Aquaculture research, tập 41, số 5, 2010

Aquaculture Research, 2010, 41, 611^612

doi:10.1111/j.1365-2109.2010.02510.x

Editorial

This issue is the result of the Special Session entitled
‘‘Basic And Applied Aspects Of Aquaculture Nutrition: Healthy Fish For Healthy Consumers’’ carried
out on 17^18 September 2008 in Krakow, Poland as
part of the Annual Meeting of the European Aquaculture Society. This session was co-sponsored by
the Organization for Economic Cooperation and
Development (OECD) Co-operative Research Programme on Biological Resource Management for
Sustainable Agriculture Systems,Trade and Agriculture Directorate. In particular we appreciate help in
the organization of the session by Dr Carl Christian
Schmidt. The European Aquaculture Society cosponsored the session and help from the Executive
Director, Dr Alistair Lane, need to be recognized.
The Ohio State University, School of Environment
and Natural Resources, contributed ¢nancially to
the session.
One of the primary goals of ¢sh nutrition is to produce healthy food for human consumption. This can
only be achieved via a thorough understanding of
¢sh nutritional requirements and the appropriate
choice of feed ingredients that will secure the economic e⁄ciency and sustainability of the industry,
i.e. decreased use of ¢shmeal and replacement
with plant protein and lipid sources. The conference
intended to promote and focus the research on
aquaculture nutrition by making a link to the human
food chain.
The conference addressed the major areas of basic
aspects of ¢sh nutrition by gathering experts in aquaculture nutrition that could outline the current state

of knowledge in the ¢eld, and present coherent perspectives on the improvements needed in ¢sh culture
to ful¢ll public expectations in terms of healthy food
and its sustainable production. Sustainable ¢sh production must concentrate on the use of plant protein
and lipid sources in ¢sh diets to replace ¢shmeal resulting in healthy seafood from the human point of
view. Consequently, pollution originating from
¢sh and other aquatic animal farming must be
decreased. This will result in higher economic
e⁄ciency of seafood production as the cost of ¢sh

r 2010 The Authors
Journal Compilation r 2010 Blackwell Publishing Ltd

food (diets) is the largest single cost (over 50%) of
production.
From the very beginning of the session concept, results of the conference were to be transferred to professionals and therefore the subject of the regular
peer-review process. The intent was also to disseminate the results to the general public and the media
in the form of a brochure that will summarize the major conclusions of the conference (abstracts) with the
reference materials to each of the contributors, its
¢eld of expertise and current professional activity.
The session was also thought to contribute to public
debate by focusing on the production of healthy human food at a time while the world ¢shery is shrinking and thus move from exploitation to sustainable
production. As outlined in the recent FAO ‘‘State of
the World Aquaculture 2006’’document, the session
intended to cross national and institutional boundaries and establish a framework for the large-scale
development of aquaculture.
That was truly a forum for multidisciplinary interactions. The programme areas that were covered by
invited speakers can be grouped into the following
major topics: (1) digestive tract morphology and regulation of nutrient uptake, (2) endocrine and neural
regulation of food intake, (3) molecular biology tools
to validate and compliment biological nutrition data,

(4) major nutrient requirements, (5) replacement of
¢shmeal protein and lipids with plant ingredients,
(6) speci¢c nutritional needs for di¡erent life stages
of ¢sh, from larvae to broodstock nutrition and (7)
management of aquaculture waste.
There are a number of challenges that must be
overcome to maintain acceptable growth rates and
feed e⁄ciency values at higher levels of substitution
of ¢shmeal. The ¢rst is cost of plant protein concentrates. The second challenge facing the aquafeed industry as it moves to substitute higher amounts of
¢shmeal with plant proteins pertains to known nutritional limitations of plant proteins. The third challenge to overcome to developing plant protein-based
aquafeeds for intensively grown ¢sh species
concerns unknown nutrients and biologically active

611


Editorial

Aquaculture Research, 2010, 41, 611^612

materials in ¢shmeal that are not present in plant
protein concentrates and that may be necessary dietary constituent for optimum growth and health of
¢sh. These three applied aspects of ¢sh nutrition were
reviewed but undoubtedly deserve special session in
the near future as the logical consequence of the basic nutrition research. These applied aspects will be

612

tackled with powerful new molecular biology tools,
not only in a descriptive manner, but also by pathway-speci¢c metabolomic markers in ¢sh nutrition

studies.
Konrad Dabrowski
Ronald Hardy

r 2010 The Authors
Journal Compilation r 2010 Blackwell Publishing Ltd, Aquaculture Research, 41, 611^612


Aquaculture Research, 2010, 41, 613^640

doi:10.1111/j.1365-2109.2009.02242.x

REVIEW ARTICLE
Live feeds for early stages of fish rearing
Lu|¤ s E C Conceic°aìo1, Manuel Yu¤fera2, Pavlos Makridis3, So¢a Morais1 & Maria Teresa Dinis1
1

Center for Marine Sciences ^ CCMAR, University of Algarve, Faro, Portugal
Instituto de Ciencias Marinas Andaluc|¤ a, CSIC, Apartado O¢cial, Puerto Real, Spain

2
3

Institute of Aquaculture, Hellenic Center for Marine Research, Heraklion, Crete, Greece

Correspondence: L E C Conceic°aìo, CCMAR, Universidade do Algarve, Campus de Gambelas, 8005-139 Faro, Portugal.
E-mail:

Abstract
Despite the recent progress in the production of inert

diets for ¢sh larvae, feeding of most species of interest
for aquaculture still relies on live feeds during the
early life stages. Independently of their nutritional
value, live feeds are easily detected and captured,
due to their swimming movements in the water
column, and highly digestible, given their lower nutrient concentration (water content480%). The present paper reviews the main types of live feeds used
in aquaculture, their advantages and pitfalls, with a
special emphasis on their nutritional value and the
extent to which this can be manipulated. The most
commonly used live feeds in aquaculture are rotifers
(Brachionus sp.) and brine shrimp (Artemia sp.), due to
the existence of standardized cost-e¡ective protocols
for their mass production. However, both rotifers and
Artemia have nutritional de¢ciencies for marine species, particularly in essential n-3 highly unsaturated
fatty acids (HUFA, e.g., docosahexaenoic acid and eicosapentaenoic acid). Enrichment of these live feeds
with HUFA-rich lipid emulsions may lead to an excess
dietary lipid and sub-optimal dietary protein content
for ¢sh larvae. In addition, rotifers and Artemia are
likely to have sub-optimal dietary levels of some amino acids, vitamins and minerals, at least for some
species. Several species of microalgae are also used
in larviculture. These are used as feed for other live
feeds, but mostly in the‘green water’technique in ¢sh
larval rearing, with putative bene¢cial e¡ects on
feeding behaviour, digestive function, nutritional value, water quality and micro£ora. Copepods and
other natural zooplankton organisms have also been
used as live feeds, normally with considerably better

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd

results in terms of larval survival rates, growth and
quality, when compared with rotifers and Artemia.
Nonetheless, technical di⁄culties in mass-producing
these organisms are still a constraint to their routine
use. Improvements in inert microdiets will likely lead
to a progressive substitution of live feeds. However,
complete substitution is probably years away for
most species, at least for the ¢rst days of feeding.

Keywords: microalgae, rotifers, Artemia, copepods, nutritional value, ¢sh larvae

Introduction
Live feeds are the main item in the diet of cultured
¢sh larvae and they are of particular importance
when rearing marine ¢sh larvae of the altricial type.
Altricial larvae are those that remain in a relatively
undeveloped state until the yolk sac is exhausted. At
¢rst-feeding the digestive system is still rudimentary,
lacking a stomach, and much of the protein digestion
takes place in hindgut epithelial cells (Govoni, Boehlert & Watanabe 1986). Such a digestive system is in
most cases incapable of processing formulated diets
in a manner that allows survival and growth of the
larvae comparable to those fed live feeds. In fact, despite recent progress in the development of inert diets
for ¢sh larvae (e.g., Lazo, Dinis, Holt, Faulk & Arnold
2000; Cahu & Infante 2001; Koven, Kolkovski, Hadas,
Gamsiz & Tandler 2001), feeding of most species of interest for aquaculture still relies on live feeds during
the early life stages. Even the ‘Artemia replacement’
products increasingly used in commercial operations
are normally used in co-feeding with live feeds (e.g.,

613


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

Curnow, King, Partridge & Kolkovski 2006; VegaOrellana, Fracalossi & Sugai 2006; Hamza, Mhetli &
Kestemont 2007; Rosenlund & Halldo¤rsson 2007).
However, the low digestive capacity of altricial larvae might not be the only aspect responsible for them
normally requiring live feed. Live preys are able to
swim in the water column and are thus constantly
available to the larvae. Most formulated diets tend to
aggregate on the water surface or, more commonly,
sink within a few minutes to the bottom, and are thus
normally less available to the larvae than live feeds.
In addition, since larvae are believed to be ‘visual feeders’, adapted to attack moving prey in nature, the
movement of live feed in the water is likely to stimulate larval feeding responses. Finally, live prey, with a
thin exoskeleton and a high water content (normally
4 80%), have a lower nutrient concentration and may
be more palatable to the larvae once taken into the
mouth, compared with the hard, dry formulated diet.
This last point is rather critical as any feed item must
enter the mouth whole, i.e., feed particles have to be
smaller than the larva’s mouth gape, and are quickly
accepted or rejected on the basis of palatability (FernaŁndez-D|¤ az, Pascual & Yu¤fera 1994; Bengtson 2003).
The present paper aims to review the main types of
live feeds used in aquaculture, their advantages and
pitfalls, with a special emphasis on their nutritional
value and the extent to which this may be manipulated. It also reviews the main concerns and potential
bene¢ts regarding the micro£ora composition of live
feed when it is added to larval tanks. Finally, it discusses the constraints of using live feeds to study ¢sh

larvae nutritional requirements, the possibilities of
using tracer studies to overcome such constraints
and assessment of the future of live feeds in ¢sh larvae
production.

Aquaculture Research, 2010, 41, 613^640

turn is used as food for the carnivorous larvae of
many of the marine ¢sh and shrimp species presently
farmed. Finally, intensive rearing of bivalves has so
far relied on the production of live microalgae, which
comprises on average 30% of the operating costs in a
bivalve hatchery.
For rearing marine ¢sh larvae according to the
‘green water technique’, algae are used directly in
the larval tanks. This technique is nowadays a normal procedure in marine larviculture, given that it
has been widely reported to improve ¢sh larval
growth, survival and feed ingestion (e.g., Òie, Makridis, Reitan & Olsen 1997; Reitan, Rainuzzo, Òie &
Olsen 1997). The observed larval quality enhancement when using microalgae in the rearing water
has been explained by di¡erent studies, which
showed that microalgae seemed to provide nutrients
directly to the larvae (Mo¡att 1981), to contribute
to the preservation of live prey nutritional quality
(Makridis & Olsen 1999), to promote changes in the
visual contrast of the medium and in its chemical
composition (Naas, Naess & Harboe 1992; Naas, Huse
& Iglesias 1996) and to play an important role in the
micro£ora diversi¢cation of both the tank and the
larval gut (Nicolas, Robic & Ansquer 1989; Reitan
et al. 1997; Skjermo & Vadstein 1999; Olsen, Olsen,

Attramadal, Christie, Birkbeck, Skjermo & Vadstein
2000). More recently, Rocha, Ribeiro, Costa and Dinis
(2008) showed that ¢sh larvae feeding ability is also
in£uenced by the presence of microalgae in the tank.
However, this e¡ect is not the same among species
and has been shown to be more pronounced with
gilthead seabream than with Senegalese sole larvae
(Rocha et al. 2008).

Main species of microalgae
Microalgae
Main utilizations of microalgae
Microalgae constitute the ¢rst link in the oceanic
food chain, i.e., the primary producer, due to its ability to synthesize organic molecules using solar energy. In aquaculture, microalgae are produced as a
direct food source for various ¢lter-feeding larval
stages of organisms such as bivalve molluscs (clams,
oysters and scallops), the larval stages of some marine gastropods (abalone) and early stages of penaeid
shrimp larvae (Yu¤fera & LubiaŁn 1990). They are also
used as an indirect food source, in the production of
zooplankton (e.g., rotifers and Artemia), which in

614

The ¢rst microalgae species produced for aquaculture were selected from those produced naturally in
pioneering ¢sh farms and were probably the easiest
ones to cultivate (Muller-Fuega, Moal & Kaasa
2004). However, other species were later investigated
based on their biological characteristics and performance under laboratory culture conditions, as well
as on their nutritional and energetic properties.
Among the most important selection criteria for

microalgae, the following can be highlighted:
1. cell size appropriate to the demands of the consumer organisms;
2. adequate nutritional value;
3. high digestibility;

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

4. ease of culture at high densities;
5. short life cycle, reproducible in captivity; and
6. tolerance to environmental variations.
Using these selection criteria, 16 genera of microalgae are generally produced nowadays. Still, some
species dominate and it is possible to relate such species to their corresponding utilization. Class Baccilariophyceae (Diatoms) are usually given to bivalve
molluscs and crustacean larvae as they are rich in
silicates, which constitute their cell walls (frustules)
and are necessary for bivalves and crustaceans for
the formation of rigid structures. Classes Prasinophyceae (e.g., Tetraselmis suecica, Tetraselmis Chuii) and
Chrophyceae (e.g., Dunaliella tertioleta, Chrorella minutissima) are ideal food for crustacean larvae, when
complemented by Baccilariophyceae for silicate supply. Class Prymnesiophyceae (e.g., Isochrysis galbana)
is widely used to culture marine ¢sh larvae (Brown
1991), while class Thraustochytriidae (e.g., Schizochtrium sp.), which consists of heterotrophic chromists,
has mostly been used as feed for live prey species
(Brachionus sp. and Artemia sp.).

Microalgae product types
A general feature of marine microalgae is their high
polyunsaturated fatty acid (PUFA) content. However,

the availability of microalgae as an a¡ordable PUFA
source is limited. At present, many ¢sh farms have
their own facilities to produce microalgae for use
during the ¢rst feeding of marine ¢sh and crustacean
larvae. The investment in such facilities is high and
can represent 30% of a hatchery operating cost
(Coutteau & Sorgeloos 1992). In addition, productivity may be variable depending on the season.
New products and methodologies with better coste¡ectiveness have been investigated and developed.
This includes microcapsules, dried microalgae,
yeasts or yeast-based diets, bacteria, thraustochytrids (Knauer & Southgate 1999; Langdon & Únal
1999) and algal pastes (Heasman, Diemar, O’Connor,
Sushames & Foulkes 2000). A fast-growing range of
such commercial products is available, including live
microalgae concentrates, and frozen and freeze-dried
microalgae. Results using these products are generally good. For instance, centrifuged concentrates of
Pavlova lutheri in combination with Chaetoceros calcitrans or Skeletonema costatum yielded 85^90% of the
growth of a mixed diet of live microalgae for oyster
larvae Saccostrea glomerata (Heasman et al. 2000).
Hatcheries that already have the infrastructure for

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

algal mass production may also prepare their own
concentrates on-site, and thus limit their algal production to less busy periods of the season, better
manage their requirements for microalgae and also
reduce costs due to over-production (Knuckey,
Brown, Robert & Frampton 2006).

Techniques for growing microalgae
Various techniques have been developed to grow microalgae on a large scale, ranging from less controlled extensive to monospeci¢c intensive cultures.

However, the controlled production of microalgae is
still a complex and expensive procedure. Culture of
microalgae for aquaculture purposes (rearing of mollusc, shrimp and ¢sh larvae) takes place mostly onsite, i.e., in the ¢sh farms where they are utilized,
although a new industry is emerging for the production of microalgae and delivery in lyophilized, frozen
or other form to the farms (Navarro & Sarasquete
1998; Muller-Fuega et al. 2004). There are various
methods for the culture of microalgae, but the induction of ‘blooms’ by fertilization (addition of nutrients
to the culture medium) is the most usual. In ¢sh
farms, stock cultures are kept under controlled conditions and protected from contamination by other microalgae, ciliates and potentially harmful bacteria.
Up-scaling of the microalgae cultures takes place in
several steps, and the ¢nal large-scale cultures are often located outdoors under natural light conditions.
Cultures may be monospeci¢c or polyspeci¢c. Polyspeci¢c cultures are normally carried out in open air
tanks, or ponds, with water volumes exceeding
100 m3, while monospeci¢c cultures are carried out
in medium (o100 m3) or small (o5 m3) containers,
in which environmental variables are controlled.
Microalgae cultures may be divided according to
the technology applied in the following types, listed
according to the ascending intensity of the culture
and average cell density: earth ponds, raceways, plastic polyethylene bags (100^500 L), open cylindrical
tanks constructed of polymer ¢breglass and tubular
and £at-plate photo-bioreactors (Zmora & Richmond
2004). Earth ponds and raceways are more open systems, exposed to weather conditions, and where
there is a higher risk of contamination. In addition,
the cell density remains at relatively low levels in
comparison with the other culture systems. Bags
and cylindrical tanks are quite common approaches
for the production of microalgae in ¢sh farms in the
Mediterranean region.

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

615


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

Photo-bioreactors are a relatively recent advance
for the production of microalgae in aquaculture
farms. They enable the production of large biomasses
of microalgae in high-cell-density cultures, in a small
area and with a low input of labour. Tubular (serpentine and manifold) and £at-plate bioreactors are
commonly used for this purpose. Small volumes of
high-density microalgae are easier to handle by ¢sh
farmers but present some problems as well, as they
require the use of experienced personnel and all
microalgae production is dependent on a few cultures. Such devices may need a cooling system, as
heat is accumulated in the system, and automated
addition of carbon dioxide.
In relation to the timing of harvest of microalgae
during the culture period, culture strategies can be
divided into: (i) batch cultures, where no growth
medium is added after initial inoculation; and semicontinuous cultures, where a part of the culture is
harvested and new growth medium is added subsequently, several times during the culture period (Ii,
Hirata, Matsuo, Nishikawa & Tase 1997). According
to Fogg (1975), the limiting factors for microalgae production by photoautotrophy, resulting from the production method, are the exhaustion of nutrients,
reduction in illumination due to the increase in cell
density (shadowing) and the inhibition of cell division due to the accumulation of catabolites. However,
the various factors may be interdependent and a

parameter that is optimal for one set of conditions is
not necessarily optimal for another.
Algal cultures need to be enriched with nutrients
in order to overcome the nutrient de¢ciencies of seawater. This includes the macronutrients nitrate and
phosphate (in an approximate ratio of 6:1), silicate (if
growing diatoms) and micronutrients, comprising
various trace metals and the vitamins thiamin (B1),
cyanocobalamin (B12) and sometimes biotin. The
Walne medium and the F/2 medium are the two most
extensively used enrichment media and are suitable
for the growth of most algae. There are also commercially available nutrient solutions that are suitable for
mass production of microalgae in large-scale extensive systems. These solutions contain only the most
important nutrients and are made of agriculturegrade rather than laboratory-grade fertilizers.
Heterotrophic culture may provide a cost-e¡ective,
large-scale alternative method of cultivation for some
microalgae that utilize organic carbon substances as
their sole carbon and energy source. This mode of
growth eliminates the requirement for light and,
therefore, o¡ers the possibility of considerably in-

616

Aquaculture Research, 2010, 41, 613^640

creasing the microalgal cell concentration and,
hence, volumetric productivity in batch systems. In
the last decade, knowledge of the cultivation of heterotrophic marine algae that accumulate PUFAs has
increased. However, knowledge and production of
such products are limited and restricted only to very
few companies (Muller-Fuega et al. 2004). Furthermore, as heterotrophic algae are not used directly as

feed in the aquaculture industry, their performance
is not known and is yet to be determined. Two heterotrophic species that are commonly used in live food
enrichment emulsions or in ¢sh diets, due to their
high levels of docosahexaenoic acid (22:6n-3; DHA),
are the dino£agellate Crypthecodinium cohnii and the
fungal thraustochytrid, Schizochytrium sp. (De Swaarf,
Pronk & Sijtsma 2003).

Nutritional value of microalgae
Whenever microalgae are used as a direct food
source or as an indirect food source, in the production of rotifers, Artemia or copepods, growth of the
animals is usually superior when a mixture of several microalgal species is used (Becker 2004). This
probably occurs as di¡erent species compensate one
another for eventual de¢ciencies in given nutrients.
Special care is needed when selecting microalgae for
ongrowing live feeds for marine ¢sh larvae, in order
to avoid the nutritional de¢ciencies of the latter, in
particular in terms of n-3 highly unsaturated fatty
acids (HUFA).
The dry matter composition of microalgae is highly
variable, even within a given species, with protein
contents ranging from 12% to 35%, lipid from 7.2%
to 23% and carbohydrates from 4.6% to 23% (Becker
2004). The protein content of microalgae is a major
factor in determining its nutritional value and may
change considerably with the composition of the
growing medium (Becker 2004). Nitrogen concentration seems to be particularly important in this respect.
De¢ciencies in the n-3 HUFA contents of microalgae may cause severe mortalities and/or quality problems in shrimp, mollusc and marine ¢sh larvae. In
addition, such de¢ciencies may also cause reduced
fecundity of rotifer and copepod cultures. Signi¢cant concentrations of eicosapentaenoic acid (EPA;

20:5n-3) are normally present in diatom species
(C. calcitrans, Chaetoceros gracilis, S. costatum andThalassiosira pseudonana), Nannochloropsis sp., T. suecica,
Tetraselmis chuii, D. tertioleta and C. minutissima
(Brown 1991; Becker 2004). High concentrations of

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

DHA are found in I. galbana and P. lutheri (Brown
1991; Becker 2004) and particularly in Thraustochytriidae (e.g., Schizochtrium sp.), which can contain
over 70% of its weight as lipids and have a DHA content up to 35% of their total fatty acids (Sijtsma & de
Swaarf 2004).

Microbial aspects in phytoplankton cultures
During the culture of microalgae, a high organic load
is progressively accumulated, which becomes the
substrate for the proliferation of bacteria. Bacterial
cells may attach to microalgae cells or may proliferate
in the water. Bacteria associated with microalgae
may reach very high values of culturable bacteria,
such as 108 mL À 1 or 103 cell À 1 (Salvesen, Reitan,
Skjermo & Òie 2000). In batch cultures, the level of
bacteria is lower during the exponential phase and
reaches a maximum during the stationary phase
(Salvesen et al. 2000; Makridis, Alves Costa & Dinis
2006). In semi-continuous cultures, where there is a
periodic replacement of part of the culture by new

growth medium, numbers of bacteria tend to stabilize at a value lower than that observed in similar
batch cultures. Unsuccessful culture of microalgae,
characterized by low growth rates and an extended
lag phase of the cultures, may result in a high bacterial load (Nicolas et al. 1989; Salvesen et al. 2000). The
cultured microalgae cells and the bacterial communities associated with the microalgae cultures are in
constant interaction, resulting either in the suppression of growth of speci¢c groups of bacteria or in decreased growth of the cultured microalgae (Munro,
McLean, Barbour & Birkbeck1995; Fukami, Nishijima
& Ishida 1997; Suminto & Hirayama 1997; Kokou,
Ferreira, Tsigenopoulos, Makridis, Kotoulas, Magoulas & Divanach 2007). The outcome of these interactions may depend on the method of microalgae
production, the microalgae species grown, the
growth media used, the quality of seawater and the
growth phase of the culture (Salvesen et al. 2000;
Makridis et al. 2006).
Antimicrobial activity has been detected in extracts of microalgae (Du¡ & Bruce 1966; Austin &
Day 1990; Austin, Baudet & Stobie 1992; Tendencia &
dela Penìa 2003) and in bacteria isolated from microalgae (Makridis et al. 2006). This antimicrobial activity can be caused by
(i) associated microbiota (Makridis et al. 2006);
(ii) antimicrobial proteins or fatty acids produced by
the microalgae cells (Kokou et al. 2007); or

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

(iii) free oxygen radicals produced due to the photosynthetic activity of the microalgae cells
(Marshall, de Salas, Oda & Hallegraef 2005).
Examination of the bacterial populations present in
microalgae cultures using molecular approaches
able to detect non-culturable and culturable bacterial strains revealed a di¡erent picture than studies
of the culturable microbiota. Cultures of P. lutheri,
I. galbana, C. calcitrans, S. costatum, C. gracilis and
Chaetoceros muelleri harboured a broad spectrum

of species belonging to the groups of a-Proteobacteria, b-Proteobacteria, g-Proteobacteria, Cytophaga^
Flavobacterium^Bacteroides (CFB) bacteria group,
Actinobacteria and Bacillus. Members of the Roseobacter clade and the CFB group were dominant in
the microalgae cultures. In microalgae cultures, culturable members of the Vibrio group were absent or
present in very low numbers (Salvesen et al. 2000;
Tendencia & dela Penìa 2003; Sainz-Hernandez &
Maeda-Martinez 2005; Makridis et al. 2006).
Addition of microalgae to the rearing tanks of marine ¢sh larvae has a positive e¡ect on the growth and
survival of the larvae. It has been suggested that this
positive e¡ect may be due to the bacteria associated
with the microalgae cultures (Reitan, Rainuzzo, Òie
& Olsen 1993). However, evidence for this is still weak
and further studies are needed.

Rotifers
Main utilization of rotifers
Since the 1970s, the rotifers and more speci¢cally
Brachionus plicatilis constitute an essential part of
the feeding during the larval stages of marine ¢sh
and crustaceans (Yu¤fera 2001; Lubzens & Zmora
2003). Its body size (between 70 and 350 mm depending on the strain and age) makes this organism an appropriate prey to start feeding after the resorption of
vitelline reserves of many species. In fact, Brachionus
is widely used as ¢rst food during a period of days or
weeks depending on the reared species, being replaced afterwards by a larger prey species, usually
Artemia nauplii (Yu¤fera, Rodr|¤ guez & LubiaŁn 1984;
Polo, Yu¤fera & Pascual 1992; Olsen et al. 2000). Obviously, ¢sh species having a wider mouth gape at the
onset of feeding may start directly on Artemia nauplii.
Besides the above-mentioned body size feature, the
main advantages of this organism to be used as live
prey in hatchery large-scale production are the following: (i) a high population growth rate, (ii) feeding

by ¢ltration of particles in suspension, being able to

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

617


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

ingest microalgae, yeast, bacteria and organic particles, (iii) a good tolerance to culture conditions
and handling, something common in euryhaline
animals, and (iv) to exhibit an appropriate energy
content and reasonable nutritional value that, in addition, is relatively modi¢able by dietary manipulation during routine feeding and/or by means of
speci¢c post-culture enrichment. Another advantage
of this prey is that it can be made permanently available. Rotifers are a renewable resource and the production chain can be completely established in the
hatchery, avoiding the dependence on external supplies. The major inconveniences arise from the occurrence of episodic collapses as well as from the e¡ort
required for maintaining the whole plankton chain.
Raising rotifers for larviculture requires the production of large amounts of organisms with appropriate size and nutritional quality. In order to meet
these objectives, many di¡erent strains have been isolated from nature and acclimated to the laboratory
conditions. In addition, di¡erent mass culture techniques have been developed, and a variety of microalgae species, yeast and commercial products have
been tested and used as food during the routine culture and the following enrichment period.

Main species and strains of rotifers
Variations in female’s body size were already observed
in the late 1970s, but were noted mainly during the
1980s when di¡erent morphotypes and strains were
described according to body size and spine shape. The
di¡erent strains were grouped into large (L-type),
medium (SM-type) and small (S-type) Brachionus plicatilis (Yu¤fera 1982; Fukusho & Okaushi 1983; Snell &

Carrillo1984; Fu, Hirayama & Natsukari 1991; Go¤mez
& Serra 1995). Currently, this group is considered to
be a multi-species complex of 9^15 di¡erent species
and biotypes. This complex includes species formally
described as Brachionus plicatilis sensu stricto,
Brachionus rotundiformis, Brachionus ibericus and
Brachionus manjavacas, together with a series of
lineages discernible by molecular techniques (Segers
1997; Ciros-Pe¤rez, Go¤mez & Serra 2001; Go¤mez, Serra,
Carvalho & Lunt 2002; Suatoni,Vicario, Rice, Snell &
Caccone 2006; Fontaneto, Giordani & Serra 2007;
Mills, Lunt & Go¤mez 2007). Besides the formal species, at least the lineages Brachionus sp. Nevada, Brachionus sp. Cayman and Brachionus sp. Austria have
been identi¢ed as common in hatcheries (Papakostas,
Dooms, Triantafyllidis, Deloof, Kappas, Dierckens,

618

Aquaculture Research, 2010, 41, 613^640

De Wolf, Bossier, Vadstein, Kui, Sorgeloos & Abatzopoulos 2006; Dooms, Papakostas, Ho¡man, Delbare,
Dierkens, Triantafyllidis, De Wolf, Vadstein, Abatzopoulos, Sorgeloos & Bossier 2007; Kostopoulou &
Vadstein 2007; Baer, Langdon, Mills, Schulz & Hamre
2008). The clari¢cation of the taxonomical situation
will continue in the coming years. From a practical
point of view, as a prey for larviculture, it is still useful
to use the identi¢cation of large (L), medium (SM) and
small (S and SS) morphotypes referring to the relative
body size (Table 1).

Rearing techniques for rotifers

Since the establishment of the basis for the culture of
Brachionus (Ito 1960), di¡erent techniques of mass
culture have been developed in order to obtain high
and constant productions (see for instance: Hirata &
Mori 1967; Theilacker & McMaster 1971; Hirata 1974;
Yu¤fera & Pascual 1980; Gatesoupe & Robin 1981;
Yoshimura, Hagiwara, Yoshimatsu & Kitajima 1996;
Suantika, Dhert, Nurhudah & Sorgeloos 2000; Dhert,
Rombaut, Suantika & Sorgeloos 2001; Park, Lee, Cho,
Kim, Jung & Kim 2001; Lubzens & Zmora 2003; Olsen
2004).
The Brachionus plicatilis complex of species and
biotypes reproduce mostly by parthenogenesis,
although a sexual phase may occur under speci¢c
environmental conditions. The most relevant aspect
is its high fecundity, which allows a population duplication time of 24^48 h (Hirayama & Kusano 1972;
Hirayama, Watanabe & Kusano 1973; Yu¤fera, LubiaŁn
& Pascual 1983; Korstad, Olsen & Vadstein 1989). Provided that the appropriate abiotic and feeding conditions are supplied, a single parthenogenetic female
Table 1 Accepted species (in bold) and other biotypes
belonging to Brachionus plicatilis species complex grouped
according to former size-related classi¢cation (Go¤mez et al.,
2002; Baer et al., 2008)
L---biotypes
220^340 lm

SM---biotypes
150^220 lm

S and SS biotypes
100^150 lm

B. plicatilis
B. manjavacas
B. ‘Nevada’
B. ‘Austria’

B. ibericus
B. ‘coyrecupiensis’
B. ‘almenara’
B. ‘tiscar’
B. ‘cayman’
B. ‘towerinniensis’

B. rotundiformis
B. ‘lost’

L, large; SM, medium; S and SS, small and super small. The numbers indicate the approximate range of adult’s body length in
each group.

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

may generate a large o¡spring in a few days. The rearing techniques developed for rotifers take advantage
of this characteristic of their reproductive biology.
The Brachionus population exhibits an exponential
growth while the favourable conditions persist, followed by a decrease and cessation in growth when
the food is exhausted or the chemical and microbiological conditions fall out beyond tolerance ranges.

Therefore, all the di¡erent culture systems that have
been developed attempt to maintain the exponential
growth by supplying high food levels and by preventing excessive accumulation of nitrogenous waste.
The best descriptor of the health and growth status
of the cultured population is the egg ratio. This index
is a direct indicator of fecundity, and therefore of the
potential for growth in the following hours (Snell &
Carrillo1984;Yu¤fera et al.1984). The best growth rates
occur approximately at a temperature between 20
and 35 1C, a salinity of 10^35 and a pH of 7.0^9.0
(Hirayama & Kusano 1972; Pascual & Yu¤fera 1983;
Miracle & Serra 1989; Yu¤fera & Navarro 1995). The renovation pattern of the culture media (water plus algal cells) determines the basic methods, from batch
culture (no renovation) to semi-continuous culture
(partial renovation) and continuous culture (permanent renovation). In addition, independent of the culture system, to establish a complete production chain,
three basic volume scales have to be considered: the
stock culture (50^500 mL) aiming to maintain the
genetic strain under optimal conditions; the starter
culture (from 5 to 50 L) proceeding directly from the
genetic strain and used to inoculate large volume cultures; and the mass culture (from 50 L to several m3).
Obviously, this basic chain may change according to
the operating conditions of each hatchery.
In the batch culture system, an inoculum with a
relatively low individual density is seeded in a dense
microalgae suspension and the rotifer population
grows for several days until the exhaustion of the algal cells. The growth pattern follows a sigmoid curve
and the maximum rotifer density attained increases
with an increase in the initial food concentration following a saturation response. The whole production
is harvested at the end of the exponential phase and
used. Commonly, a small part of this production is
used as a starter in the next production cycle two or

three times. The quality of this starter (inoculum) is
decisive for the success of the rotifer culture. The periodic use of the starter coming directly from the stock
culture guarantees the health of the cultured population and the stability of the production. This is a
very common system in hatcheries, and many exam-

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

ples of its application may be found in the literature
(Lubzens 1987; Dhert et al. 2001).
In the semi-continuous system, harvesting and
media renovation are frequent, and account for a notable part of the total volume. Harvested volume
ranges from 10% to 50% and the harvesting frequency ranges from 1 to 3 days. The culture may last
several weeks. When the renovation frequency is
every 24 h or less and the renovated volume is constant, the culture can be associated with a continuous culture (Boraas 1983; Schluter, Soeder &
Growneweg 1987). Like the batch system, the semicontinuous culture is widely used in larviculture
and is usually combined with batch culture (Hirayama & Nakamura1976; Suantika et al. 2000; Lubzens &
Zmora 2003; Olsen 2004).
The continuous rotifer culture systems are based
on the chemostat methodology (Droop & Scott 1978;
James & Abu Rezeq 1989). In chemostat-like systems,
the daily renovation rate is constant. After the initial
growth, the population reaches a steady state, maintaining an almost constant rotifer density during
weeks. During the steady state, the growth rate is
equal to the renovation rate. The female density attained during the steady state, and consequently the
production, depends on the food level (Boraas 1983;
Schluter et al.1987;Walz1993; Navarro & Yu¤fera1998).
The use of a concentrate microalgae paste, freezedried microalgae or commercial feeds allows higher
levels of cell and particle concentration to be attained
in the culture media than that normally obtained
with microalgae suspension cultures (Hirayama &

Nakamura 1976; Yu¤fera & Navarro 1995). This has
been the basis for the development of super-intensive
culture techniques in which the rotifer density and
production are notably higher than that obtained
with traditional methods (Yoshimura et al. 1996; Fu,
Ada, Yamashita, Yashida & Hino 1997; Suantika et al.
2000; Park et al. 2001; Suantika, Dhert, Rombaut,
Vandenberghe, De Wolf & Sorgeloos 2001;Yoshimura,
Tanaka & Yoshimatsu 2001). This kind of mass culture is associated with continuous or similar systems, with permanent addition of a high-quality diet
that ensures rapid rotifer growth. Rotifers may invest
a large amount of energy in relation to their body biomass in reproduction. Therefore, failure in food supply following a period of large investment in egg
production may result in collapse of the rotifer
culture. Another important issue in relation to highdensity production of rotifers is the removal of ammonia. Several techniques have been suggested, such
as lowering of pH, ion exchanger, membrane ¢lter

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

619


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

units or recirculation systems. The daily harvest depends considerably on the culture system and ranges
from 50 to 100 rotifers mL À 1 of culture in the batch
system up to 10,000 mL À 1 in the super-intensive systems. Given that these culture systems are quite variable but also somewhat £exible, each hatchery
usually develops a tailored system adapted to its speci¢c characteristics. For instance, tanks may be quite
di¡erent in volume and shape, even in the same facility; the water pre-treatment varies depending on the
seawater source and quality; the Brachionus strain is
selected according to the ¢sh species being cultivated

and to the environmental conditions; and the kind of
food for mass rearing and for the ¢nal nutritional
enrichment is selected according to the production
necessities and the desired ¢nal biochemical composition, as well as the routine procedures for harvesting and culture renovation.

Feeding and enrichments for rotifers
Brachionus is a suspension feeder that ingests cells
and particles available in the water column by ¢ltration. Many di¡erent potential feeds have been tested
in order to have a high fecundity and/or to achieve an
appropriate biochemical composition.Yeast, microalgae and commercial feeds are regularly used either
for population growth or nutritional enrichment.
Many microalgae species have been tested for rearing
Brachionus (Hirayama,Takaga & Kimura 1979;Yu¤fera
et al. 1983; Lubzens 1987; Yu¤fera & LubiaŁn 1990). The
most commonly used are Nannochloris spp., Nannochloropsis spp.,Tetraselmis spp., P. lutheri and I. galbana. Baker’s yeast Saccaromices cereviceae is also a
common and inexpensive food source for rotifer production in hatcheries that started being used early on
(Hirata & Mori 1967; Yu¤fera & Pascual 1980; James,
Dias & Salman 1987; Nagata & White 1992), usually
in combination with microalgae. Sprayed and freezedried microalgae and concentrated microalgae
pastes are also commonly used (Hirayama & Nakamura 1976; Gatesoupe & Luquet 1981; Dhert et al.
2001; Lubzens & Zmora 2003; Olsen 2004).
As in other zooplanktonic prey, rotifer biochemical
composition is of primary importance for larval nutrition. The lipid and essential fatty acids pro¢le is
relatively modi¢able by dietary manipulation (BenAmotz, Fishler & Schneller 1987; Lubzens, Tandler &
Minko¡ 1989; Rainuzzo, Reitan & Olsen1997). Baker’s
yeast and some microalgae species support high population growth but the rotifers produced are nutri-

620

Aquaculture Research, 2010, 41, 613^640

tionally de¢cient or poorly balanced in the lipid
fraction as larval food. In order to achieve an appropriate content of essential fatty acids, rotifers can be
enriched with microalgae (Òie, Reitan & Olsen 1994;
FernaŁndez-Reiriz & Labarta 1996; Hamre, Srivastava,
RÖnnestad, Mangor-Jensen & Stoss 2008), handmade marine oil emulsions (Watanabe, Kitajima &
Fujita 1983; Rodr|¤ guez, Pe¤rez, Bad|¤ a, Izquierdo, HernaŁndez-Palacios & Lorenzo 1998; Bell, McEvoy, Estevez, Shields & Sargent 2003; Hamre, Srivastava et al.
2008), prepared microparticulated feeds (Walford &
Lam 1987) or commercial products (FernaŁndezReiriz, Labarta & Ferreiro1993; Faulk & Holt 2005;Villalta, Este¤vez & Bransden 2005; Hamre, Srivastava
et al. 2008). A variety of enrichment protocols have
been described, but overall a re-feeding or an enrichment period of 8^24 h results in the desired biochemical composition. Nevertheless, feeding regimes
based on a combinetion of baker’s yeast and microalgae or an adequate combination of two microalgal
species (for instance: Tetraselmis sp. or Nannochloropsis sp.1I. galbana) during rotifer rearing can result in
an appropriate composition, with no need for further
enrichment (Nagata & White 1992; Klaoudatos, Iakovopoulos & Klaoudatos 2004). Furthermore, supplying microalgae directly to the larval rearing tank
contributes towards maintaining the nutritional
quality of rotifers up to their ingestion by the larvae.
Rotifers can also be enriched with speci¢c compounds to perform experiments in ¢sh larvae nutrition, such as vitamins, iodine or selenium (Gime¤nez,
Kotzamanis, Hontoria, Estevez & Gisbert 2007; Hamre,
Mollan, S×le & Erstad 2008). The protein content and
amino acid pro¢le is less modi¢able but may change
depending on the nutritional condition and reproductive status (Szyper 1989; Yu¤fera & Pascual 1989;
Yu¤fera, Parra & Pascual1997; Makridis & Olsen1999).

Nutritional value
Dry matter content, caloric value and chemical composition establish the nutritional value of rotifers,
and are determined by size and nutritional state
(Lubzens et al. 1989; Go¤mez et al. 2002; Lubzens &
Zmora 2003; Baer et al. 2008). According to Lubzens
and Zmora (2003), rotifer’s protein content ranges between 28% and 63%, lipid from 9% to 28%, and carbohydrate from 10.5% to 27% of the dry weight (DW).

The importance of rotifer HUFA contents in marine ¢sh larvae nutrition has long been established,
with numerous studies being available on rotifer lipid

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

enrichment procedures (see Lubzens et al.1989; Rainuzzo et al. 1997; Lubzens & Zmora 2003, for detailed
reviews). Rotifer lipids have a high phospholipid content (34^43% of total lipid), and 20^55% of triacylglycerols (Lubzens & Zmora 2003). However, when
cultured on yeast alone, rotifers are poor in HUFAs
required for normal development and good survival
of marine ¢sh larvae. DHA, EPA and arachidonic acid
(20:4n-6; ARA) contents can, however, be manipulated by short- or long-term feeding with selected
microalgae species, lipid emulsions, lipid-rich microcapsules and other feed types, as mentioned above
(Lubzens et al. 1989; Rainuzzo et al. 1997; Lubzens &
Zmora 2003).
Despite the high crude protein content of rotifers,
the availability of rotifer protein to marine ¢sh larvae
is of concern (Srivastava, Hamre & Stoss 2006). Rotifers are a common diet for ¢rst-feeding many marine
¢sh species, when the larval digestive system is still
largely immature (Govoni et al. 1986; RÖnnestad &
Conceic°aìo 2005). It has been proposed that soluble
protein may be more easily digestible in young ¢sh
larvae (Carvalho, SaŁ, Oliva-Teles & Bergot 2004). Srivastava et al. (2006) established that soluble protein
makes up to 50.6% of the rotifer crude protein.
Furthermore, growth is essentially protein deposition, and ¢sh larvae have very high amino acid requirements to support fast growth and high energy
demands (RÖnnestad, Tonheim, Fyhn, Rojas-Garcia,
Kamisaka, Koven, Finn, Terjesen, Barr & Conceic°aìo

2003; RÖnnestad & Conceic°aìo 2005). In addition, the
amino acid pro¢le of rotifer protein has been shown
to be unbalanced for several ¢sh larval species
(Conceic°aìo, Grasdalen & Ronnestad 2003; Aragaìo,
Conceic°aìo, Dinis & Fyhn 2004; Saavedra, Conceic°aìo,
Pousaìo-Ferreira & Dinis 2006; Saavedra, Beltran,
Pousaìo-Ferreira, Dinis, Blasco & Conceic°aìo 2007).
Such unbalances may a¡ect larval performance and
quality.
Rotifers seem to have high concentrations of vitamins C, E, B1 and B2 (van der Meeren, Olsen, Hamre
& Fyhn 2008). Rotifers cultured on commercial enrichment products seem to meet the larval requirements for all the B-vitamins, with the possible
exception of thiamine (Hamre, Srivastava et al.
2008). However, rotifers are possibly de¢cient in several minerals, including copper, iodine, zinc and, in
particular, manganese and selenium (Hamre, Srivastava et al. 2008). The last two were found in considerably lower amounts compared with copepods (Table 2;
Hamre, Srivastava et al. 2008), the natural diet of
marine ¢sh larvae.

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

Microbial aspects in rotifer cultures
The microbiota of cultured rotifers is similar to the
microbiota of the water in the cultures (Skjermo &
Vadstein1993). Rotifers are ¢lter-feeders, and are able
to ¢lter bacteria from the surrounding medium (Vadstein, Òie & Olsen 1993; Makridis, Fjellheim, Skjermo
& Vadstein 2000). The bacteria associated with rotifers grown on fresh baker’s yeast plus oil varied from
1.6 to 7.6 Â 103 bacteria per rotifer, whereas the
numbers of culturable bacteria in the culture water
microbiota ranged from 0.6 to 25 Â 107 CFU mL À 1
(Skjermo & Vadstein1993). The numbers of culturable
bacteria in rotifer cultures grown on microalgae

were very low (about 100 CFU rotifer À 1) in comparison with rotifer cultures grown on yeast (Òie et al.
1994). The culturable microbiota in samples from
rotifers fed baker’s yeast with added capelin oil
was dominated by members of the Pseodomonas/Alcaligenes group, Cytophaga/Flavobacterium group, Alteromonas and Vibrio (Skjermo & Vadstein 1993).
Culture-independent characterization of bacteria
associated with rotifer cultures showed a di¡erent
picture where members of Rhodobacteraceae and
Arcobacter were dominant, whereasVibrio, Alteromonas and Roseobacter were also detected in rotifers
grown on commercial diets (McIntosh, Ji, Forward,
Puvanendran, Boyce & Ritchie 2008). Analysis of mis
crobiota in rotifer cultures fed with Culture Selco by
denaturing gradient gel electrophoresis showed that
batch cultures showed continuous shifts in the dominant bands, whereas microbiota in rotifers produced
in a recirculation system was far more stable (Rombaut, Suantika, Boon, Maertens, Dhert,Top, Sorgeloos
& Verstraete 2001). The dominant bacterial species
present in this recirculation system belonged to
the genus Marinomonas and Pseudoalteromonas.
The high bacterial load present in rotifers during
the culture and enrichment process may cause problems in the rearing of marine ¢sh larvae. Opportunistic bacteria present in the rotifers may be
detrimental for the larvae (Skjermo & Vadstein 1999).
Rinsing of rotifers removes a large part of the bacteria
they carry. However, bacteria brought into the rearing system may become a problem in later stages of
the rearing. Several approaches have been suggested
for the removal of opportunistic bacteria, such as the
use of:
1. various chemicals and disinfectants (Tanasomwang & Muroga 1992);
2. ultraviolet radiation (Munro, Henderson, Barbour
& Birbeck 1999);

r 2009 The Authors

Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

621


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

Aquaculture Research, 2010, 41, 613^640

Table 2 Average biochemical composition of copepod nauplii, copepods, enriched rotifers and enriched Artemia, based on
values published by (1) van der Meeren et al. (2008), (2) Moren et al. (2006), and (3) Hamre, Srivastava et al. (2008)
Parameter

Unit

Source

Copepod nauplii

Rotifers

Copepods

Artemia M24

DW
Protein
FAA
Total lipids
Polar Lipids

DHA
EPA
DHA/EPA
EPA/ARA
Vit E
Vit C
Vit B1
Iodine
Se

(mg/individual)
(%/DW)
(%/DW)
(%/DW)
(%/DW)
(%/DW)
(%/DW)

1
1
1
1
1
1
1
1
1
1
1
1

2.3
3

0.6
30.3
8.6
8.6
5.4
3.5
1.4
2.8
27.7
–
–
–
–
–

0.6
25
1.7
15
6.1
1.9
1.1
1.7
3.7
513
220
49

4.7
0.08

8.8
43
6.0
11
6.2
3.5
1.8
2.2
24.0
113
515
23
121
4.00

2.1
29
3.3
25
4.5
2.7
1.9
1.4
4.0
456
446
16

1.1
–

(mg/g DW)
(mg/g DW)
(mg/g DW)
(mg/g DW)
(mg/g DW)

Note: Values presented are average values from single studies. Caution is needed in using these data, as a wide variation for many of
these parameters has been reported in di¡erent studies, and also within the studies considered here.
DW, dry weight; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; ARA, arachidonic acid; FAA, free amino acids.

3. ozone (Suantika et al. 2001); and
4. live bacterial additives (probiotics).
Addition of probiotics in live food may have several
purposes: (i) inoculation of bacteria that have a positive e¡ect on the growth and survival of the live feed
organisms in culture (Ii et al. 1997; Rombaut, Dhert,
Vandenerghe, Verschuere, Sorgeloos & Verstraete
1999; Douillet 2000; Dhert et al. 2001); addition of
bacteria that limit the growth of opportunistic bacteria harmful for the Artemia and the larvae (Gatesoupe, Arakawa & Watanabe 1989; Makridis 2000)
and (iii) addition of bacteria that may colonize the
¢sh larvae and may have a positive e¡ect on the
growth and survival rates of the larvae (Makridis,
Martins, Reis & Dinis 2008).

Artemia
Main utilizations of Artemia
The brine shrimp Artemia, together with rotifers, are
the most widely used live preys in aquaculture. As a

consequence of its high cost (due to low and unreliable natural resources and increased demand) and
low DHA content, as will be discussed below, a popular strategy in larviculture of many marine species
has been to attempt early weaning in conjunction
with a prolonged rotifer feeding period to eliminate
the need to use Artemia. Nonetheless, this is not always possible and in some species whose larvae are
relatively larger at hatching, Artemia might even be

622

the only live prey used in larviculture. Its high popularity in both aquaculture and aquarium pet trade
stems mostly from its ease of handling and mass culture. One of the most interesting features of this organism is its ability to form dormant cysts that are
highly resistant to adverse environmental conditions, can be kept viable for years (remarkable ‘shelflife’) and are extremely convenient to transport, store
and use. They are normally kept under dry (vacuum)
and cool conditions and, when needed, they can be
simply rehydrated in water, under favourable environmental conditions, and hatch as a nauplii in
o24 h. The ease and simplicity of hatching brine
shrimp nauplii makes them the most convenient and
least labour-intensive live foods available for aquaculture (Lavens & Sorgeloos 2000). The nauplii stage (instar I) is still dependent on its endogenous reserves,
its digestive tube is not yet completely formed and
the mouth and anus are closed. After about 8 h, the
animal moults into the metanauplius I (instar II)
stage, when it starts ¢lter-feeding small food particles
(6.8^27.5 mm, with an optimum of16.0 mm) (Van Stappen 1996a; Fernandez 2001). Metanauplius are continuous non-selective ¢lter-feeding organisms, just
as rotifers. This is another important characteristic
of these organisms that was the basis for the development of the bioencapsulation process (Van Stappen
1996a), which enables enrichment with nutrients
(mostly lipids, fatty acids and vitamins) or any other
substances (hormones, chemotherapeutics or prophylactic products and vaccines, particularly if

r 2009 The Authors

Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

lipo-soluble), as long as the particles are of an adequate dimension. This characteristic is quite important, given the biochemical composition of this prey,
which was selected more due to its convenience of
use than for its nutritional value for marine ¢sh larvae. The term bioencapsulation has, however, been
reviewed and should be avoided, as it later became
evident that Artemia are not mere passive carriers of
fatty acids, as will be discussed below.

Main species of Artemia
The brine shrimp Artemia is a relatively primitive
form of crustacean of the class Branchiopoda. The
lack of a true carapace places them in the suborder
Anostraca, and further in the family Artemidae.
There is still disagreement and controversy regarding the taxonomy of the di¡erent Artemia species.
Originally, taxonomists gave species names to populations with di¡erent morphologies, collected at different temperatures and salinities (Van Stappen
1996a), but this was later abandoned and most aquaculture-related literature only uses the genus designation Artemia spp. Molecular techniques have been
recently used to clarify the phylogenetic relationships of bisexual and parthenogenic Artemia species
and populations but this has not yet been completely
resolved (Baxevanis, Kappas & Abatzopoulos 2006;
Hou, Bi, Zou, He, Yang, Qu & Liu 2006). A multitude
of di¡erent species and strains of Artemia cysts can be
harvested along the shorelines of hypersaline lakes,
coastal lagoons and solar saltworks from all over the
world (Van Stappen 1996a), and eight species have
been documented in scienti¢c literature: three from
the NewWorld ^ Artemia franciscana, Artemia persimilis and Artemia monica ^ and ¢ve in the Old World ^

Artemia salina, Artemia urmiana, Artemia sinica,
Artemia sp. and Artemia tibetiana (Hou et al. 2006).
These species and strains may di¡er considerably in
the diameter of cysts and the length of hatched nauplii (varying between 400 and just over 500 mm),
hatching synchrony and e⁄ciency (normally 150^
250 000 nauplii are produced per gram of cysts),
growth rate, biochemical composition (essential fatty
acids, ascorbic acid, pigments and trace elements)
and thus nutritional value, but important di¡erences
can also be found within strains, depending on the
batch and harvesting, processing or storage conditions (Van Stappen 1996a). Broadly speaking, and
from a nutritional point of view, Artemia spp. may be
divided into a marine or a freshwater origin and these

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

two groups may be distinguished by their lipid and
fatty acid composition, with marine-type strains presenting higher levels of total lipid and triacylglycerols, as well as higher levels of EPA and ARA and
lower levels of linolenic acid (LNA; 18:3n-3) than
freshwater-type strains (Navarro, Amat & Sargent
1993).
Commercial harvesting of Artemia cysts has historically been performed in two major areas in the
United States ^ Great Salt Lake (GSL), UT, and San
Francisco Bay, CA. However, from the mid-1980s onwards, GSL cysts dominated the world market, and by
2000 over 90% of the cysts sold commercially were
harvested in this lake, creating a problem of overdependence of the world aquaculture industry on a single (and unreliable) source of this valuable resource.
Other natural sources then started to be evaluated
for their commercial potential, being located mainly
in Central Asia ^ Iran, China, Siberia and Turkmenistan ^ and in Argentina. Quite a few semi-natural or
managed (resulting from deliberate inoculation of

Artemia in solar salt works) production sites are also
being exploited worldwide but their production is typically quite low, only meeting the local demand (Van
Stappen 1996a; Lavens & Sorgeloos 2000).
A common protocol in many marine hatcheries
worldwide is to feed the younger larvae with newly
hatched nauplii from selected strains and batches
that produce small nauplii with a high EPA content,
replacing them, as the ¢sh larvae are able to accept
larger prey, by enriched Artemia metanauplii (GSL).
After 12 and 48 h of enrichment, GSL Artemia metanauplii are typically around 660 mm and 790 mm long
respectively (Merchie 1996).

Feeding and enrichments for Artemia
An important limitation of Artemia spp. enrichment
as a tool to study larval nutritional quantitative and
qualitative requirements is the notorious lack of consistency in this procedure, as considerable variability
has been reported in the essential fatty acid content
after Artemia spp. enrichment despite attempts to
standardize protocols (Merchie 1996). An important
fraction of the ¢ltered lipids is digested, assimilated
into the Artemia body and metabolized, and not just
simply accumulated in the gut. In addition to the differential metabolism of certain fatty acids, incorporated fatty acids redistribute themselves among lipid
classes with high unpredictability, both during enrichment and particularly under starving conditions,

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

623

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

after being added to the larval rearing tanks (Navarro, Henderson, McEvoy, Bell & Amat 1999). The high
variability in the ¢nal composition stems from the
fact that we are dealing with live organisms with an
active metabolism and physiology, particularly in the
early stages of development, which may also show
asynchronous development. Their physiology can be
a¡ected by a multitude of exogenous (e.g., environmental, such as temperature, aeration, hydrodynamics in the enrichment media, or size of micelles
in the emulsion) and endogenous (e.g., stage of development and intrinsic metabolic rate) factors that cannot be fully controlled or standardized (Navarro et al.
1999).
A large variety of lipid enrichment products, commercially available or ‘home made’, have been used
over the years. Classically, lipid enrichment protocols
were developed using marine oil (mostly ¢sh oils)
emulsions, where tiny micelles (droplets) of triacylglycerols are produced and can be ¢ltered from
the water. This, however, tends to result in a prey
with high levels of neutral lipids and potentially imbalanced protein/lipid ratios and an excess of triacylglycerols (Sargent, McEvoy, Estevez, Bell, Bell,
Henderson & Tocher 1999; Morais, Conceic°aìo, RÖnnestad, Koven, Cahu, Infante & Dinis 2007). On the
other hand, a well-documented high larval requirement has been established for phospholipids and a
bene¢cial e¡ect of supplying essential fatty acids as
phospholipids, rather than as neutral lipids, has also
been recognized (Coutteau, Geurden, Camara, Bergot & Sorgeloos 1997; Gisbert, Villeneuve, Zambonino-Infante, Quazuguel & Cahu 2005). In addition, in
an attempt to produce preys with high levels of PUFA
to meet essential fatty acid requirements, but that are
not good substrates for energy-generating fatty acid
oxidation systems, the necessity to provide a correct
balance between energy (i.e., saturated and monounsaturated fatty acid supply) and essentiality (highly
unsaturated fatty acids; HUFA) should not be overlooked (Sargent, McEvoy et al. 1999). Finally, high levels of highly digestible proteins and peptides, which
are the building blocks for growth, are most probably
required during fast larval growth, although macronutrient requirements have seldom been determined

in marine ¢sh larvae.
Advancements in the knowledge of marine ¢sh larvae lipid requirements and in marine biotechnology
and industrial feed processing technologies led to
the development of alternative enrichment products
and additives based in unicellular organisms, such as
yeast, moulds, bacteria and microalgae. Some exam-

624

Aquaculture Research, 2010, 41, 613^640

ples are spray-dried single-cell algal and fungal heterotrophic and phototrophic organisms, such as
Schizochytrium sp. and Crypthecodinium sp., Nanochloropsis sp., Mortierella alpina or Haematococcus pluvialis, for instance, that have very high contents of
DHA, EPA, ARA or astaxanthin, respectively, and
high levels of polar lipids (Harel, Koven, Lein, Bar,
Behrens, Stubble¢eld, Zohar & Place 2002; Dominguez, Ferreira, Coutinho, Fabregas & Otero 2005). Inactivated yeast, for instance, can also serve as an
additional source of amino acids and micronutrients,
such as nucleic acids, vitamins and b-glucans. These
single-cell products have several advantages in relation to emulsi¢ed oil products: PUFA are less exposed
to the atmosphere and more protected against oxidation by the cells, their use minimizes contamination
of the enrichment media with bacteria that thrive in
emulsi¢ed oils and they provide a broader supply of
other natural nutrients, besides lipids (e.g., protein,
xanthophylls, vitamins, sterols and other trace elements) (Song, Zhang, Guo, Zhu & Kuang 2007; Yamasaki, Aki, Mori, Yamamoto, Shinozaki, Kawamoto &
Onu 2007). Another very important characteristic is
the possibility to combine di¡erent microorganisms
providing rich sources of individual fatty acids,
which allows the production of a much larger range
of products capable of meeting the species-speci¢c
nutritional requirements of marine ¢sh larvae. This

is a major advantage, compared with ¢sh oil-based
products, which are limited in terms of the absolute
levels and relative proportions of DHA:EPA:ARA that
can be supplied. Even though high DHA-containing
¢sh oils can be obtained, either from extracting lipids from speci¢c tissues (cod liver oil, tuna orbital
oil) or using special extraction procedures (silage,
cold acetone), the availability of these products is
very limited and often prohibitively expensive to
produce (Harel et al. 2002). Given the present situation of the natural ¢sheries resources, the variable
quality, lowering supply and increasing costs of ¢sh
oil-based products are likely to drive forward the industrial production of the above-mentioned unicellular organisms. In particular, the production of
heterotrophic algae and fungi in conventional fermentors may become an important cost-e¡ective
alternative (Harel et al. 2002). Still, emulsion-based
commercial products using marine ¢sh oils are
widely used but they now tend to include other ingredients from an extensive list of other stabilized
supplements, such as free fatty acids, phospholipids,
plant and algae extracts, vitamins, minerals, carotenoids and other pigments, antioxidants, proteolytic

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

enzymes, immunostimulants and bacteriostatic
additives.
Apart from essential fatty acids, whose e¡ects have
been extensively studied and their presence judged to
be critical in marine ¢sh larvae diets, other nutrients,
such as other lipid classes, certain peptides, free amino acids, pigments, sterols, minerals and vitamins,

might be equally important. Enrichment procedures
for vitamins A, C and E have been developed since
the 1990s (Merchie 1996; Sorgeloos, Dhert & Candreva 2001) but research is required to identify some
of these other dietary components and their nutritional requirement levels, and strategies need to be
devised to include them in an Artemia-based larval
feeding regime (e.g., iodine; Moren, Opstad, Van der
Meeren & Hamre 2006). Liposomes and lipid microbeads have high potential as delivery systems of
some nutrients, not only of lipid and lipid-soluble nutrients but particularly of water-soluble components,
that may not be incorporated e⁄ciently into emulsions (Tonheim, Koven & Roennestad 2000; Monroig,
Navarro, Amat, GonzaŁlez, Amat & Hontoria 2003;
Barr & Helland 2007; Nordgreen, Hamre & Langdon
2007).

Rearing techniques for Artemia
Protocols for hatching and enriching brine shrimp
Artemia have been optimized over the years and are
now routinely used in a standardized manner in
commercial hatcheries and research institutes all
over the world. A large amount of literature and practical manuals are widely available (e.g., Lavens & Sorgeloos 1996; Sorgeloos et al. 2001), and most scienti¢c
research uses standard protocols. Therefore, rearing
conditions will only be mentioned brie£y here.
Hatching of brine shrimp Artemia is performed in
funnel-shaped containers aerated from the bottom,
at a temperature between 25 and 28 1C, a salinity of
15^35, pH over 8, oxygen levels close to saturation,
cyst densities of not 42 g L À 1 and with strong illumination (around 2000 lx at the water surface) (Van
Stappen 1996b). These conditions will trigger the
start of the Artemia hatching metabolism and have
been optimized to ensure optimal hatching synchrony and e⁄ciency. In some hatcheries, bleach is
used to decapsulate the cysts before hatching, to not

only facilitate hatching but also as a prophylactic
measure (to disinfect, i.e., remove bacteria and fungi
that are normally present in cyst shells). In this case,
care should be taken to avoid prolonged exposure to

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

hypochlorite solution and chlorine then needs to be
deactivated in hydrochloric acid. After a period of
around 24 h, the newly hatched nauplii are separated
from unhatched cysts, empty cyst shells (if not decapsulated) and dead nauplii usually by taking advantage of the positive phototactic behaviour of the
nauplii, and are thoroughly rinsed with freshwater,
preferentially in submerged ¢lters to prevent physical
damage of the nauplii. These are then supplied directly to the larval rearing tanks, when instar I are
required, or transferred to new clean tanks, for
further enrichment.When feeding larvae with instar
I nauplii, care should be taken not to supply a surplus
of prey and leave uneaten nauplii for a long time in
the larval tank, as they are consuming their own energy reserves and their level of free amino acids is reducing and therefore their nutritional value and
digestibility is quickly decreasing. In addition, they
rapidly moult into the second-instar metanauplii,
which are larger, faster swimming and more transparent (due to the consumption of the brownish orange yolk reserves), thus being less accessible preys
(Navarro, Amat & Sargent 1991; Merchie 1996; Sorgeloos et al. 2001). Storing the newly hatched nauplii at
a temperature below 10 1C in densities of up to 8 million nauplii L À 1, for periods up to 24 h, and distributing them more sparingly and frequently is a way to
minimize this problem (Le¤ger,Vanhaecke & Sorgeloos
1983; Merchie 1996).
Enrichment is typically conducted at around 28 1C
(up to 30 1C) for 24 h in clean, ¢ltered seawater but in
some cases manufacturers of speci¢c commercial enrichment products may advise lower temperatures
and enrichment times (e.g., 23^24 1C for 16^20 h).

Some of the advantages of reducing temperature
and enrichment time are a reduction in Artemia metabolism and DHA catabolism (Navarro et al. 1999),
thus reducing energy, phospholipid and essential
fatty acid losses, a reduction in the risk of autoxidation of HUFA during enrichment (McEvoy, Navarro,
Bell & Sargent 1995) and the production of smaller
prey sizes, which might be advantageous for some
species or younger larvae. Artemia density is normally 100 000^300 000 nauplii l À 1 of water and
vigorous aeration is used, to maintain oxygen levels
above 4 ppm. The enrichment product is typically
blended well with water and added to the enrichment
at a variable dose depending on the manufacturer’s
instructions (often around 0.2^0.3 g L À 1 for 8^12 h).
Usually, two doses are used, with a second dose being
added to the tank after 8^12 h. At the end of the enrichment period, metanauplii are rinsed thoroughly,

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

625


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

sometimes with warm water to help eliminate oily
residues, and are fed to the larvae.

Nutritional value of Artemia
According to Garcia Ortega, Verreth, Coutteau,
Segner, Huisman and Sorgeloos (1998), the proximate
composition (in DW) of newly hatched Artemia is

56.2% protein, 17.0% lipid, 3.6% carbohydrate and
7.6% ash. Decapsulated cysts can be used directly as
food for smaller larvae of some freshwater ¢sh and
marine shrimp but its use is much more limited. The
composition of decapsulated Artemia cysts is globally
the same as newly hatched nauplii, with about 50^
57% protein, 13^14% lipid, 6^7% carbohydrate and
5^9% ash, but their DW and energy content is on
average 30^40% higher than instar I nauplii. On the
other hand, compared with instar I nauplii, decapsulated cysts have a lower ratio of free amino acids to
total protein and of polar lipid and free fatty acids relative to total lipid, vitamin C is present as ascorbic
acid-2 sulphate, which has lower bioavailability and
carotenoids as a less stable form, cis-canthaxanthin
(Van Stappen 1996a; Garcia Ortega et al. 1998).
A huge disadvantage of brine shrimp is their inherent de¢ciency in essential fatty acids. They lack
DHA and have low levels of EPA (even the so-called
marine type) and are richer instead in LNA and to a
lesser extent in linoleic acid (LA;18:2n-6). As already
discussed above, enrichment techniques and products have been developed in order to overcome this
essential fatty acid de¢ciency and are generally quite
e¡ective in boosting the brine shrimp levels of EPA
and ARA. However, there still remain di⁄culties in
achieving high levels of DHA and a correct balance
between DHA and EPA, given the natural tendency
of Artemia to retroconvert DHA into EPA (Navarro
et al. 1999), leading to a product with a low DHA:EPA
ratio. Furthermore, the main PUFA found in Artemia
phospholipids is LNA, followed by EPA, meaning
that very high dietary levels of DHA are required for
the ¢sh larvae to be able to outcompete and replace

the LNA in the ingested phospholipids (in addition to
the high levels found in triacylglycerols), thus enabling the larvae to produce DHA-rich phospholipids
required for tissue growth and development (Sargent,
McEnvoy et al.,1999). This has classically been one of
the major drawbacks in the use of Artemia for the larval rearing of most marine ¢sh species. Even if ¢sh
larvae have some capacity to reconvert a fraction of
the EPA back into DHA and thus reverse the undesir-

626

Aquaculture Research, 2010, 41, 613^640

ably high EPA:DHA ratio in the live prey, studies conducted with marine ¢sh indicate that this capacity is
extremely limited and certainly insu⁄cient to meet
the high DHA requirements of rapidly growing and
developing ¢sh larvae (Sargent, Bell, McEvoy, Tocher
& Estevez 1999). Given the extremely important function of DHA as one of the main components of biological membranes, particularly in neural tissue, and
the inherent incapability of marine larvae to biosynthesize it ‘de novo’, DHA nutritional de¢ciencies
have major e¡ects on larval growth and development
of neural and visual systems (with major impacts on
the ability of larvae to capture their prey) (Sargent,
Bell et al. 1999; Sargent, McEvoy et al. 1999).
One way to minimize the problem of the low
DHA:EPA ratio in Artemia and to attempt to elevate
it over the 2 or a higher ratio that is advisable for marine ¢sh larvae (Sargent, Bell et al. 1999) is to use ‘speciality oils’or marine products with very high levels of
DHA and high DHA:EPA ratios. Some of the products
that have been classically used are tuna orbital oil
and either oil extracts or whole-dried or freezethawed cells of DHA-rich algae (e.g., Schizochytrium
sp. and C. cohnii). It should, however, be kept in mind
that the use of these ‘speciality products’ not only has

high associated costs but they also have limited availability, due to competition with the human consumption market, where they are also extensively used
(e.g., to supplement infant formulas and used in the
human food processing industry). Therefore, they
are being commonly used in research, in dose^response studies (as will be discussed below), but not
as much in routine commercial practices.
The protein contents of Artemia are low compared
with the presumptive requirements of marine ¢sh
larvae (Conceic°aìo et al. 2003), and with the values
normally observed in their natural diet ^ copepods
(Table 2; van der Meeren et al. 2008). This is particularly true as Artemia are normally enriched with lipid
emulsions, in an attempt to meet marine ¢sh larvae
DHA requirements. This results in a high Artemia lipid content (mainly neutral lipids) at the expense of
protein. It should be noted that excessive neutral lipid
in Artemia has been shown to a¡ect larval gut function (see Morais et al. 2007 for a review). Moreover,
growth is essentially protein deposition, and ¢sh
larvae have very high amino acid requirements to
support fast growth and energy demand (RÖnnestad
et al. 2003; RÖnnestad & Conceic°aìo 2005). In addition, the amino acid pro¢le of Artemia protein has
been shown to be unbalanced for several larval
species (Conceic°aìo, Dersjant, Li & Verreth 1998;

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

Conceic°aìo et al. 2003; Aragaìo et al. 2004; Saavedra
et al. 2006, 2007). Such unbalances may cause
higher mortalities (Aragaìo, Conceic°aìo, Lacuisse,

Yu¤fera & Dinis 2007) or a higher incidence of skeletal
malformations and lower nitrogen utilization
(Saavedra, Pousaìo-Ferreira,Yu¤fera, Dinis & Conceic°aìo
2009) in ¢sh larvae.
Artemia has high contents of canthaxantin, comparable to the astaxanthin contents of copepods
(van der Meeren et al. 2008). These pigments may
have important roles as antioxidants and sources of
vitamin A in ¢sh larvae nutrition (van der Meeren
et al. 2008). Artemia also seem to have high concentrations of vitamins C, E, B1 and B2 (van der Meeren
et al. 2008). However, iodine, and possibly other
minerals, contents in Artemia may be below the requirements (Moren et al. 2006). Iodine limitation has
been identi¢ed as a possible cause of malpigmentation and other metamorphosis problems in £at¢sh
larvae (Moren et al. 2006) due to its precursor role
for thyroid hormones.

Microbial aspects in Artemia
The initial content of bacteria in dryArtemia cysts depends on the origin and on the treatment of the cysts
during harvest and packaging. Quanti¢cation of the
numbers of aerobic heterotrophic bacteria in homogenates of dry Artemia cysts showed that the numbers of bacteria associated with the cysts were
equivalent to less than one bacterium per cyst
(Austin 1982). During hatching and enrichment of
Artemia, there is an overload of organic material in
the incubation water, which becomes the substrate
for the proliferation of opportunistic bacteria. The
numbers of bacteria increase exponentially and may
reach values as high as 108 colony-forming units
(CFU) mL À 1 in water and 104 CFU per Artemia after
enrichment (Villamil, Figueras, Toranzo, Planas &
Novoa 2003). These values may be even higher in the
case of on-grown Artemia (Olsen et al. 2000). Qualitative analysis of the Artemia micro£ora indicated that

the largest proportion of bacteria belongs to theVibrio
group, whereVibrio alginolyticus is the dominant species (Villamil, Figueras, Toranzo et al. 2003). A high
proportion of the bacteria associated with Artemia
metanauplii are haemolytic bacteria (Olsen et al.
2000). In the case of newly hatched nauplii, a high
percentage of bacteria are easily removed by rinsing.
In contrast, in the case of 24-h enriched Artemia
metanauplii and on-grown Artemia, it is more di⁄-

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

cult to remove bacteria e⁄ciently by simple rinsing.
Metanauplii are able to ¢lter bacteria and accumulate
them in large numbers (Makridis & Vadstein1999). In
the case of enrichment with Algamac, a commercial
spray-dried microalgae, there is even more rapid proliferation of bacteria than after incubation with oil
emulsions (Ritar, Dunstan, Nelson, Brown, Nichols,
Thomas, Smith, Crear & Kolkovski 2004). Several
strategies have been suggested to limit the presence
of opportunistic bacteria in Artemia used for feeding
of ¢sh larvae, which include the use of probiotic bacteria that limit the growth of opportunistic bacteria
or have a positive e¡ect on the growth and survival
of Artemia (Verschuere, Rombaut, Huys, Dhont,
Sorgeloos & Verstraete 1999; Makridis et al. 2000;
Gatesoupe 2002; Villamil, Figueras, Planas & Novoa
2003), incubation in microalgae (Olsen et al. 2000;
Makridis et al. 2006), or use of disinfectants, such as
formaldehyde (Tovar, Zambonino, Cahu, Gatesoupe,
VaŁzquez-JuaŁrez & Le¤sel 2002), short-chain acids
(Defoirdt, Crab,Wood, Sorgeloos,Verstraete & Bossier

2006) and quorum-sensing-disrupting furanones
(Defoirdt, Halet, Sorgeloos, Bossier & Verstraete
2006; Defoirdt, Boon, Sorgeloos, Verstraete & Bossier
2007).

Copepods and other natural zooplankton
Main utilizations of copepods
Copepods and other natural zooplankton organisms
have also been used as live feeds, in particular for
marine ¢sh larvae. In fact, they are the diet for ¢sh
larvae of most species in nature. Despite signi¢cant
progress in copepod cultivation methods (Payne &
Rippingale 2001; StÖttrup 2003; Lee, O’Bryen &
Marcus 2005), establishing cost-e¡ective protocols
for mass production is still a challenge. Copepods
have mostly been used at a pilot scale or in locations
where abundant collection of natural (or induced)
zooplankton blooms is possible. Still, and as could be
expected, the use of copepods as live feeds for marine
¢sh larvae has generally led to considerably better results in terms of larval performance and quality,
when compared with rotifers and/or Artemia (N×ss,
Germain Henry & Naas 1995; van der Meeren & Naas
1997; StÖttrup, Shields, Gillespie, Gara, Sargent, Bell,
Henderson, Tocher, Sutherland, N×ss, MangorJensen, Naas, van der Meeren, Harboe, Sanchez,
Sorgeloos, Dhert & Fitzgerald 1998; Shields, Bell, Luizi, Gara, Bromage & Sargent 1999; Hamre, Opstad,
Espe, Solbakken, Hemre & Pittman 2002; Rajkumar

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

627


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

& Vasagam 2006), in particular for Atlantic halibut
and Atlantic cod larvae. Therefore, and whenever
copepods are available, it might be a good strategy
to develop a baseline performance curve using
copepods as live feeds, in particular when developing
larval rearing protocols for ‘new species’. In situations where copepods are available in limited
amounts, their use as a fraction of the daily ration
for marine ¢sh larvae, in particular during the ¢rst
days of feeding, has also proven to improve larval
growth and survival (Conceic°aìo, van der Meeren,Verreth, Evjen, Houlihan & Fyhn 1997; Toledo, Golez, Doi
& Ohno 1999). This seems particularly interesting for
species requiring very small prey at ¢rst feeding, such
as grouper (Toledo et al. 1999; Toledo, Golez & Ohno
2005) and red snapper (Ogle, Lemus, Nicholson,
Barnes & Lotz 2005). Copepod nauplii blooms may
be induced in the rearing tanks or in separate tanks/
ponds. It has also been proposed that mass production of copepod resting eggs could facilitate availability of copepod nauplii for aquaculture (N×s & Bergh
1994; Marcus 2005). However, research is needed on
storage conditions of resting eggs in relation to the
survival and nutritional value of nauplii.

Main species of copepods
Free-living copepods belonging to the orders Calanoida, Harpacticoida and Cyclopoida are the most commonly used in aquaculture (StÖttrup 2003).
Calanoids, e.g., Acartia sp., Eurytemora a⁄nis, Centropages hamatus and Gladioferens imparipes, are by far
the most used for marine ¢sh larvae (see StÖttrup

2003 for an overview). However, their maximum
density in culture is an important limitation, and
most Calanoid uses in aquaculture result from collection of wild populations or blooms induced in con¢ned areas, such as lagoons or large ponds.
Harpacticoids, e.g., Tisbe sp., Euterpina acutifrons,
Tigriopus japonicus and Nitokra sp., are easier to culture compared with Calanoids. Cyclopoids, e.g., Apocyclops sp. and Oithona sp., have been used in only a
few studies (StÖttrup 2003). Calanoids are planktonic
in the whole life cycle, while harpacticoids are mostly
benthonic, with a few species having planktonic naupliar stages.

Rearing techniques for copepods
Harvest of wild zooplankton has been the most common source of copepods for ¢sh larviculture. How-

628

Aquaculture Research, 2010, 41, 613^640

ever, extensive and intensive production methods
have also been used (see StÖttrup 2003 for a more
complete overview).
Extensive copepod production can be performed in
outdoor tanks or ponds, and in con¢ned areas such
as lagoons or enclosed fjords. Fish larvae may be
reared directly in such enclosures/ponds (StÖttrup
2003) or copepods may be concentrated in the target
size fractions using ¢ltering devices with di¡erent
mesh sizes (van der Meeren & Naas 1997), and later
fed to larvae in tanks, £exible plastic enclosures or
ponds. Extensive production is normally based on
microalgae blooms induced by an agricultural fertilizer. Nitrogen concentration can be manipulated to favour blooms of certain microalgae types. When
nitrogen and oxygen are non-limiting, larger diatoms

are normally favoured, leading to higher copepod
productions (StÖttrup 2003). Signi¢cant quantities
of marine ¢sh juveniles produced based on extensive
copepod production have been described for grouper
in Philipines (Toledo et al.1999;Toledo et al. 2005) and
Taiwan (Liao, Su & Chang 2001; Su, Cheng, Chen &
Su 2005), red snapper in the United States (Ogle et al.
2005), £ounder in France, cod in Norway and turbot
in Norway and Denmark (StÖttrup 2003). However,
copepods produced extensively in ponds may cause
mass mortalities in grouper, through transmission
of viruses (VNN) and parasites (Amyloodinium sp.
and gill £ukes) (Su et al. 2005).
Intensive mass culture of copepods has been attempted with several species, with varying success,
and has resulted in rearing protocols for a number
of them (see StÖttrup 2003 for further details). A major problem with copepod intensive production is the
long generation time of most species. Species with
shorter generation times and with a wider tolerance
to temperature and salinity changes (normally coastal species) are preferred for aquaculture (Payne &
Rippingale 2001; StÖttrup 2003). Culture density is
also a constraint, with many calanoid species su¡ering from decreased fecundity due to overcrowding
(Miralto, Ianora, Poulet, Romano & Laabir 1996).
Thus, harpacticoid copepod species are often mentioned as the best candidates for mass production.
Harpacticoids have the advantages of a high tolerance to temperature and salinity, ability to feed
on a wide range of live and inert feeds, high fecundity,
relatively short life cycles (8^29 days), ability to be
cultured at high densities (may exceed 100 000 individuals L À 1), requirement for surface area rather
than volume due to their benthonic nature, some
have planktonic naupliar stages and own capability

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

to clean larvae and copepod rearing tanks surfaces
(StÖttrup 2003; Fleeger 2005). In Taiwan, the cyclopoid Apocyclops royi has been cultured in the
laboratory fed with microalgae, developing from
nauplius I to copepodite I in 5^6 days with a survival
rate of 89.6% (Liao et al. 2001). This makes it a species
with high potential for mass production. In contrast,
the European copepod species, Tisbe sp., takes 12^25
days to reach the copepodite I stage with a survival
rate of o50% (Liao et al. 2001).
Intensive production of copepods normally relies
on feeding a combination of at least two species of microalgae, and ensuring a high n-3 PUFA content
(StÖttrup 2003; Knuckey, Semmens, Mayer & Rimmer
2005). However, the calanoid Acartia tonsa has been
successfully reared on defatted rice bran alone (Turk,
Krejci & Yang 1982), and harpacticoids have been
shown to grow on a variety of inert diets (StÖttrup
2003; Rhodes & Boyd 2005). Microalgae species fed
to copepods should be chosen depending on the species. Cell size and concentration is a key factor, and so
is the sinking speed. Microalgae such as T. suecica,
S. costatum and Rhodomonas baltica, which quickly
sediment, are very adequate for benthic harpacticoid
copepods (StÖttrup 2003), while I. galbana would be
more suitable for calanoids.
Aeration is required in copepod intensive production, in order to help maintain microalgae in suspension, promote better algae distribution and avoid

anoxic areas in the tank (StÖttrup 2003). Calanoid
culture tanks have a higher requirement for cleaning
by siphoning, because harpacticoids have a selfcleaning capability. Cleaning is essential to avoid
bacterial blooms and ciliate infections, while rotifer
contamination is the most common cause for culture
collapse in large-scale production facilities (see
StÖttrup 2003 for a more detailed review on copepod
husbandry techniques).

Nutritional value of copepods
A recent study by van der Meeren et al. (2008) comprehensively compared the biochemical composition,
as well as the seasonal variation, of copepods extensively produced in a natural enclosure, with that of
rotifers and Artemia enriched with state-of-the-art
technology. This study provides detailed information
on major nutrient composition (lipids and amino
acids), as well as micronutrients (pigments and vitamins). The biochemical composition of copepods
showed a high stability both between years and

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

between seasons (van der Meeren et al. 2008), and,
compared with previous studies, seems to be representative for the mode for neritic calanoid species.
The superior nutritional value of copepods compared with Artemia and rotifers has traditionally
been attributed to their high PUFA, and particularly
HUFA, contents (e.g., Kanazawa 1993; Reitan, Rainuzzo & Olsen 1994; Bell et al. 2003). In fact, despite
having moderate levels of lipids (6.9^22.5% of DW),
copepods have high contents of EPA (8.3^24.6% of total lipid), DHA (13.9^42.3%) and low amounts of
ARA (0^2.6%) (van der Meeren et al. 2008). This
means that copepods have much higher contents of
DHA compared with both enriched rotifers (0.8 fold)

or enriched Artemia (0.3 fold), but only about half the
amount of total lipids and ARA (Table 2). In other
words, in an attempt to meet marine ¢sh larvae DHA
requirements, enriched Artemia are loaded with a
very high lipid content (mainly neutral lipids). These
n-3 HUFA contents of copepods also imply EPA/ARA
ratios usually above 20, and DHA/EPA ratios mostly
above 2. Such ratios have been shown to be crucial
for optimal marine ¢sh larval performance and quality (Sargent, Bell et al. 1999; Bell et al. 2003), in particular in £at¢sh, and are very di⁄cult to attain in
rotifers and especially in Artemia. In fact, copepods
clearly outperform enriched rotifers and enriched Artemia in terms of meeting ¢sh larval HUFA requirements (Bell et al. 2003; Rajkumar & Vasagam 2006;
van der Meeren et al. 2008). The higher contents
in phospholipids (37.9^70.2% of the total lipids) (van
der Meeren et al. 2008) compared with Artemia (see
Table 2) have also been pointed out as a bene¢t of
copepods. Marine ¢sh larvae have been shown to
have a requirement for dietary phospholipid supply
(Geurden 1996; Bell et al. 2003; Cahu, Infante &
Barbosa 2003). Furthermore, it should be noted that
the bene¢ts of copepods over rotifers and Artemia are
not just their higher phospholipid content, but that
the HUFA are predominantly located in the phospholipid fraction, while HUFA enrichment in Artemia results in their incorporation largely in the neutral lipid
fraction, namely in triacylglycerols (Coutteau &
Mourente 1997). This improves the biovailability of
HUFA in copepods compared with cultured live
prey, since HUFA in phospholipids are much more
readily available, digestible and retained in tissue
phospholipids, compared with HUFA supplied via a
neutral lipid (Izquierdo, Socorro, Arantzamendi &
HernaŁndez-Cruz 2000; Gisbert et al. 2005).

Copepods also have higher protein and free amino
acid contents compared with Artemia and rotifers

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

629


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

(Table 2; Fyhn, Finn, Helland, RÖnnestad & Lomsland
1993; Conceic°aìo et al. 1997; Helland, Terjesen & Berg
2003; van der Meeren et al. 2008). Free amino acids
and protein constituted between 4.3% and 8.9%,
and 32.7^53.6% of copepod DW respectively (van der
Meeren et al. 2008). This means that protein contents
of copepods are on average 0.5 fold higher than those
of Artemia, and copepod nauplii 0.2 fold higher than
rotifers (Table 2). This di¡erence is even higher for
free amino acid contents, with copepods and copepod
nauplii having on average 0.8 and 4.2 fold higher
contents, respectively, compared with Artemia and
rotifers (Table 2). Considering the immature digestive
system of ¢sh larvae, high free amino acid copepod
contents during ¢rst feeding may improve protein
utilization and growth performance (Fyhn et al.
1993; RÖnnestad et al. 2003; RÖnnestad & Conceic°aìo
2005). In addition, the superior protein contents of
copepods will contribute to a better use of the high

growth potential of ¢sh larvae (Conceic°aìo et al.
2003; RÖnnestad et al. 2003). Furthermore, the amino
acid pro¢le of copepods seems to better meet larval
grouper amino acid qualitative requirements, and
possibly also those of other species, compared with
rotifers and Artemia (Lacuisse, Conceic°aìo, Lutzki,
Koven,Tandler & Dinis 2005).
Astaxanthin is abundant in copepods (413^
1422 mg g À 1 DW), while Artemia seem to have no astaxanthin, but rather comparable amounts of
canthaxantin (Table 2; van der Meeren et al 2008).
These pigments may have important roles as antioxidants and sources of vitamin A in ¢sh larvae nutrition (van der Meeren et al. 2008). Copepods, Artemia
and rotifers have high concentrations of vitamins C,
E, B1 and B2 (Table 2; van der Meeren et al. 2008).
However, copepods have average iodine contents 109
fold higher compared with Artemia (Table 2; Moren
et al. 2006). Iodine limitation has been identi¢ed as a
possible cause of malpigmentation and other metamorphosis problems in £at¢sh larvae (Moren et al.
2006), due to its precursor role for thyroid hormones.
In fact, superior pigmentation as well as survival and
retinal morphology has been documented for halibut
larvae cultured on copepods compared with Artemia
(Shields et al. 1999).

Microbial aspects in copepods and other
zooplankton
The numbers of culturable bacteria associated with
copepods are generally much lower than in Artemia,

630

Aquaculture Research, 2010, 41, 613^640

ranging in values from 5.36 Â 102 nauplius À 1 (Verner-Je¡reys, Shields & Birkbeck 2003). The main species of bacteria associated with marine copepods
belong to the groups of Vibrio, Pseudomonas and Cytophaga (Sochard,Wilson, Austin & Colwell 1979).
The use of harvested copepods presents a risk of
introducing harmful bacteria to larvae and Artemia,
as well as for humans, such as Vibrio cholerae, Vibrio
parahaemolyticus,Vibrio vulini¢cus,Vibrio alginolyticus
and Aeromonas hydrophila (Rawlings, Ruiz & Colwell
2007; Gugliandolo, Irrera, Lentini & Maugeri 2008).

Nutritional studies with live feeds
Dose^response studies with live feed
Studies designed to determine the quantitative requirements for speci¢c nutrients require the ability
to analyse dose^response e¡ects. However, the extreme importance of the relative levels of DHA:
EPA:ARA supplied in the diet, and not just the
absolute requirements for essential fatty acids, has
long been recognized. These ratios between HUFA
have major implications in physiological mechanisms in which these fatty acids have a competitive interaction, such as in eicosanoid production or
membrane incorporation (Sargent, Bell et al.
1999;Sargent, McEnvoy et al., 1999). Originally, the
approach has been to try and simulate the biochemical composition of the egg or early larval stages vitelline reserves or, alternatively, the composition of the
larva’s natural preys (Sargent, Bell et al. 1999). However, this approach is quite limited, as the vitelline reserves are probably only adequate for the
development of the ¢rst early stages of development
and mimicking the composition of a natural multispeci¢c zooplankton population is extremely challenging. In order to address the question of nutrient
requirements, a tight control of the enrichment in individual fatty acids is required, which, as discussed
above, is not an easy task in Artemia. However, over
the last few years, advances in the production of speciality oils and the increasing availability of puri¢ed
sources of fatty acids have enabled the enrichment
of live preys, including rotifers and Artemia, with

graded levels of speci¢c fatty acids, thus enabling
dose^response designs to study the essential fatty
acid requirements of marine ¢sh larvae (Bransden,
Cobcroft, Battaglene, Morehead, Dunstan, Nichols &
Kolkovski 2005; Villalta, Este¤vez & Bransden 2005;
Villalta, Este¤vez, Bransden & Bell 2005; Lund,
Steenfeldt, Banta & Hansen 2008; Villalta, Este¤vez,

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

Bransden & Bell 2008a, b). Studies examining the
biological response of ¢sh larvae to key indicators
of larval performance such as survival, growth,
pigmentation and stress resistance, to feeding live
preys enriched with graded concentrations of the
essential fatty acid of interest or with a gradient of
essential fatty acid ratios (e.g., DHA:EPA or EPA:ARA)
will likely become more common.

Live feed labelling
Labelling of live prey has been used to quantify larval
feed intake and/or to characterize digestibility, energetic use or retention of dietary nutrients. Both stable
and radioactive isotopes have been used to label both
rotifers and Artemia (e.g., Sorokin & Panov 1966;
Govoni, Peters & Merriner1982; Boehlert & Yoklavich
1984; Conceic°aìo et al. 1998; Conceic°aìo, Skjermo,

Skjak-Braek & Verreth 2001; Morais, Torten, Nixon,
Lutzky, Conceic°aìo, Dinis, Tandler & Koven 2006). A
more comprehensive review of the methods and applications of tracer studies using live prey can be
found elsewhere (Conceic°aìo, Morais & RÖnnestad
2007). The method and tracer molecule that is used
to label the live prey will depend on the objective of
the studies being conducted.
Quanti¢cation of food intake has been performed
using either radiolabelled (Sorokin & Panov 1966;
Govoni et al. 1982; Boehlert & Yoklavich 1984; Morais
et al. 2006), stable isotope-labelled (Conceic°aìo et al.
2001) or colour-labelled (Yu¤fera 1987; Parra & Yu¤fera
2000) live preys. More recently, small di¡erences in
isotopic signatures (natural abundances of stable isotopes) have also been used to study the selectivity of
dietary items by ¢sh larvae (Schlechtriem, Focken &
Becker 2005; Gamboa-Delgado, Canìavate, Zerolo &
Le Vay 2008; Jomori, Ducatti, Carneiro & Portella
2008). Artemia incubated in a [U-14C] protein hydrolysate has been used to study ontogenetic changes
(Morais, Lacuisse, Conceic°aìo, Dinis & RÖnnestad
2004; Engrola 2008) and the impact of the feeding
regime (Engrola 2008) on the protein digestibility of
Senegalese sole early larval stages. These studies established that Senegalese sole larvae, even at young
stages, have a high capacity for digesting live preys.

Co-feeding live feeds with inert diets
The weaning onto commercial feeds usually starts at
the last stages of the larval period, after several weeks
of live prey feeding. However, mixed feeding on live

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

prey plus inert diets during earlier larval stages
has been tested since the 1980s. These co-feeding
assays have been performed with di¡erent aims: to
know to what degree larvae accept, digest and tolerate inert diets in order to advance the complete replacement of live prey; to ¢nd a way to deliver speci¢c
compounds into the larval gut; and mainly to perform an early weaning to reduce dependence on
live prey.
Di¡erent commercial and experimental microdiets
have been tested in di¡erent relative proportions with
rotifers and Artemia (Kanazawa, Koshio & Teshima
1989; Holt1993; Hart & Purser1996; Rosenlund, Stoss
& Talbot 1997; Canìavate & FernaŁndez-D|¤ az 1999;
Alves, Cerqueira & Brown 2006; Chang, Liang,Wang,
Chen, Zhang & Liu 2006; Aristizabal & Suarez 2007;
Fletcher, Roy, Davie, Taylor, Robertson & Migaud
2007; Engrola 2008). In general, these studies show
that ¢sh larvae of di¡erent species grow very well in
co-feeding when the live prey substitution level is not
excessive. More rarely, a high substitution ratio
(Kanazawa et al. 1989; Yu¤fera, Kolkovski, FernaŁndezD|¤ az, Rinchard, Lee & Dabrowski 2003) or even the
complete substitution (Fontagne¤, Robin, Corraze &
Bergot 2000) also yielded good growth results. An
early co-feeding period seems to prepare the gut for
accepting and processing inert diets, allowing earlier
weaning and with better growth performances than
when weaning starts at the end of the larval stage
(Rosenlund et al. 1997; Canìavate & FernaŁndez-D|¤ az
1999; Curnow, King, Bosmans & Kolkovski 2006; Engrola 2008). Fish larvae clearly prefer live prey to
inert diets (FernaŁndez-D|¤ az et al. 1994) and in some
species the acceptance of inert diets is worse after a

long feeding period on live prey exclusively. In addition, the co-feeding regime may contribute to correct
potential nutrient de¢ciencies.

Conclusions
Live feeds are still the main feed item in commercial
larviculture, despite their nutritional composition often being sub-optimal. Further improvements in inert microdiet technology and formulation will likely
lead to a progressive substitution of live feeds. However, this substitution will be gradual and is most
likely far from being complete for many species, at
least for the ¢rst days of feeding. Co-feeding of small
amounts of live feeds with high-quality inert microdiets will likely lead to major improvements in larval
performance. Such feeding strategies may take ad-

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

631


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

vantage of the strong points of both live feeds (stimulation of feeding behaviour) and inert diets (optimized nutritional composition).

Acknowledgments
This review was partially supported by projects:
POCI/MAR/61623/2004 ^ SAARGO, ¢nanced by program POCI 2010 (FCT, Portugal), which is co-¢nanced by FEDER; and project P06-AGR-01697
funded by Consejer|¤ a Innovacio¤n, Ciencia y Empresa
^ Junta de Andaluc|¤ a (Spain)1FEDER.

References
Alves T.T., CerqueiraV.R. & Brown J.A. (2006) Early weaning

of fat snook (Centropomus parallelus Poey 1864) larvae.
Aquaculture 253, 334^342.
Aragaìo C., Conceic°aìo L.E.C., Dinis M.T. & Fyhn H.J. (2004)
Amino acid pools of rotifers and Artemia under di¡erent
conditions: nutritional implications for ¢sh larvae. Aquaculture 234, 429^445.
Aragaìo C., Conceic°aìo L.E.C., Lacuisse M., Yu¤fera M. & Dinis
M.T. (2007) Do dietary amino acid pro¢les a¡ect performance of larval gilthead seabream? Aquatic Living Resources 20, 155^161.
Aristizabal E.O. & Suarez J. (2007) E⁄ciency of co-feeding
red porgy (Pagrus pagrus L.) larvae withy live and compound diet. Revista de biolog|¤ a marina y oceanograf|¤ a 41,
203^208.
Austin B. (1982) Taxonomy of bacteria isolated from a coastal marine ¢sh-rearing unit. Journal of Applied Bacteriology
53, 253^268.
Austin B. & Day J.G. (1990) Inhibition of prawn pathogenic
Vibrio spp. by a commercial spray-dried preparation of
Tetraselmis suecica. Aquaculture 90, 389^392.
Austin B., Baudet E. & Stobie M. (1992) Inhibition of bacterial ¢sh pathogens byTetraselmis suecica. Journal of Fish Diseases 15, 55^61.
Baer A., Langdon C., Mills S., Schulz C. & Hamre K. (2008)
Particle size preference, gut ¢lling and evacuation rates
of the rotifer Brachionus ‘‘Cayman’’ using polystyrene latex beads. Aquaculture 282,75^82.
Barr Y. & Helland S. (2007) A simple method for mass-production of liposomes, in particular large liposomes, suitable for delivery of free amino acids to ¢lter feeding
zooplankton. Journal of Liposome Research 17,79^88.
Baxevanis A.D., Kappas I. & Abatzopoulos T.J. (2006) Molecular phylogenetics and asexuality in the brine shrimp
Artemia. Molecular Phylogenetics and Evolution 40, 724^
738.
Becker W. (2004) Microalgae for aquaculture. The nutritional value of microalgae for aquaculture. in Handbook
of Microalgal Culture. Biotechnology and Applied Phycology

632

Aquaculture Research, 2010, 41, 613^640

(ed. by A. Richmond), pp. 380^391. Blackwell Science,
Ames, IA, USA.
Bell J.G., McEvoy L.A., Estevez A., Shields R.J. & Sargent J.R.
(2003) Optimising lipid nutrition in ¢rst-feeding £at¢sh
larvae. Aquaculture 227, 211^220.
Ben-Amotz A., Fishler R. & Schneller A. (1987) Chemical
composition of dietary species of marine unicellular algae
and rotifers with emphasis on fatty acids. Marine Biology
95, 31^36.
Bengtson D.A. (2003) Status of marine aquaculture in relation to live prey: past, present and future. in Live Feeds in
Marine Aquaculture (ed. by J.G. StrÖttrup & L.A. McEvoy),
pp.1^16. Blackwell publishing, Oxford, UK.
Boehlert G.W. & Yoklavich M.M. (1984) Carbon assimilation
as a function of ingestion rate in larval Paci¢c herring,
Clupea harengus pallasi Valenciennes. Journal of Experimental Marine Biology and Ecology 79, 251^262.
Boraas M.E. (1983) Population dynamics of food-limited rotifers in two-stage chemostat culture. Limnology and Oceanography 28, 546^563.
Bransden M.P., Cobcroft J.M., Battaglene S.C., Morehead
D.T., Dunstan G.A., Nichols P.D. & Kolkovski S. (2005)
Dietary 22:6n-3 alters gut and liver structure and behaviour in larval striped trumpeter (Latris lineata). Aquaculture 248, 275^285.
Brown M.R. (1991) The amino acid and sugar composition
of 16 species of microalgae used in mariculture.
Journal of Experimental Marine Biology and Ecology 145,
79^99.
Cahu C. & Infante J.Z. (2001) Substitution of live food by formulated diets in marine ¢sh larvae. Aquaculture 200,
161^180.
Cahu C.L., Infante J.L.Z. & BarbosaV. (2003) E¡ect of dietary
phospholipid level and phospholipid: neutral lipid value
on the development of sea bass (Dicentrarchus labrax) larvae fed a compound diet. British Journal of Nutrition 90,
21^28.

Canìavate J.P. & FernaŁndez-D|¤ az C. (1999) In£uence of cofeeding larvae with live and inert diets on weaning the
sole Solea senegalensis onto commercial dry feeds. Aquaculture 174, 255^263.
Carvalho A.P., SaŁ R., Oliva-Teles A. & Bergot P. (2004) Solubility and peptide pro¢le a¡ect the utilization of dietary protein by common carp (Cyprinus carpio) during early larval
stages. Aquaculture 234, 319^333.
Chang Q., Liang M.Q., Wang J.L., Chen S.Q., Zhang X.M. &
Liu X.D. (2006) In£uence of larval co-feeding with live
and inert diets on weaning the tongue sole Cynoglossus
semilaevis. Aquaculture Nutrition 12,135^139.
Ciros-Pe¤rez J., Go¤mez A. & Serra M. (2001) On the taxonomy
of three sympatric sibling species of the Brachionus plicatilis (Rotifera) complex from Spain, with the description of
B. ibericus n. sp. Journal of Plankton Research 23, 1311^
1328.
Conceic°aìo L.E.C., van der Meeren T., Verreth J.A.J., Evjen
M.S., Houlihan D.F. & Fyhn H.J. (1997) Amino acid meta-

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640


Aquaculture Research, 2010, 41, 613^640

bolism and protein turnover in larval turbot (Scophthalmus maximus) fed natural zooplankton orArtemia. Marine
Biology 129, 255^265.
Conceic°aìo L.E.C., Dersjant-Li Y. & Verreth J.A.J. (1998) Cost of
growth in larval and juvenile African cat¢sh (Clarias
gariepinus) in relation to growth rate, food intake and oxygen consumption. Aquaculture 161, 95–106.
Conceic°aìo L.E.C., Skjermo J., Skjak-Braek G. & Verreth J.A.J.
(2001) E¡ect of an immunostimulating alginate on protein turnover of turbot (Scophthalmus maximus L.) larvae.
Fish Physiology and Biochemistry 24, 207^212.
Conceic°aìo L.E.C., Grasdalen H. & Ronnestad I. (2003) Amino

acid requirements of ¢sh larvae and post-larvae: new
tools and recent ¢ndings. Aquaculture 227, 221^232.
Conceic°aìo L.E.C., Morais S. & RÖnnestad I. (2007) Tracers in
¢sh larvae nutrition: a review of methods and applications. Aquaculture 267, 62^75.
Coutteau P. & Sorgeloos P. (1992) The use of algal substitutes
and the requirement for live algae in the hatchery and
nursery rearing of bivalve molluscs: an international survey. Journal of Shell¢sh Research 11, 467^476.
Coutteau P. & Mourente G. (1997) Lipid classes and their
content of n-3 highly unsaturated fatty acids (HUFA) in
Artemia franciscana after hatching, HUFA-enrichment
and subsequent starvation. Marine Biology 130, 81^91.
Coutteau P., Geurden I., Camara M.R., Bergot P. & Sorgeloos
P. (1997) Review on the dietary e¡ects of phospholipids
in ¢sh and crustacean larviculture. Aquaculture 155,
149^164.
Curnow J., King J., Bosmans J. & Kolkovski S. (2006) The effect of reduced Artemia and rotifer use facilitated by a new
microdiet in the rearing of barramundi Lates calcarifer
(BLOCH) larvae. Aquaculture 257, 204^213.
Curnow J., King J., Partridge G. & Kolkovski S. (2006) E¡ects
of two commercial microdiets on growth and survival of
barramundi (Lates calcarifer Bloch) larvae within various
early weaning protocols. Aquaculture Nutrition 12, 247^
255.
De Swaarf M.E., Pronk J.T. & Sijtsma L. (2003) Fed-batch cultivation of the docosahexaenoic-acid-producing marine
alga Crypthecodinium cohnii on ethanol. Applied Microbiology and Biotechnology 61, 40^43.
Defoirdt T., Crab R., Wood T.K., Sorgeloos P., Verstraete W. &
Bossier P. (2006) Quorum sensing-disrupting brominated
furanones protect the gnotobiotic brine shrimp Artemia
franciscana from pathogenic Vibrio harveyi, Vibrio campbellii, andVibrio parahaemolyticus isolates. Applied and Environmental Microbiology 72, 6419^6423.
Defoirdt T., Halet D., Sorgeloos P., Bossier P. & Verstraete W.

(2006) Short-chain fatty acids protect gnotobiotic Artemia
franciscana from pathogenic Vibrio campbellii. Aquaculture 261, 804^808.
Defoirdt T., Boon N., Sorgeloos P., Verstraete W. & Bossier P.
(2007) Alternatives to antibiotics to control bacterial infections: luminescent Vibriosis in aquaculture as an example. Trends in Biotechnology 25, 472^479.

Live feeds for ¢sh larvae L E C Conceic° aìo et al.

Dhert P., Rombaut G., Suantika G. & Sorgeloos P. (2001) Advancement of rotifer culture and manipulation techniques in Europe. Aquaculture 200, 129^146.
Dominguez A., Ferreira M., Coutinho P., Fabregas J. & Otero
A. (2005) Delivery of astaxanthin from Haematococcus
pluvialis to the aquaculture food chain. Aquaculture 250,
424^430.
Dooms S., Papakostas S., Ho¡man S., Delbare D., Dierkens
K.,Triantafyllidis A., De Wolf T.,Vadstein O., Abatzopoulos
T.J., Sorgeloos P. & Bossier P. (2007) Denaturing gradient
gel electrophoresis (DGGE) as a tool for the characterisation of Brachionus sp strains. Aquaculture 262, 29^40.
Douillet P.A. (2000) Bacterial additives that consistently enhance rotifer growth under synxenic culture conditions.
1. Evaluation of commercial products and pure isolates.
Aquaculture 182, 249^260.
Droop M.R. & Scott J.M. (1978) Steady-state energetics of a
planktonic herbivore. Journal of the Marine Biological Association of the United Kingdom 58,749^772.
Du¡ D.C.B. & Bruce D.L. (1966) The antibacterial activity of
marine planktonic algae. CanadianJournal of Microbiology
12, 877^884.
Engrola S. (2008) Improving growth performance of Senegalese
sole postlarvae. PhD Thesis. University of Algarve, Faro,
167pp.
Faulk C.K. & Holt G.J. (2005) Advances in rearing cobia Rachycentron canadum larvae in recirculating aquaculture
systems: live prey enrichment and greenwater culture.
Aquaculture 249, 231^243.

FernaŁndez-D|¤ az C., Pascual E. & Yu¤fera M. (1994) Feeding behavior and prey size selection of gilthead seabream,
Sparus aurata, larvae fed on inert and live food. Marine
Biology 118, 323^328.
FernaŁndez-Reiriz M.J. & Labarta U. (1996) Lipid classes and
fatty acid composition of rotifer (Brachionus plicatilis) fed
two algal diets. Hydrobiologia 330,73^79.
FernaŁndez-Reiriz M.J., Labarta U. & Ferreiro M.J. (1993) Effects of commercial enrichment diets on the nutritional
value of the rotifer (Brachionus plicatilis). Aquaculture
112,195^206.
Fernandez R.G. (2001) Artemia bioencapsulation I. E¡ect of
particle sizes on the ¢ltering behavior of Artemia franciscana. Journal of Crustacean Biology 21, 435^442.
Fleeger J.W. (2005) The potential to mass-culture harpacticoid copepods for use as food for larval ¢sh. in Copepods
in Aquaculture (ed. by C.-S. Lee, P.J. O’Bryen & N.H. Marcus), pp. 11^24. Blackwell Publishing, Oxford, UK.
Fletcher R.C., Roy W., Davie A.,Taylor J., Robertson D. & Migaud H. (2007) Evaluation of new microparticulate diets
for early weaning of Atlantic cod (Gadus morhua): implications on larval performances and tank hygiene. Aquaculture 263, 35^51.
Fogg G.E. (1975) Algal Cultures and Phytoplankton Ecology.
University of Wisconsin Press, Madison,WI, USA,175pp.
Fontagne¤ S., Robin J., Corraze G. & Bergot P. (2000) Growth
and survival of European sea bass (Dicentrarchus labrax)

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640

633


Live feeds for ¢sh larvae L E C Conceic° aìo et al.

larvae fed from ¢rst feeding on compound diets containing mediumçchain triacylglycerols. Aquaculture 190,
261^271.

Fontaneto D., Giordani I. & Serra M. (2007) Disentangling
the morphological stasis in two rotifer species of the
Brachionus plicatilis species complex. Hydrobiologia 583,
297^307.
Fu Y., Hirayama K. & Natsukari Y. (1991) Genetic divergence
between S and L type strains of the rotifer Brachionus plicatilis O.F. Mˇller. Journal of Experimental Marine Biology
and Ecology 151, 43^56.
FuY., Ada A.,Yamashita T.,YashidaY. & Hino A. (1997) Development of a continuous culture system for stable mass
production of rhe rotifer marine Brachionus. Hydrobiologia 358, 141^151.
Fukami K., Nishijima T. & Ishida Y. (1997) Stimulative and
inhibitory e¡ects of bacteria on the growth of microalgae.
Hydrobiologia 358,185^191.
Fukusho K. & Okaushi M. (1983) Sympatry in natural distribution of the two strains of a rotifer, Brachionus plicatilis.
Bulletin of National Research Institure of Aquaculture 4,
135^138.
Fyhn H.J., Finn R., Helland S., RÖnnestad I. & Lomsland E.R.
(1993) Nutritional value of phyto- and zooplankton as live
food for marine ¢sh larvae. In: Fish Farming Technology
(ed. by H. Reinertsen, L.A. Dahle, L. Jørgensen & K.
Tvinnerheim), pp. 121–126. Balkema, Rotterdam, the
Netherlands.
Gamboa-Delgado J., Canìavate J.P., Zerolo R. & Le Vay L.
(2008) Natural carbon stable isotope ratios as indicators
of the relative contribution of live and inert diets to
growth in larval Senegalese sole (Solea senegalensis).
Aquaculture 280,190^197.
Garcia Ortega A.,Verreth J.A.J., Coutteau P., Segner H., Huisman E.A. & Sorgeloos P. (1998) Biochemical and enzymatic characterization of decapsulated cysts and nauplii
of the brine shrimp Artemia at di¡erent developmental
stages. Aquaculture 161, 501^514.
Gatesoupe F.J. (2002) Probiotic and formaldehyde treatments of Artemia nauplii as food for larval pollack, Pollachius pollachius. Aquaculture 212, 347^360.

Gatesoupe F.J. & Luquet P. (1981) Practical diet for mass culture of the rotifer Brachionus plicatilis: application to larval rearing of sea bass, Dicentrarchus labrax. Aquaculture
22,149^163.
Gatesoupe F.J. & Robin J.H. (1981) Commercial single-cell
proteins either as sole source or in formulated diets for intensive and continuous production of rotifer Brachionus
plicatilis. Aquaculture 25,1^15.
Gatesoupe F.J., Arakawa T. & WatanabeT. (1989) The e¡ect of
bacterial additives on the production rate and dietary value of rotifers as food for Japanese £ounder, Paralicthys olivaceus. Aquaculture 83, 39^44.
Geurden I. (1996).The role of phospholipids in diets of larval and
postlarval ¢sh. PhD Thesis. Universiteit Gent, Landbouwkunde, Belgium, 239pp.

634

Aquaculture Research, 2010, 41, 613^640

Gime¤nez G., KotzamanisY., Hontoria F., Estevez A. & Gisbert
E. (2007) Modelling retinoid content in live prey: a tool for
evaluating the nutritional requirement and development
studies in ¢sh larvae. Aquaculture 267,76^82.
Gisbert E.,Villeneuve L., Zambonino-Infante J.L., Quazuguel
P. & Cahu C.L. (2005) Dietary phospholipids are more e⁄cient than neutral lipids for long-chain polyunsaturated
fatty acid supply in European sea bass Dicentrarchus labrax larval development. Lipids 40, 609^618.
Go¤mez A. & Serra M. (1995) Behavioural reproductive isolation among sympatric strains of Brachionus plicatilis Muller 1786: insights into status of this taxonomic species.
Hydrobiologia 313/314,111^119.
Go¤mez A., Serra M., Carvalho G.R. & Lunt D.H. (2002) Speciantion in anvcient cryptic species complexes: evidence
from the molecular phylogeny of Brachionus plicatilis (Rotifera). Evolution 56, 1431^1444.
Govoni J.J., Peters D.S. & Merriner J.V. (1982) Carbon assimilation during larval development of the marine teleost
Leiostomus xanthurus Lacepede. Journal of Experimental
Marine Biology and Ecology 64, 287^299.
Govoni J.J., Boehlert G.W. & Watanabe Y. (1986) The physiology of digestion in ¢sh larvae. Environmental Biology of
Fishes 16, 59^77.

Gugliandolo C., Irrera G.P., Lentini V. & Maugeri T.L. (2008)
Pathogenic Vibrio, Aeromonas and Arcobacter spp. associated with copepods in the Straits of Messina (Italy). Marine Pollution Bulletin 56, 600^606.
Hamre K., Opstad I., Espe M., Solbakken J., Hemre G.-I. &
Pittman K. (2002) Nutrient composition and metamorphosis success of Atlantic halibut (Hippoglossus hippoglossus, L.) larvae fed natural zooplankton or Artemia.
Aquaculture Nutrition 8, 139^148.
Hamre K., Mollan T.A., S×le Ò. & Erstad B. (2008) Rotifers
enriched with iodine and selenium increase survival in
Atlantic (Gadus morhua) cod larvae. Aquaculture 284,
190–195.
Hamre K., Srivastava A., RÖnnestad I., Mangor-Jensen A. &
Stoss J. (2008) Several micronutrients in the rotifer Brachionus sp may not ful¢l the nutritional requirements of
marine ¢sh larvae. Aquaculture Nutrition 14, 51^60.
Hamza N., Mhetli M. & Kestemont P. (2007) E¡ects of weaning age and diets on ontogeny of digestive activities and
structures of pikeperch (Sander lucioperca) larvae. Fish
Physiology and Biochemistry 33,121^133.
Harel M., KovenW., Lein I., BarY., Behrens P., Stubble¢eld J.,
ZoharY. & Place A.R. (2002) Advanced DHA, EPA and ArA
enrichment materials for marine aquaculture using single cell heterotrophs. Aquaculture 213, 347^362.
Hart P.R. & Purser G.J. (1996) Weaning of hatchery-reared
greenback £ounder (Rhombosolea tapirina Gˇnther) from
live to arti¢cial diets: e¡ects of age and duration of the
changeover period. Aquaculture 145,171^181.
Heasman M., Diemar J., O’Connor W., Sushames T. & Foulkes
L. (2000) Development of extended shelf-life micro-algae

r 2009 The Authors
Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 613^640