贵州8种杜鹃花属植物的花粉形态及计量分析

Pollen Morphology and Quantitative Analysis of Eight Species of Rhododendron in Guizhou

本文以贵州毕节百里杜鹃为采集地,采集马缨杜鹃、迷人杜鹃等8种杜鹃属为实验材料,通过扫描电镜观察花粉的形态及外壁纹饰,并利用SPSS软件对其进行量化分析,可为杜鹃花属植物的分类鉴定和识别提供基础的参考数据。8种杜鹃花属植物的花粉均为四合体花粉,呈正四面体排列,具三孔沟,其中百合花杜鹃和毛杜鹃的形状呈扁球形,其他杜鹃属介于近球形与球形之间,表面附着粘丝情况不一。经SPSS软件量化,结果表明:(1)8种杜鹃属的极轴长、赤道轴长、四合体直径、周长、面积、萌发孔长、萌发孔宽的平均值范围分别为(13.81~21.21)μm,(17.02~29.96)μm,(28.35~42.75)μm,(101.93~170.25)μm,(673.24~1648.76)μm^(2),(7.67~16.59)μm,(0.57~1.77)μm,P/E的范围为0.684~0.839;(2)对极轴长、赤道轴长等指标进行相关性分析,极轴长、四合体直径、面积是最能影响花粉形态的指标,萌发孔宽指标对花粉形态无显著影响;(3)主成分分析极轴长、四合体直径、面积组成的因子成分可替代以上8个指标,且贡献率排名为极轴长、四合体直径、面积,可以花粉极轴长、四合体直径、面积因子成分作为分类鉴别的依据之一;(4)聚类分析显示欧式距离10时可将8种植物分为3类,这与主成分分析的结果相对应。综上,将电镜观察花粉形态的定性方面与SPSS软件定量分析结合为分类鉴别提供了参考价值。

Eight species of Rhododendron such as Rhododendron delavayi and R.agastum Franch were collected as experimental materials in Bijie,Guizhou.The morphology and outer wall decoration of pollen were observed by SEM,and the quantitative analysis was carried out by SPSS software,which can provide basic reference data for the classification and identification of Rhododendron.The pollens of eight species of Rhododendron are all tetrazygous pollens with tetrahedral arrangement and three furrows.R.liliiflorum and R.pulchrum are oblate spherical in shape,while other Rhododendrons are between subspherical and spherical,with different adhesion and adhesion on the surface.Quantified by SPSS software,the results show that:(1)The average values of polar axis length,equatorial axis length,pollen diameter,perimeter,area,germinal apertures length,and germinal apertures width of the eight Rhododendron were(13.81-21.21)μm,(17.02-29.96)μm,(28.35-42.75)μm,(101.93-170.25)μm,(673.24-1648.76)μm^(2),(7.67-16.59)μm,(0.57-1.77)μm,respectively.P/E range is 0.684-0.839.(2)The correlation analysis of the polar axis length and equatorial axis length showed that the polar axis length,the diameter and the area of the tetrazygote were the most influential indicators of pollen morphology,while the germinal apertures width had no significant effect on pollen morphology.(3)The results of PCA showed that the factor components composed of polar axis length,tetrazygote diameter and area could replace the above eight indexes,and the contribution rates were ranked as polar axis length,tetrazygote diameter and area,indicating that polar axis length,tetrazygote diameter and area factor components could be used as one of the basis for classification and discrimination.(4)Cluster analysis showed that eight species could be divided into three groups at Euclidean distance of 10,which corresponded to the results of PCA.In conclusion,the combination of qualitative aspects of pollen morphology observed by electron microscopy and quantitative analysis by SPSS software provides reference value for classification and identification.