| 摘要: |
| 本文根据2013 年8 月于舟山市沈家门东河菜场采集的5 种近海常见经济头足类(剑尖枪乌 贼、杜氏枪乌贼、金乌贼、曼氏无针乌贼和短蛸), 利用逐步判别和主成分分析法对其上、下角质颚 的各5 种长度指标(喙长、头盖长、脊突长、侧壁长、翼长)进行分析。结果显示, 这5 种头足类角质 颚长度差异显著(ANOVA: P<0.001), 其中以金乌贼的角质颚尺寸最大, 短蛸的角质颚尺寸最小。判 别分析显示, 角质颚长度适合用来划分头足类的种类, 综合判别成功率为96.2%, 其中以上颚侧壁 长、上颚头盖长和下颚脊突长对判定的贡献率最高。然而, 与之相比, 标准化后的角质颚长度则更适 合用来划分头足类的种类, 综合判别成功率达到100%, 其中以标准化下颚喙长和标准化下颚翼长对 判定的贡献率最高。主成分分析显示, 角质颚长度及标准化后角质颚长度对枪乌贼类和乌贼类的判 定成功率均达到100%.本文建立了一种基于角质颚长度判别分析法的头足类种类判定的新方法, 丰 富了头足类种类鉴定的技术手段, 为我国学者在相关领域的研究提供了基础。 |
| 关键词: 角质颚 头足类 逐步判别分析 主成分分析 种类判别 |
| DOI：10.11693/hyhz20150100011 |
| 分类号: |
| 基金项目:国家自然科学基金项目, 41306127 号, 41276156 号, 41476129 号; 上海市自然科学基金项目, 13ZR1419700 号; 上海市教委 创新项目, 13YZ091 号; 教育部博士点基金项目, 20133104120001 号。 |
|
| SPECIES IDENTIFICATION OF CEPHALOPODS BASED ON BEAK MORPHOMETRIC VARIABLES |
|
LIU Bi-Lin1,2,3,4, CHEN Xin-Jun1,2,3,4, FANG Zhou1, JIN Yue1
|
|
1.College of Marine Sciences, Shanghai Ocean University, Shanghai 201306, China;2.National Distant-water Fisheries Engineering Research Center, Shanghai Ocean University, Shanghai 201306, China;3.Key Laboratory of Sustainable Exploitation of Oceanic Fisherise Resources, Ministry of Education, Shanghai 201306, China;4.Collaborative Innovation Center for Distant-water Fisheries, Shanghai 201306, China
|
| Abstract: |
| We studied beak morphometric variables (rostrum, hood, crest, lateral wall and wing length of upper and lower beaks) of five economic cephalopods (Uroteuthis edulis, U. duvauceli, Sepia esculenta, Sepiella maindroni, and Octopus ocellatus) from the Donghe market in Zhoushan City, using stepwise discriminant analysis (SDA), principal component analysis (PCA), and univariate analyses of variance (ANOVA). The results show that all measured beak variables are species-specific in significant difference (P<0.001). S. esculenta have the largest beak while O. ocellatus the smallest. SDA results indicate that beak variables are useful for species identification. Of all the variables, the upper lateral wall length (ULWL), upper hood length (UHL), and lower crest length (LCL) contribute to the most of the identification for 96.2% in overall cross-validation rate. In contrast, standardized beak variables are more effective than original ones for having 100% reclassification rate, in which standardized lower rostrum length (LRLs) and standardized lower wing length (LWLs) contributed the most. PCA results show 100% successful discrimination identifying species of loliginids and cuttlefishes using original and standardized beak variables. Therefore, the new method can be applied to determine cephalopod species, which provides useful information for the research in relevant fields. |
| Key words: beak cephalopod stepwise discriminant analysis principal component analysis species identification |
