SEED PRODUCTION AND QUALITY OF PINUS DURANGENSIS MART., FROM SEED AREAS AND A SEED STAND IN DURANGO, MEXICO

摘要

The role of edaphic and topographic variables on the distribution pattern of conifer species in moist temperate area ofwestern Himalayan and Hindukush region of Pakistan is investigated. Arboreal vegetation of 41 sites was analyzed usingpoint-centered quarter method. Soil samples from 41 different sites were collected to determine the edaphic characteristicsand two stratifying variables i.e., elevation and slope were noted. Both classification and ordination methods were used todetermine the underlying group structure and vegetation composition. For the purpose of ordination principal componentanalysis (PCA) and for classification Ward’s cluster analysis was employed. Four groups of species were separated on thebasis of topographic and edaphic variables by Ward’s cluster analysis. Group 1 was dominated by Pinus wallichiana, groupII & III were dominated by Abies pindrow while group IV was compose of two dominant species Cedrus deodara and Pinuswallichiana. The four groups also differentiated with respect to elevation, slope, salinity and electrical conductivity. Thefirst component of PCA ordination of vegetation was correlated with elevation, soil organic matter (OM), pH, water holdingcapacity and soil conductivity; the second component exhibited association with elevation, slope, OM and conductivitywhile the third with slope, soil salinity and conductivity. Four principal component axes were well correlated withtopographic and edaphic factors. Axis 1 was highly (p0.001) significantly correlated with pH, organic matter andconductivity while axis 2 showed significant correlation (p0.001) with elevation, slope, pH and conductivity. Axis 3significantly correlated (p0.001) with salinity and conductivity. Axis 4 exhibited marked relationship (p0.001) withorganic matter of soil and slope angle. Correlation coefficient between PCA vegetation and PCA environmental variableswas also sought that exhibited significant correlation between components 2.