基于角质颚的东、黄海头足类科类判别

Preliminary study of species identification of cephalopod in East China Sea and Yellow Sea based on beak morphology

角质颚具有稳定的结构特征,是头足类分类的良好材料之一。根据在东、黄海采集的分属3个科6个种类样本(中国枪乌贼Loligo chinensis、杜氏枪乌贼Uroteuthis duvauceli、虎斑乌贼Sepia pharaonic、神户乌贼Sepia kobiensis、短蛸Octopus ocellatus和长蛸Octopus variabilis),通过测量其角质颚的形态参数,对不同科类的角质颚进行判别分析,并将原始数据和标准化数据的分析结果进行对比分析。结果发现:中国枪乌贼和虎斑乌贼的角质颚参数原始数据分别大于杜氏枪乌贼和神户乌贼,而标准化后的结果正好相反;短蛸的角质颚参数无论在标准化前后均大于长蛸;在角质颚原始数据中选出10个用于最终的判别分析,上头盖长、下脊突长、下翼长贡献了绝大部分差异。3个科总判别成功率为92.8%;在角质颚标准化数据中选出7个用于最终的判别分析,上头盖长/胴长、下侧壁长/胴长、下翼长/胴长贡献了绝大部分差异,3个科总判别成功率为98.1%。由此可见,对角质颚数据进行标准化处理,能够有效地提升判别正确率。

Cephalopod beak has a stable structure, and it is a good material for species identification. According to the six species of cephalopod samples(Loligo chinensis,Uroteuthis duvauceli,Sepia pharaonic,Sepia kobiensis,Octopus ocellatus and Octopus variabilis) caught from East China Sea and Yellow Sea, the beak morphometrics are measured, discriminant analysis is made at family level, the results are compared between non standardization data and standardization data. The results showed that non standardization beak morphometrics of Loligo chinensis and Sepia pharaonic is bigger than Uroteuthis duvauceli and Sepia kobiensis, respectively,but the results are reversed after standardization data. Beak morphometrics of Octopus ocellatus is biggger than Octopus variabilis for both non standardization data and standardization data. Ten non standardization beak morphometrics were selected for the final discriminant analysis, and upper hood length(UHL), lower crest length(LCL), lower wing length(LWL) contribute most to the morphological variation,total successful discriminant rate is 92.8% for the three families;seven standardization beak morphometrics were selected for the final discriminant analysis,and upper hood length/mantle length(UHL/ML), lower crest length/mantle length(LCL/ML), lower wing length/mantle length(LWL/ML) contribute most to the morphological variation, and total successful discriminant rate is 98.1% for the three families. In conclusion, it can effectively improve the discriminant rate after data standardization for beak morphometrics.