Abstract. The vessel areas of ten beech trees growing on a dry site were measured separately for all tree rings using automatic image analysis. These data were correlated with the monthly amount of precipitation from the July prior to the growing season until the August of the current growing season. It is evident that vessel formation at the beginning of cambial activity is mainly controlled by internal factors. The rainfall in the previous summer and autumn and in the contemporary May had only a slight influence. Vessel formation towards the end of the cambial activity is strongly influenced by the July rainfall and is thus determined to a greater degree by external factors. These results are discussed on the basis of hypotheses of tree physiology.

From a network of teak chronologies in northern Thailand, 75 trees within one province were evaluated regarding their climatic signal. The raw tree-ring series revealed a high mean sensitivity of 0.50 and a moderate first-order autocorrelation of 0.48. The first principal component of the standardized data explained 44% of the total variation in the tree-ring data, indicating a considerable climatic influence on tree growth. The climate-growth relationship suggested that growth of teak in this study area is mainly controlled by rainfall from April to June. Thus, there is some promise that the whole network of teak chronologies in northern Thailand can contribute to reconstructing climate over at least the last three centuries.

The wood formation of kapur (Dryobalanops sumatrensis) and tembaga (Shorea leprosula), growing under a weak seasonal climate in West Malaysia was studied over a four-year period using cambium marking. Technical problems arose from the heavy callus formation due to the wounding of the cambium, the small radial increment, and the high variability of the cambial activity around and along the stem. Wood formation in the two tree species appears to be a continuous process and not related to seasonality in rainfall and phenology.

Information on the vegetation and landscape history of a region is often limited, and available data are hard to interprete. A concept is presented here on how a more comprehensive picture of the structure and development of landscapes and vegetations of the past can be gained by integrating the information of several disciplines. Archaeological field methods have been combined with methods used in landscape studies (geology, soil science, micromorphology) and vegetation studies (ecology, palynology and dendrochronology).This concept has been applied and tested during an integrated study of a buried woodland at Zwolle-Stadshagen (Province of Overijssel, the Netherlands). Many large wood remnants were found in a peat layer preserved below a thick clay deposit. The wood remnants were dated by using dendrochronology to the period between ca. 150 BC and AD 580 (ca. 2200 - 1400 cal. BP). Two phases could be distinguished in the development of the peat. The woodland consisted of a closed stand with ash, alder and oak as main species, in the first phase mostly resembling an alder carr, and in the second one the near-extinctFilipendulo-AlnetumPassage et Hofmann 1968. No evidence of exploitation of the woodland by man nor of animal foraging was found.The followed integrated procedure has led to a more substantiated reconstruction of the palaeo-environment with its wetland wood, but also of the influence of human activities on the palaeo-landscape and its woodlands, that could not have been obtained otherwise.