In combining several pieces of wood for works in carpentry and cabinet-making, the different circumstances of the plank as respects its length and width should be always borne in mind. Provision must be made that the shrinking and swelling are as little restrained as possible, otherwise the pieces may split and warp with an irresistible force: and the principal reliance for permanence or standing, should be placed on those pieces, (or lines of the work,) cut out the lengthway of the plank, which are, as before explained, much less disposed to break or become crooked, than the crossway sections: these particulars will be more distinctly shown by one or two illustrations.

Let a, b, c, d, represent the flat surface of a board: e,f, the edge of the same, and g, h, the end; no contraction will occur upon the line e,f, or the length, and in the general way, that line will remain pretty straight and rigid; but the whole of the shrinking will take place on g, h, the width, which is slender, flexible, and disposed to become curved from any unequal exposure to the air; the four marginal lines of a, b, c, d, are not likely to alter materially in respect to each other, but they will remain tolerably parallel and square, if originally so formed.

A dove-tailed box consists of six such pieces, the four sides of which, A, B, C, D, fig. 19, are interlaced at the angles by the dove-tails, so that the flexible lines, as g, h, on B, are connected with, and strengthened by, the strong lines, as c, d on A, and so on: the whole collectively form a very rigid frame, the more especially when the bottom piece is fixed to the sides by glue or screws, as it entirely removes from them the small power of racking upon the four angles, (by a motion like that of the joined parallel rule,) which might happen if the dove-tails, shown on a larger scale in fig. 20, were loosely fitted

Fig. 18.

Section II Combining Different Pieces Of Wood 10016

Fig. 19.

Section II Combining Different Pieces Of Wood 10017

Fig. 20

Section II Combining Different Pieces Of Wood 10018

When the grain of the four sides, A, B, C, D, runs in the tame direction, or parallel with the edges of the box or drawer, at shown by the shade lines on A and B, and the pieces are equally wet or dry, they will contract or expand equally, and without any mischief or derangement happening to the work; to ensure this condition, the four sides are usually cut out of the same plank. But if the pieces had the grain in different directions, as C and D, and the two were nailed together, D would entirely prevent the contraction or expansion of C, and the latter would probably be split or cast, from being restrained. When admissible, it is therefore usual to avoid fixing together those pieces, in which the grain runs respectively lengthways and crossways, especially where apprehension exists of the occurrence of swelling or shrinking.

A wide board, fig. 21, composed of the slips, A, B, C, D, E,

(reversed as in diagram, fig. 17, page 51,) is rendered still more permanent, and very much stronger, when its ends are confined by two clamps, such as G, H, (one only seen;) the shade lines represent the direction of the grain. The group of pieces, A to E, contract in width upon the line A, E, and upon it they are also flexible, whereas the clamp G, H, is strong and incapable of contraction in that direction, and therefore unless the wood is thoroughly dry the two parts should be connected in a manner that will allow for the alteration of the one alone. This is effected by the tongue and groove fitting as represented; the cud piece, G. H, is sometimes only fastened by a little glue in the center of its length, but in cabinet-work, where the seasoning of the wood is generally better attended to, it is glued throughout.

If the clamp G, H, were fixed by tenons, (one of which i, j, is shown detached in fig. 22,) the contraction of the part of the board between the tenons might cause it to split, the distance between the mortises in G, H, being unalterable: or the swelling of the board might cause it to bulge, and become rounding; or the entire frame would twist and warp, as the expansion of the center might be more powerful than the resistance to change in the two clamps, and force them to bend.

Fig. 21.

Section II Combining Different Pieces Of Wood 10019

Fig. 22.

Section II Combining Different Pieces Of Wood 10020

It is therefore obvious that if any question exist as to the entire and complete dryness of the wood, the use of clamps is hazardous; although in their absence, the shrinking might tear away the wood from the plain glue joint, even if it extended entirely across, without causing any further mischief, but more generally the shrinking would split the solid board.

Another mode of clamping is represented at K; it is there placed edgeways, and attached by an undercut or dove-tailed groove, slightly taper in its length, and is fixed by a little glue at the larger end, which holds the two in firm contact: each of these modes, and some others, are frequently employed for the large drawing boards required by architects and engineers for the drawings, made with squares and instruments.

From a similar motive, the thin bottom of a drawer is grooved into the two sides and front, and only fixed to the back of the drawer by a few small screws or brads, so that it may swell or shrink without splitting, which might result were it confined all around its margin. It is more usual, however, to glue thin slips along the sides of large drawers, as in fig. 23, which strengthen the sides, and being grooved to receive the bottom, allow it to shrink without interfering either with the front or back of the draw

In an ordinary door with two or more panels, all the marginal pieces ran lengthways of the grain: the two sides, called the stiles, extend the whole height, and receive the transverse pieces or rails, now mortised through the stiles, and wedged tight, but without risk of splitting, on account of their small width; every panel is fitted into a groove within four edges of the frame. The width of the panel should be a trifle less than the extreme width of the grooves, and even the mouldings, when they are not worked in the solid, are fixed to the frame alone, and not to the panel, that they may not interfere with its alterations; therefore in every direction we have the frame-work in its strongest and most permanent position as to grain, and the panel is unrestrained from alteration in width if so disposed.

This system of combination is carried to a great extent in the tops of mahogany billiard tables, which consist of numerous panels about 8 inches square, the frames of which are 3 1/2 in. wide and 1 1/2 in. thick; the panels are ploughed and tongued, so as to he level on the upper side, and from their small size the individual construction of the separate pieces is insignificant, and consequently the general figure of the table is comparatively certain. Of late years, I am told *, that slate, a material uninfluenced by the atmosphere, has been almost exclusively used; the top of a full sized table, of 12 by 6 feet, consists of four slabs one inch thick, ground on their lower, and planed by machinery on their upper surfaces: the iron tables are almost abandoned for several reasons. Large thin slates, from their permanence of form, are sometimes used by engineers and others for drawing upon, and also in carpentry for the panels of superior doors.