Abstract: The phenotypic correlations，path coefficients， direct and indirect correlations between the body weight and several body sizes of Gansu alpine fine-wool sheep were analyzed by SPSS 190 software. Then the optimum regression equation between the body weight and body sizes was established. The results indicated that the body height （X1）， body length （X2）， chest circumference （X3） and chest depth （X4） showed significant correlations with the body weight at the 001 level. The phenotypic correlations were 0.766， 0.837， 0.834， 0.801， respectively. The body length （X2） and chest circumference （X3） had significant influences on the body weight directly and indirectly. However， the body height （X1） and chest depth （X4） showed effects on the body weight indirectly by assisting other characters. The optimal regression model of the body sizes （Xi） and body weight （Y） was established by the stepwise regression as Y=-43.851+0425X2+0.587X3. （R2=0.836， P<0.01）.

Key words: Gansu alpine fine-wool sheep, body weight, body size, correlation analysis, path analysis, stepwise analysis