ture of the timber and has not had the opportunity of investigation from that standpoint, except in a superficial manner, but has had ample opportunity to observe the practical effects.

The effects of different arrangements of apparatus, such as the distance from the sources of steam to the retort, the prevention of radiation from the retort, as affecting the character of the steam, has probably never been investigated, though the difference must be considerable. It is certain that steam passed through a superheater is more destructive than steam as it passes from the boiler. Why, then, should not the rapidity of condensation in the retort cause a variation of effect. It is obvious that a proper test of steaming can not be made without an examination of the structural change in the wood. The writer will venture the opinion. that the pressure and, consequently, the temperature, is not the direct destructive agency, the greatest damage coming from the volatilization of the moisture in the wood cells rupturing the cell walls in its effort to equalize the pressure. Referring to the discussion regarding the structure of timber, it will be noted that there are certain openings between the cells called "simple pits," "bordered pits," and "resin ducts." When the moisture in the cells is vaporized it causes a pressure on the walls sufficient to rupture at some point, and, as the membrane covering the bordered pits is the weakest point, that portion is the first to give way. The resin ducts, principally the cross ones, also furnish a means of exit for the expanding moisture, as do the transverse cells of the medullary rays. If the steam is applied so as to heat the wood gradually, causing an expansion of the contained moisture, air, and resins only sufficient to break open the "pits" and force freedom through the resin ducts and medullary rays, the writer will venture the opinion that the structure will not be seriously injured regardless of the pressure finally attained.

The moisture, saps, and resins forced out of the wood evidently come from the cells alone, the moisture remaining, or even increasing, in the cell walls. Many recorded experiments show an increase in the weight of the wood after the steaming process, which may be accounted for by this increase of the moisture in the cellular structure. Most of this increase in weight in unseasoned timber must take place within the heartwood, as the sapwood is already nearly saturated. To say that steaming green timber fills the cells and increases the weight of wood is certainly erroneous in the class of work with which the writer has been engaged. The

following case, being one of many recorded, would no doubt serve to illustrate the case at hand.

A charge of 2,000 feet of green loblolly piling weighing 65 pounds per cubic foot, just a little too heavy to float, was steamed as ordinarily done for 24-pound treatment, and the vacuum applied in the usual manner. There was collected during the vacuum process about 1 pound per cubic foot of timber of condensed moisture in the bottom of the cylinder, a small amount, of course, coming through the vacuum pump as a vapor, but this quantity must have been considerably less than that collected from the bottom of the cylinder. The 24 pounds of oil per cubic foot were injected, and on being removed the piling weighed practically the same as when put into the cylinder. What became of 22 and a fraction pounds of moisture per cubic foot? This is not an isolated case, but is the case whenever waterlogged or sap-filled timber is treated, and to a marked degree whenever less moist material is given treatment to the same extent. This is the result of the actual treatment on a commercial basis, and is incontrovertible evidence that steaming as actually done in commercial operations does remove moisture.

If steaming sufficient to cause great loss of strength is applied to timber-piling, for instance, the heat very gradually penetrates to the center of the wood, as the wood is a rather poor conductor of heat, and as it becomes more heated it is poorer yet. There is no circulation of particles within, but on the contrary the moisture is usually outbound. The experiments recorded by Dr. Hatt, as previously referred to, seemed to indicate a difference in temperature of 40 to 50 degrees between the interior and exterior after some hours of steaming. The loss of strength would therefore be confined to the outer layers of wood and the central portion would not be injured for many hours. This is contrary to the facts as exhibited by specimens, which the writer has had the opportunity to see, said to have been “burned” by live steam, the wood just within the outer ring usually being brittle, as is the heart, which fact confirms the opinion that the damage, with the exception of a small percentage practically negligible, is caused by the sudden application and release of pressure.

In computing the strength of timber, it must be reduced to some standard percentage of moisture. Steaming timber removes much of the resinous matter, as evidenced by the discharge from the blowoff of a treating cylinder, so that the weight of the materials other than water remaining in the timber is not so much as before. Fur

thermore, the fact that a piece of timber, an adjacent section of which has been steamed and then reduced to the same moisture content, is stronger than the steamed portion, is not infallible evidence of damage by the steaming process. Steaming removes the saps and resins from the cells and probably increases the moisture of the cell walls-the very moisture which affects the strength so much. In previous investigations as to strength, did the investigators ascertain that the percentage of moisture remaining in the cell walls was the same?

The conclusions to be drawn from experience are that the limit of allowable steam pressure, at least under 75 pounds, can not be set unless the conditions existing are known and a stated method of applying and cutting off of the steam is given. It is evident that the percentage of moisture in the timber, on being drawn into the cylinder, will affect the allowable pressure. Of the piling furnished by three plants for one job, two supplying the same timber, some partly seasoned, some reeking with sap, some steamed not over 30 pounds, that steamed at 60 pounds pressure continuously for ten hours and at lower pressures for an additional ten hours, showed the best under the pile hammer. One of the examples of well-treated paving blocks cited in the Yellow Pine Manufacturers' Association pamphlet on that subject, was subjected to a steam pressure of 80 pounds, corresponding to a temperature of 312° F. These cases are cited to show that high temperatures are not greatly detrimental to the timber, but that it is the wrongful application of such temperatures that cause the damage.

The final conclusion is that the specifications should not limit the steaming, but should leave this process to the judgment of the plant management, reserving the right, of course, to reject any damaged material.

VACUUM

Perhaps no part of the treating process varies so much as the application of the vacuum and, among engineers generally, there is no part of this process so little understood. Commonly, it is thought to remove the enclosed liquids by releasing the pressure and allowing them to flow out, but this part performed by the vacuum process is secondary, to say the least, and if this were all the vacuum was applied for it might be omitted in a majority of

cases.

The steaming process tends to convert the moisture into steam, leaving, at the end of thorough steaming, the cells filled with steam

instead of water as originally. Theoretically, on the release of pressure the steam contained in the cells would expand and establish an equilibrium. The cooling of the timber after the release of pressure would cause a condensation of the steam within the cells. This theoretical condition is true to a limited extent. Some of the steam does necessarily escape by expansion, but that part is on the timber surface. The fall in temperature coincident with the release of pressure causes the condensation of the vaporized moisture and consequent filling of a ring of cells which would tend to retain the pressure on the interior of the timber. It is evident therefore that the temperature in the cylinder during this process should be kept high by means of the steam coil to counteract condensation. The fall of temperature evidently can not extend very far into the timber, owing to the relatively poor conductivity of wood fiber. Consequently, we have after the release of steam pressure, the outer cells empty or nearly so; within this merges a ring of cells completely filled with condensed moisture which forms a barrier to the escape of the steam from within. The application of a vacuum, reducing still further the exterior pressure, allows the escape of the previously condensed steam by allowing a relative increase of the pressure from within. This moisture, together with that which condenses on the exterior of the timber and is collected in the bottom of the cylinder, is a relatively small portion of the total extracted from a water soaked or sappy timber-amounting to about 4 or 5 per cent in the instances investigated by the writer.

It is physically impossible to extract by a vacuum the liquids from a piece of timber which does not contain a quantity of air in the cells, as there must be some medium within to expand and force the liquids out. This is the basis of the Lowery and the Rueping patents for the empty cell processes, the former relying on the air ordinarily contained in the timber to expand under the released pressure of the vacuum, and in the latter the expansion is had by the release of the compressed air injected into the timber previously. The amount of oil to be recovered by a vacuum after the oil injection in the full cell process is a very small percentage of the total injected.

It would be safe to assume that the use of a vacuum is important in the lighter treatment, and yet more important in the treatment of timber without steaming, and where a uniform and deep penetration is desired.

The practice in applying the vacuum is extremely varied, even

for the same conditions, at different works. A variation in practice from one-half hour to six hours has been noted in the experience of the writer on work almost identical in character. Such a variation can only denote an ignorance of the real time required. The proper length of time to maintain the vacuum must vary with the extent of steaming previously given and the degree of dryness desired before the injection of the oil. If the timber is air-seasoned and no steam applied the time should be relatively short, as only the air contained in the cells is to be withdrawn, but in the case of steamed timber time must be given so that the pressure from within may force out the collected moisture in the outer cells.

On account of the decreased pressure, the evaporation from the timber surface is very rapid and, consequently, if the vacuum is maintained for a long period, much moisture is extracted and exhausted through the vacuum pump; but unless the quantity of oil proposed to be injected is very large, the time employed in this manner is needlessly extended.

It is evident from the above discussion that no specified time can be set as proper for any treatment or any timber, but would depend on the condition of the charge and the extent of treatment proposed.

The extent of the vacuum as to the height of barometer is not so varied and usually depends on the character of the machinery, the tightness of the cylinder, and last, but principally, on the location of the gauge. There is but one proper place for the vacuum gauge -on the cylinder, removed from the vacuum outlet pipe. The practice of placing the gauge on the pump, or on any pipe line to the pump is exceedingly deceiving and accounts for the high vacuum attained at many works in a short space of time. The writer has yet to see the vacuums of 28 inches which have been reported from time to time as in use.

It is not sufficient that a vacuum of 18 inches, or any other specified vacuum, be attained, but after this vacuum is attained it must not be lost. The uneven penetrations so frequently noted are not caused by lack of rapidity in filling the cylinder with oil when the application of the oil extends over even twenty or thirty minutes. The application of an 18 inch vacuum represents the extraction of about three-fifths of the air contained in the treating cylinder, consequently, if this is not maintained after the filling with oil is begun, when three-fifths of the cylinder space is filled with oil the pressure on the upper portion of the cylinder is normal and