Abstract: On the basis of catalytic polarogram analysis of 54 wild giant panda’s hair， and in accordance with the Grubbscrule and 4d method，this paper set up a background value on the microelement Cu，Zn，Pb，Cd，Co，Ni，Mo，Se，Mg and As in giant panda’s hair.and discussed relation between content of microelem ent and mountains， sxes，and correlation between the elements and its significance tests.Because of he rather significant differences of the hair's microelement in giant panda between indiVidua1s--its probability distribution were in deviate kurotosis and logarithmic normal kurotosis．The data were treated with in various ways of mathematical statistic-- t test of total mean value，when unknown，F test of varience analysis and Mann-wihtney test．As a result of the tests．the microelement contents in giant panda's hair from two mountains differed on Se，Mg in general，and Cu，Zn Contents were different with their sexes，with the significant difference more than 99% and 95% respectively．The specimen of giant pandas' hair were treated with the pattern recognition further. A doptting the nonlinear mapping，we found the count results stated that the computer's clarification conformed as highly as over90% to the factual distribution of giant pandas in different mountains．And to different sexes，83%.

Key words: Giant panda, Microelement, Hair analysis, Treatment of data, Sex

摘要： 本文根据54只野生大熊猫体毛的催化极谱分析数据，用Grubbs准则和四d法建立了大熊猫体毛中Cu、Zn、Pb、Cd、Co、Ni、Mo、Se、Mg、As等10种微量元素含量的背景值，探讨了微量元素含量水平与山系、性别等困素的关系;元素间的相关性及其差异性检验．野生大熊猫个体间微量元素含量差异较大，成偏态或对数正态分布，经未知时总体均值的检验；方差分析的F检验；曼·惠特尼检验等多种数理统计处理，表明岷山、邛崃山两山系的大熊猫体毛中Se、Mg含量水平不同，Cu、Zn含量因性别而异， 其显著性差异分别在99 %与95 %以上．进而用模式识别技术对大熊猫样本分类， 经非线性映照处理， 计算机分类结果与大熊猫按山系分的符合率达90% 以上，与雌雄划分符合率达83%。

关键词: 大熊猫, 微量元素, 毛发分析, 数据处理, 性别