alike. Carbon dioxide (CO2 ) is not currently approved

by the FDA as a sedative for fishes, but is considered a drug of
low regulatory priority, meaning that the FDA is unlikely to
enforce regulations provided that CO2 is administered according
to the stipulations made in the FDAs Enforcement Priorities
for Drug Use in Aquaculture documentation (USFDA 2011).
Although CO2 has no withdrawal period, creating and maintaining intended sedative concentrations in the field can be difficult,
and is not effective for all fishes or environments (e.g., hypercapnia tolerant species, marine environments). A number of crude
and purified drugs (e.g., clove, spearmint, and wintergreen oils,
quinaldine) are widely used to sedate fishes in the field and laboratory. These compounds, however, are not FDA-approved and
can only be used in fish for research purposes, provided that all
treated fish are destroyed via incineration, burial, or some other
method to ensure they do not enter the food chain. As such,
many fisheries professionals have desired FDA approval of a
chemical sedative that is safe, effective, and affordable, and for
which a lengthy withdrawal period is not required.
Two drugs that are not currently approved by the FDA, benzocaine (Benzoak; 20% benzocaine; manufacturer: ACD Pharmacueticals AS, Leknes, Norway; U.S. distributor : Frontier Scientific, Logan, Utah) and eugenol (AQUI-S 20E [10% eugenol];
Aqui-S New Zealand, Lower Hutt, New Zealand), may be used
as sedatives under an Investigational New Animal Drug (INAD)
authorization held by the U.S. Fish and Wildlife Service. Although a 3-d withdrawal period is currently required under the
INAD authorization, both of these drugs are being investigated
as immediate release sedatives that would allow fish to be
released immediately after sedation and could be used on fish
intended for human consumption.
Although many studies have been conducted to evaluate the
effectiveness of chemical sedatives to sedate or anesthetize fish
(Gilderhus and Marking 1987; Pirhonen and Schreck 2003;
Davis and Griffin 2004), few studies have compared chemical sedatives side by side in terms of their efficacy and effects
on fishes (Sattari et al. 2009; Trushenski et al. 2012). Although
there is considerable interest in the use of drugs or chemicals
to sedate fish during some stage of production, the need for
an effective sedative is particularly great in triploid grass carp
production because of blood testing and ploidy verification requirements. Accordingly, we evaluated the effect of different
chemical sedatives (MS-222, benzocaine, eugenol, and CO2 )
on induction and recovery times, and on postsedation survival
and blood chemistry of grass carp.
METHODS
Sedation procedures.A reference population of triploid
grass carp (301 8 g and 30.9 0.3 cm total length, mean
SE) was held in an outdoor raceway configured as a partial
flow-through system (static raceway, periodically flushed with
screened surface water) with supplemental aeration at Keo Fish
Farm, Keo, Arkansas. Feed was withheld for 24 h prior to
561
sampling. Groups of 15 fish were randomly collected from the

reference population and transferred into a sedation chamber
(114-L cooler) filled with 70 L of culture water (water depth
of 10 cm) containing a sedative solution. Sedatives were applied under static conditions as follows: CO2 , 400-mg/L solutions prepared according to the sodium bicarbonatesulfuric
acid method described by Post (1979) (analytically verified as
360 mg/L); 150 mg/L benzocaine (Benzoak; 20% benzocaine;
manufacturer: ACD Pharmacueticals AS, Leknes, Norway; U.S.
distributor: Frontier Scientific, Logan, Utah); 60 mg/L eugenol
(AQUI-SE; 50% eugenol; manufacturer: Aqui-S New Zealand,
Lower Hutt, New Zealand); and a 150-mg/L solution of MS222 (Finquel; Argent Chemical, Redmond, Washington). Culture water used to prepare all baths was aerated before use, but
baths were not aerated after the addition of the chemical sedative
or during use.
Composite water samples, collected by combining aliquots
collected from the sedative baths before and after use, were
analyzed in duplicate along with water collected from the holding system for the following: temperature, dissolved oxygen
(YSI 550 meter, Yellow Springs Instruments, Yellow Springs,
Ohio) conductivity, pH, salinity (Multi-Parameter PCSTestr
35, Eutech/Oakton Instruments, Vernon Hills, Illinois), hardness, alkalinity (digital titrator and reagents, Hach, Loveland,
Colorado), and total ammonia nitrogen, nitrite-nitrogen, and
nitrate-nitrogen (DR 2800 spectrophotometer and reagents,
Hach). All measured water quality characteristics were within
ranges appropriate for grass carp (Masser 2002) at the start of
the experiment (Table 1).
Fish were monitored during sedation to determine induction
to Stage IV of anesthesia (Summerfelt and Smith 1990; although we have elected to use the term sedation, anesthesia
is the term used by these authors). Stage IV is associated with
the total loss of equilibrium, muscle tone, and responsiveness
to visual and tactile stimuli, but maintenance of a slow, steady,
opercular ventilation rate. After the loss of equilibrium, sedation
was verified by slight manual pressure along the trunk and caudal peduncle as a tactile stimulus. Fish were considered sedated
when they no longer responded to this stimulus, but the opercular ventilation rate remained steady, albeit reduced, relative to
unsedated fish. After induction, blood samples were collected
from three fish (t = 0; see below) from the caudal vasculature using heparinized, evacuated, blood collection assemblies
(Vacutainer, Becton Dickinson, Franklin Lakes, New Jersey).
The remaining 12 fish were then monitored to determine recovery of normal equilibrium and tactile responsiveness. When all
fish were able to maintain equilibrium and were responsive to
tactile stimulus, the group was considered fully recovered (i.e.,
recovery time = time for last fish to recover). Since assessment
of induction and recovery can be somewhat subjective, bias was
minimized by having the same observer apply all stimuli and assess when fish were sedated and recovered. Recovered fish were
transferred to a second raceway configured in the same manner
as the one housing the reference population; fish in different
562
GAUSE ET AL.
TABLE 1. Water quality characteristics measured during the trial. Values represent means of water samples analyzed in duplicate.
Characteristic
Temperature ( C)
Total ammonia nitrogen (mg/L)
Nitrite-nitrogen (mg/L)
Nitrate-nitrogen (mg/L)
Alkalinity (mg/L, as CaCO3 )
Hardness (mg/L)
Salinity ()
Conductivity (S/cm)
pH
Holding system
CO2
MS-222
Benzocaine
Eugenol
16.1
9.62
0.00
0.004
0.75
228
450
0.425
876
8.22
15.8
8.73
0.02
0.003
0.95
248
488
0.834
1,681
6.32
16.0
9.57
0.04
0.003
0.95
208
458
0.432
897
7.24
15.9
9.66
0.09
0.004
0.9
230
468
0.426
882
8.27
15.8
9.46
0.71a
0.008
1.6
226
440
0.426
880
8.18
a
The presence of eugenol results in a yellowgreen color to the water, which can interfere with the Nessler ammonia method used in this trial. Similar results were observed in
previous trials using eugenol (Trushenski et al. 2012).
treatment groups were separated by means of raceway dividers

crowders positioned at 1-m intervals along the length of the
recovery raceway.
Sample collection and analysis.Blood samples were then
collected from three fish per group at t = 0.5, 1, 2, and 6 h
postsedation (three fish per group per time point, individuals were only sampled once). To facilitate handling, all fish
were immersed in a bath of metomidate hydrochloride (Aquacalm, Western Chemical, Ferndale, Washington; 510 mg/L
for 30 s), a fish sedative known to minimize corticosteroid
increase during sampling (Olsen et al. 1995; Davis and Griffin
2004), before sampling; although additional sedation was not
necessary for fish sampled at t = 0, these fish were also treated
with metomidate hydrochloride to ensure consistent treatment
of all fish. Once fish were sedated to a stage where they were
easily handled, they were weighed (to the nearest gram) and
measured (total length to the nearest 0.5 cm), and a blood sample was collected as described previously. Although metomidate
hydrochloride was used to sedate fish before collecting blood
samples, all samples were collected within 5 min of capture
to minimize the possibility of other confounding responses of
handling and blood sample collection as additional stressors. To
establish resting blood chemistry characteristics of nonsedated
fish, two fish from the reference population were sampled every
hour during the course of the experiment (n = 14). After blood
collection, fish were returned to a separate area in the recovery
raceway and monitored for 24 h.
Blood samples were kept on wet ice (<36 h) until analysis. Although this is a somewhat lengthy period of time to
hold blood samples prior to analysis, some assays could not be
immediately conducted in the field and samples had to be transported back to the Fisheries and Illinois Aquaculture Center,
Carbondale, Illinois. It is possible that levels of metabolically relevant molecules (e.g., glucose and lactate) could
have changed slightly during this holding period; however, all
samples were treated in the same manner to ensure validity
of comparisons among treatments. Hematocrit (Statspin cen-
trifuge, Fisher Scientific, Pittsburgh, Pennsylvania) and glucose

(Freestyle Freedom Lite glucose meter, Abbott Laboratories,
Abbott Park, Illinois) were determined using aliquots of whole
blood, and then the remaining whole blood was centrifuged
(3000 g, 4 C, 45 min). Resultant plasma was collected and
stored at 80 C until further analysis. Plasma samples were
analyzed to determine lactate (Accutrend lactate meter, Roche,
Mannheim, Germany), osmolality (Vapro 5520, Wescor, Logan,
Utah), and cortisol (EIA 1887, DRG International, Mountainside, New Jersey). Although portable lactate and glucose meters such as those used in this study can slightly underestimate
metabolite levels in fish blood relative to laboratory methods,
they are considered precise and reliable for use in generating
comparative data (Wells and Pankhurst 1999; Venn Beecham
et al. 2006). The cortisol kit used has a range of 0800 ng/mL
with a sensitivity of 2.5 ng/mL for human samples, and has
been validated and used successfully to measure cortisol in
samples from a variety of fish species including Nile tilapia
Oreochromis niloticus (Delaney et al. 2005), tench Tinca tinca
(Owen et al. 2009), common carp Cyprinus carpio (SepiciDincel et al. 2009), cobia Rachycentron canadum (Trushenski
et al. 2010), striped bass Morone saxatilis (Woods et al. 2008),
and hybrid striped bass (female white bass M. chrysops male
striped bass) (Trushenski et al. 2012).
Although multiple fish were sampled from each treatment
group at each time point, individuals were group-sedated and
housed together after sedation. Therefore, it was determined that
individuals did not represent independent observations. Since
the experiment lacked true replication, no quantitative statistical
analysis was performed and only qualitative comparisons were
made from summary statistical values.
RESULTS
All fish were successfully induced to Stage IV sedation; however, the observed induction and recovery times varied among
sedatives (Figure 1). With the exception of CO2 (induction
563
TECHNICAL NOTE
E R
E R
Benzocaine
CO2
Eugenol
MS-222
10
12
14
16
18
20
Time (min)
I = Induced to Phase IV Sedation
FIGURE 1.
E = Maintain Equilibrium
R = Recovery; Respond to Tactile Stimulus
Schematic illustrating induction and various stages of recovery of grass carp sedated to Stage IV sedation using various chemical sedatives.
time = 14.9 min), all sedative treatments were successful in

inducing Stage IV sedation within approximately 2 min (mean
excluding CO2 = 1.8 min; range, 1.52.3 min). Once sedation was achieved, time to regain equilibrium (mean = 4.5 min
postinduction; range, 2.58.0 min) was less variable than time to
recover tactile responsiveness. With the exception of benzocaine
(regained tactile responsiveness = 15.4 min), fish regained tactile responsiveness within 5.8 min (range, 2.88.3 min). All fish
ultimately recovered and no fish died during the 24-h postsedation observation period.
Physiological responses varied among the sedatives evaluated at different time points following sedation. Plasma lactate
varied among treatments at each time point from t = 0 to t =
2 h by as much as 8.5 mmol/L (t = 2 h,) with maximum concentrations observed at t = 0.5 h for some sedatives (eugenol >
MS-222) and t = 0 h for the others (CO2 > benzocaine). The
range of plasma lactate concentrations at t = 6 was considerably narrower (1.12.7 mmol/L) than that observed at other
time points. Plasma osmolality varied at each time point with
an overall range of 278361 mOsm/kg (reference population =
296 mOsm/kg). Peak plasma cortisol concentrations occurred
at t = 0 for CO2 (164 ng/mL) and at t = 0.5 h for all other sedatives (73166 ng/mL). Cortisol levels in sedated fish appeared
to be approaching the levels observed in fish from the reference
population (22 ng/mL) by t = 6 h, except for those in fish sedated with MS-222, in which levels increased slightly from t =
2 h to t = 6 h. Hematocrit (range, 2029%; reference population = 22%) and blood glucose (range, 6198 mg/dL; reference
population = 79 mg/dL) did not vary much among sedative
treatments at any time point. No fish died during the study.
During sedation with CO2 , fish were observed piping at the
surface (appeared to be gasping for air) and had to be routinely
pushed back down into the water to prevent attempts to avoid
CO2 narcosis via air breathing. Slight petechial hemorrhaging
was observed along the lower flank and opercular area in a
few fish before and after sedation, but the occurrence of these
hemorrhages did not appear to be related to the sedative used.
DISCUSSION
Induction times for three of the four sedatives were considered relatively rapid and would probably be considered acceptable to fisheries professionals for sedation of grass carp.
Induction time for the CO2 dose used was nearly seven times
longer than that for the other chemical sedatives. This lengthy
induction time was probably due to the low oxygen demands
and metabolic rates of grass carp (Fu et al. 2009) and their ability to avoid CO2 narcosis via air breathing. Furthermore, grass
carp used in the current study were held at cooler water temperatures (15.816.1 C) than the water temperature that grass
carp tend to prefer (2130 C, Masser 2002). The cooler water
temperature associated with the present study probably reduced
GAUSE ET AL.
564
FIGURE 2. Time course of hematological responses (A = plasma cortisol, B = blood glucose, C = hematocrit, D = plasma lactate, and E = plasma osmolality)
of grass carp following chemical sedation. Points represent means SE; grey reference bars represent means of values observed for fish sampled from the
reference population throughout the course of the experiment.
TECHNICAL NOTE
the resting metabolic rate and oxygen demand of our test fish
even further. Reduced oxygen demand combined with the relatively high environmental oxygen concentrations (>8 mg/L),
may have reduced the need for respiratory gas exchange, opercular ventilation, or gill perfusion rates (Itazawa and Takeda
1978), which may have affected CO2 uptake and the subsequent
times to sedation. Additionally, CO2 sedation times may have
been slowed because fish were observed piping at the surface
and had to be routinely pushed back below the surface of the
water to ensure that fish were constantly exposed to the CO2 treated water.
Grass carp treated with CO2 and eugenol regained tactile
responsiveness and recovered quickly after first regaining equilibrium (<0.3 min). Grass carp in the benzocaine and MS-222
treatments, on the other hand, required additional time (10.4
and 3.7 min, respectively) to recover from sedation after gaining equilibrium. The chemical similarity of benzocaine (ethyl
para-aminobenzoate) and MS-222 (ethyl meta-aminobenzoate;
Kiessling et al. 2009) may partially explain why grass carp sedated with these chemicals had similar patterns in which full recovery was preceded by regaining equilibrium by at least several
minutes. Recovery results observed for benzocaine and MS-222
in grass carp were considerably longer than that observed in a
previous study conducted by Trushenski et al. (2012) using hybrid striped bass, in which hybrid striped bass fully recovered in
less than 2 min after regaining equilibrium. The pattern of recovery observed when fish were sedated with CO2 and eugenol, in
which fish fully recovered within seconds of regaining equilibrium, was also observed in a study conducted concurrently with
ours by Bowzer et al. (2012, this issue), in which grass carp were
electrosedated with varying voltages and exposure durations. It
is unclear whether the observed differences in recovery times
between grass carp and hybrid striped bass were influenced by
differences in resting metabolic rate or other intertaxonomic
differences, differential rates of chemical sedative metabolism
and excretion during recovery, or some combination of these
factors.
Physiological responses generally followed the pattern of the
generalized stress response (Barton 2002) suggesting that sedation should be considered as a stressor (Zahl et al. 2010). Plasma
cortisol in fish sedated with all sedatives except CO2 peaked at
0.5 h postinduction, then dropped steadily over the next 6 h, but
was still elevated compared with that in the reference population.
Cortisol levels in fish from the CO2 treatment, however, peaked
at t = 0, and were probably due to the lengthy time required to induce sedation before sampling at t = 0. Plasma lactate increased
rapidly with all sedative options, peaking at 0.5 or 1 h postinduction and steadily decreasing after that. Peak lactate levels
were somewhat lower than that observed in hybrid striped bass
(Trushenski et al. 2012) and may be due to the decreased oxygen
demand observed in grass carp. Grass carp have a resting oxygen
demand of 56 mg O2 /kg per hour (Fu et al. 2009) while hybrid
striped bass have a resting oxygen demand of 132 mg O2 /kg
per hour (Tuncer et al. 1990; Brougher et al. 2005). Another
565
factor that may have contributed to the differences observed

in our study and that reported by Trushenski et al. (2012) was
that hybrid striped bass were sedated at higher temperatures
(21 C) than were grass carp (16 C). We speculate that the
higher water temperature under which hybrid striped bass were
tested and their greater overall metabolic demands could have
resulted in a more rapid depletion of available oxygen within
the tissues, an increase in anaerobic respiration, and ultimately
the greater increase in plasma lactate observed in hybrid striped
bass (maximum observed value: 17.2 0.2 mmol/L in hybrid
striped bass versus 12.4 mmol/L in grass carp). Conversely, the
cooler water temperatures under which grass carp were tested in
our study, combined with the lower overall metabolic demands
of grass carp, may have resulted in lower peak lactate values.
Blood glucose and hematocrit values of sedated fish were similar to those of the reference population regardless of sedative
used or time of sample collection. Plasma osmolality appeared
to vary somewhat among sedatives and over time; however, the
observed values did not differ greatly from those observed in the
reference fish. The lack of substantial change in these physiological characteristicsglucose, hematocrit, and osmolalitymay
suggest a relatively minor and brief acute stress response following sedation.
Although differences in the magnitude of blood chemistry responses were observed, the responses to sedation we noted are
generally consistent with the results of concurrent work involving sedation of grass carp with pulsed DC electricity (Bowzer
et al. 2012). Maximum plasma lactate observed in our study
was slightly higher (12.4 mmol/L in CO2 treated fish) than that
observed by Bowzer et al. (2012) when fish were sedated with
pulsed DC electricity at 150 V for a 10-s exposure (9.4 mmol/L),
which was probably due to the prolonged exposure to CO2 and
the resultant increase in anaerobic metabolism. Peak plasma cortisol generally occurred 0.5 h postsedation in both studies (CO2
being the exception) with slightly higher peak cortisol occurring
in electrosedated grass carp (162288 ng/mL) than in chemically
sedated grass carp (73166 ng/mL). Although a noticeable peak
in blood glucose concentration was not observed in either study,
a slight increase in concentration between 2 and 6 h postsedation was detected. Overall, glucose levels were slightly higher
in the electrosedation study (73124 mg/dL) than in the current
study (6198 mg/dL) while hematocrit (current study: 2029%;
electrosedation study: 2231%) and osmolality (current study:
278361 mOsm/kg; electrosedation study: 301375 mOsm/kg)
did not appear to vary greatly from resting levels in either
study.
In conclusion, each of the sedatives was effective in sedating
grass carp with no apparent negative effects on selected blood
chemistry characteristics or survival. Use of CO2 as a sedative
option for grass carp at the concentration used in this study
would not likely be recommended owing to the lengthy time required to reach sedation and the difficulty of maintaining steady
and effective CO2 concentrations in water. Although benzocaine
rapidly induced sedation, the lengthy recovery time should be
566
GAUSE ET AL.
taken into consideration. Eugenol and MS-222 would probably

be considered the most effective chemical sedatives (at the doses
tested) owing to relatively rapid induction and recovery times.
Lastly, since sedating fish can also act as a stressor, consideration should be given to reduce other potential stressors when
conducting ploidy testing in grass carp.
ACKNOWLEDGMENTS
We thank Mike Freeze, Mike Clark, and the staff of Keo Fish
Farm for their assistance and for space, fish, and water for the
conduct of our experiment.
REFERENCES
Barton, B. A. 2002. Stress in fishes: a diversity of responses with particular reference to changes in circulating corticosteroids. Integrative and Comparative
Biology 42:517525.
Benfey, T. J. 1999. The physiology and behavior of triploid fishes. Reviews in
Bowzer, J. C., J. T. Trushenski, B. R. Gause, and J. D. Bowker. 2012. Efficacy
and physiological responses of grass carp to different sedation techniques:
II. Effect of pulsed DC electricity voltage and exposure time on sedation and
blood chemistry. North American Journal of Aquaculture 74:567574.
Brougher, D. S., L. W. Douglass, and J. H. Soares Jr. 2005. Comparative oxygen
consumption and metabolism of striped bass Morone saxatilis and its hybrid
M. chryshops M. saxatilis. Journal of the World Aquaculture Society
36:521529.
Davis, K. B., and B. R. Griffin. 2004. Physiological responses of hybrid striped
bass under sedation by several anesthetics. Aquaculture 233:531548.
Delaney, M. A., P. H. Klesius, and R. A. Shelby. 2005. Cortisol response of
Nile tilapia, Oreochromis niloticus (L.), to temperature changes. Journal of
Applied Aquaculture 16:95104.
Fu, S.-J., L.-Q. Zeng, X-.M. Li, X. Pang, S.-D. Cao, J.-L. Peng, and Y.-X.
Wang. 2009. The behavioural, digestive, and metabolic characteristics of
fishes with different foraging strategies. Journal of Experimental Biology
212:22962302.
Gilderhus, P. A., and L. L. Marking. 1987. Comparative efficacy of 16 anesthetic
chemicals on rainbow trout. North American Journal of Fisheries Management 7:288292.
Itazawa, Y., and T. Takeda. 1978. Gas exchange in the carp gills in normoxic
and hypoxic conditions. Respiration Physiology 35:263269.
Kelly, A. M., C. R. Engle, M. L. Armstrong, M. Freeze, and A. J. Mitchell.
2011. History of introductions and governmental involvement in promoting
the use of grass, silver, and bighead carps. Pages 163174 in D. C. Chapman
and M. H. Hoff, editors. Invasive Asian carps in North America. American
Fisheries Society, Symposium 74, Bethesda, Maryland.
Kiessling, A., D. Johansson, I. H. Zahl, and O. B. Samuelsen. 2009. Pharmacokinetics, plasma cortisol, and effectiveness of benzocaine, MS-222 and
isoeugenol measured in individual dorsal aorta-cannulated Atlantic salmon
(Salmo salar) following bath administration. Aquaculture 286:301308.
Masser, M. P. 2002. Using grass carp in aquaculture and private impoundments. Southern Regional Aquaculture Center, Publication 3600, Stoneville,
Mississippi.
Olsen, Y. A., I. E. Einarsdottir, and K. J. Nilssen. 1995. Metomidate anaesthesia

in Atlantic salmon, Salmo salar, prevents plasma cortisol increase during
stress. Aquaculture 134:155168.
Owen, M. A. G., S. J. Davies, and K. A. Sloman. 2009. Light colour influences
the behaviour and stress physiology of captive tench (Tinca tinca). Reviews
in Fish Biology and Fisheries 20:375380.
Pirhonen, J., and C. B. Schreck. 2003. Effects of anaesthesia with MS-222,
clove oil and CO2 on feed intake and plasma cortisol in steelhead trout
(Oncorhynchus mykiss). Aquaculture 220:507514.
Post, G. 1979. Carbonic acid anesthesia for aquatic organisms. Progressive
Sattari, A., S. Mirzargar, A. Abrishamifar, R. Lourakzadegan, A. Bahonar,
H. E. Mousavi, and A. Niasari. 2009. Comparison of electroanesthesia with
chemical anesthesia (MS222 and clove oil) in rainbow trout (Oncorhynchus
mykiss) using plasma cortisol and glucose responses as physiological stress
indicators. Asian Journal of Animal and Veterinary Advances 4:306313.
Sepici-Dincel, A., A. Caglan Karasun Benli, M. Selvi, R. Sarikaya, D. Sahin,
I. Ayhan Ozkul,
and F. Erkoc. 2009. Sublethal cyfluthrin toxicity to carp
(Cyprinus carpio L.) fingerlings: biochemical, hematological, histopathological alterations. Ecotoxicology and Environmental Safety 72:14331439.
Summerfelt, R. C., and L. S. Smith. 1990. Anesthesia, surgery, and related
techniques. Pages 213272 in C. B. Schreck and P. B. Moyle, editors, Methods
for fish biology. American Fisheries Society, Bethesda, Maryland.
Trushenski, J. T., M. Schwarz, R. Takeuchi, B. Delbos, and L. A. Sampaio. 2010.
Physiological responses of cobia Rachycentron canadum following exposure
to low water and air exposure stress challenges. Aquaculture 307:173177.
Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Comparative energetics of
striped bass (Morone saxatilis) and hybrid (M. saxatilis M. chrysops)
juveniles. Aquaculture 86:387400.
USFDA (Food and Drug Administration). 2011. Enforcement priorities for
drug use in aquaculture. USFDA, Center for Veterinary Medicine, Program
Policy and Procedures Manual 1240.4200, Silver Spring, Maryland. Available: http://www.fda.gov/downloads/AnimalVeterinary/GuidanceCompliance
Enforcement/PoliciesProceduresManual/UCM046931.pdf. (June 2012).
Venn Beecham, R., B. C. Small, and C. D. Minchew. 2006. Using portable lactate
and glucose meters for catfish research: acceptable alternatives to established
laboratory methods? North American Journal of Aquaculture 68:291295.
Wattendorf, R. J. 1986. Rapid identification of triploid grass carp with a coulter
counter and channelizer. Progressive Fish-Culturist 48:125132.
Wells, R. M. G., and N. W. Pankhurst. 1999. Evaluation of simple instruments
for the measurement of blood glucose and lactate, and plasma protein as stress
indicators in fish. Journal of the World Aquaculture Society 30:276284.
Woods, L. C., D. D. Theisen, and S. He. 2008. Efficacy of Aqui-S as an anesthetic for market-sized striped bass. North American Journal of Aquaculture
70:219222.
Zahl, I. H., A. Kiessling, O. B. Samuelsen, and R. E. Olsen. 2010. Anesthesia induces stress in Atlantic salmon (Salmo salar), Atlantic cod (Gadus
morhua), and Atlantic halibut (Hippoglossus hippoglossus). Fish Physiology
and Biochemistry 36:719730.
Zajicek, P., A. E. Goodwin, and T. Weier. 2011. Triploid grass carp: triploid
induction, sterility, reversion, and certification. North American Journal of


Efficacy and Physiological Responses of Grass Carp to

Different Sedation Techniques: II. Effect of Pulsed DC
Electricity Voltage and Exposure Time on Sedation and
Blood Chemistry
a
John C. Bowzer , Jesse T. Trushenski , Brian R. Gause & James D. Bowker
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125
Lincoln Drive, Life Science II, Room 173, Carbondale, Illinois, 62901-6511, USA
b
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program, 4050
Bridger Canyon Road, Bozeman, Montana, 59715, USA
To cite this article: John C. Bowzer, Jesse T. Trushenski, Brian R. Gause & James D. Bowker (2012): Efficacy and Physiological
Responses of Grass Carp to Different Sedation Techniques: II. Effect of Pulsed DC Electricity Voltage and Exposure Time on
Sedation and Blood Chemistry, North American Journal of Aquaculture, 74:4, 567-574


DOI: 10.1080/15222055.2012.690830
TECHNICAL NOTE
Efficacy and Physiological Responses of Grass Carp

to Different Sedation Techniques: II. Effect of Pulsed DC
Electricity Voltage and Exposure Time on Sedation
and Blood Chemistry
John C. Bowzer, Jesse T. Trushenski,* and Brian R. Gause
Fisheries and Illinois Aquaculture Center, Southern Illinois University Carbondale, 1125 Lincoln Drive,
Life Science II, Room 173, Carbondale, Illinois 62901-6511, USA
James D. Bowker
U.S. Fish and Wildlife Service, Aquatic Animal Drug Approval Partnership Program,
4050 Bridger Canyon Road, Bozeman, Montana 59715, USA
Abstract
Owing to the current absence of an approved immediaterelease chemical sedative for use on fish, researchers have been
exploring alternative methods that would allow treated fish to be
released immediately after sedation, including the use of electrosedation. To address the efficacy of this approach, we evaluated induction and recovery times, survival, and postsedation hematology
of grass carp Ctenopharyngodon idella (291 6.7 g, 30.6 0.3 cm
TL, mean SE) sedated by exposure to 100, 150, or 200 V of
pulsed DC (30 Hz and 25% duty cycle) for 5 or 10 s. Regardless of voltage strength or exposure time, all fish were sedated to
Stage IV sedation within 0.75 min and recovered within 1.5 min.
Although recovery times for fish exposed to electrosedation for 10 s
were longer than those for fish electrosedated for 5 s using 100
and 150 V, the opposite trend was observed among fish sedated using 200 V. Overall, induction and recovery times were short: total
time elapsed from induction to full recovery ranged from 1.0 to
2.1 min (mean, 1.6 min). No mortalities were observed 24 h postsedation. Hematological changes observed were consistent with an
acute stress response, but these effects were transient and few differences were observed among the electrosedation protocols used.
Our results indicate that pulsed DC electrosedation is an effective
strategy for quickly and easily sedating grass carp.
Grass carp Ctenopharyngodon idella are an herbivorous

species introduced from China to the United States as a biological control for aquatic vegetation in aquaculture ponds and
other private and public waters (Masser 2002). As grass carp

are a nonnative species, there is concern that these fish may reproduce and establish selfsustaining populations in U.S. waters.
This has led officials from many states to ban stocking fertile
(diploid) individuals (Kelly et al. 2011), whereas triploid grass
carp may be allowed in some cases (Masser 2002). Triploidy
induction results in functional sterility (Benfey 1999; Zajicek
et al. 2011) and is recognized as an effective strategy to control undesired reproduction. A rapid detection method to verify
triploidy was developed by Wattendorf (1986), and this simple
blood test is used to verify the chromosome number of every
grass carp prior to sale and transfer to states with diploid grass
carp bans (Zajicek et al. 2011).
To facilitate testing, fish are typically sedated so that they
can be easily handled during blood sampling. Currently, there
are very few practical and effective chemical sedative options
available to fish culturists to facilitate sample collection, and
none of the sedative products available in the United States
are legal (i.e., approved by the U.S. Food and Drug Administration [FDA]) for use on food fish (including fish that may
be consumed after stocking in U.S. waters) without adhering
to a lengthy withdrawal period (321 d) following exposure
(see Gause et al. 2012, this issue). Although tricaine methanesulfonate (MS-222), benzocaine, eugenol, or carbon dioxide
(CO2 ) may be used as fish sedatives under certain circumstances,
none of these are fully suitable for procedures such as blood
sampling and triploidy verification of grass carp because they
*Corresponding author: saluski@siu.edu

Received October 24, 2011; accepted April 13, 2012
567
568
BOWZER ET AL.
require lengthy withdrawal times, are difficult to use, or are not

legally available for such use.
Given the constraints associated with chemical sedatives and
the amount of time and resources required to gain FDA-approval
of these compounds for use in fish, fish culturists have been exploring alternative sedation techniques, including electrosedation. Although electrofishing has been used as a field sampling
technique for decades by fisheries professionals, the technique
has recently been modified specifically for the purpose of sedating fish (Zydlewski et al. 2008; Trushenski et al. 2012). Electroanesthesia, or more accurately, electrosedation, can immobilize fish via electronarcosis (stunning) or electrotetany (tetanic
muscle contraction) caused by electrically induced interference
with neurotransmission. Although present electrosedation technology may be somewhat limited in fish with comparatively
fragile vertebrae (e.g., salmonids; C. V. Burger, Smith-Root,
personal communication), in many species it may offer several
advantages over chemical sedatives in terms of withdrawal periods, chemical disposal, and potentially, ease of use. Perhaps
more importantly, electrosedation of fish is currently not subject
to FDA regulation and can be used legally without having to go
through the arduous, multiyear, multimillion dollar process of
getting a chemical sedative approved for use in fish. However, it
is important to develop use protocols to ensure that fish can be
effectively sedated with minimal risk of adverse postsedation
outcomes, including mortality. Although preliminary experimentation suggested pulsed-DC electricity is suitable for sedating grass carp (described below), it is unclear whether different
waveforms (e.g., different voltage strengths, frequencies) or exposure durations affect induction or recovery times, blood chemistry responses, or overall efficacy. Accordingly, we evaluated
electrosedation effects (induction and recovery times, survival,
and postsedation blood chemistry) on grass carp (a representative warmwater fish) using three different voltage strengths and
two different exposure durations.
METHODS
Electrosedation procedures.A reference population of
triploid grass carp (291 6.7 g, 30.6 0.3 cm total length,
mean SE) was held at Keo Fish Farm, Keo, Arkansas, in an
outdoor raceway configured as a partial flow-through system
(static raceway, periodically flushed with screened surface water) with supplemental aeration. Prior to experimentation, fish
were fasted for a minimum of 24 h. To determine an appropriate
control waveform from which to derive other waveforms for
experimentation in the principal investigation, a preliminary experiment was conducted to determine whether a relatively mild
waveform (based on our previous experience with electrosedation) would effectively sedate grass carp to Stage IV of sedation
(see below). A group of 15 fish were randomly collected from
the reference population and transferred into a 142-L cooler
prefilled with 70 L of aerated culture water to achieve a depth of
approximately 8 cm and equipped with an electrosedation unit

(PES Portable Electroanesthesia System, Smith-Root, Vancouver, Washington). These fish were exposed to 100 V of pulsed
DC (60 Hz, 25% duty cycle, 5-s exposure). Using this waveform, mean sedation and recovery times were 0.5 and 1.8 min,
respectively, and no overt signs of postsedation distress were
observed. Accordingly, we determined that this waveform was
effective and would serve as the lowest intensity waveform in
the subsequent, principal investigation with one modification:
given that higher voltages and exposure durations were to be
investigated, we decided that the experimental protocols would
use a lower frequency of 30 Hz.
For the principal investigation, groups of 15 fish were randomly collected from the reference population and transferred
into the electrosedation unit configured as described above and
filled with fresh culture water from the holding system. Fish
groups were exposed to 100, 150, or 200 V of pulsed DC (30 Hz
and 25% duty cycle) for 5 or 10 s in a 3 2 factorial design (100 V for 5 s, 100 V for 10 s, 150 V for 5 s, 150 V
for 10 s, 200 V for 5 s, 200 V for 10 s). Culture water in
the sedation chamber was aerated after sedating each group
of fish but was not exchanged over the course of the experiment. A water sample was collected from the holding system at
time (t) = 0, and the sample was analyzed in duplicate for the
following: temperature and dissolved oxygen (YSI 550 meter,
Yellow Springs Instruments, Yellow Springs, Ohio), conductivity, pH, salinity (Multi-Parameter PCSTestr 35, Eutech/Oakton
Instruments, Vernon Hills, Illinois), hardness, alkalinity (digital
titrator and reagents, Hach, Loveland, Colorado), total ammonia
nitrogen, nitrite-nitrogen, and nitrate-nitrogen (DR 2800 spectrophotometer and reagents, Hach). All measured water quality
characteristics were within ranges appropriate for grass carp
(Masser 2002) (Table 1).
Fish were monitored during sedation to determine induction
to Stage IV of anesthesia (Summerfelt and Smith 1990; although anesthesia is the term used by Summerfelt and Smith,
we have used the term sedation throughout the manuscript to
better reflect the behavioral responses we observed), which is
TABLE 1. Holding system water quality measured at the beginning of the
experiment to examine electrosedation in grass carp.
Characteristic
Value
Temperature ( C)
Total ammonia nitrogen (mg/L)
Nitrite-nitrogen (mg/L)
Nitrate-nitrogen (mg/L)
Alkalinity (mg/L)
Hardness (mg/L)
Salinity ()
Conductivity (S/cm)
pH
18.6
7.98
0
0.003
0.9
240
374
0.427
877
7.5
TECHNICAL NOTE
associated with the total loss of equilibrium, muscle tone, and

responsiveness to visual and tactile stimuli, but maintenance
of a slow, steady, opercular ventilation rate. After the loss of
equilibrium, slight manual pressure was applied along the trunk
and caudal peduncle as a tactile stimulus. Fish were considered induced to Stage IV when they no longer responded to this
stimulus, but the ventilation rate remained steady albeit reduced
relative to unsedated fish. A tremor was observed immediately
following electrosedation. Although fish were not responsive
during this tremor (and were perhaps momentarily in Stage V
or VI of sedation), induction was considered complete after
the tremor had ceased and fish resumed ventilating. Immediately after induction, blood samples were collected from three
fish using procedures described below. The remaining 12 fish
were returned to holding tanks and monitored to determine recovery from sedation (return of normal equilibrium and tactile
responsiveness). The group was considered recovered when the
last fish recovered, (i.e., recovery time = time for last fish to
recover). Assessment of induction and recovery can be somewhat subjective so bias was minimized by having the same observer apply all stimuli and assess when fish were sedated and
recovered. Recovered fish were transferred to a second raceway (configured in the same manner as the one housing the
reference population) and kept separated by raceway dividers
positioned at 1-m intervals along the length of the recovery
raceway.
Sample collection and analysis.In addition to collecting
blood from fish sampled immediately after sedation (t = 0),
blood samples were collected from three fish per group at
0.5, 1, 2, and 6 h postsedation (three fish per group per time
point, individuals sampled once). Prior to sampling, all fish
were immersed in a bath of metomidate hydrochloride (Aquacalm, Western Chemical, Ferndale, Washington; 510 mg/L
for 30 s) to facilitate handling. Although additional sedation
was not necessary for fish sampled at t = 0, these fish were also
exposed to a metomidate hydrochloride bath to ensure consistent treatment of all fish. Once handleable, fish were weighed
(to the nearest gram) and measured (total length to the nearest 0.5 cm), and a blood sample was collected from the caudal
vasculature using heparinized, evacuated blood collection assemblies (Vacutainer, Becton Dickinson, Franklin Lakes, New
Jersey). Although metomidate hydrochloride was selected, in
part, because it limits or prevents corticosteroid increase during
sampling (Olsen et al. 1995; Davis and Griffin 2004), all blood
samples were collected within 5 min of capture to minimize
the possibility of other confounding responses of handling and
venipuncture. In addition to fish sampled at set time points after
sedation, two fish from the reference population were also sampled every hour over the course of the experiment (experiment
duration, 7 h; no fish were sampled more than once). After blood
collection, all fish were returned to a separate area in the recovery system and monitored for adverse behavior and survival for
24 h. Blood samples were kept on wet ice (<36 h) until analysis for glucose, lactate, cortisol, and osmolality as described
569
by Gause et al. (2012). Although 36 h might be considered a

lengthy period of time to hold blood samples prior to analysis,
some assays could not be immediately conducted in the field
and samples had to be transported back to the Fisheries and
Illinois Aquaculture Center, Carbondale, Illinois. It is possible
that levels of metabolically relevant molecules (e.g., glucose and
lactate) could have changed slightly during this holding period;
however, all samples were treated in the same manner to ensure
validity of comparisons among treatments. Briefly, hematocrit
(Statspin centrifuge, Fisher Scientific, Pittsburgh, Pennsylvania) and glucose (Freestyle Freedom Lite glucose meter, Abbott Laboratories, Abbott Park, Illinois) were determined using
aliquots of whole blood, and then the remaining whole blood
was centrifuged (3,000 g, 4 C, 45 min). Resultant plasma
was collected and stored at 80 C until further analysis. Plasma
samples were analyzed to determine lactate (Accutrend lactate
meter, Roche, Mannheim, Germany), osmolality (Vapro 5520,
Wescor, Logan, Utah), and cortisol (EIA 1887, DRG International, Mountainside, New Jersey). Although portable lactate
and glucose meters such as those used in this study can slightly
underestimate metabolite levels in fish blood relative to laboratory methods, they are considered precise and reliable for use in
generating comparative data (Wells and Pankhurst 1999; Venn
Beecham et al. 2006).
Although multiple fish were sampled from each treatment
group at each time point, they were group-sedated and cohoused
after sedation. Therefore, we determined that individual fish did
not represent truly independent observations. Since the experiment lacked replicate experimental units, qualitative comparisons of within-group mean values were assessed rather than
using statistical analysis to try to make quantitative comparisons.
RESULTS
All fish were successfully induced to Stage IV sedation in less
than 1 min (mean = 0.6 min, range = 0.50.7 min), regardless
of voltage strength or exposure duration (Figure 1). Recovery
times were more variable, but recovery of equilibrium (mean =
0.7 min; range, 0.31.3 min) and tactile responsiveness (mean
= 0.9 min; range, 0.51.4 min) were achieved in less than 2 min
postsedation. Although a positive relationship was evident between longer exposure durations and increasing recovery times
in fish sedated using 100 and 150 V, recovery times were shorter
among fish sedated using 200 V. Overall, induction and recovery times were short, total time elapsed from induction to full
recovery ranged from 1.0 to 2.1 min (mean = 1.6 min), there
was minimal group-to-group induction or recovery variability,
and no mortalities were observed 24 h postsedation.
Although most blood chemistry parameters did not vary substantially by electrosedation protocol at any single timepoint,
hematocrit, blood glucose, and plasma cortisol, lactate, and osmolality varied over time following sedation (Figure 2AE).
Plasma cortisol and lactate concentrations initially increased
570
BOWZER ET AL.
100 V for 5 s
100 V for 10 s
150 V for 5 s
0.5
150 V for 10 s
200 V for 5 s
200 V for 10 s
1.5
2.5
Time (min)
I = Induced to Phase IV Sedation
E = Maintain Equilibrium
R = Recovery; Respond to Tactile Stimulus
FIGURE 1. Schematic illustrating induction and various stages of recovery of grass carp electrosedated to Stage IV sedation using various pulsed DC voltage
strengths and exposure durations.
after sedation and then decreased over time. The cortisol response was rapid and transient, peaking (162288 ng/mL) at
0.5 h postsedation and returning to resting levels (0100 ng/mL)
between 2 and 6 h postsedation. The peak lactate response developed more slowly than cortisol, reaching maximum levels
(69 mmol/L) between 0.5 and 2 h postsedation, but dropped
below resting levels by 6 h postsedation. Peak levels of blood
glucose (98124 mg/dL) observed within the first 0.5 h postsedation, decreased slightly over the next 0.5 h and then increased
slightly from 1 to 6 h postsedation. Hematocrit and plasma osmolality fluctuated near resting levels throughout the sampling
period. There was no indication that any one combination of
voltage strength and exposure duration consistently produced
the highest or lowest blood chemistry responses.
During electrosedation, fish exhibited opercular flaring, fin
extension, and body rigidity but regained normal posture after
resolution of the postsedation tremor. Slight petechial hemorrhaging was observed along the lower flank and opercular area
in a few fish during electrosedation.
DISCUSSION
Pulsed DC, applied at voltage strengths of 100200 V and exposure durations of 5 or 10 s, was effective in sedating grass carp
to Stage IV sedation. Voltage strength and exposure duration had

little effect in terms of time to induction and recovery from sedation. In addition, there was little variability among electrosedation protocols on the effects on blood chemistry responses
following sedation. Although slight numeric differences were
noted for induction and recovery times, the maximum difference
observed (0.23 and 1.12 min, respectively) would probably not
be considered practically relevant to most fisheries professionals. All fish recovered within 2 min of induction, which would
probably be considered adequate for sedating grass carp to facilitate procedures such as triploidy verification. The observed
changes in blood chemistry were consistent with an acute stress
response (Barton 2002), but these physiological responses were
resolved or nearly resolved within 6 h of sedation, and no postsedation mortality was observed. Slight epidermal hemorrhaging
was observed in some electrosedated fish. However, we concluded that these relatively uncommon lesions were probably
unrelated to electrosedation since petechiae were also observed
in some fish prior to electrosedation. Based on these data, it
would appear that juvenile grass carp are resilient to electrosedation at the range of voltage strengths and exposure durations
tested in this experiment, and that the electrosedation protocols
tested are reasonably safe with respect to postsedation survival
and physiological status of these fish. In the present work, we
571
TECHNICAL NOTE
FIGURE 2. Time course of blood chemistry responses (A = plasma cortisol, B = blood glucose, C = hematocrit, D = plasma lactate, E = plasma osmolality) of
grass carp following electrosedation using various pulsed DC voltage strengths and exposure durations. Points represent mean values; grey reference bars represent
means of values observed for fish sampled from the reference population throughout the course of the experiment.
572
BOWZER ET AL.
did not examine fish for vertebral or other internal injuries following electrosedation. These types of injuries have been observed following exposure to pulsed DC electrosedation in some
(Gaikowski et al. 2001; Zydlewski et al. 2008) but not all fishes
(Vandergoot et al. 2011). Electrically induced injury and mortality rates are a function of the type and strength of the waveform
used, as well as the fish involved (Snyder 2003). In general,
short duration exposure to low-intensity, pulsed DC waveforms
is considered less risky than longer duration exposure to high
intensity, AC waveforms, though the reported effects of these
and other factors are sometimes sparse, difficult to compare,
and often questionable (Snyder 2003). Regardless, the absence
of direct or delayed mortality and overt signs of injury suggest that each of the protocols assessed in the present work are
reasonably safe when used to sedate grass carp.
Blood chemistry responses observed in this experiment were
comparable with those reported in two experiments conducted
on hybrid striped bass (white bass Morone chrysops striped
bass M. saxatilis) using the same Portable Electroanesthesia
System and in the experiment conducted by Gause et al. (2012)
in which grass carp were sedated using a variety of chemical
sedatives. Trushenski et al. (2012) reported a slightly greater
plasma cortisol pulse and greater plasma glucose and lactate
pulses, but similar hematocrit and osmolality levels, in hybrid
striped bass (510 12 g, 33.7 0.2 cm, mean SE) electrosedated at 100 V, 30 Hz, and 25% duty cycle for 3 s than we
observed in grass carp in this experiment. In another experiment,
Trushenski and Bowker (in press) used similar electrosedation
protocols to sedate smaller hybrid striped bass (211 4 g,
26.1 0.1 cm total length, mean SE), and found that grass
carp plasma cortisol levels were lower than or comparable with
those observed in the smaller hybrid striped bass at t = 0, 1,
2, and 6 h. However, the plasma cortisol levels observed in
grass carp were much lower (150300 ng/mL) at t = 0.5 h than
observed in the smaller hybrid striped bass (400650 ng/mL)
(Trushenski and Bowker, in press). Responses of lactate, hematocrit, and osmolality noted for grass carp were of a comparable
or smaller magnitude than those reported for smaller hybrid
striped bass by Trushenski and Bowker (in press), but followed
the same basic patterns of acute response and resolution within
6 h of sedation. The somewhat attenuated lactate response observed in grass carp, in comparison with hybrid striped bass, is
probably the result of the comparatively lower metabolic rate
of grass carp (Tuncer et al. 1990; Brougher et al. 2005; Fu
et al. 2009). Increased lactate formation results from anaerobic
metabolic activity occurring during periods of limited or no oxygen availability, such as when environmental oxygen availability
is limiting or during exhaustive physical activity when respiration is insufficient to meet tissue oxygen demand for aerobic
metabolism (Bennett 1978; Burton and Heath 1980). Sedated
fish exhibiting reduced ventilation rates may accumulate lactate,
particularly if their metabolic rate and oxygen demand is high.
Juvenile hybrid striped bass have a considerably higher oxygen demand at rest (132 mg O2 kg1h1) (Tuncer et al. 1990;
Brougher et al. 2005) than some other fish, including grass carp
(56 mg O2 kg1h1; Fu et al. 2009); for purposes of comparison, Clarke and Johnston (1999) modeled the metabolic rate
of 55 species of fish, and estimated the resting oxygen consumption of a 50-g fish at 15 C to range from 27 to 133 mg
O2 kg1h1 depending on species. This may explain why electrosedated hybrid striped bass experience greater postsedation
lactate pulses than do electrosedated grass carp. The blood glucose response of grass carp observed in this experiment fluctuated near 100 mg/dL compared with the smaller hybrid striped
bass in the previous experiment by Trushenski and Bowker
(in press); in that study blood glucose displayed a pulse from
resting levels of approximately 55 mg/dL to a peak at t = 1 h of
180220 mg/dL. The reduced metabolic rate and lower plasma
cortisol levels of grass carp may explain the relatively minor glucose response observed in the current experiment. Cortisol can
activate glycogenolysis and gluconeogenesis processes in fish,
which cause increases in substrate levels (glucose) in the blood
to produce enough energy to meet the demand of the organism
(Barton and Iwama 1991; Martnez-Porchas et al. 2009), and
since grass carp have lower metabolic demands and experienced
a lower cortisol response to electrosedation than hybrid striped
bass (and possibly a lower catecholamine response), a lower
glucose response may be expected. This and other research with
chemical sedatives and various methods of electrosedation (AC,
continuous DC, pulsed DC) or electroshock (Schreck et al. 1976;
Mesa and Schreck 1989; Barton and Grosh 1996; Barton and
Dwyer 1997) have demonstrated that fish undergo the generalized stress response (Barton 2002) following sedation (Bourne
1984; Bernier and Randall 1998; Davidson et al. 2000; Davis
and Griffin 2004; Woods et al. 2008; Feng et al. 2009; Neiffer and Stamper 2009; Sattari et al. 2009; Carter et al. 2011;
Trushenski et al. 2012; Gause et al. 2012). Because these effects
also occur after exposure to sedatives in the absence of handling
further emphasizes that the sedatives themselves act as stressors (Zahl et al. 2010). Differences in the magnitude of physiological responses aside, our present results are broadly consistent with our previous work sedating adult hybrid striped bass
(Trushenski et al. 2012) and the majority of published works
on the subject.
In conclusion, pulsed DC electrosedation is an effective strategy for sedating grass carp quickly and easily for routine handling procedures. Electrosedation offers one distinct advantage
over other currently available options: fish can be released immediately after treatment. Like other sedatives, electrosedation
induces an acute stress response in fish. Although electrosedated
grass carp exhibited responses consistent with the generalized
stress response in fish, none of the protocols used elicited responses that were particularly severe in comparison with the
others or the reported effects of chemical sedatives (Gause et al.
2012), and fish were observed to recover from these effects
within 6 h. Although slight differences in induction and recovery times were associated with different voltage strengths and
exposure durations, all of the protocols used yielded sedation
TECHNICAL NOTE
patterns that would be considered acceptable for handling grass

carp and testing them for triploidy. However, to minimize the
incidence of unforeseen injuries or physiological alterations, we
recommend that users employ the lowest voltage strength and
exposure duration that yields effective electrosedation.
ACKNOWLEDGMENTS
We thank Smith-Root, Inc. for providing access to a Portable
Electroanesthesia System, and Jack Wingate and Mike Holliman for providing training and technical support in using the
electrosedation unit. We also thank Mike Freeze, Mike Clark,
and the staff of Keo Fish Farm for their assistance and for accommodating our experiment.
REFERENCES
Barton, B. A. 2002. Stress in fishes: a diversity of responses with particular reference to changes in circulating corticosteroids. Integrative and Comparative
Biology 42:517525.
Barton, B. A., and W. P. Dwyer. 1997. Physiological stress effects of continuousand pulsed-DC electroshock on juvenile bull trout. Journal of Fish Biology
51:9981008.
Barton, B. A., and R. S. Grosh. 1996. Effect of AC electroshock on blood
features in juvenile rainbow trout. Journal of Fish Biology 49:13301333.
Barton, B. A., and G. K. Iwama. 1991. Physiological changes in fish from stress
in aquaculture with emphasis on the response and effects of corticosteroids.
Annual Review of Fish Diseases 1:326.
Benfey, T. J. 1999. The physiology and behavior of triploid fishes. Reviews in
Bennett, A. F. 1978. Activity metabolism of the lower vertebrates. Annual
Review of Physiology 40:447469.
Bernier, N. J., and D. J. Randall. 1998. Carbon dioxide anaesthesia in rainbow
trout: effects of hypercapnic level and stress on induction and recovery from
anaesthetic treatment. Journal of Fish Biology 52:621637.
Bourne, P. K. 1984. The use of MS 222 (tricaine methanesulphonate) as an
anaesthetic for routine blood sampling in three species of marine teleosts.
Brougher, D. S., L. W. Douglass, and J. H. Soares Jr. 2005. Comparative oxygen
consumption and metabolism of striped bass Morone saxatilis and its hybrid
M. chrysops M. saxatilis. Journal of the World Aquaculture Society
36:521529.
Burton, D. T., and A. G. Heath. 1980. Ambient oxygen tension (PO2 ) and transition to anaerobic metabolism in three species of freshwater fish. Canadian
Journal of Fisheries and Aquatic Sciences 37:12161224.
Carter, K. M., C. M. Woodley, and R. S. Brown. 2011. A review of tricaine
methanesulfonate for anesthesia of fish. Reviews in Fish Biology and Fisheries 21:5159.
Clarke, A., and N. M. Johnston. 1999. Scaling of metabolic rate with body mass
and temperature in teleost fish. Journal of Animal Ecology 68:893905.
Davidson, G. W., P. S. Davie, G. Young, and R. T. Fowler. 2000. Physiological
responses of rainbow trout Oncorhynchus mykiss to crowding and anesthesia
with AQUI-STM. Journal of the World Aquaculture Society 31:105114.
Davis, K. B., and B. R. Griffin. 2004. Physiological responses of hybrid striped
bass under sedation by several anesthetics. Aquaculture 233:531548.
Feng, G. P., P. Zhuang, L. Z. Zhang, N. N. Chen, Z. F. Yao, and Y. Y. Men.
2009. Effects of electroanesthesia on behavior and serum iron concentration
of juvenile Acipenser baeri. Marine Fisheries 2009(1):article 6. (In Chinese
with English abstract.) Available: en.cnki.com.cn/Article en/CJFDTOTALHTYY200901005.htm. (October 2011).
Fu, S. J., L. Q. Zeng, X. M. Li, X. Pang, Z. D. Cao, J. L. Peng, and Y. X. Wang.
2009. The behavioural, digestive and metabolic characteristics of fishes with
573
different foraging strategies. Journal of Experimental Biology 212:2296

2302.
Gaikowski, M. P., W. H. Gingerich, and S. Gutreuter. 2001. Short-duration
electrical immobilization of lake trout. North American Journal of Fisheries
Gause, B. R., J. T. Trushenski, J. C. Bowzer, and J. D. Bowker. 2012. Efficacy
and physiological responses of grass carp to different sedation techniques: I.
Effects of various chemicals on sedation and blood chemistry. North American Journal of Aquaculture 74:560566.
Kelly, A. M., C. R. Engle, M. L. Armstrong, M. Freeze, and A. J. Mitchell.
2011. History of introductions and governmental involvement in promoting
the use of grass, silver, and bighead carps. Pages 163174 in D. C. Chapman
and M. H. Hoff, editors. Invasive Asian carps in North America. American
Fisheries Society, Symposium 74, Bethesda, Maryland.
Martnez-Porchas, M., L. R. Martnez-Cordova, and R. Ramos-Enriquez. 2009.
Cortisol and glucose: reliable indicators of fish stress? Pan-American Journal
of Aquatic Sciences 4:158178.
Masser, M. P. 2002. Using grass carp in aquaculture and private impoundments. Southern Regional Aquaculture Center, Publication 3600, Stoneville,
Mississippi.
Mazeaud, M. M., F. Mazeaud, and E. M. Donaldson. 1977. Primary and secondary effects of stress in fish: some new data with a general review. Transactions of the American Fisheries Society 106:201212.
Mesa, M. G., and C. B. Schreck. 1989. Electrofishing markrecapture and
depletion methodologies evoke behavioral and physiological changes in
cutthroat trout. Transactions of the American Fisheries Society 118:644
658.
Neiffer, D. L., and M. A. Stamper. 2009. Fish sedation, anesthesia, analgesia,
and euthanasia: considerations, methods, and types of drugs. ILAR (Institute
for Laboratory Animal Research) Journal 50:343360.
Olsen, Y. A., I. E. Einarsdottir, and K. J. Nilssen. 1995. Metomidate anaesthesia
in Atlantic salmon, Salmo salar, prevents plasma cortisol increase during
stress. Aquaculture 134:155168.
Sattari, A., S. S. Mirzargar, A. Abrishamifar, R. Lourakzadegan, A. Bahonar,
H. E. Mousavi, and A. Niasari. 2009. Comparison of electroanesthesia
with chemical anesthesia (MS222 and clove oil) in rainbow trout (Oncorhynchus mykiss) using plasma cortisol and glucose responses as physiological stress indicators. Asian Journal of Animal and Veterinary Advances 4:
306313.
Schreck, C. B., R. A. Whaley, M. L. Bass, O. E. Maughan, and M. Solazzi. 1976.
Physiological responses of rainbow trout (Salmo gairdneri) to electroshock.
Journal of the Fisheries Research Board of Canada 33:7684.
Snyder, D. E. 2003. Electrofishing and its harmful effects on fish. U.S. Geological Survey, Information and Technology Report USGS/BRD/ITR20030002, U.S. Government Printing Office, Denver.
Somero, G. N., and J. J. Childress. 1980. A violation of the metabolism-size
scaling paradigm: activities of glycolytic enzymes in muscle increase in
larger-size fish. Physiological Zoology 53:322337.
Trushenski, J. T., and J. D. Bowker. In press. Effect of voltage and exposure time
on fish response to electrosedation. Journal of Fish and Wildlife Management.
DOI: 10.3996/102011-JFWM-060.
Tuncer, H., R. M. Harrell, and E. D. Houde. 1990. Comparative energetics of
striped bass (Morone saxatilis) and hybrid (M. saxatilis M. chrysops)
juveniles. Aquaculture 86:387400.
Vandergoot, C. S., K. J. Murchie, S. J. Cooke, J. M. Dettmers, R. A. Bergstedt,
and D. G. Fielder. 2011. Evaluation of two forms of electroanesthesia and
carbon dioxide for short-term anesthesia in walleye. North American Journal
of Fisheries Management 31:914922.
Venn Beecham, R., B. C. Small, and C. D. Minchew. 2006. Using portable
lactate and glucose meters for catfish research: acceptable alternatives to
574
BOWZER ET AL.
established laboratory methods? North American Journal of Aquaculture

68:291295.
Wattendorf, R. J. 1986. Rapid identification of triploid grass carp with a Coulter
Counter and Channelyzer. Progressive Fish-Culturist 48:125132.
Wells, R. M. G., and N. W. Pankhurst. 1999. Evaluation of simple instruments for the measurement of blood glucose and lactate, and plasma protein
as stress indicators in fish. Journal of the World Aquaculture Society 30:
276284.
Woods, L. C., III, D. D. Theisen, and S. He. 2008. Efficacy of Aqui-S as an anesthetic for market-sized striped bass. North American Journal of Aquaculture
70:219222.
Zahl, I. H., A. Kiessling, O. B. Samuelsen, and R. E. Olsen. 2010. Anesthesia induces stress in Atlantic salmon (Salmo salar), Atlantic cod (Gadus
morhua) and Atlantic halibut (Hippoglossus hippoglossus). Fish Physiology
and Biochemistry 36:719730.
Zajicek, P., A. E. Goodwin, and T. Weier. 2011. Triploid grass carp: triploid
induction, sterility, reversion, and certification. North American Journal of
Zydlewski, G. B., W. Gale, J. Holmes, J. Johnson, T. Brigham, and W. Thorson.
2008. Use of electroshock for euthanizing and immobilizing adult spring Chinook salmon in a hatchery. North American Journal of Aquaculture 70:415
424.

North American Journal of Aquaculture

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

North American Journal of Aquaculture

Uploaded by

Copyright:

Available Formats

This article was downloaded by: [Department Of Fisheries]

On: 25 September 2012, At: 23:59