Afforestation by water-harvesting improves the soil water environment in forest owing to increase of runoff volume collected in the planting strip, so that the productivity of forests is influenced. According to the model of “the double best” and the relationship (the water productivity-WP) between the present situation of water environment improved and the forest production in the different model of afforestation by water-harvesting (forest density), the rational density of afforestation by water-harvesting has been chosen. With the decrease of forest density, the water amount harvested in the planting strip is increased and the soil moisture in available zone of root system is improved, so that the function curve of tree's water productivity changes from linear to logarithmic upward. When the forest density are respectively 1,000 per hm² (in 5 years old ) and 840 per hm² (in 10 years old), the relation curve of tree's WP(WP_t) to transpiration is parabolic, which shows that at this moment water is not a limiting factor of production. Namely, the suitable density values of 5 years old and 10 years old black locust forest are respectively 1,000 per hm² and 840 per hm². Under the suitable density, the WP values of individual tree are respectively 6.7144 g·kg^-1 and 12.1632 g·kg^-1, which accord with the standards of “the designed individual pattern with high-yield” model, and the WP values of forest are respectively 8.4103 kg·hm^-2·mm^-1 and 21.6646 kg·hm^-2·mm^-1, which can be considered as “the optimum population structure” model. It can be applied in both theory and practice to choose the model of forest based on the relationship between water and forest production in the developing of water-saving forestry and in the improvement of water use efficiency.

This is an attempt to classify trees according to their vegetative organs, aiming at the purpose of finding an easier and more practical way to identify trees than the traditional method according to their propagative organs.　Nearly 10,000 spicies from more than 1400 genera of 212 families in China have been extensively investigated in reference to the leaf blade and leaf vein, the branches, the pith structure and medium, vascular system (mainly leaf gap and leaf trace), bud body and leaf scars.　The results indicate that the vegetative organs of Chinese trees can be classified into 7 types. To make further researches for the reliabitity of the vegetative organ classification of trees, a comprehensive key for families and genera (including species groups) in each type and the key to the genera of each family has been made for ready use.　Practice shows that this attempt is completely realizable so long as we can accurately master each above-mentioned basic characters in classifying the trees.　The study finally has made some general conclusions according to the vegetative organs and put forward some valuable views on the basis of the ordinary classification system of the vegetable kingdom(Division, Subdivison, Family and Genus). The study also analyzes the indicative effects of tree venation patterns and vascular system in the systematic development of plants. Additionally there could be a supplementary value to the theory of traditional plant taxonomy at the same time.

Forest biomass is a basic quantity character of the forest ecological system. Biomass data are foundation of researching many forestry and ecology problems, thus accurate measurement of biomass is very important. Establishing biomass models is a major way to biomass estimation. There were a serious shortcomings in the models established previously, i.e.the results were incompatible for models of each component, in other words, the sum of estimated biomass of wood, bark, branches and foliage was unequal to estimated biomass of total aboveground, the sum of estimated biomass of wood and bark was unequal to estimated biomass of stem, the sum of estimated biomass of branches and foliage was unequal to estimated biomass of crown. There fore how to obtain the compatibility is stile a difficult problem for biomass estimate. A new method, nonlinear joint estimate, was proposed in this paper, and compared with method of adjustment in proportion. To the different methods of establishing models, five alternative methods were designed, then one of them was determined as a optimum method through the analysis and comparison. The optimum method took stem as a basis component and adopted two steps joint estimate, structure of models was as follows; the first step, total aboveground\;W₁=f₂(x)+f₅(x), stem W₂=f₂(x), crown W₅=f₅(x); the second step, wood W₃=f₃(x), bark W₄=f₂(x)-f₃(x); branch W₆=f6(x), foliage W₇=f₅(x)-f₆(x)In this paper, the progressive variable selection method was used to select models structure, and weighted least squares method was used to estimate parameters for reducing errors of non-homogeneous variance. At meantime, the paper used five indices to evaluate models, they were coefficient of variation for parameters C%, total relative error RS%,average relative error EE%,average absolute value of relative error RMA% and prediction precision P%. All of researches shown above in this paper took Larix olgensis as an example.

Chinese fir [Cunninghamia lanceolata(Lamb.)Hook.]plantation systems were a series experiment plots consisted of four density treatments: 1665trees/hm², 3330trees/hm²,4995trees/hm² and 6660 trees/hm² (four densities levels are taken place by A,B,C and D). There were not differences in the soil texture and topography of the experimental plots. 1-year-old Chinese fir seedlings were planted in spring of 1987. Tree growth, canopy thickness, tree height under first branch, and ecological benefit of different plantation systems were investigated at 8-year-old of Chinese fir trees. The average DBH increment decreased from 8.1cm to 6.6cm with the increase of plantation density from A to D, at the same time the average tree height increment decreased from 5.9m to 5.4m, the canopy thickness thined from 4.9m to 3.3m, the average tree height under first branch raised from 1.0m to 2.3m. The productivities of different density systems were 4.6, 7.3, 9.5, 10.3t/hm²·a from A to D respectively. The systimatic runoff coefficient, liguid runoff, and solid runoff in the A density region were 1.35%, 19.4 mm/hm²·a, and 334kg/hm²·a respectively, and those in the B density region were 0.52%, 8.1mm/hm²·a, and 138kg/hm²·a respectively, in the C density region were 0.69%, 8.9mm/hm²·a, 196kg/hm²·a respectively, and 0.75%, 11.8mm/hm²·a, 284kg/hm²·a in the D density region respectively. The ecological benefits of B and C density region were better than others. The comprehensive evaluation results based on 5 intems including 15 category values showed that in the score of C density was 50, the score of B density was 43. The score of A density was 36, that of A density was 21.

The planting density experiment of Masson pine has proceeded for 15 years at Longli Forest Farm in Guizhou Province. The result of the experiment showed that planting density had no obvious effect on height growth of the trees. Mean DBH and individual volume decreased with increase of the density, and ratio of height to DBH and growing stock increased. The differences among treatments were significant at 5% or 1% level after 6 or 8 years old. The percentage of out-turn of small log decreased obviously with increase of density, and the percentage of out turn and volume of the short small log and fuel wood all increased with increase of the density. The differences among trentments were significant at 1% level. The growth ring width of wood decreased with increase of the density, and percentage of late wood and basic density of wood increased on the whole. The differences among treatments were obvious for the ring width and the percentage of late wood. The planting density had no significant effect on the fiber morphology and chemical composition of Masson pine wood, and the both all accorded with the basic demands of pulping and papermaking at 15 years old. The economic benefits of 2500, 4444 and 10000 stem/hm², were better than others among 5 treatments. Based on comprehensive analysis the effects of the planting density on growth, log volume, papermaking characteristics of wood and economic benefit of Masson pine plantations, the initial planting densities on medium sites in the Central Part of Guizhou were as follows: 2000～2500 stem/hm² for cultivating building timber of large or middle diameter, 3500～4444 stem/hm² for pulp wood, and about 6000 stem/hm² for fiber board and particle board.

The growth of poplar (Populus tomentosa Carr.) was studied under intercropping systems by means of “time-space complementary” method. The results demonstrated that poplar-crop intercropping system carried out in plain farm region was a available way to increase poplar growth and cultivate large timber, for example, average volume growth of 11-year-old poplar trees under intercropping systems increased 30.3%～37.7% than that of grown in pure stand with the same site and at the same age. It also showed that intercropping practice, considering of the real need farm region, should leave enough space for crop growth so as to give farmer an incentive to carry out intensive management, including soil water and fertility maintaining, which will be advantageous to maintaining the fast growing of poplar trees. In fact, the average stem volume increased more than 80% after thinning, i.e.3m×14m to 3m×28m, in the tenth year of intercropping, at that stage the tree crown coverage rate on farmland was too high and heavy shading harmed crop developing. Poplar crown deflected, on the other hand, existed in all stands because of intercropping pattern (i.e. bigger spacing between-rows and smaller spacing in-row) and there was a positive proportion between the degree of crown defleeted and the intercropping year. The regression equation of crown deflected index (Y) and intercropping years (x) could be shown as belows:Y=0.0301x-0.0708 (r=0.89)

Genetic variation and linkage disequilibrium in a disease resistant seed orchard of Pinus elliottii Engelm were investigated by using RAPD markers in this paper. The results showed that the effective number of alleles per locus ranged between 1.047 and 1.996, averaging 1.449; Estimates of gene diversity per locus ranged between 0.0449 and 0.499, averaging 0.2781;and the Shannon's information index ranged between 0.1095 and 0.6922, averaging 0.4327. The order of these three measures per locus was the same. Out of 5356 pairs of RAPD loci in 104 segragating loci, 355 pairs showed significant deviations from linkage equilibrium. These results indicated that high genetic variation and linkage disequilibrium were harbored in this seed orchard.

The test result from slash pine experimental forest of 43 half-sib family's planted in 1981 shows that there are significant difference in height, DBH(diameter at breast height), volume, stemform, branch diameter and timber density among different families, which possesses some heritability and the significant family effect of the tracheid length was not found; the growth and timber characteristics are controlled to certain extend by genetic factors. There are positive correlation among height, DBH and volume while the correlation among timber density, tracheid length and volume growth isn't obvious. Therefore, with the help of constraint and free selection indexes, and according to different demand between pulpwood and building timber, combination selection of growth, quality and timber properties,was carried out. Four families, No. Ⅳ-47^#,2.46^#,0.609^#and 0.464^#,are the best for pulp wood, and three families of NO. 0.187^#,0.464^# and 0.609^# are suitable for building timber.

This paper deals with the appropriate traits and optimum age of early selection to improve individual volume of Chinese fir (Cunninghamia lanceolata Hook) through clonal selection. The estimated parameters of early selection were based on height and diameter at breast height (DBH) measured from two clonal trials, which included 18 and 17 clones respectively, and were as parts of a large scale clonal breeding program. At the same time, there were two kinds of CK in each trial, which came from a first generation seed orchard and a superior seed source respectively. In the spring of 1987 and 1988, one-year-old rooted cuttings and seedlings were planted with 5～6 trees in a row plot in randomized complete blocks, with 5 replications at 2 m×1.7 m spacing. Height was measured every year after planting for trial 1, and at the second year after planting for trial 2, while DBH was measured at the third year after planting for the two trials. The last investigation was taken in 1997 for trial 1 with 11-year-old and trial 2 with 10-year-old, and the individual volume of this year was calculated according to the standard volume table for Chinese fir. The genetic variances, heritabilities, genetic and phenotypic variation coefficients and juvenile mature correlations of height and DBH were estimated. Meanwhile, the genetic and phenotypic correlation between the height, DBH measured at different ages and volume calculated were also estimated to determine the best indicator for early selection. The result showed that there were significant differences (P<0.0001) among the test materials, and very strong and stable juvenile-mature genetic correlations, and also high clonal heritabilities for height and DBH at different ages, and significant correlations among height and DBH at 3～4 years after planting and volume calculated. So the clonal selection or elimination of Chinese fir clones might be carried out reliably arround 3～4 years after planting. Since heritabilities of DBH and juvenile-mature genetic correlations of DBH with individual volume were higher than that of height at the same year, the efficiency of early phenotypic selection based on DBH was greater than that of height. The paper also noted that some mistakes could be occured in early selection because of some clones with special growth type and growth pattern such as “slim clones" and "stout clones” or “upgrading performers” and “degrading performers”. So, the early selection could be carried out by the strategy of two stages selection, the first stage selection may be conducted with height at the age of 3-year-old and the final selection completed with DBH during the period of 4～6 years after planting.

Loblolly pine is the most widely planted southern pines for timber and pulp production in the southern United States. In China, loblolly pine was introduced more than 60 years. To identify the sources best suited for the diversity of climates and sites being planted in China, as well as to initiate tree improvement programs of genetically adapted plant material, provenance studies were started in 1981. This paper summarize the 15-year results at 6 locations of the 1981 trial. Seeds of loblolly pine was obtained from the Forest Tree Seed Center, USDA Forest Service, Macon, GA (Tab.1). They were of unimproved origin and were identified by state and county. Latitudes and longitude in the tables are for the centers of the counties; mean annual and rainfalls are from reporting stations closest to the county centers. Containerized seedlings were outplanted at about 5～8-month-old in six-tree row plots with 9 replications. Side-by side plantings of the 2 species were repeated 6 locations. Spacing was 3m×2m. One slash pine provenance (S-8, Tayor Co.FL) and a local lot of the indigenous Masson pine(P.massoniana) were included with each pine loblolly pine replication. In addition, one “land race” loblolly pine source(L-11,Minhou, Fujian) was included with each loblolly pine replication. 　Results of the provenance test showed that there were significant variation between different seed sources of loblolly pine. There were significant correlation between the height growth and the minimum temperature of the seed origin.The seed source by location interaction of the 11 provenances planted at 7 locations for loblolly pine was highly significant for height and diameter. Even though the interaction term was highly significant for height and diameter, 98% of the variation in the performance for loblolly pine was accounted for by location, 1.2% of seed source and seed source×location only 0.5%. This leaves very little of the variation (height) to be accounted for by the interaction and error terms. It means common best loblolly pine seed sources could be chosen.Coastal plains of loblolly pine seed sources grow fast, widely adapted could be used in our vast subtropical regions. In northern subtropical region and southern warm temperate region, western seed sources(Arkansas) and inland seed sources could be used, because southern seed sources are not cold resistant.Species comparison showed that in our southern subtropical, including low hills and coast sand soil, slash pine could be planted. In central subtropical region, both loblolly and slash pine could be planted. In northern subtropical region, loblolly pine could be planted, because loblolly pine grows faster and more resistant cold to than slash pine there.

Many photosynthetic property indices and nitrogen metabolic indices of 9 clones of Larix olgensis Henry were measured in phototron. The results showed that the indices varied greatly among 9 clones of Larix olgensis Henry, and correlated with growth character closely. Net photosynthetic rate, respiration rate, true photosynthetic rate, maximum net photosynthetic rate, net photosynthetic rate at half of light saturation point, light response intensity of net photosynthetic rate, nitrate reductase activity, chlorophyll content, amino nitrogen and total nitrogen content were higher in clones growing fast than those in clones growing slowly. Correlation analysis showed that correlation coefficient between net photosynthetic rate or maximum net photosynthetic rate and growth character were higher, and those between respiration rate or chlorophyll content and growth character were lower. The results of system clustering analysis for indices of 6 photosynthetic properties were similar to the results of indices for 6 growth characters. 9 clones could be divided into 3 categories of best, middle, and worst. Research showed that there was a positive correlation between the indices above-mentioned and growth in Larix olgensis Henry, which could be applied to early selection of clones.

According to 18 years located study of density and thinning experiments of Chinese fir stands,we can get some results as followings:1.At the same site index class,comparing the unthinned stand with the thinned stand which keeps the same density with the unthinned stand after thinning,their average tree height and dominant tree height have no evident difference. The diameter at breast height(DBH) of thinned stand is much lower than that of unthinned stand. The standing stock of thinned stand is lower than that of unthinnned stand.The total volume (standing stock+thinned tree volume) of thinned stand is higher than that of unthinned stand. The ratio of total volume of thinned stand with high primary density to unthinned stand is higher than that of thinned stand with low primary density to unthinnned stand.The tree height under the first branch of thinned stand is higher than that of unthinned stand,and it is especially evident when the stand is less than 16 years old.The crown diameter of thinned stand is lower than that of unthinned stand,and the difference is remarkable when the stand is 11～16 years old.2.With the same primary density,the ratio of total volume of thinned stand to that of unthinned stand at the high site index class is higher than that at the low site index class.

The article summarizes the previous research achievements on spruce-fir forest which is largely distributed in Asian-European Continent and North American,especially the responds of regeneration and habitat to different cutting methods.The authors reveal problems existing in the exploitation of spruce-fir forest and its futrue developmental orientation.

The article discusses the electric field and magnetic field effects of plant seeds,affirms the positive effects of the electromagnetic field treatment on the physiology and biochemistry indexes,which can raise the vitality and germination rate of seeds, and increase the germinating energy of seeds.The present condition and developing process of seed dielectric separation apparatus and equipment are introduced and analyzed.The mechanism of dielectric separation is further researched.It is put forward that the key factor to distinguish traditional separation is electric polarization force.While the construction of separation field and the method of calculating are presented,the state of seeds considered as separation dielectric in the electric field,dielectric property and affecting factors are instructed as well It points out that the main factor affecting the dielectric property of seed is moisture content of seeds. Further more,there is usually a linear relationship between them.Also the methods and equipment to measure the dielectric property of seeds are given.

This paper discusses the mechanism of structure characteristics and dynamics for spruce forest in Qinling Mountains through the investigation and comparison of Chinese spruce(Picea asperata) forests at different stages.The result shows that there is a large amount of seed sources and a little number of saplings at age below 20 years,but only this species is the dominant regeneration tree species in mature forest of Picea asperata. In the regeneration forest under selection cutting system,the height growth of poplar and birch is very fast during first 30 years,and dominates rapidly over the canopy.Picea asperata shows a lower growth rate and most tolerant tree species in first growth stage so that it will replace gradually all the other trees.After clear cutting,poplar and birch often occur dominant,only a small number of saplings are Picea asperata in the regeneration forest.In nature conditions,it will take a long time to recover the Chinese spruce forests.The amount of saplings and seedlings in the spruce forest is a key factor in the maintain of the steady state of the spruce forest.This dynamics process can be controlled and improved by human activity.

Betula platyphylla and Populus davidiana forests of the forest-steppe ecotone in northern Hebei and eastern Inner Mongolia are abundant in plants,which are,at least,35 families,95 genera and 117 species.The plant species decrease and a diversity reduces from east to west in the ecotone because of gradually low precipitation.β diversity index reaches peak at the different limits such as those between the ecotone and the forest,and the steppe,and between the forest meadow and the meadow steppe zone in the ecotone.The plant diversity is closely related to climate,the community organization and succession.

In this study,cowpea trypsin inhibitor (CpTI) gene, an insecticidal gene,was introduced into two Populus tomentosa clones by gene transformation mediated by Agrobacterium tumefaciens. The transformed regeneration shoots were obtained directly by leaf discs.In order to established selection condition,leaf discs were cultured in the medium with increasing concentration kanamycin.Kanamycin resistant(km^r)plantlets were obtained by 3～4 cycles screening in selective condition.Then transformed shoots were rooted in the medium containing kanamycin 50mg/L and transferred to greenhouse.The presence of CpTI gene in transgenic plants were confirmed by PCR and PCR-Southern blot.

Forkienia hodginsii is one of the most important tree species in south of China. However, the natrual forest of Forkienia hodginsii nearly does not exist now in China because of disorderly cut and bad management. For this reason, a large area of plantation of Forkienia hodginsii was planted in Fujian Province since 1950's, but the study on it was very little. In 1997, we studied the biomass of Forkienia hodginsii plantation in order to provide scientific basis for the other studies and management about it. We set the plots on the continuous slope and used the usual methods to study the biomass of Forkienia hodginsii plantation. The results showed: (1) We could estimate the biomass of F. hodginsii plantation with both models W=aD^b and W=a(D²H)^b; (2) The biomass distribution on the continuous slope decreased gradually from bottom to top of the slope and the biomass had the smallest value on the top of the slope; (3) The biomass of trunk and bark went obviously down from base to top of tree, and did the ratio of branch to leaf so; (4) The proportion of photosynthetic yield that used to make up the trunk increased with the growth of stands, however, the proportion that used for other organs of tree decreased; (5) The ratio of root to whole tree almost always was same,about 20%; (6) In accordance with the diameter grade distribution of biomass, the diameter grade of biomass maximum was always larger one diameter class than that of number maximum.

Wood basic densityies of about 100 Populus tomentosa Carr. clones grown in 4 clone testing plantations located in Hebei, Henan, Shandong and Gansu were tested and its inheritance and variation among clones and sites were studied. The results showed that the average wood basic density of the studied clones was 0.4363g/cm³ and there was significant difference in wood basic density between clones and sites. The high value of clonal repeatability(0.82～0.91)of wood basic density indicated that the character was under strong genetic control. Correlation analysis showed general significantly negative correlation between growth traits(including diameter at breast height, height, and volume) and wood basic density, although the correlation coefficients varied in a certain degree from one site to another. Studies on wood properties and relationships between tree growth and wood quality are of importance in such fields as wood properties breeding and combined improvement of both growth and wood quality.

Based on the six year localized observation data obtained at Daweishan Forest Experimental Station of Central South Forestry University, the thinning effects in Chinese fir (Cunninghamia lanceolata) plantations with various sites and thinning densities were studied using biomass increment as the quantitative indexes. Results showed that regardless of site conditions, the trees grew more vigorously after thinning was conducted, the biomass increment greater than that of the check plots at both the individual and stand level. For individual plants, the biomass increment responded positively to the thinning intensity, where the heavily thinned plots demonstrated 1.63～2.00 times more increment as against the check plots. However, at the level of stand, the biomass increment declined as the thinning intensity increased. While the biomass total in the newly thinned plots and the total net biomass increment of six years' growth post to the thinning were both greater than the increment in the check plots, no significant difference was observed for both cases with light thinning. The biomass total of heavily thinned plots was by 14.23～25.28t per hm² less than that of lightly thinned plots regardless of site index (SI) class. The stand productivity of the thinned plots followed the same pattern of change as that of the stand biomass increment, namely, the productivity increased when the thinning intensity decreased during the six years post to the thinning, with the productivity by heavy thinning by 2.38～4.21t per hm² and per year lower than that by light thinning and the productivity by light thinning having no significant difference from the check plots. In a word, thinning in general lengthened the prime stage of Chinese fir growth, and the lenthening effect positively correlated to the intensity of thinning. Besides, thinning helped to raise the stemwood percentage of stand, which increased with the increase of the thinning intensity. when the best effect of thinning was analyzed, it was found that the tenth year after planting was the potimal time for the first thinning, considering three SI (Site index) classes, the total biomass increment at the individual and stand level and the overall productivity of the stand. It follows that in intensively managed plantations, short intervaled light thinning is desirable. The time of the thinning should depend on the SI class. For the stand with the SI of 18 or over, the second thinning with moderate intensity is recommended at the 16th year, the final density is about 1500 stems per hm² which is expected to produce large timber. For the SI of 14～16, the second thinning with light intensity is recommended at the 16th year also, but the final density is 1800～2200 stems per hm² which is expected to produce middlewood. And the stand with SI of 12 or less is suitable for middle and small wood production, one moderate thinning without further operations afterwards is recommended. These thinning criteria provide the scientific basis for the rational cultration and management of Chinese fir plantations in various aspects.

The research results showed that the mean height, DBH, the individual volume and the growing stock of 4.5-year-old Eucalyptus plantation decreased with the Continuous Planting Rotation(CPR)increasing. The mean tree height of the second, the third and the fourth rotation decreased by 7.4%, 7.9% and 20.0% respectively compared with the first rotation; the mean DBH decreased by 8.4%, 15.9% and 25.8% respectively; the individual volume decreased by 20.4%,32.8% and 50.4% respectively; and the growing stock decreased by 20.9%,38.7% and 50.5%; The changes of height, DBH, the individual volume and the growing stock with CPR increasing may be simulated by linear, parabola and exponential model: (1) y=a+bx; (2)y=a+bx+cx²; (3)y=ax^b. The corwn length and volume decreased with the CPR increasing, which indicated that the more CPR increased, the worse the growth poentials became, and in which the leaf biomass of the second, the third and the fourth rotation decreased by 10.2%, 44.7% and 71.0% respectively compared with the first planting rotation. Their changes with rotations accorded with the linear model. There was a significat multiple linear regression relationship between the parameter of crown structure and the standing growth index.

The growth process of Chinese fir trees under the pattern of reserving broad leaved tree species in Chinese fir plantation and in pure Chinese fir stand(control) were studied. The results showed that comparing with the control, the tree height, D.B.H and tree volume increment of Chinese fir under the pattern was a little slower in the early seven years and higher at 8～22.5-year-old. After 22.5-year-old the increment declined hardly (though total increment was higher than the control). All these reflected the effect of tree competition and soil fertility development on the tree growth of Chinese fir. In order to cultivate large timber under the pattern, the relationship of Chinese fir and broad leaved tree species must be adjusted in time and the broad leaved trees existed in the same forest canopy with Chinese fir must be cut over at about 20-year-old.

The number of bamboo stem at different ages and the mean diameter at breast height(DBH)which are the important target in evaluating productivity of bamboo stand were investigated in 50 plots established in Jianou city, Fujian Province in this paper, and the authors selected the method of artificial neural network to biuld the simulative and predictive model of mean DBH for bamboo stands. Artificial neural network is a good method in handling the overall nonlinear mapping problems between input variables and output ones, which has a wide application in many research fields, such as system simulating, automation controlling, paralleled data processing and so on. In this paper, the input variables were the number of different age and the total number of stand, the output variable was mean DBH for bamboo stands, the number of neurons of hide-level(M) was M=2L+1=3 according to the last document (L is the number of factors of input-level), and the network activity function is Sigmiod function as follows:F(x)=1/(1+e^-x). Using the built BP network, the samples were trained until E_j(W¹_lm,W²_mn)=Nn=1(O_nj-Y_nj)²=min, where O_nj and Y_nj are output values of network and really values of DBH for bamboo stands respectively,N is the number of trained samples, and E_j is sum of square deviation of BP network. If E_j didn't converge, the weights and thresholds of BP network were adjusted as follow: ΔW_ij(n+1)=βλ_jX_i+αΔW_ij(n) and Δη_j(n+1)=-βλ_j+αΔη_j(n) .. The results showed that the mean simulative accuracy and the mean predictive accuracy of mean D.B.H BP model for bamboo stands were all satisfactory, which were 89.95% and 89.26% respectively. Therefore, it provided a scientific basis for evaluating the productivity and realizing high yield for bamboo stands.