Artificial feed plays an important role in semi-intensive fish farming, where density of fish is high. Fish feed development is the least developed sector in aquaculture, particularly in third world countries. The formulation of nutritionally balanced and acceptable diet for fish feed is possible only if information regarding its digestibility is available. It is well known that these two parameters, nutritionally balanced diet and its digestibility, play a key role for the development of artificial feed for fish. Determination of nutrient digestibility is the first step in evaluating the potential of an ingredient for use in the diet of reared species. Information on digestibility coefficients of feed ingredients is very useful not only to enable formulation of diets that maximize fish growth by providing appropriate amounts of available nutrients but also to reduce fish wastes.

The present study was therefore, designed to find out the digestibility of nutritionally balanced diet that contains cheap and easily available ingredients. The experimental and reference diets were evaluated for comparative growth and diet utilization efficiency in three Indian major carps (Catla catla, Labeo rohita and Cirrhinus rnrigala). The highest weight gain was observed in the Cirrhinus miri gala (20.57 ± 2.8 g & 0.7 ± 0.4 cm) followed by Labeo rohita (19.17 ± 3.02g & 1.1 ± 0.4 cm) and Catla catla (19.23 ± 3.Og & 0.4 ± 0.3cm), with the experimental or test diet (TD). The lowest growth was also observed in the Catla catla (14.6 ± 3.8 g & 0.2 ± 0.1cm) with reference diet (RD). Growth, FCR and digestibility of nutrients (Protein, EE, DE) were assessed. Proximate analysis (i.e., crude protein, gross energy, dry matter and marker estimation) of feed ingredients and formulated diets was also done. The highest protein digestibility was observed in Labeo rohita (83.4±3.5%) followed by Cirrhinus mn gala (82.2 ± 4.2) and Catla catla (81.8± 4.6%) The over all digestibility's of nutrients was higher for experimental diet in three fish species i.e. crude protein (84.4 ± 2.1), dry matter (33.3 ± 8.4), crude fat (82.1± 3.1) and gross energy (61.0 ± 5.1). Nutrient digestibility of reference diet were lower (crude protein 77.2 ± 1.3, dry matter 53.0 ± 3.0, crude fat 78.7 ± 0.58 and gross energy 58.3 ± 4.5) as compared to the experimental diet.

Similarly the FCR and FCE values for experimental diet (3.1 ± 0.12 and 32.4 ± 0.69) were better than the reference diet (3.8 ± 0.06 and 25.6 ± 0.58). Among three fish species the Cirrlzinus miri gala has better FCR and FCE (3.4 ± 0.57 and 29.6 ± 5.1) followed by Cat/a catla (3.5 ± 0.42 and 29 ±.0 4.2) and Labeo rohita (3.5 ± 0.49 and 28.5 ± 4.95), respectively. FCR and FCE ratios for reference diet was the lowest in the Labeo rohita (3.9 and 25.0), whereas Cirrhinus mrigala and Cat/a cat/a showed similar ratios (3.8 0 and 26.0), respectively.

Key physico-chemical parameters viz, temperature, dissolved oxygen (DO), pH, conductivity, total dissolved solids, were regularly monitored during the study period. All the parameters showed positive significant correlation with each other except DO which showed negative significant correlation. The correlation of physico-chemical parameters with growth was negatively significant whereas positive correlation was observed between DO and growths indicating that DO have a significant effect on the growth.

The over all results showed that the experimental diet (40% protein level) has better growth and nutrient digestibility as compared to the reference diet (24.5% protein level) which resulted poorer growth and digestibility values.