Study On The Evaluation Technique Of Wood Property And Multiple Trait Selection Of Pinus Elliottii

Slash pine,Pinus elliottii,has been introduced into China from the southeast United States for about 80 years.It is one of the mainly important forestation tree species of industrial plantation of timber forest in the world.The large scale genetic improvement programs for slash pine have been implemented to produce the seedlings with fast-growing and high quality of pulp wood property in southern China to address the severe shortage of pulp wood production in the last two decades.However,the improvement of wood mechanical properties with destructive sampling was neglected because of the difficulty of sampling and measurement.And this is one of the most important reasons for the high risk of bending or lodging under the extreme weather conditions,which restricts the development and application of this tree species in China.Therefore,in order to select the excellent materials with high yield?fast-growing?and quality of wood property,four main objectives were carried on:?1?the non-destructive evaluation technique of wood basic density?DEN?and modulus of elasticity?MOE?of standing tree,?2?genetic variation analysis for growth and wood properties based on the non-destructive testing technologies,?3?effect of sampling size on estimation of trait heritabilities,and?4?index selection for growth and wood properties.The main conclusions are listed as follows:?1?Pilodyn value?Pr?and the basic density???showed a significant negative correlation?P<0.01?,and the regression model was ?=0.6557-0.0778 Pr.1/Pr could be used as an effective prediction index to indirectly measure DEN of individuals or families in a same population.MOE determined by conventional methods had a very significant linear correlation with MOE??MOE?=?V2?.The regression model was MOE=2.0133+0.8286MOE?.This indicated that the modulus of elasticity calculated by MOE=?V2 can represent the determination values.A significant positive correlation?P<0.001?was observed between MOE? and MOEP?MOEP=1/Pr×V2?,and the linear regression model was MOE?=0.2018+0.9431 MOEP.It can be directly to use MOEP as the MOE value for each sample tree to participate in genetic analysis.The correlation between MOE? and V was high positive significantly?P<0.001?and the linear regression model between them was MOE?=-5.1218+3.2394 V.V value can be used as the feasible parameter of MOE in the same population.?2?DENP?DENP=1/Pr?and MOEP were controlled by a moderate degree of genetic control,and the heritabilities?0.3050 for DENP,0.2920 for MOEP?were higher than growth traits?ranged from 0.0625 to 0.2160?.Apparently a great gain in stem volume?Vol??31.47%?,DENP?6.18%?and MOEP?29.11%?could be achieved via direct selection for breeding?the selection rate was 1%?.MOEP was significant?P<0.001?positive correlated with DENP.The genetic correlations between growth traits and wood properties were not significant.In contrast,the phenotypic correlations between them were significantly positive but the correlation coefficients were very low.The improvement for these traits simultaneously was possible.?3?The estimated values of heritabilities and its standard errors by small sample size were not stable.The accurate of these values increased gradually by the increasing sample size or family size.So a large sample and as many family coefficients as possible should used in the estimation of heritability to ensure the precision and accuracy.The sample size should be increased correspondingly in the estimation of the genetic parameters of the traits with low heritabilities.For this trial,more than 39 families or 600 individuals should be used for the reliable heritability estimation.?4?The target traits could be improved to the greatest extent through the single-trait selection method.When growth trait were chosen as the target trait,the gain of wood property would be inhibited,and vise versa.So this method is not suitable for the improvement of growth and wood properties simultaneously.Independent-culling levels is one of the effective methods for multiple trait selection in the absence of effective genetic parameters because it can make each selection trait get preferable improvement at the same time.The appropriate selection index that with the expected gains of target traits was determined as the index which giving 10 times as much emphasis to economic weights?calculated by equal emphasis method?of DEN and MOE.The top 10 families selected by index selection method were Fam₁1-98,Fam₄-87,Fam_B8-16,Fam₀-247,Fam₀-933,Fam_A4-10,Fam_A4-23,Fam_A2-9,Fam_A4-19 and Fam₄-39?in no particular order?,the first 5 of which were included in the top 10 families selected by independent-culling levels and these 5 families were excellent both in growth and wood property.The last 4 families just showed excellent growth trait.