Aquaculture Research, 2010, 41, 161^181 doi:10.1111/j.1365-2109.2009.02323.x REVIEW ARTICLE Modelling growth and body composition in fish nutrition: where have we been and where are we going? AndreÔ Dumas, James France & Dominique Bureau Department of Animal and Poultry Science, Centre for Nutrition Modelling, University of Guelph, Guelph, Ontario, Canada Correspondence: A Dumas, Department of Animal and Poultry Science, Centre for Nutrition Modelling, University of Guelph, Guelph, Ontario N1G 2W1, Canada E-mail: adumas@uoguelph.ca Abstract Introduction Mathematical models in ¢sh nutrition have proven indispensable in estimating growth and feed requirements Nowadays, reducing the environmental footprint and improving product quality of ¢sh culture operations are of increasing interest This review starts by examining simple models applied to describe/predict ¢sh growth pro¢les and progresses towards more comprehensive concepts based on bioenergetics and nutrient metabolism Simple growth models often lack biological interpretation and overlook fundamental properties of ¢sh (e.g ectothermy, indeterminate growth) In addition, these models disregard possible variations in growth trajectory across life stages Bioenergetic models have served to predict not only ¢sh growth but also feed requirements and waste outputs from ¢sh culture operations However, bioenergetics is a concept based on energy-yielding equivalence of chemicals and has signi¢cant limitations Nutrient-based models have been introduced into the ¢sh nutrition literature over the last two decades and stand as a more biologically sound alternative to bioenergetic models More mechanistic models are required to expand current understanding about growth targets and nutrient utilization for biomass gain Finally, existing models need to be adapted further to address e¡ectively concerns regarding sustainability, product quality and body traits Aquaculture has become a multinational industry over the last 30 years and is expected to maintain an average annual growth rate of 44% over the period 2010^2030 (Bruge're & Ridler 2004) Greater demand for ¢sh, combined with the reduction in capture ¢sheries and more a¡ordable retail prices for several species, has contributed to foster and sustain the aquaculture industry (NRC 1999; FAO 2006) However, intensi¢cation and potential for development of the aquaculture sector have created challenges regarding pro¢tability, environmental sustainability and product quality, most of which are related ultimately to nutrition (e.g Naylor, Goldburg, Primavera, Kautsky, Beveridge, Clay, Folke, Lubchenco, Mooney & Troell 2000; Watanabe 2002) These concerns along with uncertainties surrounding production costs stress, among other things, the need to develop accurate tools to manage production and predict scenarios soundly Here, mathematical modelling ^ de¢ned as the use of equations to describe or simulate processes in a system ^ represents an e¡ective approach to taking up the challenges that aquaculture is facing Mathematical models in animal nutrition have proven indispensable in estimating growth and feed requirements that have always represented major ¢elds of interest in livestock production (Kellner 1911; Murray1914; Brody1945; Blaxter1989; Baldwin 1995; Dumas, Dijkstra & France 2008) In aquaculture, the quality, safety and health bene¢ts of ¢sh products are now of increasing interest (Hocquette, Keywords: modelling, ¢sh, growth, body composition, nutrition r 2009 Blackwell Munksgaard No claim to original US government works 161 Modelling growth and body composition in Âsh A Dumas et al Richardson, Prache, MeÔdale, DuĂy & Scollan 2005; Caswell 2006; Moza¡arian & Rimm 2006) Composition of ¢sh with reference to carcass yield, fatty acid composition and levels of lipid and contaminants has recently received further attention in studies on nutrition, genetics and health (Rasmussen 2001; Blanchet, Lucas, Julien, Morin, Gingras & Dewailly 2005; Hamilton, Hites, Schwager, Foran, Knuth & Carpenter 2005; Tobin, Kause, MÌntysaari, Martin, Houlihan, Dobly, Kiessling, Rungruangsak-Torrissen, Ritola & Ruohonen 2006) This article begins by summarizing brie£y the biological properties of ¢sh growth Thereafter, major current models applied in ¢sh nutrition are reviewed and challenged Finally, a global perspective is o¡ered and future directions in modelling are suggested to address better the concerns in ¢sh production Biological properties of fish growth Despite its complexity, growth takes place in a highly organized scheme in animals Diverse regulatory strategies exist in organisms to adjust in£ux of chemicals (amino acids, fatty acids, minerals, etc.) and excretion of waste products even in a disruptive environment in order to maintain homeostasis (Nelson & Cox 2000) As growth processes not occur in a chaotic manner, they can be generally described and predicted using conventional mathematics Growth, body composition and metabolic utilization of nutrients or allocation of resources are related to each other and change considerably during the lifespan of animals Growth trajectories of animals ^ de¢ned here as the pattern of weight gain achieved through time ^ display an almost universally sigmoidal shape with an asymptotic body size at adult stage (Fig 1a) It is well documented that growth rate increases during the juvenile stage, i.e the so-called self-accelerating phase of growth, and levels o¡ when the animal approaches the adult stage or induces reproductive growth This last portion of the growth Aquaculture Research, 2010, 41, 161^181 curve is also referred to as the self-inhibiting phase of growth (Brody 1927; Charnov, Turner & Winemiller 2001; Lester, Shuter & Abrams 2004) In contrast with birds and mammals, several species of ¢sh, molluscs, crustaceans and amphibians are capable of growing well beyond their size at sexual maturity These organisms display a much less evident self-inhibiting phase (Fig 1b) This phenomenon, also referred to as indeterminate growth, results in a debatable position of asymptotic weight at the adult stage Indeterminate growth is regulated by environment and genetics (Sebens 1987), which a¡ect the physiological capacity of an organism to synthesize muscle ¢bres throughout its life cycle (Biga & Goetz 2006) Another peculiarity of ¢sh is their ectothermic nature Growth rate of ¢sh is thus highly dependent on water temperature To date, few attempts have been made to describe ¢sh growth with an algebraic expression that accommodates their ectothermic nature and indeterminate growth Current models in fish nutrition The complexity of interactions in nutrition, the vast amount of information available nowadays and the substantial cost of experiments make the use of mathematical models appealing Models are helpful tools in that they have the ability to represent complex phenomenon (e.g growth) in a relatively simple way [e.g weight gain as a function of protein deposition (PD)] The following sections review brie£y extant models currently applied in ¢sh nutrition Simple growth functions Growth functions are any models where weight or length (dependent variable, y) is calculated using time, t, as the predictor (independent variable) taking the form y f(t), where f represents some functional relationship Growth functions are usually analytical solutions to di¡erential equations that can be ¢tted to (b) Body weight (arbitrary units) Body weight (arbitrary units) (a) Age (arbitrary units) Age (arbitrary units) Figure Typical growth trajectory of (a) terrestrial animals and (b) ¢sh 162 r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 Aquaculture Research, 2010, 41, 161^181 Modelling growth and body composition in ¢sh A Dumas et al the growth data generally by means of non-linear regression analysis (Thornley & France 2007) The sigmoidal or curvilinear shape of the growth trajectory indicates that linear regression is not suitable to describe growth, unless only small portions of the curve are considered For this reason, growth functions stand presumably as the best means of estimating animal growth Because a large number of growth functions had been proposed in the last century, only those that have been widely applied in ¢sh studies or that have considered the e¡ect of temperature on growth of ectotherms are discussed here For a broader description of extant growth functions in animal science and theories associated with them, the reader is referred to Ricker (1979), Parks (1982), Ratkowski (1990), Seber and Wild (2003) and Thornley and France (2007) von Bertalanffy equation The equation of von Bertalan¡y (1957) stands as the most studied and applied growth function to predict growth of ¢sh and other ectotherms (Ricker 1979; Hernandez-Llamas & Ratkowsky 2004; De Graaf & Prein 2005; Katsanevakis 2006) The equation was ¢rst proposed by Pˇtter (1920), a German ¢sh biologist, who conceptualized growth as anabolism prevailing over catabolism The di¡erential and integral forms of his equation, currently referred to as the von Bertalan¡y equation, are dW ¼ ZW b À kW dt h   i1=u W ¼ Wfu À Wfu À W0u eÀkt ð1Þ Parameters : Z > k ! 0; < b < 1 Z1Àb t ¼ 0; W ¼ W0 ; t ! 1; W ¼ Wf ¼ k where Z and k are rate parameters for anabolism and catabolism, respectively, k is a rate constant equal to k(1 À b), and u equals À b The allometric exponent for anabolism b is allowed to vary between and The equation has an asymptote, a £exible point of in£exion, and adheres to the law of allometry (0obo1) Various rearrangements of the von Bertalan¡y equation exist in the literature (Ricker 1979; Katsanevakis 2006) The assumption regarding an asymptotic ¢nal size led to unrealistic values for indeterminate growers and, for this reason, was regarded as a mathematical artefact rather than a fact of nature (Knight 1968; Ro¡ 1980) Parker and Larkin (1959) removed the catabolic part of Eq (1) in order to relax the constraint on the ¢nal asymptote and suggested estimating m and b by ¢tting to particular life history groups and growth stanzas: dW ¼ mW b dt ð2Þ The assumption that growth is determined by the di¡erence between anabolism and catabolism has been proven inaccurate because it overlooks the role of timing of maturation on the shape of the growth curve (Day & Taylor1997; Lester et al 2004) Evidence suggests that the change in growth rate of indeterminate growers results from the decision to allocate more resources towards gonad development rather than movement towards equilibrium between anabolism and catabolism (Day & Taylor1997; Czarnol˛eski & Kozlowski 1998; Charnov et al 2001) However, the e¡ect of reproduction is not always perceptible in ectotherms with indeterminate growth, especially in an environment with £uctuating water temperatures (Dumas & France 2008) (Correction added on September 2009, after ¢rst online publication: In the sentence containing Equation (1),‘u equals 1b’ was corrected to ‘u equals À b’.) Thermal-unit growth coefficient (TGC) The French botanist ReÔaumur laid the basis of the thermal-unit concept in1735 in an attempt to explain the time required from sowing to harvesting of crops by summing the degreÔ -chaleur over that period (Allen 1976; Bonhomme 2000) The concept was introduced in ichthyology at the turn of the 20th century (Be›lehraŁdek 1930) Although Norris (1868) noted that the development rate of trout eggs varies with temperature, Wallich (1901) apparently ¢rst applied the concept of the thermal unit to record the development of ¢sh eggs Wallich (1901) de¢ned one thermal unit as 11F above 321F during day, meaning that the mean daily water temperature of 361F is equivalent to four thermal units Krogh (1914) showed that the relationship between developmental rate (usually in % day À 1) and temperature ( 1C) exhibited a straight line (slope has degree-day as denominator) over a certain range of temperature The time and thermal summation (degree-day) needed for hatching ¢sh eggs can thus be estimated using a simple regression equation (Krogh 1914; Embody 1934; Hayes 1949) The thermal-unit concept was also applied to estimate growth of hatched ¢sh Iwama and Tautz (1981), r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 163 Modelling growth and body composition in ¢sh A Dumas et al Aquaculture Research, 2010, 41, 161^181 who did not use the concept explicitly, started from Eq (2) and related the rate parameter for anabolism to mean daily water temperature averaged over the rearing period (T): dW 3ị ẳ mTW b dt lation increased according to a geometric progression (Gilbert 1993) His model, known as Malthus’ Law or the Malthusian Model, corresponds to the exponential growth equation: where m (40) has units of g1 À b( 1C day) À 1, T (a constant) is water temperature ( 1C) and the allometric exponent b (40) is dimensionless Integrating Eq (3) yields where W is body weight, W0 is body weight at time t 0, m is a growth coe⁄cient (in units of per unit of time) and time t is measured as age The growth coe⁄cient m is better known as SGR, which is used ubiquitously in ¢sh studies The equation for SGR (m) is ZW dW ¼ mT Wb W0 Zt dt 4ị W 1b ẳ W01b ỵ mT1 bịt where W0 is the initial (time 0) value of W Starting from Iwama and Tautz (1981), Cho (1992) explicitly introduced the degree-day concept into their model and proposed, without formal mathematical derivation, a modiÂcation to Eq (4): 1=3 Wn1=3 ẳ W0 þ n c X Ti 1000 i¼1 where c [g1/3( 1C day) À 1)] is TGC and Ti ( 1C) is mean daily temperature From an inspection of Eq (1) and Eq (3), it is evident that the TGC model is a special case of von Bertalan¡y’s equation with  TGC  T m¼ ; b ¼ 2=3; l ¼ 1000 The TGC model has since been widely used in the aquaculture literature (e.g Einen, Holmefjord, —sgÔrd & Talbot 1995; Kaushik 1998; Willoughby 1999; Stead & Laird 2002; Hardy & Barrows 2002) This simple model has been adapted recently to the di¡erent growth stanzas of rainbow trout (Oncorhynchus mykiss, Walbaum) across life stages (Dumas, France & Bureau 2007) Despite its convenience, the thermal-unit approach can entail systematic errors in situations where the temperature moves too far away from the optimum for growth (Krogh 1914; Hayes 1949; Ricker 1979; Jobling 2003) W ¼ W0 emt m¼ ln Wf À ln W0 tf whereWf is the ¢nal body weight (g) and tf is the time (days) betweenW0 andWf The SGR has often been proposed as a growth model in aquaculture (Willoughby 1999; AlanÌrÌ, Kabri & Paspatis 2001) even though it gives no consideration to the e¡ect of body weight and temperature on the growth of ¢sh Keeping in mind these drivers of ¢sh growth, Brett (1974) determined di¡erent SGR for various water temperatures and body weights and entered the values (observed and extrapolated by eye) into tables that served afterwards to predict ¢sh growth according to prevailing conditions (Brett 1974; Willoughby 1999) However, the relationship between SGR and temperature can be a¡ected by the amplitude of temperature £uctuations (Brett 1979; Xu 1996) In other words, growth rates observed at constant temperature (e.g 15 Ỉ 1C) might di¡er from those at an average temperature (e.g 15 Ỉ 1C), especially when £uctuations occur over a short period of time The SGR model is based on the incorrect assumption that ¢sh growth is continually exponential This has proven not to be the case and, therefore, growth predictions have to be re-calculated every time the predicted growth curve moves too far away from the observed trajectory (Brett 1979) Unlike Brett (1974), Elliott (1975) plotted the relationships between SGR, body weight and temperature and derived the following equation to predict the growth of brown trout Salmo trutta (LinneÔ): Exponential equation or specific growth rate (SGR) dW ẳ a ỵ b2 TịW 1b1 dt The origin of SGR goes back in 1798 and was developed to address demographic concerns Reverend Thomas Malthus, a mathematician, published an essay in 1798 in which he stated that the human popu- with the integral form (provided T is assumed constant): h i1=b1 W ẳ b1 a ỵ b2 Tịt ỵ W0b1 164 5ị r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 Aquaculture Research, 2010, 41, 161^181 Modelling growth and body composition in ¢sh A Dumas et al where b1, b2 and a are weight exponent (dimensionless), slope [% (day 1C) À 1] and intercept (% day À 1) of the relationship between SGR (% day À 1) and T ( 1C) respectively The Elliott model is often used to investigate ¢sh growth, especially in the ecology literature (Craig 1982; Allen 1985; Jensen 1990) From an inspection of Eqns (1) and (5), it is evident that the Elliott model is also a special case of von Bertalan¡y’s equation with m ¼ a þ b2 T; b ¼ À b1 ; l ¼ Moreover, Eq (3) of Iwama and Tautz (1981) has many similarities to Eq (5) Therefore, Elliott (1975) introduced the e¡ect of temperature into Eq (2) of Parker and Larkin (1959) before Iwama and Tautz (1981) Equation (5) needs to be solved repeatedly over the growing period because slope and intercept change with water temperature and body weight (Fig 2) This drawback limits application of the Elliott (1975) model because predictions can be applicable only to very short intervals and preclude comparison between studies, especially under £uctuating water temperatures Elliott, Hurley and Fryer (1995) revised the Elliott model and included considerations for optimum (Topt) and limiting (Tlim) temperatures for growth (Fig 2) The resulting equation takes the form  b 1=b W0 ỵ bcT Tlim ịt Wẳ 6ị 100ðTopt À Tlim Þ Specific growth rate (%/d) where c is the SGR of a 1g ¢sh at Topt,Tlim lower (TL) or upper (TU) temperature at which SGR is 0:Tlim 5TL if T Topt or Tlim 5TU if T4Topt Equation (6) is valid as long as the water temperature does not change Under £uctuating temperature conditions, the equation needs to be extended and body weight at the end of a growing period (t1, t2, , tk), Wk, is now predicted using the Topt 2.5 g BW 50 g BW 2.0 300 g BW 1.5 1.0 g BW TL TU 0.5 0.0 10 13 15 Temperature (°C) 18 Figure E¡ects of body weight (BW) and temperature on speci¢c growth rate (SGR) Lower (TL) and upper (TU) temperatures indicate where SGR is zero (adapted from Elliott 1975) following:  bc ðT1 À Tlim Þt1 ðT2 Tlim ịt2 ỵ Topt Tlim 100 Topt Tlim  Tk Tlim ịtk ỵ ỵ Topt Tlim Wkb ẳ W0b ỵ 7ị where T1,T2, ,Tk correspond to average temperature ( 1C) for intervals 1, 2, ,k, and t1, t2, , tk are in days The authors reported that Eq (7) yields signi¢cant discrepancies when the growing period exceeded months (Elliott et al.1995) Furthermore, the assumption of a ¢xed growth rate c in Eqns (6) and (7) is contrary to the biology and growth trajectory of ¢sh Another exponential ¢sh growth model was proposed more recently by Lupatsch and Kissil (1998): Y ¼ aX b ecT where Y and X are weight gain (g ¢sh À day À 1) and body weight (g ¢sh À 1), respectively, a and c are constants, b is weight exponent (dimensionless) and T is water temperature ( 1C) This equation is also a special case of the von Bertalan¡y with Z aecT and k in Eq (1): dW ¼ a ecT W b dt Let dW ¼ Y; W ¼ X dt Therefore, Y ¼ aX b ecT This model has been used successfully to describe the growth trajectory of warmwater ¢sh species such as gilthead seabream (Lupatsch & Kissil 1998), European sea bass (Lupatsch, Kissil & Sklan 2001), white grouper (Lupatsch & Kissil 2005) and barramundi (Glencross 2006) within a relatively narrow range of temperature ($ 20^27 1C) It assumes an exponential relationship between water temperature and growth rate, which can be true only for a certain range of optimal temperature, and appears in disagreement with the thermal-unit concept and reaction kinetic models for ectotherms The latter showed that growth rate is inhibited at high temperature, and relationship between growth rate and temperature displays an asymmetric bell-shaped curve (Sharpe & DeMichele 1977; School¢eld, Sharpe & Magnuson 1981) (Correction added on September 2009, after ¢rst online publication: In the sentence ‘This equation is also a special case of the von Bertalan¡y with Z YecT .’, the symbol Y was corrected to a.) r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 165 Modelling growth and body composition in ¢sh A Dumas et al Based on visual appraisal of typical growth curves (e.g Fig.1), animals not grow geometrically, i.e exponentially, across life stages The exponential growth function is therefore not suitable for accurately predicting or describing the growth trajectory of ¢sh and other animals Furthermore, this function yields unavoidably systematic deviations (Fig 3) Growth data on Arctic charr Salvelinus alpinus (LinneÔ) obtained from Simmons (1997) are used here to compare the TGC and SGR models (constant water temperature:12 1C; duration: 112 days) Using the latter equation, growth is underestimated from 11.5 g (W0) to 174.2 g (Wf) whereas body weight increases steeply from 174.2 to 678 g over a 56-day period, which is unrealistic This is in agreement with Brett (1974,1979) and Cho (1992) who pointed out that SGR leads to underestimation of growth between values of W0 andWk used to compute SGR and to serious overestimations of weight gain beyondWk In spite of its limitations, SGR remains widely accepted by editors and recommended ubiquitously in the ¢sh literature likely because of its ease of use (Barton 1996; Willoughby 1999; AlanÌrÌ et al 2001; Stead & Laird 2002) At best, SGR can serve in comparing di¡erent performances, although comparisons using SGR are valid only if ¢sh have similar W0 and Wk and are reared at the same water temperature because, as stated earlier, growth rate of ¢sh varies with size and temperature For all the reasons mentioned above, SGR ¢nds very little biological support and is therefore largely unsuitable as a ¢sh growth model and tool to compare short-term growth performance 700 SGR Body weight (g) 600 TGC 500 Observed 400 300 200 100 0 50 100 Time (days) 150 200 Figure Comparison between observed and predicted body weight of Arctic charr (Salvelinus alpinus LinneÔ) using the thermal-unit growth coe⁄cient (TGC) and speci¢c growth rate (SGR) Growth data are from Simmons (1997) 166 Aquaculture Research, 2010, 41, 161^181 Simple models of feed conversion to biomass Goals in animal nutrition are arguably to maximize the conversion of inputs (e.g feed, investments) into high-quality outputs over a short period of time Improving the conversion of dietary inputs to lean rather than adipose tissue growth is of bene¢t to producers and consumers It can also contribute to reduced waste outputs and provide room for manoeuvre given the volatility of pro¢t margins As a consequence, several studies have turned their attention towards feed e⁄ciency, protein utilization and lipid distribution as a function of ¢sh size, feeding level and alternative ingredients for example (Aursand, Bleivik, Rainuzzo, JÖrgensen & Mohr 1994; Azevedo, Cho, Leeson & Bureau 1998; Lupatsch, Kissil, Sklan & Pfe¡er 2001; Cheng, Hardy & Usry 2003) These studies have generated a large amount of information (e.g on body composition) that still needs to be explored and synthesized Most of these studies were designed to describe animal responses (e.g weight gain) within speci¢c experimental conditions Unfortunately, their ability to describe a wide array of animal responses in varying situations is limited because their experimental designs prevent representation of the mechanisms in the internal structure of the organism that are responsible for the observed responses For this reason, several mathematical modellers have insisted on the need to move from a requirement-based (input^output) to a rate:state approach where the major variables in play can be described and related dynamically, similar to a metabolic pathway (AFRC 1991;Thornley & France 2007; LoÔpez 2008) The rate:state formalism consists of representing the rate of change of pools, referred to as state variables, using di¡erential equations (Dijkstra, Mills & France 2002) Such formalism considers the state of a pool as the result of dynamic exchanges, i.e in£ux (e.g protein synthesis) and e¥ux (e.g protein degradation) of substances Di¡erential equations are a valuable tool and have been proven essential in dynamic modelling in describing the behaviour of a system concisely and e⁄ciently (Kleiber 1961; France & Kebreab 2006) The rate:state formalism is discussed further in Nutrient-based models Bioenergetic models Animal energetics refers to the quantitative study of energy exchanges induced by metabolic processes in r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 Aquaculture Research, 2010, 41, 161^181 Modelling growth and body composition in ¢sh A Dumas et al living organisms to stay alive, grow and reproduce (Nelson & Cox 2000) Energy exists in materials of dietary and body origin and is released in the form of heat to support work (Blaxter 1989) Models constructed on the basis of bioenergetic principles utilize mathematical equations describing the heat transactions and adhere generally to a factorial scheme, also referred to as an energy budget The factorial approach follows from the metabolizable energy concept (Armsby 1903; HMSO 1975), where energy expenditures or heat production are allocated to di¡erent metabolic processes according to an order of priority (NRC 1993; Bureau, Kaushik & Cho 2002) Inspired by Ivlev (1939) and Winberg (1956), Warren and Davis (1967, 1968) adhered to the factorial approach and proposed a simple additive equation to describe the energy budget of Âsh: C ẳ F ỵ U ỵ DB ỵ R 8ị where C is intake of energy and F and U are energy losses in faeces, and urine and gills respectively (all variables in units of MJ day À 1) Variable DB represents growth (energy gain) of the ¢sh and R is energy loss through metabolic processes associated with maintenance and heat increment of feeding Each component of the equation is described using mathematical relationships derived mostly using statistical analyses Equation (8) gained acceptance in ¢sheries and was adopted by Ricker (1968), Elliott (1976a, b) and Kitchell, Stewart and Weininger (1977) A systematic terminology for the description of energy budget and metabolic processes in animal nutrition was developed later by NRC (1981), and heat losses were categorized as shown in Fig Fish growth has usually been predicted using two di¡erent approaches in bioenergetic models One way of forecasting ¢sh growth assumes that energy intake drives weight gain This assumption is encountered mostly in ¢sheries and ecology studies because availability of food in natural ecosystems often limits ¢sh growth (Elliott 1976a, b; Kitchell et al 1977; From & Rasmussen 1989) An alternative approach considers genetic or desired growth rate rather than nutrition as the factor limiting animal growth (Hubbell 1971; Calow 1973; Oldham, Emmans & Kyriazakis 1997) Here, intake of energy is a function of the requirements of the individual to achieve a given growth capability or growth target This approach was suggested byWinberg (1956, p.174) and is mostly used in aquaculture where ¢sh are generally fed to satiation with nutritionally complete diets (Cho 1990; Lupatsch, Kissil & Sklan 2001; Zhou, Xie, Lei, Zhu & Yang 2005) Genetically determined growth capability of ¢sh is assessed using simple growth functions, especially the TGC model and the Intake Energy Faeces excretions Digestible Energy Gill excretions Urine excretions Metabolizable Energy Heat Increment Waste formation and excretion Product formation Digestion and absorption Net Energy Maintenance Basal metabolism Voluntary activity Thermal regulation Recovered Energy Growth, Fat, Reproduction Figure Factorial framework of energy partitioning in typical bioenergetic models intended to evaluate feed requirements Each metabolic process results in heat loss that is determined mostly using regression equations For further information on de¢nitions of terms and mathematical description of metabolic processes, the reader is referred to NRC (1981) and Bureau et al (2002) r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 167 Modelling growth and body composition in ¢sh A Dumas et al exponential model of Lupatsch and Kissil (1998) (Cho & Bureau 1998; Lupatsch & Kissil 1998; Lupatsch, Kissil & Sklan 2001; Glencross 2006) Probably the most adapted bioenergetic model for farmed ¢sh is the FISH-PRFEQ program (Cho & Bureau 1998) The model follows a factorial scheme and estimates feed requirements and waste outputs from expected growth performance, digestible energy of the diet and body energy deposition The FISH-PRFEQ model has been used or adapted to di¡erent ¢sh species and for various purposes (Kaushik 1998; Papatryphon, Petit, van der Werf, Kaushik & Claver 2005; Zhou et al 2005) Bioenergetic models predict energy gain, but they provide little information on the chemical composition (moisture, protein, lipid and ash) of biomass gain This characteristic has two signi¢cant drawbacks Firstly, the bioenergetic models can entail systematic errors because the relationship between recovered energy and weight gain changes across life stages (Bureau et al 2002) More energy is contained per unit of biomass gain for a large ¢sh (e.g 10 kJ g À BW) than for a small ¢sh (e.g kJ g À BW) under typical rearing conditions Studies have shown that the composition of biomass gain includes more lipid and less water in a large ¢sh than in a small ¢sh (Shul’man 1974; Dumas, de Lange, France & Bureau 2007) Protein and lipid deposition (LD) are two distinct biological processes driven by di¡erent factors or determinants that are overlooked in bioenergetic models Secondly, the recovered energy can serve to determine the energy retention e⁄ciency, but it is of no utility in assessing the e⁄ciency of nutrient utilization or rates of deposition unless reliable equations are developed to describe body composition across life stages It has been shown that feed evaluation systems and animal growth models based on bioenergetics have major limitations (Birkett & de Lange 2001a; Bajer, Whitledge & Hayward 2004; Dijkstra, Kebreab, Mills, Pellikaan, LoÔpez, Bannink & France 2007) Feed evaluation systems cannot rely on bioenergetics exclusively and have to consider dietary proteins and other nutrients, especially with ¢sh that rely heavily on proteins to meet their metabolic needs Moreover, digestible proteins, along with dietary amino acids, a¡ect feed e⁄ciency and nitrogen retention e⁄ciency signi¢cantly (Azevedo, Leeson, Cho & Bureau 2004a; Encarnac°o, de Lange, Rodehutscord, Hoehler, Bureau & Bureau 2004; Booth, Allan & Anderson 2007) The e¡ect of protein intake, and not only energy, on ¢sh growth performance was soon acknowl- 168 Aquaculture Research, 2010, 41, 161^181 edged and included in models to estimate feed requirements, weight gain, and e⁄ciency of energy and protein retention of African cat¢sh (Machiels & Henken1986), tilapia (van Dam & De Vries1995), carp (Schwarz & Kirchgessner 1995), European sea bass (Lupatsch, Kissil & Sklan 2001, Lupatsch et al 2003), gilthead sea bream and white grouper (Lupatsch et al 2003; Lupatsch & Kissil 2005) Although the factorial approach assumes that energetic costs of metabolic processes are additive, evidence suggests that energy is allocated in a compensatory fashion, i.e according to the metabolic scope of the animal at a particular life stage (Wieser 1989; Rombough 1994) This particularity may explain why the concept of energy requirement for maintenance remains debatable and is a¡ected by body composition and other factors such as ambient temperature and breed (e.g Close, Mount & Brown 1978; ARC 1981; Thompson, Meiske, Goodrich, Rust & Byers 1983; Campbell, Crim, Young & Evans 1994; Knap 2000) For instance, models based on bioenergetic principles assume that growth and feed e⁄ciency will be nil when animals are fed a maintenance ration (recovered energy 0) This assumption has been proven inaccurate in ¢sh, as well as in other animals, where positive weight gain was still observed even though animals were fed at or below a maintenance ration and the whole-body energy balance was negative (Huisman 1976; Le Dividich, Vermorel, Noblet, Bouvier & Aumaitre 1980; MeyerBurgdor¡, Osman & Gˇnther 1989; Lupatsch, Kissil & Sklan 2001; Bureau, Hua & Cho 2006) Bioenergetic models have also been used to estimate feed requirements of ¢sh and waste outputs from ¢sh culture operations (Winberg 1956; NRC 1993; Cho & Bureau 1998; Lupatsch & Kissil 1998, 2005) Assessing waste outputs requires good estimates of body composition in order to compute, for example, nitrogen and phosphorus discharge into the environment Nutrient-based models Historically, animal nutritionists ¢rst considered nutrients (i.e chemicals and macromolecules that provide essential nourishment for maintenance, growth and reproduction) rather than energy to study the conversion of feed to biomass (for a review, see Dumas et al 2008) Chemical (water, nitrogen, fat, minerals and carbon) and physical (bone, muscle, adipose tissue, blood, skin, hair and o¡al) composi- r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 Aquaculture Research, 2010, 41, 161^181 Modelling growth and body composition in ¢sh A Dumas et al tions of carcass and chemical composition of feedstu¡s were estimated for farm animals before the 20th century (Wol¡ 1895) Wol¡ (1895) appears to be the ¢rst to adopt a factorial approach to describe relatively and in detail the fate of dietary nitrogen, carbon and fat with consideration of intake, losses through faeces and urine, and recovery as body fat and body £esh in the carcass In view of the limitations of bioenergetics, animal nutritionists and growth modellers have returned to more nutrient- or biochemical-oriented approaches (e.g Machiels & Henken 1986; Gerrits, Dijkstra & France 1997; Birkett & de Lange 2001b) These nutrient-based models may be de¢ned as mechanistic systems designed to simulate the fate of dietary nutrients, with consideration of utilization of amino acids, fatty acids and their precursors Similar to bioenergetics, nutrient-based models serve to predict growth, nutrient requirements and waste outputs Intake Faecal excretion Anabolism and Catabolism Urinary Excretion Basal Production Figure Example of a factorial framework of nutrient partitioning (adapted from Blaxter & Mitchell 1948; Birkett & de Lange 2001a) Flow of nutrients through each metabolic process (intake, faecal and urinary excretion, anabolism and catabolism, basal metabolism and production) is determined mostly using regression and mass balance equations However, these models further explain the processing of nutrients by considering intermediary metabolism and are therefore more mechanistic Bioenergetic models are mostly descriptive, rely on a rather simple framework of energy transaction, represent energy using units of joules or calories and overlook the stoichiometry of energy-yielding nutrients Nutrient-based models are more explanatory, rely on metabolic pathways of nutrients, represent energy in terms of ATP (e.g mol ATP per molecule substrate), and consider the stoichiometry of chemical reactions These nutrient-based models have been shown to be e¡ective for mammals and ¢sh (e.g Gill, Thornley, Black, Oldham & Beever 1984; van Dam & De Vries 1995) Partitioning of nutrients can follow either a factorial or a compartmental scheme Figures and illustrate and contrast the factorial and compartmental approaches respectively The former approach is consistent with conventional bioenergetic models and adheres to the same assumptions (e.g energy is allocated according to a hierarchy, metabolic processes are additive) The latter was introduced in the 1950s into animal nutrition by Blaxter, Graham and Wainman (1956)-these authors did not nominate it as compartmental or mechanistic modelling, though-and consists of subdividing a given level of organization (e.g whole animal, tissue, cell) into di¡erent pools (e.g amino acids in the blood, intracellular glucose) (Thornley & France 2007) Pools are referred to as state variables (i.e a quantity that de¢nes the size of the pool at a given point in time) and can be in steady state (e.g blood glucose in a fasting animal) or non-steady state (e.g muscle protein content in a growing animal) Flows of substrates (e.g lysine and other metabolites) between pools and into and out of the system are represented as terms within di¡erential equations, which are usually Blood Fatty acids Amino acids Catabolism Protein in dressed carcass Protein in viscera Lipid in viscera Lipid in dressed carcass Figure Example of a simple compartmental framework of nutrient partitioning (adapted from Gill et al 1989) Flow of nutrients between each pool (amino acids, fatty acids, protein and lipid in the viscera and dressed carcass) is determined using di¡erential and stoichiometric equations r 2009 Blackwell Munksgaard No claim to original US government works, Aquaculture Research, 41, 161^181 169 Arti¢cial spermatogenesis in grouper R Murata et al Aquaculture Research, 2010, 41, 303^308 Figure (a) An immature ovary of an initial-control ¢sh (144 DPH) (b) An ovary of a control ¢sh months after the start of the experiment (c) An ovary of an MT-treated ¢sh months after start of the experiment (d) A testis of an MTtreated ¢sh months after start of the experiment (e) An ovary of a control ¢sh,6 months after start of the experiment (f) An ovary of an MT-treated ¢sh for months and then fed a normal diet, like that of the control ¢sh DPH, days post-hatch; MT, 17a-methyltestosterone; GC, germ cell; Sc, Somatic cell; OC, ovarian cavity; OO, oocyte; SG, spermatogonia; SC, spermatocyte; ST, spermatid; SZ, spermatozoa Inset ¢gure indicates high-magni¢cation images duce precocious sex change in this species with the administration of exogenous androgen (Murata et al 2009) In the present study, all ¢sh in the initial control had immature ovaries with an ovarian cavity after the completion of ovarian di¡erentiation Here, however, we show for the ¢rst time that sex change can be induced in underyearling E malabaricus during the period of ovarian di¡erentiation through treatment with MT In addition, sex-changed males had de¢ned testes with active spermatogenic germ cells Arti¢cial sex change in various groupers usually is carried out when the female matures following the completion of ovarian di¡erentiation (Chen et al 1977; Kuo et al 1988; Chao & Chow 1990; Chao 306 & Lim 1991; Fang et al 1992; Tan-Fermin et al 1994; Glamuzina et al 1998; Yeh et al 2003; Bhandari, Higa et al 2004; Bhandari, Komuro et al 2004; Bhandari et al 2006; Alam et al 2006) However, some studies of the dusky grouper Epinephelus marginatus showed that MT implants in immature 1-year-old juveniles for 12 weeks induced complete sex change (Glamuzina et al 1998; Sarter et al 2006) These studies clearly demonstrate that in groupers, the ovaries ranging from immature to mature states have the ability to change to a testis in response to exogenous androgen An important result of the present study is the induction of active spermatogenic germ cells, including r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 303^308 Aquaculture Research, 2010, 41, 303^308 spermatozoa in the testes, after sex change in the underyearling grouper There is no previous report in hermaphrodite ¢sh In gonochoristic ¢sh, it is known that the sexually undi¡erentiated Japanese eel undergoes gonadal masculinization with active spermatogenesis in the testis upon treatment with exogenous androgen (Ohta & Takano 1996) However, there is no report of spermatogenesis in immature o1-year-old, undi¡erentiated or newly di¡erentiated grouper ¢sh Epinephelus malabaricus usually takes more than 10 years to become male with active spermatogenic tissues in the testis (H Karimata, pers comm.) Our results suggest that some germ cells in the immature ovary of protogynous groupers have the potential to di¡erentiate into spermatozoa, in response to exogenous androgen However, in present study, we detected di¡erences in heads size among the spermatozoa in the testes of ¢sh treated with MT for months This fact suggests that some spermatozoa in the testes of underyearlings have problems in precocious spermatogenesis induced by exogenous androgen We induced sex change in underyearling E malabaricus by a 6-month treatment of MT at dose of 50 mg g À diet This dose of MT is also e¡ective for inducing sex reversal in the Nile Oreochromis niloticus (Nakamura & Iwahashi 1981) The dose and duration of androgen treatments and successful induction of female to male sex change vary among various studies depending on the species and stage of sexual maturity of the grouper examined The oral administration of 120 mg MT kg À body weight (BW) for months in Epinephelus tauvina (Chao & Chow 1990) and 1.0 mg MT kg À BW for months in Epinephelus fario (Kuo et al.1988) successfully induced sex change in these two groupers Six biweekly injections of 30 mg MT kg À BW induced sex change to males in Epinephelus suillus (Tan-Fermin et al 1994) 17a-methyltestosterone (0.5 mg kg À BW) implantation for months resulted in functional males in E tauvina (Chao & Lim 1991) The implantation of an androgen mixture (MT, testosterone, testosterone propionate) at 1000 mg kg À BW was capable of producing functional males in Epinephelus coioides within months (Yeh et al 2003) The duration of the MT treatment in our present study is longer than in these reported studies This di¡erence could be explained by immaturity of germ cells in underyearling grouper In the present study we showed that sex change is possible in underyearlings of the Malabar grouper using synthetic androgen during the period of ovarian di¡erentiation.We have not shown, however, that Arti¢cial spermatogenesis in grouper R Murata et al the sex-changed ¢sh produce viable sperm or that sex inversion is permanent Additional work will be necessary to address these issues and thereby fully establish that MT treatment of underyearlings yields fertile males Acknowledgments The authors highly appreciate Dr Paul V Dunlap of University of Michigan, Department of Ecology and Evolutionary Biology for his generous help to improve the English of this manuscript This research was supported in part by a Grant-in-Aid for Science Research from the Ministry of Education, Science, Sports and Culture of Japan, and by a research grant entitled ‘Utilizing advanced technologies in agriculture, forestry and ¢sheries’, Japan References Alam M.A., Bhandari R.K., Kobayashi Y., Soyano K & Nakamura M (2006) Induction of sex change within two full moons during breeding season and spawning in grouper Aquaculture 255, 532^535 Bhandari R.K., Komuro H., Nakamura S., Higa M & Nakamura M (2003) Gonadal restructuring and correlative steroid hormone pro¢les during natural sex change in protogynous honeycomb grouper (Epinephelus merra) Zoological Science 20, 1399^1404 Bhandari R.K., Higa M., Nakamura S & Nakamura M (2004) Aromatase inhibitor induces complete sex change in the protogynous honeycomb grouper (Epinephelus merra) Molecular Reproduction and Development 67, 303^307 Bhandari R.K., Komuro H., Higa M & Nakamura M (2004) Sex inversion of sexually immature honeycomb grouper (Epinephelus merra) by aromatase inhibitor Zoological Science 21, 305^310 Bhandari R.K., Alam M.A., Soyano K & Nakamura M (2006) Induction of female-to-male sex change in the honeycomb grouper (Epinephelus merra) by 11-ketotestosterone treatments Zoological Science 23, 65^69 BrusleÔ-Sicard S., Debas L., Fourcault B & Fuchs J (1992) Ultrastructural study of sex inversion in a protogynous hermaphrodite, Epinephelus microdon (Teleostei, Serranidae) Reproduction Nutrition Development 32, 393^406 Cardwell J.R & Liley N.R (1991) Hormonal control of sex and color change in the stoplight parrot¢sh, Sparisoma viride General and Comparative Endocrinology 81,7^20 Chao T.M & Chow M (1990) E¡ect of methyltestosterone on gonadal development of Epinephelus tauvina (FORSKAL) SingaporeJournal of Primary Industries 18, 1^14 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 303^308 307 Arti¢cial spermatogenesis in grouper R Murata et al Chao T.M & Lim L.C (1991) Recent developments in the breeding of grouper (Epinephelus spp.) in Singapore SingaporeJournal of Primary Industries 19,78^93 Chen F.Y., Chow M., Chao T.M & Lim R (1977) Arti¢cial spawning and larval rearing of the grouper, Epinephelus tauvina (FORSKAL) in Singapore Singapore Journal of Primary Industries 5, 1^21 FangY., Lin Q., Qi X & Hong G (1992) E¡ects of 17a-methyltestosterone on sex reversal in Epinephelus akaara Journal of Fisheries of China (Shuichan Xuebao Shanghai) 16, 171^174 Glamuzina B., Glavic N., Skaramuca B & Kozul V (1998) Induced sex reversal of dusky grouper, Epinephelus marginatus (Lowe) Aquaculture Research 29, 563^567 Higa M., Ogasawara K., Sakaguchi A., NagahamaY & Nakamura M (2003) Role of steroid hormones in sex change of protogynous wrasse Fish Physiology and Biochemistry 28, 149^150 Kroon F.J & Liley N.R (2000) The role of steroid hormones in protogynous sex change in the Blackeye goby, Coryphopterus nicholsii (Teleostei: Gobiidae) General and Comparative Endocrinology 118, 273^283 Kroon F.J., Munday P.L., Westcott D.A., Hobbs J.P.A & Liley N.R (2005) Aromatase pathway mediates sex change in each direction Proceedings of the Royal Society B: Biological Sciences 272,1399^1405 Kuo C.M.,TingY.Y & Yeh S.L (1988) Induced sex reversal and spawning of blue-spotted grouper, Epinephelus fario Aquaculture 74,113^126 LeeY.H.,YuehW.S., Du L.L., Sun L.T & Chang C.F (2002) Aromatase inhibitors block natural sex change and induce male function in the protandrous black porgy, Acanthopagrus schlegeli: possible mechanism of natural sex change Biology of Reproduction 66,1749^1754 Li G.L., Liu X.C & Lin H.R (2005) Aromatase inhibitor letrozole sex inversion in the protogynous red spotted grouper (Epinephelus akaara) Sheng Li Xue Bao 57, 473^ 479 Marino G., Azzurro E., Finoia M.G., Messina M.T., Massari A & Mandich A (2000) Recent advances in induced breeding of the dusky grouper Epinephelus marginatus (Lowe, 1834) CIHEAM- Opt Medit 47, 215^225 Murata R., Karimata H., Alam M.A & Nakamura M (2009) Gonadal sex di¡erentiation in the Malabar grouper, Epinephelus malabaricus Aquaculture 293, 286–289 Nakamura M & Iwahashi M (1981) Studies on the practical masculinization in Tilapia niloticus by the oral administration of androgen Japanese with English abstract Bulletin of the Japanese Society of Scienti¢c Fisheries 48, 763^769 Nakamura M & Takahashi H (1973) Gonadal sex di¡erentiation in tilapia, with special regard to the time of estro- 308 Aquaculture Research, 2010, 41, 303^308 gen treatment e¡ective in inducing complete feminization of genetic males Bullettin of Faculty of Hokkaido University 24, 1^13 Nakamura M., Kobayashi T., Chang X.T & Nagahama Y (1998) Gonadal sex di¡erentiation in teleost ¢sh Journal of Experimental Zoology 281, 362^372 Nozu R., KojimaY & Nakamura M (2009) Short term treatment with aromatase inhibitor induces sex change in the protogynous wrasse, Halichoeres trimaculatus General and Comparative Endocrinology 161, 360^364 Ohta H & Takano K (1996) Testicular maturation induced by methyltestosterone in elvers of the Japanese eel Fisheries Science 62, 990^991 SadovyY (2000) Regional survey for fry/¢ngerling supply and current practices for grouper mariculture: evaluating current status and long-term prospects for grouper mariculture in South East Asia Final report to the Collaboration APEC grouper research and development network (FWG 01/99) Sadovy Y & Colin P.L (1995) Sexual development and sexuality in the Nassau grouper Journal of Fish Biology 46, 961^976 Sarter K., Papadaki M., Zanuy S & Mylonas C.C (2006) Permanent sex inversion in 1-year-old juveniles of the protogynous dusky grouper (Epinephelus marginatus) using controlled-release 17a-methyltestosterone implants Aquaculture 256, 443^456 Shapiro D.Y (1987) Di¡erentiation and evolution of sex change in ¢shes Bioscience 37, 490^496 Shapiro D.Y., SadovyY & McGehee M.A (1993) Periodicity of sex change and reproduction in the red hind, Epinephelus guttatus, a protogynous grouper Bulletin of Marine Science 53, 1151^1162 Smith C.L (1965) The patterns of sexuality and the classi¢cation of serranid ¢shes American Museum Novitates 2207,1^20 Tan S.M & Tan K.S (1974) Biology of the tropical grouper, Epinephelus tauvina (Forskal) I: a preliminary study on hermaphroditism in E tauvina Singapore Journal of Primary Industries 2,123^133 Tan-Fermin J.D., Garcia L.M.B & Castillo A.R Jr (1994) Induction of sex inversion in juvenile grouper, Epinephelus suillus, (Valenciennes) by injections of 17a-methyltestosterone JapaneseJournal of Ichthyology 40, 413–420 Yamamoto T (1969) Sex di¡erentiation In: Fish Physiology, Vol III (ed by W.S Hoar & D.J Randall), pp 117^175 Academic Press, NewYork, NY, USA Yeh S.L., Kuo C.M.,Ting Y.Y & Chang C.F (2003) Androgens stimulate sex change in protogynous grouper, Epinephelus coioides: spawning performance in sex-changed males Comparative Biochemistry and Physiology Part C: Toxicology and Pharmacology 133, 375^382 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 303^308 Aquaculture Research, 2010, 41, 309^314 doi:10.1111/j.1365-2109.2009.02333.x Comparative efficacy of MS-222 and benzocaine as anaesthetics under simulated transport conditions of a tropical ornamental fish Puntius filamentosus (Valenciennes) Padinhare Kattil Pramod1, Alappat Ramachandran1,Thavarool Puthiyedathu Sajeevan2, Sunesh Thampy1 & Srinivas Somnath Pai3 School of Industrial Fisheries, Cochin University of Science and Technology, Cochin, India Department of Marine Biology, Microbiology and Biochemistry, Cochin University of Science and Technology, Cochin, India National Centre for Aquatic Animal Health, Cochin University of Science and Technology, Cochin, India Correspondence: P K Pramod, School of Industrial Fisheries, Cochin University of Science and Technology, Cochin 16, India E-mail: pramodmohan@yahoo.com Abstract Introduction There is a growing commercial interest in the ¢sh, Puntius ¢lamentosus, in the ornamental ¢sh trade in India and elsewhere The trade is, however, hampered by severe mortalities during transport of the ¢sh owing to insu⁄cient data available on the use of anaesthetics To resolve this problem, we evaluated the e⁄cacy of two anaesthetics, MS-222 and benzocaine, in sedating P ¢lamentosus in simulated transportation experiments and used stress response parameters such as cortisol and blood glucose levels to perform assessments We observed that MS-222 at 40 mg L À and benzocaine at 20 mg L À were su⁄cient to induce sedation for 48 h Above these concentrations, both the anaesthetics adversely a¡ected the ¢sh and resulted in mortalities Both anaesthetics signi¢cantly lowered the blood cortisol and glucose levels compared with the unsedated controls Importantly, the anaesthetics treatment signi¢cantly lowered the post-transport mortality in the ¢sh The results of the study show that MS-222 and benzocaine could be used as sedatives to alleviate transport-related stress in P ¢lamentosus to improve their post-transport survival and hence reduce economic loss Ornamental ¢sh export in India is a growing industry in tune with the increased interest for ornamental ¢sh-keeping all over the world The export of ornamental ¢sh from India mainly focuses on freshwater indigenous ornamental species (Ramachandran 2002) It is an inevitable part of modern aquaculture practices to stress the animals by containment, handling, sorting, transportation, periodic low oxygen or high ammonia, etc., and the associated post-transport mortality (Donaldson1981;Wendelaar Bonga 1997; Carneiro & Urbinati 2001; Pavlidis, Angellotti, Papandroulakis & Divanach 2003; Southgate 2008) The transport logistics involved inevitably stress the animals, causing post-transport mortality, and exporters are expected to compensate customers for losses exceeding 5% death on arrival (DOA) industry standard (Lim, Dhert & Sorgeloos 2003) The bulk of post-transport stress-mediated mortality occurs during the 1-week recovery period and enhancing the stress resistance of the ornamental ¢sh is insu⁄ciently addressed In aquaculture operations, anaesthetics are commonly used to minimize stress and reduce the physical injury to ¢sh during the various handling procedures The choice of anaesthetics is often dependent on considerations such as availability, cost-e¡ectiveness, ease of use, nature of the study and user safety (Cho & Heath 2000; Mylonas, Cardinaletti, Sigelaki & Polzonetti-Magni 2005) Be- Keywords: ornamental ¢sh, Puntius ¢lamentosus, simulated transport, stress, anaesthetics, posttransport survival r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd 309 Anaesthetics in ornamental ¢sh transport P K Pramod et al Aquaculture Research, 2010, 41, 309^314 fore recommending the use of a particular anaesthetic, a range of stress-response indices must be measured to assess its e⁄cacy Until recently, the packing and live transport of ornamental ¢sh did not involve the use of any chemical anaesthetics However, the increased concern for ¢sh health and product quality makes the use of anaesthetics inevitable to reduce the stress during handling and transportation procedures To date, much of the information on the use of anaesthetics in ¢sh has been derived from studies on salmonids (Pickering 1992; Iversen, Finstad, McKinleyc & Eliassen 2003; Pirhonen & Schreck 2003; Iversen, Eliassen & Finstad 2009) and other temperate species (Mattson & Ripple 1989) Investigations on high-latitude species have also yielded valuable information (Wells, Mcintyre, Morgan & Davies 1986), and except for a few reports, the use of anaesthetics in the transport of ornamental ¢sh remains largely unaddressed (Teo, Chen & Lee 1989; Guo, Teo & Chen 1995a, b; Kaiser & Vine 1998; Crosby, Hill,Watson & Yanong 2006) In the international ornamental ¢sh market, the Indian tiger barb Puntius ¢lamentosus is very popular, known by di¡erent names such as Filament barb, DMK ¢sh, Long ¢n barb, Feather ¢n barb, etc The export ¢gures reveal that the contribution of Indian tiger barb to the total ornamental ¢sh export from India is increasing (Pramod, Mini & Ramachandran 2002) In this study, we compared the e⁄cacy of the anaesthetics, MS-222 (tricaine methanesulphonate) and benzocaine (ethyl aminobenzoate), in P ¢lamentosus, to alleviate stress during a simulated transportation experiment Biochemical stress indices such as plasma cortisol level and glucose levels were used to assess the e¡ect of the anaesthetics size Since it is readily soluble in water, MS-222 was mixed directly into the transporting medium, whereas benzocaine, being insoluble, was dissolved in a few drops of ethanol before mixing into the transporting medium Six ¢sh (approximately 72 g L À 1) were transferred into each polyethylene bag ¢lled with 1L water containing the anaesthetics at di¡erent concentrations as described earlier It was then in£ated with medical-grade oxygen and the top of the bag was tied and made airtight A control group without the anaesthetics was also similarly maintained All treatments were in triplicates The experiment was conducted at a temperature of 22 Ỉ 1C to simulate the air shipment conditions for 48 h The ¢sh were observed carefully at 30-min intervals and their behavioural responses were noted The highest concentration of anaesthetics providing sedation, but the lowest mortality at the end of 48 h was chosen as the optimal dose for transportation Materials and methods Experiment I: optimal dose of MS-222 and benzocaine as anaesthetics Indian tiger barb P ¢lamentosus (average body length 126 Ỉ mm and body weight, 12 Ỉ 1g), collected from Chalakudy river systems, were brought to the laboratory and acclimatized for week in FRP tanks with a 5000 L capacity The ¢sh were fed a commercial pellet feed for week and feeding was terminated 24 h before the experiment Two commercial anaesthetics viz., MS-222 and benzocaine (Sigma Chemicals, St Louis, MO, USA), at concentrations of 5, 10, 20, 30, 40, 50 and 60 mg L À 1, were added to water in low-density polyethylene bags (LDPE) of 22  60 cm 310 Experiment II: biochemical analysis for stress indices Seventy-two LDPE bags (22  60 cm) containing six ¢sh each in 1-L water were divided into three groups (n 24) One group served as the control while the anaesthetics MS-222 and benzocaine were added to the other two at the optimum concentration determined in the earlier experiment The bags were ¢lled with medical-grade oxygen, made airtight and each group was placed together in a styrofoam box (60  40  55 cm size) for thermal insulation to prevent sudden changes in the temperature of the transport water All the boxes were kept at a controlled temperature of 22 Ỉ 1C for 48 h Plasma cortisol and glucose levels were analysed as stress indices in ¢sh sampled at 6-h intervals from each group during the experiment Before sampling, ¢sh from both control and anaesthetic-treated groups were euthanized with MS-222 at a lethal concentration of 200 mg L À following Cho and Heath (2000), where such exposure did not produce changes in the cortisol levels (Barton, Schreck & Sigismondi 1986) Blood was collected through a cardiac puncture using a heparinized syringe and transferred to 1.5 mL tubes The plasma was separated by centrifugation at 4500 g for 10 at 1C and stored at À 20 1C until further analysis Baseline values for all parameters were obtained from blood samples were taken from ¢sh just before packing Cortisol levels in the plasma were quanti¢ed using a commercially r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 309^314 Post-transport survival After 48 h of the experiment, the remaining ¢sh in the experimental bags were released into FRP tanks containing aerated water Separate tanks were maintained for all the three experimental groups for observing post-transport mortality for days after simulated transport The water temperature in the tanks was 28 Ỉ 1C with an average dissolved oxygen level of 12 mg L À 1, and the ¢sh were fed with pellet feed Statistical analysis All results were analysed using a one-way analysis of variance and Duncan’s multiple comparison of the means using SPSS 10.0 for Windows Di¡erences were considered to be signi¢cant when Po0.05 Results % Survival The safe and optimal anaesthetic dosage required for inducing sedation with least mortality after 48 h in P ¢lamentosus was determined Fish exposed to MS222 at concentrations 40 mg L À experienced sedation with no mortality at the end of 48-h experiments but not at the higher dosages tested (Fig 1) Therefore, the concentration of 40 mg L À was judged to be optimal and safe and the concentrations below this were found to be suboptimal In benzocaine-treated groups, sedation and no mortality were observed at concentrations 20 mg L À 1, above Treatment groups Figure Percentage survival of Puntius ¢lamentosus with di¡erent doses of MS-222 and benzocaine at the end of 48 h of simulated transportation (a) Plasma cortisol (µg dL- 1) available direct ELISA kit (CAN-C-270, Diagnostics Biochem Canada, ON, Canada), whereas the plasma glucose level was measured using a glucose colorimetric assay kit (Sigma-Aldrich, St Louis, MO, USA) Anaesthetics in ornamental ¢sh transport P K Pramod et al Time (hours) (b) Plasma glucose (mg dL- 1) Aquaculture Research, 2010, 41, 309^314 Time (hours) Figure (a) Plasma cortisol level (mg dL À 1) and (b) glucose level (mg dL À 1) of Puntius ¢lamentosus during 48-h simulated transportation with 40 mg L À MS-222 or 20 mg L À benzocaine Values with the same exposure time with di¡erent superscripts are signi¢cantly di¡erent (Po0.05) which the ¢sh displayed adverse e¡ects like loss of equilibrium and death Overall, anaesthesia-induced mortality was more pronounced in benzocaine-treated ¢sh, than in MS-222 treatments When the e⁄cacy of the anaesthetics was tested, the lowest increase in plasma cortisol levels was in the MS-222-treated group, followed by the benzocaine treatment compared with the untreated control within the ¢rst h of packing (Fig 2a) This increase was signi¢cantly lower (Po0.05) in the sedated ¢sh than in the control group, which indicated that the anaesthetics used were e¡ective in lowering the stress response in the animals The mean plasma cortisol levels in ¢sh treated with the anaesthetics gradually declined after h and almost reached the baseline within 48 h Plasma cortisol concentrations in the control group remained signi¢cantly higher than that of the other treatments throughout the experiment (Po0.05) However, there was no signi¢cant di¡erence between the two anaesthetics treatments in the study The blood glucose levels increased steadily in all treatment groups up to 18 h and declined thereafter (Fig 2b) Herein too, the increase was signi¢cantly r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 309^314 311 Cumulative mortality (%) Anaesthetics in ornamental ¢sh transport P K Pramod et al Post shipment days Figure Post-transport mortality for days in Puntius ¢lamentosus after 48 h of a simulated transport experiment with 40 mg L À MS-222 or 20 mg L À benzocaine Values with di¡erent superscripts are signi¢cantly di¡erent (Po0.05) lower in the sedated groups compared with the controls but not between the anaesthetics tested The blood glucose levels, however, did not reach baseline levels at the end of 48 h in all the groups Expectedly, the post-transport mortality at the end of days was signi¢cantly higher in the unsedated ¢sh than in the sedated groups (Fig 3) This indicated that both the anaesthetics tested e¡ectively reduced the stress during transport Discussion Handling of ¢sh out of their natural environment for transportation always results in considerable stress to the animal Struggling of the ¢sh has detrimental e¡ects on their physiology and behaviour, leading to large-scale mortality and loss in product quality during the subsequent period of transportation (Ross & Ross 1999, 2008) Losses due to DOA and dead after arrival are signi¢cant problems in ornamental ¢sh trade, which is unequivocally attributed to the stress associated with packing and transportation of ¢sh (Schmidt & Kunzmann 2005) This severely impedes the ability of suppliers to meet the customers’ needs for high-quality ¢sh Use of modern packaging technology for air transport to increase the survival and product quality is a principal factor in ornamental ¢sh trade Therefore, current packaging practices focus on minimizing the stress imposed on the ¢sh by controlling the metabolic rate and removal of metabolic waste from the transport water (Lim et al 2003; Harmon 2009) Although the use of anaesthetics to reduce the stress associated with handling, transport, con¢nement, etc is well established in aquacul- 312 Aquaculture Research, 2010, 41, 309^314 ture, their use in ornamental ¢sh trade is little studied The proper dosage of anaesthetics required is critical and varies widely between the species and the size of ¢sh (Coyle, Durborow & Tidwell 2004) While low concentrations reduce activity and metabolic rate, higher dosages are routinely speci¢ed during procedures that are deemed stressful or painful for the ¢sh There are many instances where light sedation is su⁄cient and in fact desirable over deeper sedation, to facilitate handling of ¢sh for di¡erent husbandry practices or especially for the transport of ¢sh (Wedemeyer 1997; Golovanova, Nikonorov & Moise 2006) Notably, while increasing the concentration of anaesthetics decreases the induction time, it also signi¢cantly increases the recovery period (Gullian & Villanueva 2009) In this study, we found that 40 and 20 mg L À of MS-222 and benzocaine, respectively, were optimal in P ¢lamentosus to impart light sedation, above which signi¢cant loss of equilibrium and mortality resulted Although benzocaine (ethyl aminobenzoate) is an e¡ective ¢sh anaesthetic with the desirable characteristics of rapid induction and recovery times (Ross & Ross 2008), the structurally similar MS-222 (tricaine methanesulphonate) is the most frequently used and preferred anaesthetic for ¢sh A marked di¡erence between these two local anaesthetics was the delayed response to the visual stress stimulus after benzocaine anaesthesia It was also observed that benzocaine concentrations above 20 mg L À resulted in partial or complete loss of equilibrium within after application and with subsequent mortalities This emphasizes that benzocaine is a more potent anaesthetic than MS222 requiring lower concentrations and that with a quicker action, there is a narrower margin of safety The greater activity of benzocaine can be attributed to its higher lipid solubility compared with MS-222, permitting easier passage through the blood^brain barrier, resulting in more pronounced central nervous system e¡ects (Kiessling, Johansson, Zahl & Samuelsen 2009) Plasma catecholamines and cortisol levels are an indices of stress response in ¢sh (Gamperl,Vijayan & Boutilier 1994) and acute stressors cause a rapid increase in these hormones, which in turn increase blood glucose levels through rapid breakdown of glycogen (Barton & Iwama 1991) Herein, we showed that anaesthetizing P ¢lamentosus before transport resulted in signi¢cantly lower plasma cortisol and glucose levels, indicating e¡ective sedation Although the plasma cortisol and blood glucose r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 309^314 Aquaculture Research, 2010, 41, 309^314 Anaesthetics in ornamental ¢sh transport P K Pramod et al levels increased initially in the anaesthetized groups, they quickly returned to the baseline level compared with the unanaesthetized controls The MS-222-treated group recovered faster from the stress as evidenced from the reduced levels of plasma cortisol and blood glucose indices in this group The initial elevation in cortisol and glucose levels is probably due to the handling stress during the capture of the ¢sh for experiment Once the ¢sh were transferred to transporting bags and tranquillized, their levels reduced considerably and reached the basal level in the subsequent hours Recently, Iversen et al (2009) reported a similar observation, which shows the stress-reducing potential of clove oil anaesthetics in Atlantic salmon The studies by Tomasso, Davis and Parker (1980) and Carmichael, Tomasso, Simco and Davis (1984) reported a decrease in stress response in hybrid striped bass and largemouth bass when the ¢sh were anaesthetized before capture and kept sedated during transport Similarly, Wagner, Arndt and Hilton (2002) observed that when adult rainbow trout were anaesthetized with MS-222 or CO2, the cortisol level returned to the initial level within and 24 h after handling The reduction in post-transport mortality plays a central role in the management of ornamental trade and there is a strong interest, motivated by sound economic and conservational reasons, to avoid such mortalities (Schmidt & Kunzmann 2005) Presently, exporters are expected to compensate the customers for the loss exceeding 5% DOA (Lim et al 2003) These post-transport mortalities are presumably due to osmoregulatory dysfunction or stress-mediated diseases, occurring during the ¢rst week after transport In the present study, the post-transport mortality was markedly reduced in anaesthetized groups of P ¢lamentosus, which shows that their use alleviated the stress Teo et al (1989) reported zero mortality in a guppy packaging experiment using 2phenoxyethanol at 0.11 and 0.22 g L À after a 20-h simulated shipment Similarly, Guest and Prentice (1982) reported that in blue back herring, Alosa aestivalis, the anaesthetized ¢sh had higher survival than non-dosed ¢sh after transportation In conclusion, the low stress indicators and low post-transport mortalities associated with the use of anaesthetics would be desirable in the ornamental ¢sh trade to reduce the economic loss and increase product quality The present result implies that the use of anaesthetics during packing and transportation would help to reduce the transportation stress and improve the post-transport survival of P ¢lamen- tosus In ornamental ¢sh trade, the use of anaesthetics at a desired and optimal level should be encouraged and, because the dose of anaesthetics is species speci¢c; more studies are warranted in adopting this technology, particularly for high-value species Acknowledgments The ¢rst author is grateful to Cochin University of Science and Technology for the fellowship during the course of the study Our thanks are due to Mr Suresh Kumar, Research Scholar, Department of Marine Biology, Microbiology and Biochemistry, Cochin University, for assistance with the biochemical analysis of the blood samples References Barton B.A & Iwama G.K (1991) Physiological changes in ¢sh from stress in aquaculture with emphasis on the response and e¡ects on corticosteroids Annual Review of Fish Diseases 1, 3^26 Barton B.A., Schreck C.B & Sigismondi L.A (1986) Multiple acute disturbances evoke cumulative physiological stress responses in juvenile chinook salmon Transactions of the American Fisheries Society 115, 245^251 Carmichael G.J.,Tomasso J.R., Simco B.A & Davis K.B (1984) Characterization and alleviation of stress associated with hauling largemouth bass Transactions of the American Fisheries Society 113,778^785 Carneiro P.C.F & Urbinati E.C (2001) Salt as a stress response mitigator of matrinxa, Brycon cephalus (Gunther), during transport Aquaculture Research 32, 298^307 Cho G.K & Heath D.D (2000) Comparison of tricaine methanesulphonate (MS-222) and clove oil anaesthesia e¡ects on the physiology of juvenile chinook salmon Oncorhynchus tshawytscha (Walbaum) Aquaculture Research 31, 537^546 Coyle S.D., Durborow R.M & Tidwell J.H (2004) Anaesthetics in Aquaculture, Publication No 3900 Southern Regional Aquaculture Center Stoneville, MS, USA Crosby T.C., Hill J.E.,Watson C.A & Yanong R.P.E (2006) Effects of tricaine methanesulfonate, hypno, metomidate, quinaldine, and salt on plasma cortisol levels following acute stress in threespot Gourami Trichogaster trichopterus Journal of Aquatic Animal Health 18, 58^63 Donaldson E.M (1981) The pituitary^interrenal axis as an indicator of stress in ¢sh In: Stress and Fish (ed by A.D Pickering), pp.11^47 Academic Press, London, UK Gamperl A.K., Vijayan M.M & Boutilier R.G (1994) Experimental control of stress hormone levels in ¢shes: techniques and applications Reviews in Fish Biology and Fisheries 4, 215^255 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 309^314 313 Anaesthetics in ornamental ¢sh transport P K Pramod et al Aquaculture Research, 2010, 41, 309^314 Golovanova T.S., Nikonorov S.I & Moise M.A (2006) Evaluation of potential anaesthetics for the Caspian inconnu (Stenodus leucichthys G1772) juveniles Meeting Abstract 1139, World Aquaculture Society, AQUA, 2006, Florence, Italy Guest W.C & Prentice J.A (1982) Transportation techniques for blueback herring The Progressive Fish-Culturist 44, 183^185 Gullian M & Villanueva J (2009) E⁄cacy of tricaine methanesulphonate and clove oil as anaesthetics for juvenile cobia Rachycentron canadum Aquaculture Research 40, 852– 860 doi:10.1111/j.1365-2109.2009.02180.x (in press) Guo F.C.,Teo L.H & Chen T.W (1995a) E¡ects of anaesthetics on the water parameters in a simulated transport experiment of platy¢sh, Xiphophorus maculatus (Gunther) Aquaculture Research 26, 265^271 Guo F.C.,Teo L.H & Chen T.W (1995b) E¡ects of anaesthetics on the oxygen consumption rates of platy¢sh Xiphophorus maculatus (Gunther) Aquaculture Research 26, 887^894 Harmon T.S (2009) Methods for reducing stressors and maintaining water quality associated with live ¢sh transport in tanks: a review of the basics Reviews in Aquaculture 1, 58^66 Iversen M., Finstad B., McKinleyc R.S & Eliassen R.A (2003) The e⁄cacy of metomidate, clove oil, Aqui-STM and Benzoaks as anaesthetics in Atlantic salmon (Salmo salar L.) smolts, and their potential stress-reducing capacity Aquaculture 221, 549^566 Iversen M., Eliassen R.A & Finstad B (2009) Potential bene¢t of clove oil sedation on animal welfare during salmon smolt, Salmo salar L transport and transfer to sea Aquaculture Research 40, 233^241 Kaiser H & Vine N (1998) The e¡ect of 2-phenoxyethanol and transport packing density on the post-transport survival rate and metabolic activity in the gold¢sh, Carassius auratus Aquarium Sciences and Conservation 2,1261^1263 Kiessling A.A., Johansson D., Zahl I.H & Samuelsen O.B (2009) Pharmacokinetics, plasma cortisol and e¡ectiveness of benzocaine, MS-222 and isoeugenol measured in individual dorsal aorta-cannulated Atlantic salmon (Salmo salar) following bath administration Aquaculture 286, 301^308 Lim L.C., Dhert P & Sorgeloos P (2003) Recent developments and improvements in ornamental ¢sh packaging systems for air transport Aquaculture Research 34, 923^935 Mattson N.S & Ripple T.H (1989) Metomidate, a better anaesthetic for cod (Gadhus morhua) in comparison with benzocaine, MS-222, chlorbutanol, and phenoxyethanol Aquaculture 83, 89^94 Mylonas C.C., Cardinaletti G., Sigelaki I & Polzonetti-Magni A (2005) Comparative e⁄cacy of clove oil and 2-phenoxyethanol as anesthetics in the aquaculture of European sea bass (Dicentrarchus labrax) and gilthead sea bream (Sparus aurata) at di¡erent temperatures Aquaculture 246, 467^481 Pavlidis M., Angellotti L., Papandroulakis N & Divanach P (2003) Evaluation of transportation procedures on water quality and fry performance in red progy (Pagrus pagrus) fry Aquaculture 218,178^202 Pickering A.D (1992) Rainbow trout husbandry: management of the stress response Aquaculture 100, 125^139 Pirhonen J & Schreck C.B (2003) E¡ects of anaesthesia with MS 222, clove oil and CO2 on feed intake and plasma cortisol in steelhead trout (Oncorhynchus mykiss) Aquaculture 220, 507^514 Pramod P.K., Mini S & Ramachandran A (2002) Prospects of exporting Indian tiger barb Puntius ¢lamentosus (Valenciennes) and Malini’s barb Puntius mahecola (Valenciennes) as ornamental ¢shes from Kerala In: Riverine and Reservoir Fisheries of India (ed by M.R Boopendranath, B Meenakumari, J Joseph, T.V Sankar, P Pravin & L Edwin), pp 393^399 Society of Fisheries Technologists India, Cochin, India Ramachandran A (2002) Fresh water indigenous ornamental ¢sh resources in Kerala and their prospects for international marketing In: Riverine and Reservoir Fisheries of India (ed by M.R Boopendranath, B Meenakumari, J Joseph, T.V Sankar, P Pravin & L Edwin), pp 109^134 Society of Fisheries Technologists India, Cochin, India Ross L.G & Ross B (1999) Anaesthesia of ¢sh In: Anaesthetic and Sedative Techniques for Aquatic Animals (ed by L.G Ross & B Ross), pp 58^88 Blackwell Science, Oxford, UK Ross L.G & Ross B (2008) Anaesthetic and SedativeTechniques for Aquatic Animals, 3rd edn Wiley-Blackwell, Oxford, UK 222pp Schmidt C & Kunzmann A (2005) Post-harvest mortality in the marine aquarium trade: a case study of an Indonesian export facility SPC Live Reef Fish Information Bulletin 13, 3^12 Southgate P.J (2008) Welfare of ¢sh during transport In: Fish Welfare (ed by E.J Branson), 312pp Blackwell Publishing, Oxford, UK ISBN: 9781405146296 Teo L.H., Chen T.W & Lee B.L (1989) Packaging of the guppy, Poecilia reticulata, for air transport in a closed system Aquaculture 78, 321^332 Tomasso J.R., Davis K.B & Parker N.C (1980) Plasma corticosteroid and electrolyte dynamics of hybrid striped bass (white bass  striped bass) during netting and hauling Proceedings of theWorld Marine Culture Society 11, 303^310 Wagner E., Arndt R & Hilton B (2002) Physiological stress responses, egg survival and sperm motility for rainbow trout broodstock anaesthetized with clove oil, tricaine methanesulfonate or carbon dioxide Aquaculture 211, 353^366 Wedemeyer G (1997) E¡ects of rearing conditions on the health and physiological quality of ¢sh in intensive culture In: Fish Stress and Health in Aquaculture (ed by G.K Iwama, A.D Pickering, J.P Sumpter & C.B Schreck), pp 35^71 Cambridge University Press, Cambridge, UK Wells R.M.G., Mcintyre R.H., Morgan A.K & Davies P.S (1986) Physiological stress response in big game ¢sh capture: observation in plasma chemistry and blood factors Comparative Biochemistry and Physiology 84, 565^571 Wendelaar Bonga S.E (1997) The stress response in ¢sh Physiological Reviews 77, 591^625 314 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 309^314 Aquaculture Research, 2010, 41, 315^320 doi:10.1111/j.1365-2109.2008.02098.x SHORT COMMUNICATION Application of a microdiet in cobia Rachycentron canadum (Linnaeus, 1766) larvae rearing Bao G Tang1,2,3,4, Gang Chen3 & Zao H Wu4 South China Sea Institute of Oceanology, Chinese Academy of Sciences, Guangdong, China Graduate University of Chinese Academy of Sciences, Beijing, China Key Laboratory of Proliferation and Culture of Aquatic Economic Animals in South China Sea, the Education Department of Guangdong Province, Guangdong, China Guangdong Provincial Key Lab of Pathogenic Biology and Epidemiology for Aquatic Economic Animals, Guangdong, China Correspondence: Z H Wu, Guangdong Provincial Key Lab of Pathogenic Biology and Epidemiology for Aquatic Economic Animals, Zhanjiang, Guangdong 524025, China E-mail: tangbg@gdou.edu.cn Cobia (Rachycentron canadum Kaup), a large, pelagic ¢sh, is primarily distributed throughout warm temperate, subtropical and tropical waters worldwide, except for the eastern Paci¢c, with substantial concentrations located in the South China Sea and the Gulf of Tonkin Because of its prominent growth performance, cobia has been regarded as an excellent mariculture species in coastal areas of south China (Tang, Chen, Shi & Wu 2006) However, the heavy mortality of larvae frustrated its aquaculturists, especially during the yolk-sac stage, while the feeding style of larvae changes dramatically, from endogenous yolk to mixed feeding, and then to exogenous live food For many marine ¢sh larvae, rotifers are the most popular live food for initial feeding (Lubzens, Zmora & Barr 2001; Dou, Masuda, Tanaka & Tsukamoto 2005), but more and more attention has been paid to the microdiet for its availability and storability (Blair, Castell, Neil, D’Abramo, Cahu, Harmon & Ogunmoye 2003; Pousaìo-Ferreira, Santos, Carvalho, Morais & Narciso 2003; Wang, Takeuchi, Hirota, Ishida, Miyakawa & Hayasawa 2004) Although it was commonly considered to be di⁄cult to feed newly hatched marine ¢sh larvae with a compound diet due to nutritional de¢ciency, improper size or attraction absence (Khemis, Noue & Audet 2000; Cahu & Zambonino 2001), a few species have been co-fed and even fed solely dry feed from the onset of ¢rst feeding (FernaŁnr 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd dez-Diaz, Pascual & YuÔfera 1994; Luz & Filho 2001; Takeuchi, Wang, Furuita, Hirota, Ishida & Hayasawa 2003; Curnow, King, Bosmans & Kolkovski 2006) To investigate the applicability of a compound diet and the nature of endogenous nutrient utilization, cobia larvae (4.31 Ỉ 0.0191mm) were fed with live food (rotifer and Artemia nauplii) and a microdiet from day (Table 1), and starved larvae were treated as the control group s Rotifers were enriched with Super Selco (INVE , Belgium) for h at a rate of 0.25 g Super Selco per 106 rotifers, and Artemia nauplii were enriched for s h at a rate of 1.5 g o-yeast (INVE ) per gram of Artemia cyst The microdiet was approximately 150^ 250 mm in diameter, with the proximate composition marked (crude protein 57.5%, crude lipid 12.4%, crude ¢bre 1.6%, ash 13.3% and L -lysine 2.2%) The larvae were assigned to three groups and treated with starvation, feeding with a microdiet and live food, respectively, with each group having three repetitions The larvae were reared in 70 L plastic tanks at a density about 70^80 ind L À with slight aeration Food was supplied four times (at 08:00, 12:00, 16:00 and 20:00 hours) daily on average Debris and dead ¢sh were siphoned from the containers daily About 50% of the tank water was replaced with fresh¢ltered sea water To determine the e¡ect of treatment on survival of larvae, another 900 larvae were reared 315 Application of microdiet in cobia Rachycentron canadum B G Tang et al Aquaculture Research, 2010, 41, 315^320 Table Feeding regime and sources of the diets Diets Period Feeding regime Source Brachionus plicatilis Artemia nauplii Microdiet Days 1–6 Days 7–9 Days 1–9 16 mL À mL À 400 mg Live food laboratory of Guangdong Ocean University San Francisco Bay Brand, Argent, USA Shandong Shengsuo Feed Institute, China 120 Per cent yolk remaining 100 Starved Microdiet Live food 0.5 0.4 80 0.3 0.2 60 0.1 40 20 0 Age(d) Per cent oil globule remaining 120 0.6 3.5 2.5 1.5 0.5 100 80 60 40 Starved Microdiet Live food 20 0 Age(d) Figure Yolk consumption in relation to age and feeding regimes (mean Æ SE, n 30) The inset shows the remaining yolk of larvae during days 3^5 Figure Oil globule consumption in relation to age and feeding regimes (mean Ỉ SE, n 30) The inset shows the remaining oil globule of larvae during days 5^8 in 95 L plastic tanks with the treatment and management consulting 70 L tanks Ten larvae were sampled randomly from each 70 L tank daily (30 larvae of each treatment), and the body lengths (mm), average body weights (mg), yolk and oil globule volumes (mm3) were measured respectively The volumes of yolk and oil globule were measured using the methods of Blaxter and Hempel (1963) and Cetta and Capuzzo (1982) The gut contents of larvae fed with the microdiet and live food were then examined to determine the incidence of feeding Statistics were run with SPSS 11.0 software (issued by Statistical Package for the Social Science), and graphic drawings were performed using EXCEL 2003 software (issued by Microsoft) The signi¢cance of di¡erences among treatments (Po0.05) was tested using univariate GLM (LSD multiple comparison), and the signi¢cance of di¡erence in absorption between the yolk and the oil globule was tested using a paired-samplesT-test Within days post hatching, the yolk sac had almost been depleted, with very little remaining (Fig 1) Among the three treatments, larvae fed with live food had depleted the yolk completely by day 5, and the larvae of the other two groups consumed more slowly (Po0.001) with a little yolk remaining until death, implying that there was a positive rela- tionship between the absorption of endogenous nutrients and exogenous feeding (Wu, Liu, Ma, Xu & Wang 2006), whereas the relationship may be negative for some species and may re£ect an adaptation to the uncertainties associated with feeding (Yin & Blaxter 1987; Mookerji & Rao 1999) On the other hand, the oil globule was absorbed more than 96% in days (Fig 2), but evidently more slowly than the yolk (Po0.05) (Avila & Juario 1987) Because the main component of the oil globule is lipid (Silversand & Norberg 1996), it implies that the utilization of lipid by yolk sac cobia larvae is later than protein Furthermore, the absorption peak of starved larvae appeared on day 4,1 day later than those of the other two groups, and as a whole, the larvae fed with live food absorbed the oil globule more quickly than the others (Po0.01), similar to the trend of yolk consumption The result suggested that oil globule consumption could also be delayed by starvation (Liu, Shi, Chen, Luo, Fu & Luo 2006) Surprisingly, we found that the oil globule reserves of days and for larvae fed with the microdiet were higher than that of day This might be the result of passive elimination of larvae with less oil globule reserve, coinciding with the dramatic mortality on day 6, but not reported in other studies 316 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 315^320 Aquaculture Research, 2010, 41, 315^320 Application of microdiet in cobia Rachycentron canadum B G Tang et al 120 8.5 Microdiet Live food 7.5 Full length (mm) Per Cent Feeding 100 80 60 40 5.5 4.5 0 Age (d) Figure Feeding incidence of larvae fed with a microdiet and live food (mean Ỉ SE, n 3) 60 40 20 Age (d) Starved Microdiet Live food Body weight (mg) 80 3.5 Starved Microdiet Live food 100 Figure Full length of larvae with di¡erent feeding regimes (mean Ỉ SE, n 30) 120 6.5 20 Per cent survival Starved Microdiet Live food 2.5 1.5 0.5 Age (d) Figure Survival of larvae in relation to age and feeding regimes (mean Ỉ SE, n 3) Even before complete yolk exhaustion, the larvae fed with either microdiet or live food exhibited elements of foraging behaviour, including prey search, pursuit and attempts at capturing prey Some larvae started feeding successfully as early as the second day post hatching (Fig 3), with o4% of their initial yolk reserves remaining, and this was regarded as the onset of exogenous feeding by Turner and Rooker (2005) It also indicated that there was a mixed feeding stage of cobia larvae (Chu, Ye, Song, Shi & Chen 2005), which has also been observed in Epinephelus malabaricus, Gadus morhua, Rohu, Singhi and Abramis brama (Liu, Zhou, Yu, Sheng & Ma 1994; Hunt & Boutilier 1996; Mookerji & Rao 1999; Ziliukiene 2005), but not observed in larvae of Sparus aurata and Lutjanus campechanus (Y’ufera, Pascual, Polo & Sarasquete 1993; Williams, Papanikos, Phelps & Shardo 2004) The incidence of feeding was 100% on the third day after hatching for larvae fed with live food; how- 0 Age (d) Figure Average body weight of larvae with di¡erent feeding regimes (n 30) To minimize error, the total weight of 30 larvae from each group was determined ever, its maximum only reached 60% for larvae fed with the microdiet (Fig 3) The dramatic di¡erences in feeding incidence, survival (Fig 4) and growth performance (Figs and 6) showed that live food could not be completely substituted by a microdiet (Wang et al 2004) The previous analysis of gut contents showed that cobia larvae begin to prey on rotifer, protozoan and copepod nauplii as initial feeds within 48 h post hatching, and the sizes of these organisms are usually between 110 and 250 mm (Tang et al 2006), which cover the diameter of the microdiet used in the experiment Therefore, improper size could be ruled out as a major cause for the low feeding incidence and poor survival of larvae fed the microdiet, which might be the result of poor nutritional composition or low utilization of the diet (Khemis et al 2000; Cahu & Zambonino 2001) r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 315^320 317 Application of microdiet in cobia Rachycentron canadum B G Tang et al Aquaculture Research, 2010, 41, 315^320 Table Polyunsaturated fatty acid composition (percentage of total fatty acid) of the three diets (mean Ỉ SE) Fatty acid Microdiet C18:2n-6 C18:3n-3 C18:3n-6 C20:4n-6 C20:5n-3 C22:6n-3 n-3 PUFA n-6 PUFA DHA/EPA n-3/n-6 PUFA 26.46 4.15 0.56 0.84 3.47 9.40 17.02 27.86 2.71 0.61 Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ Æ Æ Artemia Rotifer 0.750 0.070 0.033 0.048 0.015 0.382 0.461 0.728 0.099 0.024 5.98 1.14 0.19 3.39 18.65 29.61 49.40 9.56 1.59 5.17 Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ Æ Æ Æ Æ 0.093 0.009 0.012 0.101 0.773 0.323 1.085 0.019 0.049 0.103 5.26 18.96 1.07 3.34 4.77 0.31 24.04 9.67 0.06 2.49 Ỉ Ỉ Ỉ Ỉ Ỉ Ỉ Æ Æ Æ Æ 0.233 0.040 0.038 0.069 0.092 0.010 0.100 0.207 0.003 0.054 DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; SE, standard error Many studies have shown that poly unsaturated fatty acids (PUFAs) are essential for the survival and normal development of marine ¢sh larvae (Faulk & Holt 2003; Morais, Narciso, Dores & Pousaìo-Ferreira 2004) To compare the nutritional di¡erence between the microdiet and live food, fatty acid analysis was performed on an Agilent 1100 gas chromatograph The result (Table 2) demonstrated that the inert diet had a higher ratio of DHA to EPA than rotifer and Artemia, but a lower ratio of n-3^n-6 PUFAs than the others The present study indicated that the ratio of n-3^n-6 PUFAs is more important than that of DHA to EPA for the survival and growth of cobia larvae However, there is no consensus on the e¡ect of n-3/ n-6 PUFAs on survival and growth of marine ¢sh larvae (Salhi, Izquierdo, Herna’ndez-Cruz, Gonzales & Fernandez- Palacios 1994; Pousaìo-Ferreira, Morais, Dores & Narciso 1999) Since day 2, increases in growth gain had been achieved for larvae fed with live food and a microdiet; however, a constant negative growth of starved larvae occurred on day until their death on day (Figs and 6), and this suggested a selfabsorption, which was also found in Clupea harengus (Yin 1991), indicating the onset of the negative growth period (Farris 1959) But the period is not prevalent in all marine ¢sh larvae, such as Pagrosomus major and Paralichthys olivaceus (Bao, Su & Yin 1998) Heavy mortality of starved larvae occurred on day 4, and the survival decreased to 34.7%; however, the survival of larvae fed with the microdiet was 64.0% then The 29.3% survival di¡erence between the two groups could be attributed to the 50% feeding incidence, which also suggested that the diet could be digested by larvae partially, although it could not support them to live beyond day 318 For marine ¢sh larvae fed with a formulated diet, the sticking points were digestion and assimilation, instead of ingestion (Wang, Wang & Hu 2003) Kolkovski, Tandler and Izquierdo (1997) and Kolkovski (2001) suggested that digestive enzymes, rich in live food, were insu⁄cient in both fry gut and formulated diet and could limit the utilization of the diet Cofeeding live food and a microdiet may solve this problem; better growth and survival and a substantial reduction in the daily supply of live food can be expected (Rosenlund, Stoss & Talbot 1997; PousaìoFerreira et al 2003; Curnow et al 2006) Acknowledgments We thank Professor Jian D Zhang for providing cobia larvae and rearing assistance, and appreciate the help of Dr Kai M Cheng, Dr Erica Vidal, Ms Jeasy, Dr Zheng Z Zhang and Professor Bin H Gu References Avila E.M & Juario J.V (1987) Yolk and oil globule utilization and developmental morphology of the digestive tract epithelium in larval rabbit¢sh, Siganus guttatus (Bloch) Aquaculture 65, 319^331 Bao B.L., Su J.X & Yin M.C (1998) E¡ect of delayed feeding on feeding ability, survival and growth of red sea bream and olive £ounder larvae during early development Journal of Fisheries of China 22, 33^38 (in Chinese, with English abstract) Blair T., Castell J., Neil S., D’Abramo L., Cahu C., Harmon P & Ogunmoye K (2003) Evaluation of microdiets versus live feeds on growth, survival and fatty acid composition of larval haddock (Melanogrammus aegle¢nus) Aquaculture 225, 451^461 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 315^320 Aquaculture Research, 2010, 41, 315^320 Application of microdiet in cobia Rachycentron canadum B G Tang et al Blaxter J.H.S & Hempel G (1963) The in£uence of eggs size on herring larvae (Clupea harengus L.) Journal du Conseil International pour l’Exploration de la Mer 28, 211^240 Cahu C & Zambonino I.J (2001) Substitution of live food by formulated diets in marine ¢sh larvae Aquaculture 200, 161^180 Cetta C.M & Capuzzo J.M (1982) Physiological and biochemical aspects of embryonic and larval development of the winter £ounder Pseudopleuronectes americanus Marine Biology 71, 327^337 Chu Q.Z.,Ye F.L., Song B.L., Shi G & Chen G (2005) Feeding and growth of Rachycentron canadum during early development Journal of Zhanjiang Ocean University 25,8^12 (in Chinese, with English abstract) 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 (in Anglais, with English abstract) Dou S.Z., Masuda R., Tanaka M & Tsukamoto K (2005) Effects of temperature and delayed initial feeding on the survival and growth of Japanese £ounder larvae Journal of Fish Biology 66, 362^377 Farris D.A (1959) A change in the early growth rate of four larval marine ¢shes Limnology and Oceanography 4, 29^36 Faulk C.K & Holt G.J (2003) Lipid nutrition and feeding of cobia Rachycentron canadum larvae Journal of the World Aquaculture Society 34, 368^378 FernaŁndez-Diaz C., Pascual E & YuÔfera M (1994) Feeding behaviour and prey size selection of gilthead seabream, Sparus aurata, larvae fed on inert and live food Marine Biology 118, 323^328 Hunt V.H.I & Boutilier R.G (1996) Activity and metabolism of larval Atlantic cod (Gadus morhua) from Scotian Shelf and Newfoundland source populations: Ocean Production Enhancement Network (OPEN) Marine Biology 124, 607^617 Khemis I.B., Noue J.D.L & Audet C (2000) Feeding larvae of winter £ounder Pseudopleuronectes americanus (Walbaum) with live prey or microencapsulated diet: linear growth and protein, RNA and DNA content Aquaculture Research 31, 377^386 Kolkovski S (2001) Digestive enzymes in ¢sh larvae and juveniles ^ implications and applications to formulated diets Aquaculture 200, 181^201 Kolkovski S.,Tandler A & Izquierdo M.S (1997) E¡ect of live food and dietary digestive enzymes on the e⁄ciency of microdiets for seabass (Dicentrarchus labrax) larvae Aquaculture 148, 313^322 Liu M.H., Shi Z.H., Chen B., Luo H.Z., Fu R.B & Luo H.J (2006) E¡ect of starvation on the growth and development of the larvae of Epinephelus malabaricus Bloch & Schneider in the process of food transformation Marine Fisheries 28, 292^298 (in Chinese, with English abstract) Liu Z.S., Zhou H.T.,Yu K.K., Sheng X.Z & Ma Y.J (1994) Studies on the nutrient requirements and mixed feeding of red seabream Pagrus major Transactions of Oceanology and Limnology 4, 48^55 (in Chinese, with English abstract) Lubzens E., Zmora O & Barr Y (2001) Biotechnology and aquaculture of rotifers Hydrobiologia 446/447, 337^353 Luz R.K & Filho E.Z (2001) Di¡erent diets in the ¢rst feeding of the yellow mandi (Pimelodus maculatus, Lacepede) Acta Scientiarum Universidade Estadual de Maringa 23, 483^ 489 (in Portuguese, with English abstract) Mookerji N & Rao T.R (1999) Rates of yolk utilization and e¡ects of delayed initial feeding in the larvae of the freshwater ¢shes Rohu and Singhi Aquaculture International 7, 45^56 Morais S., Narciso L., Dores E & Pousaìo-Ferreira P (2004) Lipid enrichment for Senegalese sole (Solea senegalensis) larvae: e¡ect on larval growth, survival and fatty acid pro¢le Aquaculture International 12, 281^298 Pousaìo-Ferreira P., Morais S., Dores E & Narciso L (1999) Eggs of gilthead seabream Sparus aurata L as a potential enrichment product of Brachionus sp in the larval rearing of gilthead seabream Sparus aurata L Aquaculture Research 30,751^758 Pousaìo-Ferreira P., Santos P., Carvalho A.P., Morais S & Narciso L (2003) E¡ect of an experimental microdiet on the growth, survival and fatty acid pro¢le of gilthead seabream (Sparus aurata L.) larvae Aquaculture International 11, 491^504 Rosenlund G., Stoss J & Talbot C (1997) Co-feeding marine ¢sh larvae with inert and live diets Aquaculture 155, 183^191 Salhi M., Izquierdo M.S., Herna’ndez-Cruz C.M., Gonzales M & Fernandez-Palacios H (1994) E¡ect of lipid and n-3 HUFA levels in microdiets on growth, survival and fatty acid composition of larval gilthead seabream (Sparus aurata) Aquaculture 124, 275^282 Silversand C & Norberg B (1996) Fatty-acid composition of ovulated eggs from wild and cultured turbot (Scophthalmus maximus) in relation to yolk and oil globule lipids Marine Biology 125, 269^278 Takeuchi T., Wang Q.R., Furuita H., Hirota T., Ishida S & Hayasawa H (2003) Development of microparticle diets for Japanese £ounder Paralichthys olivaceus larvae Fisheries Science 69, 547^554 Tang B.G., Chen G., Shi G & Wu Z.H (2006) Alimentary canal contents of cobia (Rachycentron canadum) larvae cultured in ponds Journal of Zhanjiang Ocean University 26, 12^16 (in Chinese, with English abstract) Turner J.P & Rooker J.R (2005) E¡ect of dietary fatty acids on the body tissues of larval and juvenile cobia and their prey Journal of Experimental Marine Biology and Ecology 322,13^27 Wang Q.R.,Takeuchi T., Hirota T., Ishida S., Miyakawa H & Hayasawa H (2004) Application of microparticle diets for Japanese £ounder Paralichthys olivaceus larvae Fisheries Science 70, 611^619 r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 315^320 319 Application of microdiet in cobia Rachycentron canadum B G Tang et al Wang S.F., Wang A.L & Hu J.R (2003) Progress of study on substitution of live food by formulated diets in marine ¢sh larvae Marine Science Bulletin 22, 89^96 (in Chinese, with English abstract) Williams K., Papanikos N., Phelps R.P & Shardo J.D (2004) Development, growth, and yolk utilization of hatcheryreared red snapper Lutjanus campechanus larvae Marine Ecology 275, 231^239 WuY.Y., Liu X.Z., Ma A.J., XuY.J & Wang Q.Y (2006) E¡ects of starvation on the growth and development of Cynoglossus semilaevis Gˇnther larvae Marine Fisheries Research 27, 87^93 (in Chinese, with English abstract) Yin M.C (1991) Feeding ability and growth of the yolk-sac larvae of north sea herring Oceanologia et Limnologia Sinica 22, 554^560 (in Chinese, with English abstract) 320 Aquaculture Research, 2010, 41, 315^320 Yin M.C & Blaxter J.H.S (1987) Feeding ability and survival during starvation of marine ¢sh larvae reared in the laboratory Journal of Experimental Marine Biology and Ecology 105,73^83 Y’ufera M., Pascual E., Polo A & Sarasquete M.C (1993) Effect of starvation on the feeding ability of gilthead seabream (Sparus aurata L.) larvae at ¢rst feeding Journal of Experimental Marine Biology and Ecology 169, 259^272 Ziliukiene V (2005) The diet of Abramis brama (L.) larvae reared in illuminated cages Journal of Applied Ichthyology 21, 406^409 Keywords: cobia, yolk, oil globule, microdiet, PUFA r 2009 The Authors Journal Compilation r 2009 Blackwell Publishing Ltd, Aquaculture Research, 41, 315^320