PREDICTION OF LIVE WEIGHT BY MORPHOMETRIC MEASUREMENTS IN FEMALES AND MALES OF CRIOLLO SHEEP OF THE OAXACAN MIXTECA, MEXICO

Abstract

<p><strong>Background</strong>. The live weight (LW) and morphometric variables of sheep are useful characteristics to generate adequate criteria for genetic improvement and conservation of Creole animal resources. <strong>Objective</strong>. To establish morphological relationships and through them predict LW of females and males in Creole sheep of the Oaxacan Mixteca, Mexico. <strong>Methodology</strong>. LW, thoracic perimeter (TP), height at withers (HW), back length (BL), cane perimeter (CP), head length (HL), neck length (NL), abdominal perimeter (AP) and rump width (RW) were recorded in 720 Creole sheep in 24 production units, measured in animals of both sexes. First, they were analyzed using general linear model procedures to determine the effect of sex on the variables of interest, and then, they were analyzed using discriminant analysis. Simple correlations between morphometric variables were calculated and step-by-step multiple linear regression equations were generated to predict LW. <strong>Results.</strong> Differences were found between sexes for LW and morphometric measures, which were correlated (<em>p</em> ≤ 0.05). The variables TP, AP and BL had the highest correlations (0.92, 0.90 and 0.89, respectively) and the variables CP, NL and RW, the lowest (0.75, 0.82 and 0.83, respectively). The statistical models explained a large proportion of the total variability (R<sup>2</sup> from 85 to 91 %). According to sex, males had R<sup>2</sup> (0.90 to 0.94) higher than females (0.83 to 0.89). In both cases TP and HW were the most important characteristics for predicting LW. <strong>Conclusions</strong>. Under the conditions of this work, the model that included TP and HW was the one that best predicted the LW of the Creole sheep of the Oaxacan Mixteca. It is recommended to use different equations for each sex and training of sheep producers in the use of those equations.</p>