EFFECT OF THE FINAL VACUUM.

During the process of seasoning most of the moisture in the cells and intercellular spaces in the wood is evaporated and is replaced by air. The preliminary vacuum, drawn only for a few minutes, chiefly for the purpose of filling the treating cylinder with the preservative, fails to reduce materially the amount of air in the wood structure. When the preservative enters the wood, either by its own penetrative power or by the application of pressure, it fails to replace entirely the air in the cells and intercellular spaces. If pressure is applied, the air becomes compressed, but can not all escape. When the oil is run out of the treating cylinder and a vacuum is applied the air in the wood structure not only tends to resume its original volume but by reason of the vacuum it exerts great expansive force, driving the superfluous oil out of the timber. This collects in the bottom of the treating tank, and after the vacuum has been continued for a sufficient period can be recovered by being blown back into the storage tanks.

Just as it is easier to penetrate sapwood than heartwood, the vacuum acts more sluggishly in the heartwood portions of the timber. Hence, although a greater amount of oil is forced into the sapwood, a proportionately greater amount is subsequently recovered, and the residue in the two different classes of wood tends to become fairly equal. By this method it is possible to continue the pressure until the heartwood has received the desired penetration without running the risk of leaving an unnecessary amount of oil in the more porous sapwood. The amount of oil that can be recovered by a final vacuum from pine timber is surprising. The treatments shown in Table 7 were made in the large commercial cylinders, 6 feet in diameter and 96 feet long. The amount of timber in a single load varied from 1,200 to 1,400 cubic feet.

TABLE 6.-The effect of drawing a vacuum after impregnation.

If the vacuum had been continued longer in all of the runs it is probable that even a greater amount of oil could have been extracted. The treated wood appeared dry on splitting. The cells contained little free oil, but the walls were thoroughly coated.

It is believed, therefore, that the method of drawing a vacuum after impregnation accomplishes the desired results. The heartwood, in most cases, received an absolute impregnation without the absorption of large quantities of oil. In some cases the penetration extended in the heartwood for about an inch and a half from the sides; the interior was not wholly discolored by the oil, but showed the presence of a small amount of creosote by dark spots and flecks appearing along the pores. Such a penetration is believed to be ample for timber of this character, where decay alone, and not mechanical abrasion, is to be guarded against.

The amount of oil that should be injected before drawing the vacuum depends almost wholly upon the relative amounts of heartwood and sapwood. If the timber is composed almost wholly of heartwood much less need be injected than if sapwood is present. The point of importance is not the amount of oil the timber is made to absorb, but to apply enough pressure to force the oil into the heartwood portions. Hence, if no heartwood is present, little or no pressure need be applied.

The amount of oil left in the timber must depend upon whether it is composed of heartwood or sapwood and upon the individual opinion of the owner. Until more is known of the changes which creosote undergoes in wood exposed to the full force of the sun and wind, it is recommended that 10 pounds of oil per cubic foot be allowed to remain in the sap class of arms, 8 pounds in the intermediate class, and 6 pounds in the heart class. It may be perfectly safe to reduce the oil in all classes of timbers to about 6 or 8 pounds per cubic foot, or even lower. Certain of the most abundant constituents of creosote evaporate at summer temperature on exposure to the air; hence it is prudent to obtain creosotes which have the smallest proportion of volatile oils. It is very desirable that experiments should be undertaken to ascertain definitely what changes take place in treated wood. Preliminary analyses of treated sections of several years' exposure are suggestive, and indicate a need for more complete results.

EXPERIMENTS WITH GREEN TIMBER.

In the foregoing experiments the timber weighed about 31.5 pounds per cross-arm, or between 34 and 35 pounds per cubic foot, and, if possible, the wood should always be allowed to reach this or a less weight before treatment. In order to test the impregnation of green wood, a few cross-arms of green loblolly pine were weighed and

treated. An oil pressure of 20 pounds was applied for fifteen minutes. The pressure was then increased to 40 pounds and held at that point for fifteen minutes. The superfluous oil was then blown back into the storage tank, and a vacuum of about 25 inches was drawn for half an hour. The green timber showed an average gain of only 2.2 pounds per cubic foot as compared with 12.6 pounds in dry sapwood in the same load. But by splitting the green arms it was found that a good penetration had been received. The oil had penetrated into the wood for several inches from the ends and from the insulator pin holes, and even along the sides a fairly good penetration had been obtained. The color of the treated wood was much lighter than in the dry arms, on account of the large amount of water with which the preservative was mixed.

This experiment by no means indicates that seasoning is unnecessary. The comparative lack of strength of green timber, as well as its tendency to check and split, and so to expose the unprotected wood, make thorough air seasoning imperative. And it is not known what the effect of so large an amount of water upon the permanency of the creosote in the wood will be. That such thoroughly moistened fibers can become completely saturated with creosote is doubtful. The results are of value, however, in indicating that, when necessary because of an emergency, partially seasoned timber can be used.

THE SAVING BY THE PROCESS RECOMMENDED.

The adoption of the process recommended will effect a saving both in the amount of oil required for thorough impregnation of the timber and in the time of treatment. A saving of even 2 pounds of oil per cubic foot of timber in an ordinary commercial load of about 1,400 cubic feet means a very considerable reduction in the cost of treatment. In many cases a much greater quantity can be saved.

The omission of the steam bath brings about a decided saving in fuel; of considerably more importance is the saving in time. To treat a load of cross-arms in accordance with the specifications in common use requires from six to eight hours. Under the specification recommended in this report the time will be reduced to less than two hours, and when it is remembered that the number of the quickly treated sapwood runs will greatly exceed that of the slower heartwood runs, it is evident that the average time of treatment can probably be still further reduced. The capacity of a plant, and therefore, to a lesser extent, its profits, depends primarily upon the daily number of runs.

[Cir. 151]

GRADING.

PERMANENT BASIS FOR GRADES.

At the beginning of the study the basis for the heartwood and sapwood classes was chosen somewhat arbitrarily. It provided that all arms containing at least 75 per cent of heartwood should be separated into Class A and those with at least 75 per cent sapwood into Class B. Class C, or the intermediate arms, were to contain 75 per cent of neither heartwood nor sapwood. The results of the later treatments have proved this basis correct, and it is recommended that it be followed in commercial practice.

GRADING AT THE TREATING YARD.

When the carloads of arms for the experiments arrived at the treating plant they were unloaded and piled on the yard to season. The grading was done as the arms were unloaded from the car. A force of about six men was employed for each car. Each arm had to be marked individually with crayon, to be later restamped with iron stencils, and this necessarily delayed the unloading to a greater extent than would have been the case had it been necessary merely to grade the arms and not to mark them. Even with this handicap, however, the unloading was almost as quickly accomplished as in ordinary commercial practice without grading. It is important to note that even the unskilled labor employed in unloading and piling the arms soon became fairly expert in quickly deciding into which class each arm belonged, and it soon became sufficient merely to check the judgment of the laborers. The present commercial custom requires the presence of an inspector at the unloading of each carload of lumber, in order to pass upon the soundness and other specified qualities of the timber. This in itself necessitates the individual examination of each arm as it is removed from the car; no additional time is required to indicate to the laborers in which of the three piles the particular arm is to be placed. The cost of grading, therefore, under most circumstances is negligible.

GRADING AT THE MILL.

If conditions make it advisable to avoid the grading at the treating yards and under most circumstances this will probably be the case-an excellent plan is to have the arms graded at the mill and shipped to the treating yard in separate cars.

In many localities the cross-arm mill is located on the railroad, where shipping is comparatively quick and inexpensive. The actual felling of the timber is carried on within a wide radius of the surrounding country. The logs are hauled to small portable sawmills, which constantly follow the base of supply. There they are sawed into timber of rather rough dimensions, which is next carried on

wagons to the cross-arm mill. The next step is to plane and edge the timber to the exact dimensions specified for the arms. The timber is then passed on to the boring machine, where the bolt, pin, and brace holes are bored, and finally the upper edges are rounded off and the arm is ready for shipment. During this process of manufacture each piece of timber is carefully examined by an inspector, who separates the arms into different grades-firsts, seconds, and culls.

To one as experienced in handling lumber as this inspector must necessarily be it is only the work of an instant to grade the arms into the additional classes, depending upon the relative amounts of heartwood and sapwood. His judgment can be indicated by a slight crayon mark. When the arm is finally removed from the finishing machine this mark will indicate into which car or in which pile the arm is to be placed. Then the graded arms are shipped to the treating mill in separate cars, and the work of the treating plant inspector, so far as the grading is concerned, is reduced merely to passing upon the judgment of the inspector at the mill. If the arms have reached an air-dry condition they can then be unloaded directly from the railroad cars to the cylinder buggies; otherwise they are piled in the yard to season.

SEASONING AT THE MILL.

Whether or not the arms should be seasoned at the treating yard or at the mill before shipment depends entirely upon local conditions. If convenient arrangements can be made at the mill, they should by all means be seasoned before shipment. This results in lessening the handling of the arms at the treating yard, and in considerably reducing the freight charges. A carload of green arms frequently weighs from 55,000 to 60,000 pounds. A loss in weight of 40 per cent reduces this weight to not more than 36,000 pounds, a saving of the freight charges on from 22,000 to 24,000 pounds of lumber. Even for comparatively short hauls this saving is decidedly more than sufficient to defray the slight extra cost of grading and seasoning. In most cases no extra expense is entailed in grading the arms; 50 cents per carload is believed to be a liberal average estimate. There is no extra cost to season the arms other than that entailed by one extra handling of the timber and by holding it for a comparatively few weeks. Often both these items of expense are already present because, even in the ordinary practice, arms will arrive at the treating plant more rapidly than they can be treated, and piling in the yard is necessary.

SUMMARY.

The results of the entire study are summarized in the following recommendations:

1. During the process of manufacturing the arms they should be graded into three classes: Arms that contain at least 75 per cent of