-

2. Percentage of Water contained: Moistness and Seasoning. — The green unseasoned timber of hardwoods contains on the average water to the extent of about 42 per cent of its total weight, that of softwoods about 52 per cent, and that of Conifers about 57 per cent. Hence green wood in general may be said to contain water to the extent, in round numbers, of about half its weight; but the actual quantity varies with the kind of tree, season of the year, part of the tree, soil and situation, &c. It is never pure water, but here too the kind and quantity of sap-substances it contains vary also according to the tree, season, soil, &c. This water, contained throughout the tissues and in the intercellular spaces, begins to diminish in quantity by evaporation into the air after a tree is felled; and this evaporation of the imbibition-water contained in the wood goes on, at a gradually diminishing rate, until a condition is reached when there is more or less of equilibrium between the tension of the atmosphere and the power of evaporating the water still contained in the wood. When this condition is attained the wood is said to be seasoned or air-dried, but it still contains a considerable quantity of water, varying according to the nature of the woody tissue and the other substances within the wood, and to their power of retaining water. It varies from 8 to 12 per cent of the weight, but is greater in coniferous than in broad-leaved wood owing to the resin that it contains; and the more resinous the wood, the larger is the quantity of water retained. To differentiate this ordinary residual portion from the water of imbibition originally present when the tree was alive in the full flow of sap, it has been called hygroscopic water, because its quantity constantly varies proportionately with the atmospheric moisture. This natural residuum can only be expelled from the wood by artificial dessication (see p. 533).

Wood in the round usually takes from two to four years to season, and unbarked logs still longer. Logs that have been floated absorb more water than they originally contained, but when afterwards stored in depots they dry sooner than other timber. The evaporation of the water takes place more quickly longitudinally (in the run of the fibres) than radially. Light porous wood of course seasons sooner than heavy close-grained wood, and sapwood usually sooner than heartwood. To dry thoroughly, wood should be well raised above the ground; and the freer the circulation of air around it, the more rapid will the seasoning take place. Split or sawn wood of course seasons much sooner than wood in the round; and the larger the surface exposed to the air, the quicker it parts with its moisture. To save time and money it is customary to artificially season all the finer kinds of wood in special drying-chambers by means of hot air, as this does not diminish its strength.

3. Relation towards Water: Shrinkage, Cracking, Warping, and Expansion. As it loses water in seasoning, wood shrinks or decreases in volume as soon as the water contained in the woody tissue begins to be evaporated, after that contained in the intercellular spaces has been parted with. The more water the woody tissue gives off, the greater must be the shrinkage in volume, and the

1 Hartig's investigations showed the following results as regards percentage of water by volume:

greater the tendency to warp; while the extent to which the shrinkage and warping take place depends mainly on the anatomical structure of the wood and the extent to which it parts with the water contained. As the shrinkage depends on the rate and extent of seasoning, the dense slow-drying heartwood shrinks more gradually and to a less extent than sapwood, while shrinkage is less in Conifers (and especially when very resinous) than in the wood of broad-leaved trees. But no connection can be proved between the specific weight of green or dry wood and its shrinkage, nor any usual proportion between the quantity of water contained and the shrinkage of wood generally.

Nördlinger classified wood in the following scale with regard to shrinkage while seasoning:

Shrinking most (5 to 8 per cent of the volume when green): Walnut, Lime, Beech, Hornbeam, Elm, Chestnut, Wild Cherry, Turkey Oak, Alder (?), Birch, Apple

tree.

Shrinking moderately (3 to 5 per cent of the volume when green): Maple and Sycamore, Austrian Pine, Scots Pine, Poplar, Yew, Horse-Chestnut, Ash, Aspen, Sessile Oak, Robinia.

Shrinking least (2 to 3 per cent of the volume when green): Weymouth Pine, Spruce, Larch, Silver Fir, Arborvitæ, Pedunculate Oak (?).

He also found that there was no practical difference between the shrinkage of wood that had been floated for a short time and of wood not floated at all.

The shrinkage longitudinally (in the direction of the fibre) is only nominal; but radially (in the direction of the radius) it can amount to about 6 per cent, and tangentially (in the direction of the circumference) to about 10 per cent. And, of course, the greater and the more rapid the shrinkage, the greater will be the number and the depth of the cracks thus made in the wood. Hence the timber of trees felled in summer, while the sap is in full flow, is more apt to be damaged by shrinkage than that of those felled in autumn or winter, and barked logs crack more than those on which the bark is wholly or partially left (see log-ends, Fig. 136).

The following table gives the results of investigations made regarding shrinkage in different directions (Exner) and as regards total volume (R. Hartig, Frey, Schwappach) :

:

1 An abstract of Schwappach's investigations into the Specific Gravity and Resistance to Crushing will be found in Transactions of the Royal Scottish Arboricultural Society, vol. xv., part iii., 1898, pp. 279-291.

In consequence of the want of uniformity in the shrinkage in different directions, the wood is liable to crack or split while still in bulk, and to warp when converted. The larger the logs, beams, planks, and scantlings, the more liable they are to become cracked, especially at the ends; while the smaller the pieces of wood, and the further a beam or a board is cut from the centre of the stem, the more likely it is to warp, owing to the greater amount of tangential shrinkage in its fibres. The slower the process of drying, the fewer and the slighter are the cracks formed during seasoning. Cracks in the ends of logs can be prevented by painting or smearing these with any greasy or tarry substance to check evaporation. In timber-yards the logs are sawn into planks, scantlings, sleepers, &c., of standard sizes and put under shade to season, while the cracking of the ends can be minimised by greasing them or driving S-shaped clamps into them (see p. 533). The wood of Conifers, especially when very resinous, is less liable to warp than that of broad-leaved trees, and heartwood generally warps less than sapwood. By steaming wood, besides being thereby made much more flexible, it also becomes less liable to warp, without its other technical properties being impaired. Beech, Ash, Oak, Walnut, and other furniture woods are therefore treated in this manner to make them more flexible, less liable to warp, and of a darker and better ornamental colour.

In proportion as wood gradually shrinks when seasoning, it also, in consequence of its hygroscopicity, gradually expands again by the absorption of water during a moist condition of the atmosphere. Although there are differences between the rate of shrinking and of expanding, yet for all practical purposes the co-efficient of shrinkage may also be taken as that of expansion for any given kind of wood. Converted wood therefore expands much more tangentially than radially, and very much more radially than longitudinally; and expansion must be allowed for in fixing beams and other wood-work in house-building, &c.

The absorption of water and the expansion of seasoned wood do not, however, take place at the same rate. Weissbach's investigations showed that while the expansion was completed within 1 to 2 months, the weight increased (by absorption of water) for about 6 months, and sometimes for 2 to 3 years. The variation in specific weight, volume, and absolute weight in consequence of saturation with water is thus shown by Exner :

Specific gravity.

:-

In weight in percentage of volume (Hampel).

Defects and Unsoundness.-The value of wood for technical purposes is often diminished by defects due to structural and other conditions, and by unsoundness occasioned by disease. Such defects include branch-knots, twisted fibre, old wound-surfaces, and shakes, or internal cracks usually formed in the tree before the seasoning of the timber begins; while unsoundness in the form of rot of one kind or another is usually directly due to saprophytic and parasitic fungi.

(1) Branch-knots are a common defect in many classes of British timber. As

the national system, centuries old, of growing Oak for shipbuilding was to give each tree a large growing-space, this method was (and still is) applied generally to other trees, and usually results in the formation of large branch-knots, which make the timber difficult to work. They improve the ornamental qualities of wood for furniture, &c., and increase its density, weight, and toughness; but they impair the quality and diminish the market-value of coniferous timber for general constructive purposes. Timber free from knots is therefore more useful for many, but not for all, purposes; it gives longer logs, and it involves less waste in conversion.

The more shade-enduring a tree is, the more likely it is to have snags of dead branches embedded in the wood of the stem, which are apt to fall out when the timber is converted into boards and dries thoroughly. The knots in Conifers become saturated with resin, and are then, especially in Larch, exceedingly hard to plane and work.

(2) Twisted Fibre may occur in any kind of tree, but it is commonest in Horse-Chestnut, Sweet-Chestnut, Oak, Sycamore, Elm, Beech, and White Poplar (also Walnut and Pear-tree) among broad-leaved trees, and in Scots Pine among Conifers. Usually the twist goes from left to right, especially in the Horse-Chestnut, Sycamore, and Scots Pine, though it often also runs from right to left. Planks, beams, and scantlings sawn from trees with twisted fibre are more liable to warp, and are less strong than those cut from straight-fibred stems. Sycamore boles with a slight twist are often preferred for mill-rollers.

(3) Wound-surfaces and Rind-galls that have become completely occluded by cicatrisation often fail to unite with the older wood, and then form blemishes and weak spots when the timber is converted into scantling, beams, and planks. When the defect is serious, it often necessitates the bad part being cut off altogether, as useless for timber.

(4) Shakes are sometimes to be found in the shape of radial clefts (heartshakes), or of cracks following the circumference of old internal annual rings (cup- or ring-shakes), or of longitudinal clefts (frost-shakes), and in each case they greatly diminish the technical quality and value of the wood. Heart-shakes may be either star-shaped if several clefts are produced near the centre of the stem, or simple if they occur merely singly between the centre and the circumference, though both kinds may be found together, especially in large stems. Heart-shakes are often formed, especially in old Oaks and Sweet-Chestnuts, long before the trees are felled (in which case they will be found on the stump too); but they may also be produced by the jar or shock when the tree crashes on the ground, or by shrinkage during seasoning. The liability to shake is less in winter-felled wood than in summer-felled, and less when the axe is used than when felling takes place with the saw. To what extent heart-shakes may interfere with the conversion of the wood depends of course on the nature and size of the cracks, but star-shaken logs are spoiled for the sawing of thin planks. Cup- or Ring-shakes, whether due to internal tension and rupture, wind, frost, or fungi, spoil, as far up as they extend, the wood for sawing into planks and beams, &c. This class of defect occurs chiefly in old trees, and especially among Sweet-Chestnut, Beech, Oak, Elm, Willows, and Poplars. Frost-shakes occur oftenest on Sweet-Chestnut, Oak, Elm, HorseChestnut, Beech, and Lime, and seldom become occluded without enclosing rot within them, so that a considerable portion of the wood is thereby rendered useless.

Unsoundness or Rot in timber is a diseased condition, due to the injurious action of fungi (see pp. 182-4, 187, 190). Whereas saprophytic fungi only attack lifeless wood (e.g., branches, dead-wood at wound-surfaces, &c.), they sometimes develop into parasites and attack the living tissues in the same way as the specially parasitic

1 See also Part IV., Protection of Woodlands, p. 206.

fungi which effect their entrance at wound-surfaces, however minute. Even in the incipient stages of discase the unsoundness of the tissue is generally sufficient to discolour the wood, and often to affect it as timber, while of course in all the advanced stages of decay up to actual rottenness it is useless as timber, and often worth little even as fuel. In many kinds of wood incipient decomposition produces a dark-blue, a coppery-green, or a very dark colour near the centre of the stem, or along star-shakes (as in Maple, Sycamore, and Elm), or along the outer edge of the heartwood (as in Elm). From deep, old, open wounds, too, a dark-coloured product of decomposition often percolates downwards and produces "false heartwood" of specially dark tint in the lower portions of the tree. A somewhat more advanced stage of decomposition sometimes causes the wood to become greenish (as in Oak, Birch, and Beech), but the most destructive of all the diseases of timber-trees are those occasioning the various kinds of canker and of red- and white-rot, which may occur on or in the stem, the roots, and the branches of all kinds of trees. The fungous diseases of trees have been fully dealt with in Part IV., Chap. V., and it need only be remarked here that

Canker is caused chiefly by Peziza (Dasyscypha) Willkommii (Larch); Nectria ditissima (Ash and Beech mostly, but also other broad-leaved trees); Aecidium elatinum (Silver Fir); Cronartium (Peridermium Pini corticola: Scots Pine).

Red-rot, owing to the decomposition of the cellulose, is caused chiefly by Trametes Pini (Scots Pine), Fomes annosus (Trametes radiciperda: Scots and Weymouth Pines, Spruce, Silver Fir, and also broad-leaved trees); Polyporus sulphureus (Oak and Birch chiefly); P. vaporarius (Spruce, Silver Fir); P. betulinus (Birch); P. sistotrematis (Scots and Weymouth Pines).

White-rot, owing to the decomposition of the lignin, is caused chiefly by Fomes (Polyporus) igniarius (Oak and Willows mostly, but also other broad-leaved trees); F. fomentarius (Beech; Oak and Elm); F. fulvus (Plum-trees; Aspen, Hornbeam); Polyporus dryadeus (Oak); P. squamosus (most broad-leaved trees); Agraricus melleus (all kinds of trees); Hydnum diversidens (Oak, Beech).

White-piping, or rings of white decomposing wood spreading from the branches round the stem and appearing as long whitish or yellowish stripes on a longitudinal section, is caused by Stereum hirsutum (Oak).

Blueing of Conifer timber is caused by Ceratostoma piliferum (especially common in Scots Pine sapwood).

Root-rot is often produced in consequence of unsuitable soil (e.g., as in Larch on moist or imperfectly drained land, and often even on gravelly soil), but it is not of much consequence technically unless it spreads up into the bole and makes the wood "foxy" or "dosed," or still worse, "pumped" and hollow. Branch-rot is also in itself of comparatively little consequence unless it affects the wood of the stem, as it is very apt to do; hence the advantage of treating all such woundsurfaces antiseptically, after pruning broken branches, whenever necessary. It is sometimes impossible to tell from the outward appearance of a tree whether or not it be sound in the stem; but if tapped sharply with the back of an axe, extensive stem-rot will make its presence known by a somewhat hollow sound quite different from the dull, solid tone emitted from a sound, healthy trunk. Very old trees are of course much more likely to be unsound than younger trees still in vigorous growth.

Dry-rot is often occasioned, both in wood in the round and in converted timber, by Merulius lacrymans and other saprophytic fungi; and the more the wood is exposed to warmth and moisture simultaneously, the greater is the chance of its becoming attacked. (See p. 187, also Agricultural Leaflet No. 113—Dry-Rot.)

III. Mechanical Properties or Relation towards External Influences.— In many respects these are by far the most important of all the factors in determining the technical properties or quality of timber required for con