黄海、东海近海七个棘头梅童鱼地理群体的形态差异分析

对2013年采自连云港、大丰、崇明、舟山、宁德、厦门和温州的7个棘头梅童鱼(Collichthys lucidus)群体共238个个体进行形态学比较研究,运用单因素方差分析、主成分分析、聚类分析和判别分析等多元统计分析方法对13项可量性状以及20项框架参数进行形态差异分析。单因素方差分析结果显示,7个地理群体之间确实存在显著性差异;主成分分析得到的4个主成分对不同群体间总差异的贡献率分别为53.131%、9.328%、4.249%和3.406%,累积贡献率为70.113%,躯干部最能用以区分不同的地理种群;主成分分析和聚类分析结果基本一致,认为7个棘头梅童鱼地理群体在形态上不存在明显的南北分化现象;通过判别分析得到14个比例变量并构建判别函数,将7个群体的数据代入函数,得到群体的综合判别率为85.3%,判别准确率P1为62.5%~96.7%(群体判别正确的尾数/群体的实际尾数),P2为78.1%~90.6%(判别正确的尾数/判入该群体的尾数),舟山群体由于其特殊的地理因素,判别准确率较低(P1=62.5%,P2=78.1%)。本研究为棘头梅童鱼地理种群判别、种质资源保护以及良种选育等提供了基础的形态学数据。

Morphology of seven Collichthys lucidus populations near the coast of the Yellow Sea and East China Sea

Collichthys lucidus, which occurs in the coastal waters of China, including the Yellow, East China, and South China Seas, is a favorite seafood of coastal residents and has great economic value. However, because C. lucidus resources have been rapidly reduced by overfishing and artificial breeding technology is immature, breeding high-quality fish fry and germplasm conservation and utilization are critical. The aim of the present study was to analyze and compare the morphology of 238 C. lucidus individuals that were collected from seven locations, including Lian Yungang, Dafeng, Chongming, Zhoushan, Ningde, Xiamen, and Wenzhou. More specifically, one-way ANOVA, principle components analysis, cluster analysis, and discriminant analysis were used to investigate whether the seven geographic populations differed in regards to 13 morphological traits and 20 truss network characters. Because individual size variation could have affected the results, all data were calibrated using body length (BL). One-way ANOVA revealed significant differences between the seven C. lucidus geographical populations, and the four principal components identified using principal component analysis accounted for 53.131%, 9.328%, 4.249%, and 3.406%, respectively, of the variation among populations and 70% in total. The first principal component reflected trunk characteristics-body depth/BL, D6 (origin of pelvic fin to the front edge of head occipital spine)/BL, D7 (the front edge of head occipital spine to origin of anal fin)/BL, D8 (origin of pelvic fin to origin of dorsal fin)/BL, D9 (the front edge of head occipital spine to end point of anal fin)/BL, D11 (origin of anal fin to origin of dorsal fin)/BL, and D14 (origin of dorsal fin to end point of dorsal fin)/BL, which indicates that trunk characteristics can be used as a basis for judging population differences. Both the cluster and principle components analyses showed that there was no obvious north-south differentiation in the species'' morphology. Many studies have reported opposite patterns, possibly owing to different sampling times, locations, quantities, and analyses. Discrimination analysis indicated that the comprehensive discrimination rate of the seven C. lucidus populations was 85.3% and that the discriminant accuracy ratio was 62.5%-96.7% and 78.1%-90.6% for P1 (Correct discriminant number of one group/the actual number of the group) and P2 (Correct discriminant number of one group/discriminant number entering the group) respectively. However, the discrimination accuracy of the Zhoushan population was low, owing to its unique geographical factors, including ocean current and freshwater flow. The present study provides basic C. lucidus morphological data that can be used for identifying C. lucidus populations, protecting germplasm resources, and selecting fine varieties.