| Abstrakt: |
Abstract: This paper presents the results of a multivariate morphometric study of leaf anatomical characters in different populations of Sesleria rigida sensu lato in Serbia. Morphometric analyses were performed on cross-sections of 521 tiller leaves collected from 21 populations of S. rigida. Principal component analysis and canonical discriminant analysis were used to identify the structure of variability and to determine which variables discriminate between the groups. The UPGMA (unweighted pair-group average linkage) clustering analyses based on leaf anatomical characters and habitat climatic characteristics were employed to explore whether the observed anatomical differences are a result of adaptive responses. Regression analysis (linear regression) was performed to identify the level of correlation between leaf anatomical characters and basic orographic, geological, and bioclimatic habitat characteristics. DIVA-GIS software was used to extract 19 bioclimatic parameters from the WorldClim set of global climate layers. Our detailed analyses have shown that the studied populations are anatomically very well differentiated into the serpentine populations from western Serbia and carbonate populations from eastern Serbia. The regression analysis has shown that the geological substrate (serpentinites vs. carbonates) represents the most significant abiotic factor that is correlated with the anatomical differentiation of the populations in the studied area. Besides the geological substrate, bioclimatic parameters such as precipitation and habitat humidity are also highly correlated with the leaf anatomical characters, which was not the case with temperature. |
Full text from SpringerLink