Trees (2006) 20: 34–41 DOI 10.1007/s00468-005-0010-x ORIGINAL ARTICLE Johannes Sorz · Peter Hietz Gas diffusion through wood: implications for oxygen supply Received: 1 February 2005 / Accepted: 25 May 2005 / Published online: 23 August 2005 C  Springer-Verlag 2005 Abstract Living tissue in tree stems has to be supplied with oxygen, which can be transported upwards with the transpiration stream; but in times of zero sapflow, the only source is the oxygen stored or diffusing radially through bark and xylem. We measured radial and axial diffusion of oxygen against nitrogen gas in wood of coniferous (Picea abies (L.) Karst. and Taxus baccata L.), ring-porous (Quer- cus robur L. and Fraxinus excelsior L.) and diffuse-porous (Fagus sylvatica L. and Carpinus betulus L.) trees at dif- ferent water and gas contents in the laboratory. The diffu- sion coefficient (D) in radial direction was mostly between 10 −11 and 10 −7 m 2 s −1 and was strongly related to the gas content. At 40% gas volume, D increased 5–13-fold in Picea, Taxus and Quercus, 36-fold in Fraxinus, and about 1000-fold in Carpinus and Fagus relative to D at 15% gas volume. In the axial direction, diffusion was 1 or 2 orders of magnitude faster. Between-species differences in diffu- sion velocities can largely be explained by wood structure. In general, D was lowest in conifers, highest in diffuse- porous and intermediate in ring-porous hardwoods, where the large vessels were mostly blocked by tyloses. Model calculations showed that at very high water content, radial diffusion can be too low to ensure the supply of respiring sapwood with sufficient oxygen and an important function of gas in living stems appears to be the supply of oxygen through storage and diffusion. Keywords Gas diffusion . Oxygen supply . Wood structure J. Sorz · P. Hietz () Department of Integrative Biology, Institute of Botany, University of Natural Resources and Applied Life Sciences (BOKU), Gregor Mendel-Str. 33, 1180 Vienna, Austria e-mail: peter.hietz@boku.ac.at Tel.: +43-1-47654-3154 Fax: +43-1-47654-3180 Introduction Oxygen is required for oxidative respiration, which un- der most conditions provides the energy for plant cells. Plants growing in submerged or waterlogged soil often show anatomical adaptations for the transport of oxygen to below-ground parts, and gas flow can be substantially en- hanced by Venturi-ventilation (Strand and Weisner 2002) and thermo-osmosis (Buchel and Grosse 1990; Grosse et al. 1992). Tree stems are normally exposed to light and air, but unless the bark is transparent, they do not produce oxygen. Stem tissue respiration in the growing season can be sub- stantial (Stockfors and Linder 1998; Teskey and McGuire 2002; Gansert 2004) and bark, cambium and wood may pose considerable, but mostly unquantified, barriers to gas diffusion. To supply live cells in the sapwood, oxygen can either diffuse radially through periderm, phloem, cambium and wood, or be taken up by the roots and transported up- wards with the transpiration stream. In some tree species adapted to waterlogged soil, the cambium has small in- tercellular spaces permitting oxygen supply through the bark (Hook and Brown 1972; Buchel and Grosse 1990). In other species lacking such intercellular spaces, the cam- bium plus bark appear to be quite impermeable to gases and the transpiration stream is supposed to be the main source of oxygen for the xylem (Hook et al. 1972). More recently, direct measurement of oxygen in the stem support this idea (Eklund 2000; del Hierro et al. 2002; Mancuso and Marras 2003; Gansert 2003), though in times of zero sapflow, such as during the night and on wet and cool days, only the oxygen either diffusing radially or present in gas spaces or dissolved in water is available. All sapwood contains, by definition, living cells but not all sapwood conducts water (Phillips et al. 1996). Thus, whether oxygen passes through the cambium or is trans- ported via the transpiration stream, it has to diffuse through wood to reach parts of the inner sapwood with no sapflow. Indeed, the oxygen deficiency in the innermost part of the sapwood has been implicated in heartwood formation (Eklund and Klintborg 2000). By contrast, Knoke (2003) suggested that dark heartwood in Fagus sylvatica forms