Study On Morphological Variation And Optimal Selection Of Cornus Walteri

Cornus walteri belongs to cornaceae which is a deciduous tree, it also is an important ecological oil materials species used along with timber in temperate regions. On the base of extensive field investigation of resources, this paper took Corner walteri as research object. Combined with statistical analysis, characteristics of morphological variation were sudied, mean while, to propose the method of tree selection and planting regionalization for providing a more solid foundation for the future development of fuel energy forest and the process of high-quality seeding.Based on the extensive survey of Cornus walteri resources in the main distribution. The method of nested analysis of variance, correlation analysis, multiple comparison and hierarchical cluster analysis were used for analysis of phenotypic diversity among populations and within populations and geographic variation. The criterions and methods of plus tree selection were carried out by the principal component index and comprehensive scoring method. According to the suitability and eco-geographic condition in different areas, the silviculture division in the main distribution area was put forward. The main results are as follows:(1) By the comparative analyses of 17 morphological traits, it shows that there is extremely rich morphological variation existed either among populations or within populations. The mean phenotypic differentiation coefficient for 17 traits is 19.88%, and the variation between populations (80.12%) is slightly higher than that within populations (19.88%), which indicate that the variation within population is the main part of the phenotytic variation of Cornus walteri. The CVs of morphological traits shows that, the extent of pericarp thickness, seed-shell thickness, was relatively rich, but seed diameter was poor; variation of leaf area was higher while leaf length variation was smaller; The shape variation of fruiting-cyme was higher than the width of fruiting-cyme. The variation of various types of morphological traits in order from big to small: fruiting-cyme>leaf>seeds and fruits>tree bark. The comparative analysis of average value of morphological variation in different groups: Qingzhou group has a high comprehensive variation, however, Jingxing was low. In summary, the central group of morphological variation in leaf is higher than marginalized groups, while the shape of fruiting-cyme in the center of groups is greater than the marginalized groups, The differences in morphological traits between groups large with the geographical span.(2) Correlation analysis showed that there is a significant positive correlation between fruit horizontal diameter and 100-fruit weight (r = 0.811^*); seed diameter and species of shell thickness there was a significant positive correlation (r = 0.892^**). In addition, there are good relationship among diferent morphological traits in various types, which, bark index were significant correlation associated with fruit shape index, 100-fruit weight, fruit horizontal diameter, fruiting-cyme width; fruiting-cyme height were significantly correlated with seed diameter, seed shell thickness (r=0.860^*,0.846^*); fruiting-cyme width were highly significant correlated with 100-fruit weight, fruit horizontal diameter (r= 0.927^**, 0.876^**); there is a significant negative correlation between fruiting-cyme index and pericarp thickness; leaf length has a significant negative correlation with fruiting-cyme height, fruiting-cyme index, seed diameter and seed shell thickness (r=- 0.850^*,-0.777^*,-0.771^*); leaf width has a significant negative correlation with fruiting-cyme index and seed shell thickness (r =- 0.793^*,-0.854^*), while has a significant positive correlation with pericarp thickness and solid-shell ratio (r = 0.867^*, 0.889^**); leaf index was significantly negatively correlated with fruit horizontal diameter (r =-0.806^*); leaf area was negatively correlated with fruiting-cyme height, fruiting-cyme index, seed diameter, seed shell thickness (r=-0.833^*,-0.774^*,-0.765 ^*,-0.830^*), while was a significant positive correlation with solid-shell ratio (r = 0.788^*).(3) According to analysis between eco-geographical factors and morphological traits, it shows that, the leaf width and leaf area has a significant positive correlation with longitude (r =0.812^*, 0.782^*), seed diameter has a significant negative correlation with longitude (r=-0.759^*), seed-shell thickness has not only a significant correlation with latitude (r=-0.892^*), but also with the annual precipitation and average temperature (r=0.787^*,-0.772^*); pericarp thickness was significantly related to latitude (r=0.945^**); solid-shell ratio was significantly correlated with longitude and latitude (r=0.925^**, 0.865^*).(4) In aspect of seed quality, the results of multiple comparisons showed that the significant differences were obvious in different regions. Seed quality in Yangling was the best, which kernel weight and rate, 100-seed weight, volume of 100-seed weight was the maximum among regions. Seed quality in Lushi was the worst, which it has the lowest rate of live embryos. Comparison of seed oil content showed Lushan is the highest and Boshan is the lowest. Analysis of correlations among the seed quality characteristic showed that, the correlations of 100-seed weight and volume of 100-seed was a highly significant. Besides, there was a significant positive correlation between kernel rate and full kernel rate. Eight provenances were divided into three levels by clustering analysis: Yangling was the first level. Lushan and Boshan were the second level. Other provenances were the third level. The January average temperature was significantly positively related to seed-shell thickness and volume of 100-seed.(5) In this study, based on the optimal indicators of the candidate tree, The criterions and methods of plus tree selection were carried out by the principal component index and comprehensive scoring method, two methods of selection rate were 14.5% and 55.13%. The research achieved the national primary selection of Cornus walteri resources in the conditions of individual plant and plantation. Moreover, it provided a scientific basis for plus tree selection and achieved good results.(6) In aspect of cultivation division, according to biogeographic characteristics with the existing distribution, we draw up cultivation division in the main distribution area of Cornus walteri in china. Meanwhile, 4 divisions and 7 sub-divisions were divided in these places: the best suitable division consisting of Weihe plain sub-divisions and Shanxi southern basin, suitable division consisting of Shanxi southwest plateau sub-divisions and Hebei western mountain subregion and low mountain-hilly subregion of Shandong mid-south, sub-suitable division consisting of Shanxi basin sub-district and Huang-Huai-Hai plain subregion, non-suitable division consisting of western mountainous sub-divisions in Henan.