22 TRCHNIOAL PHOrKliTIIOS OP WOOD. 



uniform throughout its mass, the change of volume is not accom- 

 panied by any change of form. But if some parts shrink or expand 

 more or less than others, a change of form necessarily occurs, or, in 

 technical language, the wood warps. It hence follows that the 

 extent to which a wood is liable to warp is in direct relationship 

 with the extent to which it shrinks or expands with loss or gain 

 of moisture, and we thus find that the softer and lighter woods 

 warp less than those which are harder and heavier. Boards sawn 

 parallel to a radius, since the tissues are thus uniformly distributed, 

 are less disposed to warp than boards sawn parallel to a tangent, 

 which no amount of care will prevent from warping; and similarly, 

 among the latter class of boards, those taken off furthest away from 

 the centre of the log warp most. Boards and scantlings cut out 

 of trees with twisted fibre always warp very badly. Even-grained 

 wood will warp less than wood wanting in uniformity of structure ; 

 bamboos and canes are examples of wood possessing conspicuously 

 uniform structure. Warping may be prevented or minimised by 

 steaming the wood (this, however, reduces its strength), or by 

 impregnating it with oil, or, instead of making an article of a single 

 piece of wood, by composing it of several pieces so as to secure 

 every possible direction for the run of the fibres, and thus coun- 

 teract any tendency to warp in any one direction. 



Cracking and splitting. — If in unequal contraction the different 

 parts of a piece of a wood cannot move and keep together, and the 

 force with which they are drawn apart from one another is great 

 enough to overcome the cohesion between them, one or more 

 cracks result. Such cracks are most numerous along radii or lines 

 of easiest fission, and least so parallel to the concentric rings of 

 growth or lines of most difficult fission. The size of the cracks 

 increases (a), with the rapidity with which the wood dries and 

 shrinks (timber felled during the rains or winter has fewer cracks 

 than timber felled at any other season) ; (&), with the extent of the 

 shrinkage (c) ; with the removal of the bark before seasoning has 

 made any progress ; (d), with the diameter of the log or breadth 

 and thickness of the scantling ; and (e), with the want of uniformity 

 in the structure of the wood (the uniform-textured sounding board 

 of musical instruments, looked after properly, scarcely ever cracks). 

 Wood in the round is most of all subject to cracks. This tendency 

 in logs may be diminished by rough-squaring, so as to leave conti- 

 nuous strips of bark at the corners ; such treatment, although not 

 preventing the formation of numerous little cracks, checks that of 

 large ones, which often render the wood useless for many purposes. 

 If rough-squaring cannot be resoi'ted to, then in dry climates it 



