The chief species studied were paper birch, poplar (Populus tremuloides), jack pine, white spruce, and balsam fir.Methods for the study of water contents are described. Determinations of densities and swelling percentages are summarized. Conversion factors that may be employed to convert moisture contents based on dry weight into percentages of original volume are:—for jack pine 0.38, for balsam 0.315, for poplar 0.42 and for birch 0.49.The hardwoods examined show a maximum water content in spring and a sharp drop in the summer. This appears to vary from year to year and the possible reasons for this variation are discussed. In 1931 birch and poplar lost half their total (spring) water during the summer months. In birch this may not be made up until the following spring. The softwoods show no marked seasonal changes in water content.The distribution of water is characteristic for each species. Changes in distribution throughout the year have been followed. In birch all parts of the wood (there is no heartwood) join in the seasonal changes; in poplar only the sapwood varies in water content. The results of individual year-ring analyses and of borings at different heights point to uniform water content in corresponding parts of the tree.Diurnal changes in water content have been investigated and rapid fluctuations recorded. These point to a decrease during the morning followed by an increase later in the day. These variations are correlated with tension changes and no doubt also with transpiration. It seems certain that the actual amount of gas in the tree varies but little during the diurnal changes, though it does vary with the seasonal fluctuations in water content.Girdling of birch, balsam and spruce is described and the effects on water contents are followed. It is shown that in the case of birch, removal of wood to a depth of more than one inch leads to little change during two seasons. This is correlated with the continued activity of all parts of the wood. In balsam, almost complete drying of the sapwood within two or three months follows girdling through the sapwood. The characteristic wet patches of balsam heartwood, however, are unaffected, and it is concluded that these have no connection with the sapwood and so play no part in water conduction. The results from spruce are irregular.Experiments on seasoning and flotation in the field and in the laboratory are described. The summer seasoning of "sour-felled" birch is more rapid than that of normal or of peeled logs or the normal water loss of living standing trees, and this must be due to evaporation from the leaves.The effects of log length, of barking, and of end and/or side painting on rate of penetration of water have been investigated. While penetration of water is chiefly through the ends of logs, escape of dissolved air is largely in the radial direction, and so end penetration is less important than might be expected. There is considerable top drying from unseasoned floating logs (in laboratory tanks), which may assist in solution and removal of air and so hasten rather than slow up sinkage of the log. Seasoning followed by end painting results in very slow entry of water and so is excellent in flotation.