Lorenz曲线及其在植物生态研究中的应用

由Lorenz曲线派生的基尼系数和洛伦茨不对称系数在植物生态研究中可用来描述植物大小或其它数量性状的不均匀性，以揭示不同环境条件下植物数量性状在群体中的分布特性和生态学意义。笔者介绍了这两个参数以及用Jackknife和Bootstrap两种重新取样法来估算该统计量的精度和误差的数学计算方法，并采用作者自行开发的统计软件，对盐胁迫条件下Iris hexagona天然群体生长和繁殖性状进行了不均匀性分析。结果表明，Iris hexagona植株的地上部分生物量、地下部分生物量和总生物量随盐浓度的增加均呈递减趋势，但不均匀性均呈递增趋势。盐胁迫对植株个体间花量的不均匀性影响不大，而植株在浓度为2和4μg／g的盐胁迫条件下，个体间种子数量的不均匀性下降。另外，Iris hexagona两个繁殖性状的不均匀性均远大于生长性状的不均匀性。基尼系数揭示了均值背后隐藏的生物学意义，它直接反映了植物数量性状在植株间的不均匀性的程度，但产生不均匀性的原因可能包括遗传、生理和(或)生态等因素。若要从遗传角度弄清其本质，需要进行分子生物学特别是有关植物数量性状基因位点及其与环境互作等方面的研究。

Lorenz Curve and Its Application in Plant Ecology

The Lorenz curve is widely used in economics to describe inequality in wealth.In recent years,the Lorenz asymmetry coefficient and Gini coefficient derived from Lorenz curve have been used to describe inequality in plant size and other characteristics in plant ecology instead of Skewness coefficient.Like other important ecological indexes, the precision and bias of Lorenz asymmetry coefficient and Gini coefficient can be estimated using the Jackknife and Bootstrap resampling techniques.The estimating of bias and standard error of Gini coefficient by the Jackknife and Bootstrap procedures and five different methods used to construct bootstrap confidence intervals were summarized in this paper.We took the experiment results of 10 natural populations of Iris hexagona under salinity stress from University of Louisiana at Lafayette as an example to illustrate the biological meaning of Gini coefficient.Gini coefficients and it's bias and standard errors of fives quantitative traits including aboveground biomass, belowground biomass,total biomass,number of flowers and number of seeds were estimated using GINI2003(1.0 for WINDOWS) developed by us.We found that salinity decreased the aboveground biomass,belowground biomass and total biomass of Iris hexagona(by the mean value) but increased the inequality of them(by the Gini coefficient) under salinity stress.However,salinity decreased the inequality of the number of seeds.The number of seeds increased under salinity stress(by mean),but Gini coefficient illustrated that the increase of the number of seeds in total due to the increase of the equality.We also found that the inequality coefficients of reproduction traits were larger than that of growth traits.Changes in the inequality of plant size involve many genetic and ecological factors.Modern molecular techniques like quantitative traits Loci especially the interaction with environment can be used for further exploring the essence of inequality of plant quantitative traits.