Abstract: About 5000 individual weights issued from periodical survey of 722 indigenous kids in 8 herds and during 4 years were analyzed in the present study. The aim is to identify the non genetic effects upon the growth curve of local goat population raised under arid conditions as a necessary step to improve kids` meat production on range lands. The Gompertz model was adopted to assess individual kids` growth curves. A non linear regression iterative procedure allows estimating 688 kids curves parameters and convergence criteria was 10-8. For each kid, curve parameters, asymptotic weight, inflexion age and weight, were considered as quantitative traits. A GLM procedure was applied to decompose the total variance of the kid`s traits. A means comparison test (SNK, α = 0.05) was applied to identify homogenous class by factor. Results show that the GLM determination coefficient remains lower than 67% of all studied traits due to the observed data structure. All traits seem to be affected by the significant effects (p< 0.001 or 0.05) of the factors related to the restrictions and the irregularity of the technical and natural environment of pastoral husbandry. Specially, growth traits varied with year, herd, month of birth and natural region. Herd mobility contributes to stamp climatic stresses and resources irregularities. Equally, the year factor plays an important role upon the kids weight and it evolution till 6 months age. Kids growth curve loses its S-shape and becomes linear under harsh conditions of dry years. Also, the non-genetic factors impact increases with kids age and requirements. The sex, birth mode and mother age acts only during first 2 months age. So, arid environment affects both quantitatively and qualitatively individual kid`s growing behaviour and have to be considered for local goat rational genetic improvement modeling and planning.