Yield models for Eucalyptus globulus fuelwood plantations in Ethiopia

Based on 53 tree analyses and 105 sample plots of Eucalyptus globulus, models for volume and biomass at single tree and stand levels were developed. The possible growing sites were divided into four site classes. In seedling stands, the site class I corresponds to yield class 44 m³ ha⁻¹ year⁻¹, in coppice stands to yield class 46 m³ ha⁻¹ year⁻¹. The site class IV corresponds in seedling stand to yield class 9 m³ ha⁻¹ year⁻¹, in coppice stands to yield class 13 m³ ha⁻¹ year⁻¹. The maximum mean annual increment was reached in seedling stands at the age of 18–19 years, in coppice stands at the age of 14 years.     

Species screening and biomass trials of woody plants in the semi-arid southwest United States

Species screening and selection trials are being conducted to identify appropriate woody species and technology for biofuel farming in the southwest United States. During the initial phase of the program, 26 native and exotic species have been tested at sites in Texas, New Mexico and Arizona. Second season biomass estimates indicate that initial yields of 3·2–8·5 dry Mg ha⁻¹ year⁻¹ are possible without post-establishment irrigation. The species Atriplex canescens (Pursh) Nutt., Prosopis alba Griseb. and Leucaena leucocephala (Lam.) de Wit reached canopy closure within two seasons of growth and are identified as rapid initial biomass producers under the semi-arid conditions of the study area.     

Carbon of Chaillu forests based on a phytosociological analysis in Republic of Congo,more than meets the eye?

The objectives were to quantify aboveground, belowground and dead wood carbon pools near Mayoko in the Chaillu massif of Republic of Congo and explore relationships between carbon storage and plant diversity of all growth forms. A total of 190 plots (25 m by 25 m) were sampled (5072 stems, 211 species) and data analysed using recommended central-African forest allometric equations. Mean stem diameter at breast height was 33.6 cm, mean basal area 47.7 m² ha⁻¹ and mean density of individuals 407 ha⁻¹. Mean aboveground carbon (AGC) ranged from 13.93–412.66 Mg C ha⁻¹, belowground carbon from 2.86–96.97 Mg C ha⁻¹ and dead wood from 0.00–7.59 Mg C ha⁻¹. The maximum AGC value recorded in a plot was 916 Mg C ha⁻¹. The analysis performed using phytosociological association as basis rather than broad vegetation type is unique. AGC values for undisturbed terra firme forest sites featured among the highest recorded for African tropical forests. Considering only tree diversity, a weak, yet significant, relationship existed between AGC and species richness, Shannon-Wiener index of diversity and Fisher's alpha. However, if diversity of all plant growth forms is considered, no relationship between carbon and plant diversity existed.     

Tree regeneration before and after restoration treatments in managed boreal Picea abies stands

Questions: What are the short-term effects of restoration treatments, including fire and/or partial cutting with dead wood creation, on seedling density and distribution among microsites, and do they differ between upland and paludified biotopes? Location: Mature managed Picea abies-dominated stands in southern Finland. Methods: The restoration treatments consisted of four levels of cutting, with and without fire: uncut, low-CWD (partial cutting leaving 50 m³ ha⁻¹ of standing retention trees and 5 m³ ha⁻¹ of down retention trees, DRT), intermediate-CWD (as previous but with 30 m³ ha⁻¹ of DRT), and high-CWD (as previous but with 60 m³ ha⁻¹ of DRT). Results: The cutting treatment alone did not affect seedling density. Fire had an immediate effect, both by itself in upland biotopes (decreasing effect on P. abies density) and in combination with cutting in paludified biotopes (increasing effect on Betula spp. density). The density of other deciduous species (Sorbus aucuparia, Populus tremula, and Salix spp.) was not affected. Before treatments, seedlings grew predominantly on level ground and mounds. After treatments, more Betula, but fewer P. abies and other deciduous species, grew on these microsites. More Betula and other deciduous species grew next to CWD and under a fallen crown; more P. abies grew under a fallen crown in unburned stands. Conclusion: The post-treatment tree seedling density and distribution among microsites can be affected by restoration treatments. However, knowledge of local site characteristics and their interaction with different restoration treatments is needed to achieve the goals set for restoration at the stand and landscape levels.     

Short‐term changes affecting regeneration of Fagus crenata after the simultaneous death of Sasa kurilensis

Abstract. Question: The aim of the present study is to determine whether seed/seedling predation will increase and Fagus survival will decline with the recovery of the Sasa cover. Methods: We examined Fagus crenata regeneration for seven years in an old‐growth Fagus‐Sasa forest near Lake Towada, northern Japan, by examining the effects of simultaneous death of Sasa, tree canopy gap formation, mast seeding of Fagus and seed and seedling predation by rodents on the survival of Fagus seeds and current year seedlings. We established four types of sites differing in forest canopy (closed or gap) and Sasa status (dead or alive) following the simultaneous flowering and death of Sasa kurilensis (dwarf bamboo) in 1995. Results: Fallen Fagus seed was abundant in 1997 and 2000 (mast years). In sites with alive Sasa, survival from the first growing season was low due to high seed and seedling predation. In sites with dead Sasa, seed survival under the canopy was high for both mast years, but in gaps it varied between years. Seedling survival was highest in canopy gaps with dead Sasa (gap‐dead) in 1998, because of higher light levels and lower predation by rodents. However, seedling survival in these plots was low in 2001, apparently because rapid Sasa recovery favoured rodent predation. In both mast years, Sasa die‐back had significant positive effects on seed and seedling survival under closed canopies because the seedlings there were more successful in escaping predation. Conclusion: The change in successful sites for the early stage of regeneration of Fagus appears to reflect the combined effects of canopy gap, seed/seedling predation and revegetation of Sasa.     

Nitrogen aboveground turnover and soil stocks to 8 m depth in primary and selectively logged forest in southern Amazonia

Extensive areas of Amazonia undergo selective logging, modifying forest structure and nutrient cycles. Anthropogenic‐accelerated rates of nitrogen (N) turnover could increase N loss and affect regeneration, carbon sequestration and timber production. We quantified leaf area reduction, canopy opening and downed biomass and resultant N flux from reduced impact logging (RIL) activities. We compared canopy reduction, surface soil moisture and nitrate to 8 m depth between logged gaps and intact primary forest to determine if logging activities increase subsoil nitrate. To test long‐term logging effects, we evaluated surface N stocks along a 12‐year postlogging chronosequence. At the harvest rate of 2.6 trees ha^?1, total N additions in logging gaps, including leaves and wood from felled crowns (24.8 kg N ha^?1) and other killed trees (41.9 kg N ha^?1), accounted for over 80% of the total N addition to aboveground necromass from all logging activities (81.9 kg N ha^?1). Despite this N turnover by logging, belowground nitrate storage to 8 m depth did not differ between logging gaps and primary forest at the low harvest rate and disturbance intensity of this study. Soil water depletion also did not differ between gaps and primary forest over 1 year, indicating the impact on belowground inorganic N was low. Compared with primary forest, nitrate concentrations to 8 m depth in logging gaps were only significantly higher at 60–100 cm, suggesting some N redistribution beyond the bulk of the fine roots in logging gaps. Extrapolated to the Amazon Basin scale, we provide a conservative estimate that logging damage and bole export under RIL would turn over 0.14 ± 0.07 to 0.23 ± 0.12 Tg N yr^?1 based on 1999–2002 selective logging rates. Greater damage during conventional selective logging would cause higher N turnover throughout the Amazon Basin than our results based on RIL.     

Tree species regeneration in a mid-elevation,temperate rain forest in Isla de Chiloé, Chile

The regeneration of canopy and subeanopy species in a mid-elevation, primary rain forest in the Coastal Range of Isla de Chiloé (42°30S), in the cold-temperate region of Chile, was studied by comparing seedling and sapling abundances under the forest canopy, and within 36 tree-fall gaps. The forest was dominated byAmomyrtus luma andLaurelia philippiana (33 and 32% of the main canopy individuals), and two subcanopy species (Myrceugenia ovata, andMyrceugenia planipes) were also important. Uncommon species in the canopy wereDrimys winteri, Amomyrtus meli, andRaphithamnus spinosus. Tree-fall gaps were created generally by the fall of several trees, and the main canopy species were the principal gap-makers. Gap sizes varied between 28 and 972 m², with a mean of 197 m². Seedling and sapling abundances indicate that the dominant species are capable of regenerating below the canopy, but they also germinate and show enhanced growth within small light gaps. For one of the common subcanopy species (M. planipes) and the two infrequent canopy species (D. winteri, andA. meli) regeneration seems to depend entirely on tree-fall gaps. Thus, in this forest, light gaps allow the persistence of infrequent canopy species, but seem less important for the regeneration and maintenance of dominant canopy species.     

The regeneration of pioneer tree species under browsing pressure of Sika deer in an evergreen oak forest

The effects of Sika deer (Cervus nippon) browsing on the regeneration of pioneer species were studied in relation to canopy gaps in a warm temperate evergreen oak forest in Kasugayama, Nara City. Four study sites, three in canopy gaps and one under a closed canopy, were selected and each divided into fenced and unfenced plots. Under the closed canopy, seedlings of all the pioneer species died irrespective of browsing pressure. However, in the canopy gap sites, seedlings of the pioneer species could establish and grow well. The seedling survival ratio in the fenced plots in the canopy gaps was >60% 1 year after germination. However, in the unfenced plots, only <20% of the seedlings survived 1 year, with all dying within 3 years after germination. Thus, the regeneration of pioneer trees in this forest was strongly inhibited by deer browsing. Successful regeneration of a pioneer,Zanthoxylum ailanthoides, occurred for several years even after two major wind disturbances during the past 90 years. This may be due to less browsing pressure from the deer.     

Spontaneous Return of Biodiversity in Restored Subtropical Thicket: Portulacaria afra as an Ecosystem Engineer

An accepted criterion for measuring the success of ecosystem restoration is the return of biodiversity relative to intact reference ecosystems. The emerging global carbon economy has made landscape‐scale restoration of severely degraded Portulacaria afra (spekboom)‐dominated subtropical thicket, by planting multiple rows of spekboom truncheons, a viable land‐use option. Although large amounts of carbon are sequestered when planting a monoculture of spekboom, it is unknown whether this is associated with the return of other thicket biodiversity components. We used available carbon stock data from degraded, restored, and intact stands at one site, and sampled carbon stocks at restored stands at another site in the same plant community. We also sampled plant community composition at both sites. The total carbon stock of the oldest (50 years) post‐restoration stand (250.8 ± 14 t C ha^?1) approximated that of intact stands (245 t C ha^?1) and we observed a general increase in carbon content with restoration age (71.4 ± 24 t C ha^?1 after 35 and 167.9 ± 20 t C ha^?1 after 50 years). A multiple correspondence analysis separated degraded stands from stands under restoration based on ground cover, floristic composition, and total carbon stock. Older post‐restoration and intact stands were clustered according to woody canopy recruit abundance. Our results suggest that spekboom is an ecosystem engineer that promotes spontaneous return of canopy species and other components of thicket biodiversity. The spekboom canopy creates a cooler micro‐climate and a dense litter layer, both likely to favor the recruitment of other canopy species.     

The effects of Sasa and canopy gap formation on tree regeneration in an old beech forest

Abstract. We examined the response of tree seedling emergence and survival to the dieback of Sasa and canopy gap formation in an old‐growth forest near Lake Towada, northern Japan. Synchronous death of Sasa occurred in 1995. We established four types of sampling sites differing in forest canopy conditions (Closed or Gap) and Sasa status (Dead or Live). Gap‐Dead sites had the highest light levels and the greatest fluctuation in soil temperatures. The death of Sasa alone facilitated the emergence (Acer japonicum, Fagus crenata, Fraxinus lanuginosa, and Tilia japonica) and survival (Acanthopanax sciadophylloides, F. crenata, F. lanuginosa, Kalopanax pictus, and Sorbus commixta) of species with a seedling bank strategy. Cercidiphyllum japonicum grew at all sites at a higher density than other species, but survived well only in Gap‐Dead sites. This behaviour was associated with a seed rain strategy. The additive effects of Sasa death and canopy gap formation promoted seedling emergence of pioneer tree species (Betula maximowicziana, Lindera umbellata, and Magnolia obovata), probably through break of dormancy by the large temperature fluctuation. In addition, the scarcity of advance regeneration in canopy gaps due to Sasa cover facilitates the regeneration of pioneer species. The dominance and dieback cycle of Sasa contributes to species diversity in this forest.     

Mangrove forests in a peri-urban setting: the case of Mombasa (Kenya)

The structure and regeneration patterns of the peri-urban mangrove vegetation of Mombasa at Tudor creek were studied along belt transects at two forest sites of Kombeni and Tsalu. Based on the species importance values, the dominant mangrove species were Rhizophora mucronata Lam. (Rhizophoraceae) and Avicennia marina (Forssk.) Vierh. (Avicenniaceae). Lumnitzera racemosa Willd., reported in an earlier floristic survey, was not encountered. Tree density varied from 1,264 trees ha^–1 at Kombeni to 1,301 trees ha^–1 at Tsalu and mean tree height was higher at the former site compared to the latter. The size-class structure at both localities showed the numerical dominance of small trees over larger trees. The spatial distribution pattern of adults and juveniles varied greatly between sites and showed a close to uniform pattern (Morisita’s Index I _δ ≪ 1) for adult trees, but a tendency to clustered distribution (I _δ ≫ 1) for juveniles. The present paper shows that unmanaged but exploited peri-urban mangroves are structurally stressed, having enlarged canopy gaps that are characterised by spatial and temporal site heterogeneity that influences regeneration, implying longer periods for canopy closure. Diversifying uses of mangrove products and establishing reserves as no cut zones with regulated harvesting will minimise canopy gap sizes, and promote conservation practices. The proposed management strategy shall boost the ecosystem resilience to both anthropogenic and natural stressors expected in the peri-urban setting in the long run.     

A Neighborhood Analysis of the Consequences of Quercus suber Decline for Regeneration Dynamics in Mediterranean Forests

In forests, the vulnerable seedling stage is largely influenced by the canopy, which modifies the surrounding environment. Consequently, any alteration in the characteristics of the canopy, such as those promoted by forest dieback, might impact regeneration dynamics. Our work analyzes the interaction between canopy neighbors and seedlings in Mediterranean forests affected by the decline of their dominant species (Quercus suber). Our objective was to understand how the impacts of neighbor trees and shrubs on recruitment could affect future dynamics of these declining forests. Seeds of the three dominant tree species (Quercus suber, Olea europaea and Quercus canariensis) were sown in six sites during two consecutive years. Using a spatially-explicit, neighborhood approach we developed models that explained the observed spatial variation in seedling emergence, survival, growth and photochemical efficiency as a function of the size, identity, health, abundance and distribution of adult trees and shrubs in the neighborhood. We found strong neighborhood effects for all the performance estimators, particularly seedling emergence and survival. Tree neighbors positively affected emergence, independently of species identity or health. Alternatively, seedling survival was much lower in neighborhoods dominated by defoliated and dead Q. suber trees than in neighborhoods dominated by healthy trees. For the two oak species, these negative effects were consistent over the three years of the experimental seedlings. These results indicate that ongoing changes in species’ relative abundance and canopy trees’ health might alter the successional trajectories of Mediterranean oak-forests through neighbor-specific impacts on seedlings. The recruitment failure of dominant late-successional oaks in the gaps opened after Q. suber death would indirectly favor the establishment of other coexisting woody species, such as drought-tolerant shrubs. This could lead current forests to shift into open systems with lower tree cover. Adult canopy decline would therefore represent an additional factor threatening the recruitment of Quercus forests worldwide.     

Effects of supplemental nitrogen rate and source on biomass production by three weed species in fallow vegetable land

In 1983, a study was conducted to evaluate three weed species for their biomass yield potential on a fallowed tomato land with various nitrogen (N) rates and sources. Dogfennel (Eupatorium capillifolium (Lam.)Small), sida (Sida rhombifolia L.), and pigweed (Amaranthus hybridus L.), were grown with three N rates: residual, 1x and 2x (1x = 56 kg ha⁻¹ N) and with two N sources: NH₄NO₃ and isobutylidene diurea (IBDU). Sida produced the greatest fresh and dry biomass yields and percentage dry matter, 43·3, 23·3 Mg ha⁻¹ and 54·1%, respectively. Yields of the three crops increased with increasing N rates and were higher with NH₄NO₃ than with IBDU. Overall, sida at the 2x N rate with NH₄NO₃ had the highest fresh, 73·8 Mg ha⁻¹, and dry biomass, 37·9 Mg ha⁻¹, yield. Lowest residual soil total soluble salt concentration, 170 ppm, was found after pigweed, in plots which did not receive additional N. Additional N application increased residual soil total soluble salts after each of the three weeds. Thus, additional N, in NH₄NO₃ form, was necessary for high yields in a fallowed vegetable land, but the additional fertilizer also increased soil total soluble salts, regardless of the plant species.     

Comparison of the effects of canopy and understory nitrogen addition on xylem growth of two dominant species in a warm temperate forest,China

Understanding the effect of increasing atmospheric nitrogen (N) deposition on xylem growth of trees is critical to predict tree growth and carbon sequestration under global change. Canopy N addition (CAN) is generally believed to realistically simulate atmospheric N deposition on terrestrial ecosystems given it takes all processes of N deposition from forest canopy to belowground into account. However, whether CAN is more effective in reflecting the effect of atmospheric N deposition on xylem growth of trees than understory N addition (UAN) has been rarely reported. To address the question, we conducted a CAN vs. UAN experimental study to weekly monitor xylem growth of two dominant broadleaf species (Quercus acutissima Carruth. and Quercus variabilis Blume) in a warm temperate forest of China during 2014–2015. Weekly xylem increment during the two years was measured. Mixed-effects models were used to quantify the effects of N addition on xylem growth and detect the differences among treatments. We found that CAN of 50 kg N ha⁻¹ yr⁻¹ plays a more significant role in promoting xylem growth of Q. acutissima than UAN of 50 kg N ha⁻¹ yr⁻¹, and significantly enhanced the formation of differentiating xylem (zones of radial enlarging and wall-thickening cells) of Q. acutissima in the early growing season (April-June) and the rate of xylem increment, but no significant difference in xylem increment of Q. variabilis was detected between CAN50 and UAN50. This is the first study to quantitatively demonstrate that previous UAN studies may have underestimated the effects of atmospheric N deposition on tree growth by ignoring the N interception through forest canopy. Furthermore, our study also suggested a species-specific response of xylem growth to N addition. Under a certain amount of atmospheric N deposition in the future, the xylem increment of Q. acutissima may be superior to that of Q. variabilis.     

Vegetation patterns and environmental gradients in tropical dry forests of the northern Yucatan Peninsula

Patterns of plant species composition and their relationships to soil and topographic variables were investigated in tropical dry forests across the north central Yucatan, Mexico. Seven sites were studied in the oldest accessible forests along a 200–km transect oriented northwest to southeast; an eighth site was located in a little‐disturbed area located 75 km northeast of the transect. Two of the sites were on Mayan ruins. All sites were sampled using 9–24, 10m × 20m plots (<n= 132) for woody stems ≥ 3.0 cm diameter breast height. The important natural forest species were Bursera simaruba, Caesalpinia gaumeri, Gymnopodium floribundum, Piscidia piscipula, and Thouinia paucidentata. The two most important woody species in ruin woodlands were Brosimum alicastrum and Croton lundellii. Forest plots (n=108) had 17 species on average, ruin plots (n= 24) nine species. Mean basal area of stems at the forest plots (20.7 m².ha^‐1) was lower than in ruin plots (28.4 m².ha^‐1). Detrended Correspondence Analysis generally placed plots by site along the geographic transect. Natural forest plots and sites were separated from the plots on ruin sites. The five soil and topographic variables (slope, soil depth, percent surface rock, soil pH, total soil organic matter) differed significantly among sites. Plot values were correlated with DCA axe scores. Intersite floristic variation reflects an overall west to east environmental gradient affected by climate.     

小兴安岭阔叶红松混交林林隙特征

对小兴安岭阔叶红松混交林林隙基本特征进行了研究。结果表明:林隙的线状密度为31.78个/km,冠空隙和扩展林隙所占的面积比例分别为15.71%和30.78%;冠空隙的年干扰频率为0.46%,干扰轮回期约为434.8a。冠空隙的大小变化在42.12—372.52m2之间,平均为153.37m2;扩展林隙的大小变化在98.65m2—633.10m2之间,平均为300.44m2。冠空隙和扩展林隙面积分布格局均符合Weibull分布。林隙形成方式主要为干基折断,占总形成木总数的35.29%,其次为掘根风倒,占28.43%。平均每个林隙的形成木为4.98株,由红松、白桦、枫桦、冷杉形成,径级在20—30 cm之间,高度在15—30 m之间。冠空隙的直径与高度比值的相对频率的分布呈单峰型曲线,当比值为0.30—0.45时,出现峰值;而扩展林隙的直径与高度比值的相对频率的分布呈双峰型曲线,当比值分别为0.75—0.90和1.05—1.15时,出现峰值。林隙边缘木胸径级的多度分布和高度级多度分布符合Weibull分布,但不符合正态分布。约13.41%的边缘木未出现偏冠现象,偏冠率在0.5—0.7之间的边缘木占70.49%。     

Savanna fires and their impact on net ecosystem productivity in North Australia

Savannas comprise a large area of the global land surface and are subject to frequent disturbance through fire. The role of fire as one of the primary natural carbon cycling mechanisms is a key issue in considering global change feedbacks. The savannas of Northern Australia burn regularly and we aimed to determine their annual net ecosystem productivity (NEP) and the impact of fire on productivity. We established a long‐term eddy covariance flux tower at Howard Springs, Australia and present here 5 years of data from 2001 to 2005. Fire has direct impacts through emissions but also has indirect effects through the loss of productivity due to reduced functional leaf area index and the carbon costs of rebuilding the canopy. The impact of fire on the canopy latent energy exchange was evident for 40 days while the canopy was rebuilt; however, the carbon balance took approximately 70 days to recover. The annual fire free NEP at Howard Springs was estimated at −4.3 t C ha⁻¹ yr⁻¹ with a range of −3.5 to −5.1 t C ha⁻¹ yr⁻¹ across years. We calculated the average annual indirect fire effect as +0.7 t C ha⁻¹ yr⁻¹ using a neural network model approach and estimated average emissions of fine and coarse fuels as +1.6 t C ha⁻¹ yr^‐1. This allowed us to calculate a net biome production of −2.0 t C ha⁻¹ yr^‐1. We then partitioned this remaining sink and suggest that most of this can be accounted for by woody increment (1.2 t C ha⁻¹ yr^‐1) and shrub encroachment (0.5 t C ha⁻¹ yr^‐1). Given the consistent sink at this site, even under an almost annual fire regime, there may be management options to increase carbon sequestration by reducing fire frequency.     

Understorey gaps influence regeneration dynamics in subtropical coastal dune forest

Dominant understorey species influence forest dynamics by preventing tree regeneration at the seedling stage. We examined factors driving the spatial distribution of the monocarpic species Isoglossa woodii, a dominant understorey herb in coastal dune forests, and the effect that its cover has on forest regeneration. We used line transects to quantify the area of the forest understorey with I. woodii cover and with gaps in the cover. Paired experimental plots were established in semi-permanent understorey gaps with I. woodii naturally absent and in adjacent areas with I. woodii present to compare plant community composition, soil, and light availability between the two habitats. Isoglossa woodii was widespread, covering 65–95% of the understorey, while gaps covered the remaining 5–35% of the area. The spatial distribution of this species was strongly related to tree canopy structure, with I.␣woodii excluded from sites with dense tree cover. Seedling establishment was inhibited by low light availability (<1% of PAR) beneath I.␣woodii. When present, I. woodii reduced the density and species richness of tree seedlings. The tree seedling community beneath I. woodii represented a subset of the seedling community in gaps. Some species that were found in gaps did not occur beneath I. woodii at all. There were no significant differences between the sapling and canopy tree communities in areas with I. woodii gaps and cover. In the coastal dune forest system, seedling survival under I. woodii is dependent on a species’ shade tolerance, its ability to grow quickly during I. woodii dieback, and/or the capacity to regenerate by re-sprouting and multi-stemming. We propose a general conceptual model of forest regeneration dynamics in which the abundant understorey species, I. woodii, limits local tree seedling establishment and survival but gaps in the understorey maintain tree species diversity on a landscape scale.     

Gap characteristics and gap regeneration in primary evergreen broad-leaved forests of western Japan

Gap characteristics and regeneration in gaps were studied in some primary evergreen broad-leaved forests of the warm temperate zone in western Japan. Total observed 161 gaps covered 15.7% of the total land area of 8.23 ha. Gap density was 19. 6 gaps ha⁻¹ and mean gap size was 80.3 m². Smaller gaps (<80 m²) were much more frequent than larger ones, and gaps larger than 400 m² were rare. Gaps created by the death or the injury of single trees were 79.5%. Canopy trees died most often with broken trunks and not so often by uprooting or leaving standing-dead. Different types of gap regeneration behaviour were recognized among the major canopy tree species, though gap regeneration of the common evergreen broad-leaved tree species did not clearly depend on a species-specific gap size.Distylium racemosum, which occurred in equal importance (about 25%) in the canopy layer of each study stand, regenerates in gaps from saplings recruited before gap creation and can replace not only its own gaps but also gaps of other species. Therefore, it can be considered a typical climax species in this forest type of western Japan.Persea thunbergii, which can reproduce vegetatively, showed a similar type of gap regeneration behaviour.Castanopsis cuspidata can replace itself with relatively higher frequency by means of vegetative reproduction (stump sprouting) after gap creation.Quercus acuta andQuercus salicina did not regenerate under the current gap-disturbance regime. Though the frequency of uprooting is low, soil disturbance by uprooting seems to be important for the perpetuation of the pioneer tree species,Fagara ailanthoides, which recruits from buried seeds in the soil