引用本文:	马爱军,王新安,孙志宾,赵艳飞,孙建华,王广宁,孟雪松,刘圣聪,张涛.红鳍东方鲀(Takifugu rubripes)三个不同群体的形态差异分析.海洋与湖沼,2016,47(1):166-172.

【打印本页】【HTML】【下载PDF全文】【查看/发表评论】【下载PDF阅读器】【关闭】

过刊浏览高级检索

本文已被：浏览 1575次下载 1841次	码上扫一扫！
分享到：微信更多字体:加大+\|默认\|缩小-
*红鳍东方鲀(Takifugu rubripes)三个不同群体的形态差异分析*
马爱军,王新安,岳亮,赵艳飞,孟雪松,刘圣聪,张涛
中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室青岛;中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室青岛;中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室青岛,中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室,中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室,中国水产科学研究院黄海水产研究所农业部海洋渔业可持续发展重点实验室青岛市海水鱼类种子工程与生物技术重点实验室,大连天正实业有限公司,大连天正实业有限公司,大连天正实业有限公司

摘要:

对3个不同群体红鳍东方鲀(Takifugu rubripes)进行16项形态性状的测定,采用3种多元统计分析方法和方差分析方法,比较了3个群体的外部形态特征。聚类分析结果表明,3个红鳍东方鲀群体中,日本群体(J)和养殖群体(R)的差异最大,野生群体(Y)和养殖群体(R)的差异最小。主成分分析构建了3个反映形态特征信息的综合性指标——主成分1、主成分2和主成分3,三者的贡献率分别为54.961%、12.129%和6.459%,三个主成分的累积贡献率为73.549%,3群体之间存在明显的偏离,形成三个不同的类群,其中,养殖群体(R)和日本群体(J)间的形态差异最大,野生群体(Y)和养殖群体(R)的形态差异最小。判别分析表明,3个红鳍东方鲀群体的形态差异基本上达到差异显著(P<0.05)或极显著水平(P<0.01),利用贡献率最大的6个参数建立了3个群体的判别函数,其判别准确率分别为97.69%-100%(P₁)、97.39%-100%(P₂),综合判别率为98.62%,可以认为逐步判别法对于红鳍东方鲀不同群体的初步鉴定是可行的。方差分析多重比较结果表明,3个群体在部分形态特征上表现出明显差异,其中,养殖群体(R)和日本群体(J)之间形态特征差异显著指标最多,养殖群体(R)和野生群体(Y)之间形态特征差异显著指标最少。显然,聚类分析、主成分分析、判别分析和方差分析的结论基本上是类似的,它们从不同角度反映了群体间的形态学差异。

关键词: 红鳍东方鲀形态差异聚类分析判别分析主成分分析

DOI：10.11693/hyhz20150300083

分类号:

基金项目:中央级公益性科研院所基本科研业务费专项资金资助,20603022012005号;大连金州新区科技计划项目资助,2012-B1-012号。

附件

MORPHOLOGICAL VARIATIONS AMONG THREE DIFFERENT POPULATIONS OF TAKIFUGU RUBRIPES

MaAijun,Wangxinan,Yueliang,Zhaoyanfei,MENG Xue-song,Liu Shengcong and ZhangTao

Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences; Key Laboratory of Sustainable Development of Marine Fisheries, Ministry of Agriculture; Qingdao Key Laboratory for Marine Fish Breeding and Biotechnology,Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences; Key Laboratory of Sustainable Development of Marine Fisheries, Ministry of Agriculture; Qingdao Key Laboratory for Marine Fish Breeding and Biotechnology,,Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences; Key Laboratory of Sustainable Development of Marine Fisheries, Ministry of Agriculture; Qingdao Key Laboratory for Marine Fish Breeding and Biotechnology,,Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences; Key Laboratory of Sustainable Development of Marine Fisheries, Ministry of Agriculture; Qingdao Key Laboratory for Marine Fish Breeding and Biotechnology,,Dalian Tianzheng Industrial Co. Ltd.,Dalian Tianzheng Industrial Co. Ltd.,Dalian Tianzheng Industrial Co. Ltd.

Abstract:

Morphological comparison using16 morphological indices among three populations of Takifugu rubripes from group R (farmed in Dalian, NE China), J (Japanese) and Y (wild one from the Yellow Sea) was conducted statistically in hierarchical cluster analysis, principal component analysis, discriminatory analysis, and one-way ANOVA analysis. The hierarchical cluster analysis showed that the morphological difference between J and R was small, while that between Y and R was the biggest. Principal component analysis revealed three principal components, contributing 54.961%, 12.129%, and 6.459%, and reaching cumulatively 73.549%. Therefore, the morphological variations among the three populations were significant (P<0.05), and come mainly from the first principal component. The results of discriminant analysis revealed also a significant (P<0.05) or very significant (P<0.01) difference among the three groups. Three discriminant functions were obtained, the discrimination accuracy being 97.69%-100% (P₁) and 97.39%-100% (P₂), the integrated discrimination accuracy being 98.62%, which shows that it is feasible to adopt stepwise discriminatory analysis to distinguish the populations. The results of One-way ANOVA analysis and multiple comparison are in accordance with the conclusion of other methods, and they reflect morphological variations among the populations from different points of view.

Key words: Takifugu rubripes morphological difference cluster analysis discriminant analysis