Past global climate changes had strong regional expression. To elucidate their spatio-temporal pattern, we reconstructed past temperatures for seven continental-scale regions during the past one to two millennia. The most coherent feature in nearly all of the regional temperature reconstructions is a long-term cooling trend, which ended late in the nineteenth century. At multi-decadal to centennial scales, temperature variability shows distinctly different regional patterns, with more similarity within each hemisphere than between them. There were no globally synchronous multi-decadal warm or cold intervals that define a worldwide Medieval Warm Period or Little Ice Age, but all reconstructions show generally cold conditions between AD 1580 and 1880, punctuated in some regions by warm decades during the eighteenth century. The transition to these colder conditions occurred earlier in the Arctic, Europe and Asia than in North America or the Southern Hemisphere regions. Recent warming reversed the long-term cooling; during the period AD 1971-2000, the area-weighted average reconstructed temperature was higher than any other time in nearly 1,400 years

This manuscript consists of 21-typed pages including figure legends, 1 table and 3 figures. Takahashi et al. 2 To show what climatic conditions affect the growth of tree species in the upper and lower distribution limits of an altitudinal ecotone, tree-ring width chronologies were developed for Abies veitchii, Betula ermanii and Betula platyphylla var. japonica in their altitudinal ecotone (about 1600 m above sea level) on Mt. Norikura, central Japan.This altitude was the lower distribution limit for A. veitchii and B. ermanii in the subalpine zone, and was the upper distribution limit for B. platyphylla var. japonica in the montane zone on Mt. Norikura. The tree-ring widths of the two Betula species and A. veitchii were positively correlated with the August precipitation of the current year and that of the previous year, respectively. Precipitation in the hottest month of August was less compared with other months during summer. The tree-ring width of B. platyphylla var. japonica showed no correlation with temperatures of any month in its upper distribution limit. In contrast, the tree-ring widths of B. ermanii and A. veitchii were negatively correlated with the August temperatures of the current year and of the previous year, respectively, in the lower distribution limit of these species. Therefore, the two Betula species and A. veitchii responded to climatic conditions of the current year and those of the previous year, respectively. This study also suggests that water deficit in August reduces the growth of the three species in this altitudinal ecotone, irrespective of the upper or lower distribution limit, and that a high August temperature is more detrimental to the growth of A. veitchii and B. ermanii in their lower distribution limit. Thus, the examined three species with different altitudinal distributions differently responded to climatic conditions in this altitudinal ecotone on Mt. Norikura.

An investigation was made of the effects of tracheid dimensions on variations in the maximum density of Picea glehnii Mast., which were associated with climatic changes. Radial cell diameter and the thickness of the tangential cell walls of the last-formed cells in 90 annual rings of nine trees with different annual ring widths were analyzed by image analysis. Correlations between maximum density and tracheid dimensions indicated that changes in maximum density were due mainly to changes in cell wall thickness of the last-formed cells in annual rings and were not due to changes in radial cell diameter. The effects of climatic factors on tracheid dimensions were examined by application of dendroclimatological techniques. A chronology of cell wall thickness that represented common signals among trees was established. Simple correlation and response function analyses of the chronology revealed that cell wall thickness was influenced positively by summer temperature and negatively by precipitation in August, and these responses were similar to those of maximum density. The study demonstrated that variations in maximum density were due to variations in the cell wall thickness of the lastformed cells, which varied depending on the weather in summer.

The anatomical characteristics and density of wood were examined in 23-year-old Acacia mangium trees that had been planted in Yogyakarta, Indonesia. The seeds had been collected from trees of five different provenances. The distance from the pith of the boundary between juvenile and mature wood was also examined to clarify the maturity of the wood. Lengths of wood fibers near the pith and the distance from the pith of the boundary between juvenile and mature wood differed significantly among provenances. By contrast, other anatomical characteristics of the wood such as fiber wall area, fiber wall thickness, fiber diameter, vessel lumen area, vessel diameter, vessel frequency and wood density did not differ significantly among provenances. Wood density was strongly correlated with the area of fiber walls. Our observations suggest that Sidei and Daintree might be more appropriate provenances among those examined for the Acacia mangium treebreeding programs in Indonesia that are aimed at improving wood quality, because these provenances are associated with longer initial wood fibers and narrower juvenile areas than the other provenances studied.

Takahashi et al. 2Tree-ring width chronology of Betula ermanii was developed at the timberline (2400 m a.s.l.) on Mount Norikura in central Japan, and climatic factors affecting the tree-ring width of B. ermanii were examined. Three monthly climatic data were used for the analysis (i.e., mean temperature, insolation duration, and sum of precipitation). The tree-ring width of B. ermanii was negatively correlated with December and January temperatures and with January precipitation prior to the growth. However, why high temperatures and much snow in winter had negative effects on the growth of B. ermanii is unknown. The tree-ring width was positively correlated with summer temperatures during June to August of the current year. The tree-ring width was also positively correlated with the insolation duration in July of the current year. On the contrary, the tree-ring width was negatively correlated with summer precipitation during July to September of the current year. However, these negative correlations of summer precipitation do not seem to be independent of temperature and insolation duration, i.e., much precipitation reduced the insolation duration and temperature. Therefore, it is suggested that much insolation duration and high temperature due to less precipitation in summer of the current year increase the radial growth of B. ermanii at the timberline.The results were also compared with those of our previous study conducted in the lower altitudinal limit of B. ermanii (approximately 1600 m a.s.l.) on Mount Norikura. This study suggests that climatic factors increasing the radial growth of B. ermanii differ between its upper and lower altitudinal limits.Key words: Betula ermanii; climatic conditions; dendrochronology; Mount Norikura; timberline; tree-ring width chronology. Takahashi et al. 3 INTRODUCTIONTimberlines are the most severe climatic conditions such as low temperature, strong winds and short growing season for plants along altitudinal gradients, and therefore, the regeneration of trees at the timberlines is supposed to be sensitive to climatic changes (e.g., Briffa & Osborn 1999;Kusnierczyk & Ettl 2002;Lloyd & Fastie 2002;Chapin et al. 2004;Daniels & Veblen 2004;Dullinger et al. 2004). For example, some researchers have discussed that frequent seedling establishment occurred in warmer and/or more mesic years near the timberlines (Kullman 1986;Taylor 1995;Kajimoto et al. 1998;Camarero & Gutiérrez 1999;Gervais & MacDonald 2000). In addition, many dendrochronological studies have revealed that radial growth of trees near the timberlines increased in ameliorated years with higher temperatures than the average (Ettl & Peterson 1995;Gostev et al. 1996;Buckley et al. 1997;Peterson & Peterson 2001;Wilson & Hopfmueller 2001). Trees may respond to climatic conditions more quickly in growth rather than seedling establishment. Thus, the growth is a useful indicator to examine how climatic conditions affect regeneration of trees at the timberlines. Although many researchers investigated the effects of climat...