太湖秀丽白虾形态性状对体重影响的通径分析

随机选取了521尾太湖秀丽白虾，分别测定腹部长（X1）、腹部宽（X2）、体长（X3）、头胸甲长（X4）、头胸甲宽（X5）、尾扇长（X6）、额剑长（X7）、尾扇宽（X8）和体重（Y）9个指标。采用相关分析和通径分析的方法，计算了秀丽白虾以形态性状为自变量对体重(y)的相关系数、通径系数和决策系数，进而对各性状对体重影响进行分析。结果表明，太湖秀丽白虾各形态性状与体重的相关性均达到极显著(P<0.01)水平。秀丽白虾的腹部长（X1）、体长（X3）和额剑长（X7）的通径系数达到了极显著的水平(P<0．01)；除体长（X3）外，各形态性状对体重（Y）的间接影响均大于直接影响。体长（X3）对体重（Y）的直接影响最大，其次为腹部长（X1）和额剑长（X7），其它形态性状对体重（Y）的直接影响很小，主要通过体长（X3）和腹部长（X1）间接影响体重； 决策系数分析结果表明，体长（X3）、腹部长（X1）和额剑长（X7）的决策程度最大。决策系数分析结果与通径分析的结果一致。剔除与体重多元相关不显著的形态性状，建立了腹部长（X1）、体长（X3）、额剑长（X7）与体重（Y）的多元回归方程Y=-2.768+0.046X1+0.066X3+0.025X7。

Path analysis of effects of morphometric traits on body weight of Exopalaemon modestus in Taihu Lake

521 white shrimps （Exopalaemon modestus） in Taihu Lake were randomly sampled and nine morphometric traits including abdominal length（X 1）,abdominal width（X 2）,body length（X 3）,carapace length（X 4）,carapace width（X 5）,tail length（X 6）,length of upper frontal eminence（X 7）,tail width（X 8） and body weight （Y） were measured.The morphometric traits were used as independent variables,and body weight （Y）was used as dependent variable.The correlation coefficients,path coefficients and decision coefficients were calculated by correlation analysis and path analysis,then effects of morphometric traits on body weight was analyzed.The results showed that path coefficients of abdominal length,body length and length of upper frontal eminence achieved significant （P0.01） levels.Indirect effects of morphometric traits on body weight（Y） were higher than that of direct effect except body length.Body length,abdominal length and length of upper frontal eminencehad the top three direct effects on body weight（Y） in turn.Other morphometric traits had a little effect on body weight and they mainly affected body weight indirectly through body length and abdominal length.The result of decision coefficients revealed that body length,abdominal length and length of upper frontal eminence had the most effect on body weight（Y）,which was consistent with the results of path analysis.The morphometric traits which did not reach the significance level （P0.05） were eliminated and others were used to establish the multiple regression equation,which was Y=-2.768＋0.046X 1 ＋0.066X 3 ＋0.025X 7.