rapidly and begin to warp in advance of the rest of the lumber. If stain conditions are severe, stain will appear in the crossings, unless stickers are inserted between the boards at this point. Lumber is left only a few days in such racks and when taken down is flatpiled and allowed to finish seasoning. This method of handling, thoroughly carried out, reduces stain very materially with some species at some seasons; in other cases considerable stain occurs. It costs $1 a thousand or more additional to put lumber through this

process.

The report on Air Seasoning of Oak and Gum in the Southern Hardwood Region devotes considerable space to various processes for steaming lumber before piling for air seasoning as quite effective in reduction of stain, providing they are so conducted as to secure rapid surface drying of the lumber; where moisture stands on the lumber local areas of severe stain may occur. The lumber is sometimes steamed in pressure cylinders (Kraetzer process), live steam at 20 to 30 pounds pressure being applied for 30 to 45 minutes, or sometimes steamed in a hot box or kiln at atmospheric pressure, usually for 12 hours, though the report suggests that 4 hours would probably be sufficient. The report also mentioned that the practice of one company in kiln-drying product 36 to 48 hours at temperatures of 140° to 180° F. previous to piling for air drying was effective in preventing stain and gave excellent results in reducing the time for air seasoning.

In general, the most common method of treating lumber for air drying to prevent the development of blue stain is dipping in a chemical solution, usually of soda carbonate or bicarbonate. This is more common for softwoods than for hardwoods; in the above report only two hardwood firms were mentioned as using dipping, and one of these was using a solution of mercury bichloride instead of soda. The usual method is to provide a tank in the conveying chain at the tail of the mill, allowing the lumber to run into it and out again. A by-pass is usually arranged so that lumber to be kiln dried is carried over the tank instead of through it, though sometimes the tank is set outside the main line of lumber travel and lumber intended for dipping is diverted to it. Dipping usually heavily reduces the amount of stain, which it does not entirely prevent but is chiefly effective in preventing surface infection; if the stain is already inside the lumber, it may continue to develop in the interior. Dipping slightly discolors the lumber with the light-brown stain of the chemicals used, but buyers usually accept this in preference to the more objectionable blue stain, and it mostly disappears when the lumber is dressed. The cost of dipping has been variously estimated as 16 cents to $1 a thousand, and the additional value produced or conserved in the lumber by efficient dipping may amount to $3 to $6. Inefficient dipping consists in using cheap chemicals, or too weak a solution, or applying it at too low a temperature. In the National Committee on Wood Utilization's publication entitled "The Sap Stains of Wood and Their Prevention will be found a detailed description of sap stain on lumber.

Whether the lumber is end-racked, dipped, steamed, or partially kiln dried before being flat piled to finish seasoning, the nature of the piling has much to do with the amount of stain that subsequently

develops and should be such as to favor air circulation. The Inland Empire air-seasoning study, referring to the wide piles used in that

region, says:

Four to five chimneys, 7 to 8 inches in width, with the stock piled almost tight together in the sections between, constitute the best practice.

This refers more especially to random-width stock, and it is stated elsewhere that this method results in less surface checking as well as less stain. It should be thoroughly tested in connection with stain in other regions.

The southern hardwood air-seasoning study gives some data in connection with tests of 2-inch stickers in comparison with 1-inch thickness. The wood was sap gum, and the 2-inch stickers did not keep stain out of the pile, though it dried with less degrade than two other piles on 1-inch stickers, one of them being of stock which had been given a preliminary kiln drying. The total degrade, including stain, was much less in the pile which had received the preliminary kiln drying.

Blue stain reduces the value of lumber even in those species and grades where it is not considered a defect, and bright lumber commands a premium over that which shows serious stain. It affects not only hardwoods but a number of species of softwoods.

Replies to the questionnaire included 17 that indicated blue stain as a source of material damage, while 26 stated that some blue stain occurred but that the damage was slight, partly because of the admission of blue stain in the grades which were dried on the yard at these mills. Of the 12 replies, 6 were in California pine while the other 6 were scattered among 6 other species, redwood, North Carolina pine, and cypress not being included. As to expedients used to counteract blue stain, 17 mills used dipping, mostly with a hot soda solution, 13 of these being southern pine mills; 8 used very open piling, 5 of them being in California pine; 4, all in southern hardwoods, used pole stacking; 1 southern hardwood manufacturer presteams stock subject to stain before piling on the yard; and 1 manufacturer of Douglas fir, during the months when that species is liable to stain, ships the product green to local factories which presumably kiln-dry it before stain can affect it.

DEFECTS PRODUCED IN LUMBER BY AIR SEASONING

It may be useful to discuss further various kinds of defects and the causes that produce them. It is not possible to avoid such defects entirely, because it is only by accident that air seasoning of lumber occurs under ideal conditions, which depend largely upon the weather. Much can be done by intelligent piling and in some cases by seasonal variation of piling in accordance with the sort of weather that may be expected.

SURFACE CHECKING AND CASEHARDENING

Whenever lumber is air-drying the surface is always somewhat drier than the interior, and indeed the interior moisture will not otherwise be drawn to the surface. If, however, the weather is such as to surface dry the lumber at a faster rate, it will become so dry

near the surface as to create a zone through which interior moisture can pass but slowly, thus retarding interior drying. Such a condition is reached eventually anyway, but if it occurs while there is considerable moisture left in the interior, unfortunate results are likely. The thicker the lumber, the more is the difference between

FIGURE 26.-The upper left-hand section shows internal honeycombing in oak, and the other sections show checking, much of it too deep to be called surface check. Photograph courtesy of Forest Products Laboratory

surface zone and interior likely to develop to an undesirable degree, and fast-drying weather is apt to encourage it.

When the wood near the surface becomes dried to fiber saturation point of 25 per cent moisture it starts to shrink; but the interior has not yet reached its own point of shrinkage, and thus a tension stress develops near the surface. If this becomes strong enough to tear the fibers apart, surface checks develop; this depends upon the

species, as some species will hold together under the stress while others check. If the wood checks, the portions between the checks are then able to take their natural amount of shrinkage; if the wood holds together, it remains in expanded form until it dries to the point where it sets and fails to shrink farther. In time the interior dries out and starts to shrink, but it may now be resisted by the set outer shell. The interior shrinkage may produce a tension stress inside the piece, so that, if resawed in that condition, the two halves will warp inwardly. It may develop interior checks, which will relieve the stress; and these checks may be numerous and open so wide as to produce the interior damage known as honeycombing, though not extending through to the outside. If, however, the outer zone of wood has surface-checked, these checks will close up again when the interior shrinkage occurs and perhaps be invisible in the seasoned rough board. The checks are there, however, and are apt to become visible later on the finished surface of the wood; surface checking in seasoning is a defect to be avoided as far as possible. (Fig. 26.)

FIGURE 27.-Section sawed through an oak plank, revealing internal honeycombing not evident on the surface. Photograph courtesy of Forest Products Laboratory

The other results of this unequal shrinkage-casehardening and honeycomb-are of more importance in hardwoods than softwoods, because they develop to a greater degree in many hardwoods. (Fig. 27.) Casehardening stresses mostly die out of softwoods, though, if severe, they may remain for some months. All these effects are more marked in kiln-drying than in air-drying because of the faster drying rate; but they can be almost entirely avoided by provision for adequate humidity, which is not controlled in air-drying.

The only way to reduce surface checking in the pile is to reduce the drying rate, and the practical way to do this is to reduce the air circulation by closer piling. Nor can much be accomplished in this direction, because the lumber might take longer than is desired to reach seasoned condition, or because at the slower rate other damage might result, as from stain or decay developing. In the southern hardwood air-seasoning study tests were made of use of 1/2-inch stickers in place of inch, on red oak and white oak, for the purpose of reducing honeycomb. It did this, but the total degrade from all sources was greater than with the 1-inch stickers.

The practice of using in wide piles four or five vertical chimneys, piling the lumber between these chimneys with practically no spacing between boards, is said to decrease surface checking and to result

in less stain, whereas most expedients that decrease surface checking tend to increase stain.

Stickers are sometimes responsible for surface checking on either side of them, where the lumber dries and shrinks more rapidly than under them and checks result from the stresses produced by the unequal shrinkage. Another class of checks are sometimes found in wide stock extending under the stickers. This occurs when the lumber has dried enough to shrink under the stickers, thus tending to decrease the width of the board at that point; but the weight on the board is such as to resist the movement of the edges of the board, developing a tension stress that results in checking or splitting. Lower piles for specially wide stock, or distributing the load between more lines of stickers, will reduce this defect.

Questionnaire replies to the inquiry about occurrences of splitting in piles of wide lumber seem to indicate that it does not occur very extensively. Only 7 of the replies indicated material damage of this sort, while 10 others stated that occasional splitting has been noticed but was unimportant in extent. The use of more stickers, so that

FIGURE 28. Showing somewhat the usual appearance of end checking, though the shake of tangential checking is not usual. This occurred in 2-inch plank of western larch during air-drying. Photograph courtesy of Forest Products Laboratory

each sticker carries less weight and holds the lumber against shrinkage less firmly, was reported as effective by 2 mills, 1 in southern hardwoods and the other in northern hardwoods and hemlock; 2 other mills, 1 in Douglas fir and 1 in western (Pondosa) pine, reported that wide lumber was piled in lower piles.

LOOSENING OF KNOTS

This defect occurs almost entirely in softwoods, the common grades of which are used full length with whatever knots may be in them; the common grades of hardwoods are mostly used for cutting up, during which process knots are largely eliminated.

For common softwood uses a sound knot may be tolerated where a checked knot or a loosened knot would be objectionable. In other words sound tight construction is desired even when appearances are unimportant.

In southern pine and Douglas fir, and many minor softwooods, knots that are foosely ingrown or not interlaced in the grain of the wood are frequent and tend to fall out as a result of either air-drying or kiln-drying; this often develops when knives hit the knot in its passage through the planer. The impact of a planer knife at high speed on the hard wood of a knot is heavy, but the more knives there are on a head the smaller the individual blow delivered by each, if they are set true.