SHORT COMMUNICATION Influence of dietary protein levels on growth and egg quality in broodstock female bagrid catfish (Mystus nemurus Cuv. & Val.)

Bagrid cat¢sh, Mystus nemurus (Cuv. & Val.) is a popular cat¢sh currently being produced in both semiintensive and intensive culture systems in Malaysia fetching a higher price than tilapia (Oreochromis sp.) and clariid cat¢sh (Clarias gariepinus). Although the technique for induced breeding of M. nemurus has been developed, the problems of low egg hatchability and larval survival rate remain a major hindrance to further increase the production of this species. Furthermore, there is no information on the reproductive biology of this species in natural habitats. Egg quality is a useful indicator to the viability and subsequent quality of newly hatched larvae (Brooks, Tyler & Sumpter 1997).Various reports have emphasized the importance of broodstock nutrition to enhance reproductive performance of cultured species (Cerda, Carillo, Zanuy, Ramos & Higuera 1994; Coward & Bromage 2000). Among dietary nutrients, protein has been reported to in£uence reproductive parameters in both marine and freshwater broodstock ¢sh (Watanabe, Itoh, Satoh, Katajima & Fujita 1985; Eskelinen 1989; El-Sayed, Mansour & Ezzat 2003; Chong, Ishak, Osman & Hashim 2004). These studies mainly showed that dietary protein levels in£uenced parameters such as weight gain and proximate composition of brood ¢sh, quantity and quality of eggs and larval viability. As diierences exist in nutrient requirements between grow-out and brood ¢sh, its desirable to ensure that nutritional requirements of broodstock be ful¢lled to optimize reproductive performance (Izquierdo, Fernandez-Palacios & Tacon 2001). As a prerequisite to develop a cost-eiective diet that ful¢lls the requirements of M. nemurus broodstock, we report here ¢ndings froma studyconducted to determine the eiects of diierent dietary protein levels on growth, body composition and egg quality in female broodstock. Three semi-puri¢ed isoenergetic diets (395 kJg ) containing 30%, 35% and 40% crude protein were formulated using casein, gelatin and ¢shmeal as protein sources (Table 1). Ingredients were mixed in a Hobart mixer with the resulting dough being extruded through a pelleting machine to produce 3^4mmdiameter pellets. Proximate analysis of diets, ¢sh carcass and eggs was conducted according to AOAC (1990) for veri¢cation of nutrient levels. Twelve-month-old M. nemurus females (average weight 620.3 3.7 g) were collected from a local farm and evenly distributed in nine outdoor £owthrough concrete raceway tanks (2 1.2 1m and water level 0.90m) at a density of 4 ¢sh tank with three replications per treatment. Feeding was done once a day at 3% body weight day 1 at 18:00 hours. This schedule is based on our unpublished observation on feeding behaviour of this species in our laboratory conditions (Z. A. Muchlisin, unpubl. obs.). The feeding experiment lasted 16 weeks. At the end of the feeding trial, all females were sacri¢ced for further analyses. Eggs were removed from the ovary and mixed with working buier solution comprising of 60mL ethanol, 30mL formaldehyde and 10mL glacial acetic acid. Fecundity, percentage of ripe eggs, mean egg diameter (mm), and mean dry weight of egg (mg) were measured. Proximate analysis of eggs, ¢sh carcass (without eggs) and diets was conducted according toAOAC (1990). All data were subjected to Aquaculture Research, 2006, 37, 416^418 doi:10.1111/j.1365-2109.2005.01382.x