摘要
为了研究半刺厚唇鱼雌雄个体的形态性状特征,运用主成分分析、R-聚类分析、逐步判别分析和T检验等方法,对60尾半刺厚唇鱼17个直接测量性状和14个标准化性状进行分析。主成分分析结果显示:提取的3个主成分的贡献率分别为70.056%、8.284%、4.674%,3个主成分累积贡献率达83.014%;通过逐步判别分析法,从60尾半刺厚唇鱼的14个标准化性状中筛选出4个性状,建立性别判别方程:(雄)F 1=917.211X_(1)+78.772X_(2)+1898.439 X_(3)+814.049X_(4)-621.398、(雌)F 2=976.079X_(1)+71.879X_(2)+1809.187 X_(3)+774.359X_(4)-610.235,经判别方程验证,雌、雄个体综合判别成功率为81.7%。对作为判别方程的4个性状进行T检验,结果显示半刺厚唇鱼躯干长/体长、尾长/体长、尾柄高/体长、头高/头长4个标准化性状在雌雄群体间差异极显著(P<0.01),表明半刺厚唇鱼雌性个体躯干、尾部较长,雄性个体尾柄、头部较高。
The purpose of this study was to investigate the morphological characters of male and female Acrossocheilius hemispinus.Principal component analysis,R-cluster analysis,Stepwise discriminant analysis and T-test were used to analyze 17 measured characters and 14 standardized character traits of 60 individuals.The result of principal component analysis showed that the contribution rates of the three extracted principal components were 70.056%,8.284%and 4.674%,respectively,while the cumulative contribution rates reached 83.014%.4 characters were screened out from 14 standardized morphological characters of 60 A.hemispinus by stepwise discriminant analysis to establish the sex discriminant equation(Male:F 1=917.211X_(1)+78.772X_(2)+1898.439 X_(3)+814.049X_(4)-621.398,Female:F 2=976.079X_(1)+71.879X_(2)+1809.187X_(3)+774.359X_(4)-610.235).The comprehensive discriminant accuracy rate for male and female population was 81.7%.The T-test showed that the trunk length/body length,caudal peduncle length/body length,caudal peduncle depth/body length and head depth/head length were significantly different between male and female populations(P<0.01).Female had longer trunk and tail,and male had higher caudal peduncle and head.
作者
赖铭勇
LAI Mingyong(Freshwater Fisheries Research Institute of Fujian Province,Fuzhou 350002,China)
出处
《渔业研究》
2023年第2期119-126,共8页
Journal of Fisheries Research
基金
福建省科技计划项目-省属公益类科研院所基本科研专项(2020R1014001)
福建省科技计划项目-省属公益类科研院所基本科研专项(2021R1014002)
福建省种业创新与产业化工程项目(2021FJSCZY04)。
关键词
半刺厚唇鱼
形态特征
主成分分析
判别分析
R-聚类分析
Acrossocheilius hemispinus
morphological characteristics
principal component analysis
discriminant analysis
R-cluster analysis