不同地理种群野生大黄鱼部分生长相关性状间的通径分析

Path analysis of partial growth-related traits for wild large yellow croaker (Larimichthys crocea) of different geographical populations

  • 摘要: 大黄鱼 (Larimichthys crocea)不同地理种群的多样性体现在形态性状上,依据不同地理种群选育出的大黄鱼的形态特征也略有不同。为给养殖大黄鱼的品种选育和人工增殖放流过程中的苗种选择提供表型性状参考依据,分别从广东湛江、福建厦门和浙江舟山3个地点采集野生大黄鱼,测定其全长、体长、体高和尾柄高,并对表型性状进行相关性分析、通径系数分析和逐步回归分析。结果表明,3个地理种群的大黄鱼体质量间的变异系数介于17.44%~19.79%;不同种群间各表型性状的相关系数也均达到极显著水平 (P<0.01);对湛江、厦门和舟山种群起主要作用的表型性状分别有2个 (体长、体高)、 3个 (全长、体长和体高) 和2个 (全长、体长);3个种群对体质量有重要影响的个体表型特征复相关系数 (r) 分别为0.962、0.915和0.964。经多元回归方程的建立,去除与体质量相关性不显著的表型性状,得到的回归方程中的形态指标和通径分析结果相同。

     

    Abstract: Morphological traits reflect the diversity of different of large yellow croaker (Larimichthys crocea) of different geographic populations, so the morphological traits of L. crocea selected on the basis of different geographic populations are also slightly different. In order to provide references of phenotypic traits for the breeding of L. crocea and the selection of seedlings during artificial proliferation and release, we collected wild L. crocea from three locations (Zhanjiang, Guangdong Province; Xiamen, Fujian Province; Zhoushan, Zhejiang Province), measured their total length, body length, body height as well as caudal stalk height, as well as conducted correlation analysis, path analysis and multiple stepwise regression analysis. The coefficient of variation in body mass ranged from 17.44% to 19.79% among the three L. crocea populations. There was an extremely significant correlation among all the phenotypic traits (P<0.01). The numbers of phenotypic traits which played a major role in the populations of Zhanjiang, Xiamen and Zhoushan were 2 (Body length, body height), 3 (Total length, body length, body height), and 2 (Total length, body length), respectively. The multiple correlation coefficients (R) of phenotypic traits with significant effects on body mass for Zhanjiang, Xiamen and Zhoushan populations were 0.962, 0.915 and 0.964, respectively. After eliminating morphological traits with insignificant coefficients to body mass, we established the multiple regression equation by multiple regression analysis. The morphological traits retained by multiple regression analysis were consistent with those identified by path analysis.

     

/

返回文章
返回