ICAR-Central Avian Research Institute, Izatnagar-243 122, India
*Corresponding Author Email: latainbsp78@gmail.com
Online published on 12 August, 2020.
Three chicken population namely Aseel Peela (AP), Kadaknath (KN) and CARI-Red (CR) were utilized for making full 3 × 3 diallel cross experiment which, resulted into nine genetic groups. Body weights of all the genetic groups were recorded biweekly from 0 to 20 weeks of age. The growth rates of resulted genetic groups were compared with nonlinear optimization technique by using four models viz., Monomolecular, Logistic, Modified logistic and Gompertz. The best fitting of model to growth data was judged by four measures of goodness of fit i.e. Adjusted R2, Mean absolute error (MAE), Mean squares error (MSE) and Akaike information criterion (AIC). The randomness of the error was tested by Run-test (Z) and Durbin-Watson test (d) and used to test the auto- correlation among errors at different time points. It was observed that the Gompertz model was the best fitted model to describe the growth pattern of male and female of all genetic groups under study. From the observed growth curve of male and female of their respective genetic groups it was also observed that at brooding age of birds, both male and female were grows almost at the same rate, thereafter male grow faster than female. The growth curve corresponding to cross AP × CR (male and female) lies above than all other growth curve therefore for better body growth cross of AP × CR would be consider best among AP × AP, KN × AP, CR × AP and AP × KN. Similarly male and female of cross CR x KN were consider best among KN × AP, AP × KN, KN × KN and KN × CR while, cross AP × CR consider best among CR × AP, CR × KN, KN × CR and CR × CR.
Chicken, Diallel cross, Body weight, Growth models, Growth curve