Effects of different sources of dietary fibre on the length-weight relationship of lemon fin barb hybrid (Barbonymus gonionotus ♀ (Bleeker, 1850) ×Hypsibarbus wetmorei ♂ (H.M. Smith, 1931)) fingerlings

• Main Authors: Jimoh Wasiu A., Kamarudin Mohd S., Dauda Akeem B., Nwachi Oster F.
• Format: Article
• Language: English
• Published: University of Belgrade - Faculty of Agriculture, Belgrade 2020-01-01
• Series: Journal of Agricultural Sciences (Belgrade)
• Subjects: ancova; dietary fibre; regression model; lemon fin barb hybrid
• Online Access: https://scindeks-clanci.ceon.rs/data/pdf/1450-8109/2020/1450-81092004391J.pdf
• ISSN: 1450-8109; 2406-0968

The effect of different sources of dietary fibre on the length-weight relationship of lemon fin barb hybrid (Barbonymus gonionotus ♀ × Hypsibarbus wetmorei ♂) was examined in a 56-day study using five dietary treatments; control and test dietary treatments containing different leaf meals of high dietary fibre added at the 10% inclusion level to a basal diet containing fish meal, soybean meal and rice bran. The weight of fish in each treatment was measured using a sensitive weighing balance while the total length was measured with a 15-cm-long ruler following standard procedures. The logarithmic transformation of the values obtained from the weight and length measurements was depicted using scatter diagrams. SPSS version 17.0 was used in regressing length against the weight to obtain the various components of the regression equations. The Levene test of homogeneity and interaction between covariate LogL and independent variable (treatments) was not significant - F(4.85)=0.838 and F(4.84)=1.345 respectively indicating that our data have not violated the assumption of homogeneity of variance. The various regression models developed for fish across various dietary treatments were statistically significant (p<0.01). The t-values for components of the regression equation such as intercept/constant (a) and slopes were all statistically significant except the t-value for the intercept/constant (a) of control. There was no significant effect of dietary treatments on LogW after controlling for logL, F(4.84)=1.296, revealing that the regression lines were not significantly different from each other. This shows that the regression models could be used interchangeably.