| 摘要: |
| 为了解翘嘴鳜(Siniperca chuatsi)主要形态性状与体重的相关性，以及雌雄形态性状的差异，本研究测量了同塘养殖的2837尾翘嘴鳜的7个生长相关性状，包括全长、体高、头长、眼径、尾柄全长和尾柄高与体重。采用相关分析、通径分析和多元线性回归分析方法，分析各形态性状与体重的相关关系，并从中随机取雌雄个体各150尾建立多元回归方程，并另随机取120尾进行雌雄判别验证。结果显示，在各形态性状中，全长、体高与体重的相关系数最大。通过建立回归方程组并进行偏回归系数检验，发现头长和尾柄高与体重的相关性不显著，故剔除这2个性状后作进一步分析。通径分析显示，全长和体高对体重的直接作用最大，且大于间接作用；单性状及两两性状协同对体重的决定系数显示，全长及体高对体重的单独决定程度最高，且二者的协同作用也最大；4个性状对体重的决定系数总和(∑d)达0.896，说明这4个性状与体重有较大的相关性，拟合的多元回归方程具有较大的参考意义。建立雌雄鱼判别回归方程并分析验证显示，对于体重小于200 g的个体其雌雄性别判别的准确率较高。本研究可为翘嘴鳜生长性状选择评价指标的确定提供参考，将有助于提高选育种效率。 |
| 关键词: 翘嘴鳜 形态性状 体重 相关关系 多元回归 通径分析 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Correlation Analysis of the Main Morphological Traits and Body Weight of Mandarin Fish (Siniperca chuatsi) and Morphological Traits Between Males and Females |
|
DONG Junjian1, SUN Chengfei2, TIAN Yuanyuan3,4, ZENG Qingkai3,4, SHI Hongya3,4, LU Maixin3,4, YE Xing3,4
|
|
1.Key Laboratory of Tropical & Subtropical Fishery Resources Utilization & Cultivation, Ministry of Agriculture,Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences, Guangzhou 510380;2.Key Laboratory of Tropical & Subtropical Fishery Resources Utilization & Cultivation, Ministry of Agriculture,Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences, Guangzhou 510381;3.Key Laboratory of Tropical & Subtropical Fishery Resources Utilization & Cultivation, Ministry of Agriculture,Pearl River Fisheries Research Institute, Chinese Academy of Fishery Sciences, Guangzhou 510382;4.College of Fisheries and Life Science, Shanghai Ocean University, Shanghai 201306
|
| Abstract: |
| To assess the correlation between the main morphological traits and body weight of Siniperca chuatsi as well as the morphological differences between males and females, 2837 S. chuatsi individuals from the same farm pond were used to measure seven growth-related morphological traits including total length, body height, head length, eye diameter, caudal peduncle length, caudal peduncle depth, and body weight. Correlation analysis, path analysis, and multivariate linear regression analysis were used to analyze the correlation between morphological traits and body weight. A total of 150 males and 150 females were randomly selected to establish multiple regression equations, and 120 individuals were used for gender verification. The results showed the correlation coefficients of body weight to total length and body height are the highest among all the morphological traits. The established regression equations and partial regression coefficient tests indicated no significant correlation of head length or caudal peduncle depth to body weight, which were excluded for further analysis. Path analysis revealed the largest direct effect of total length and body height upon body weight, which was greater than their indirect effect. The highest determinant coefficients of single and pairwise traits to body weight were total length and body height in both single trait determination and pairwise determination. The determination coefficient (∑d) of the four traits was 0.896, suggesting a greater correlation for the four traits to body weight and a higher reference value for the fitted multiple regression equation. A regression equation for the gender was established. Verification results indicated that higher verification accuracy of gender was obtained for individuals with body weight less than 200 g. This study provides an essential reference for determining the evaluation indices for the morphological growth traits of S. chuatsi, and would facilitate the breeding selection. |
| Key words: Siniperca chuatsi Morphological traits Body weight Correlation coefficient Multiple regression Path analysis |
