the traffic strains. After five years' service in the city of Minneapolis, Norway pine appears to be wearing as well as longleaf. The action upon a pavement is so complex and is distributed over so long a time that it is difficult to correlate it with laboratory tests. The significance of these can be determined only by a careful comparison with the service results actually shown on a pavement of some of the standard woods. Yet nowhere in the United States have wooden pavements of the modern type been down long enough to warrant final conclusions.

The values on the mechanical tests of wood in the standard engineering manuals do not generally specify the moisture conditions under which the tests were made. It is only within recent years that the enormous influence of this factor upon the strength of timber has been appreciated. It is now proved, for instance, that the average compression strength of longleaf pine, parallel to the grain, when green or water soaked, is only about 30 per cent of that of the absolutely dry wood, and about 60 per cent of that of wellseasoned wood. It is obvious, therefore, that tests upon different woods in which the moisture condition is not taken into account are not safe guides.

CREOSOTING.

Creosoting has generally been supposed to weaken timber. The results of recent experiments conducted by the Forest Service indicate, however, that the creosote itself does not lessen the strength of timber, but that this result in practice has been due chiefly to certain methods used in the impregnation."

The Rush Street Bridge in Chicago, Ill., is said by the chief engineer of streets to carry a traffic as heavy as any in the city. It has two 20-foot roadways, which were paved in 1899, one with creosoted longleaf pine blocks, the other with uncreosoted blocks. The creosoted pavement, after a service of seven years, was still in good condition and is expected by the chief engineer to last several years more, while the uncreosoted pavement was renewed in 1902.

The explanation of these facts lies in the antiseptic qualities of the creosote and its physical action in filling the pores of the wood and decreasing its absorption of water. Every pavement is wet a large part of the time, and wood, when saturated with water, has generally less than 40 per cent of its kiln-dry strength. The creosoting treatment, by lessening the absorption of water, maintains the strength at much nearer the dry-wood value, and thereby increases the actual service strength.

a Bulletin 70, Forest Service, U. S. Department of Agriculture, "The Effect of Moisture on the Strength and Stiffness of Wood."

Circular No. 39, Forest Service, U. S. Department of Agriculture, "Experiments on the Strength of Treated Timbers."

This action is most effective only when the wood is well seasoned before being creosoted. In the United States most creosoters attempt. to dry the wood by applying steam and vacuum. Recent experiments have shown that wood will, except when very green, weigh more after such treatment than before. That is, by this process, it will take up moisture rather than lose it. This would destroy any possible gain in strength through the creosote treatment. Not only is the steaming in most cases valueless, but if carried too far, as it often is, it seriously injures the wood." Natural seasoning, which in Europe is used almost entirely, would certainly contribute to increased excellence of results in the United States.

The common opinion among engineers in regard to the creosoting of wooden pavements is expressed by a prominent paving engineer, who says, "For wood paving creosoting is only of contingent value. If the traffic is heavy enough to wear out the pavement before it decays, which it will do on most city streets, it is a waste of time. and money to creosote the blocks. Creosoting is only economically desirable, therefore, when the traffic is so light that decay will reduce the block before it fails from wear." This would be true if creosoting treatment had no effect on wood pavements other than the preservation of the wood from decay. But creosoting effects, through the increased service strength of the blocks, an additional and quite different gain. It is believed that there is already enough evidence, such as that offered by the Rush Street Bridge, to overthrow the traditional doctrine of the limited value of creosote preservative treatment for wood paving.

The best preservative.-Preservative treatment for wood paving, then, must accomplish the following results:

(1) Preservation from decay.

(2) Mechanical filling of the pores, to prevent the absorption of fluids; which further accomplishes:

(a) Elimination of expansion.

(b) Increase of resistance to wear.

(c) Maintenance of sanitary value.

Creosote oil seems to meet these requirements better than any other preservative commonly used. The other preservatives are mostly water solutions, which are easily leached out of the wood and thus lose their preservative effect. Moreover, being themselves largely composed of water, they do not prevent expansion in the blocks, do not increase their resistance to wear, and do not maintain them in a sanitary condition.

The same requirements which have caused creosote to supersede other preservatives in wood paving call for a larger proportion of

a Circular No. 39, Forest Service, U. S. Department of Agriculture, Experiments on the Strength of Treated Timbers."

high distilling constituents in the creosote than is necessary in the preservation of timber for many other purposes.

Impregnation. In the United States the timber for paving purposes is universally impregnated by the vacuum-pressure method. Closed cylinders, usually about 6 feet in diameter by 100 feet in length, are used. The amount of oil usually injected into the wood is 16 pounds per cubic foot, though in some cases it reaches 22 pounds. To many municipal engineers this has seemed an unnecessarily large amount, constituting an item of expense in the already high cost of wood paving which might profitably be reduced. But the necessity for the prevention of moisture absorption under conditions more exacting than those to which almost any other creosoted wood product is subjected makes imperative the impregnation of wood paving blocks with an amount of oil per cubic foot greater than that ordinarily given to railroad ties or construction timbers. The tendency is toward heavier charges rather than the reverse, to secure for wood pavement a service so greatly enhanced as to yield better returns upon the investment, even at the higher cost.

In Paris, the blocks are usually impregnated by merely soaking for a short period in an open tank, and the amount of oil injected per cubic foot is small. The life of Paris pavements so treated is from six to fourteen years only, according to the traffic; and, because of the small impregnation, expansion in the pavements is large and calls for the laying of longitudinal courses of blocks at the curbs, which are removed as it becomes necessary. The vacuum-pressure method for the treatment of wood paving blocks is only now being introduced." This is the reason for the statement made on page 10 that American creosote treatment of wood paving is superior to that of Europe. The excellence of European wood pavements, despite this defect, is largely due to better construction, maintenance, and repair.

Inspection. Our cities, which are the chief consumers of wood paving blocks, perhaps fail more than any other consumers of creosoted products to inspect properly the product and its manufacture. Cities are often dependent, even for their specifications, upon the companies which are seeking contracts. This is largely due to the technical nature of the subject, both as to processes and as to the chemical analysis of the oils, and largely also to the almost entire lack of reliable information. The Forest Service is carrying on investigations of

a M. Mazerolle, Ingénieur, Service Technique de la Voie, Paris, 1906.
[Cir. 141]

wood preservatives and wood preservation, and results will be placed at the disposal of engineers as fast as they become available."

LAYING THE PAVEMENT.

During the progress of this study the practice in wood-pavement construction was carefully observed in many cities. Some of the points are here discussed.

FOUNDATION AND CUSHION.

For the most satisfactory service, wood-block pavement requires a concrete foundation. This is usually made from 5 to 6 inches thick, although some engineers reduce it to 4 inches on lightly traveled residence streets. As a top cushion for the foundation either Portland-cement mortar or sand is used. The former is considered superior by most engineers. The bearing for the blocks is permanent, and, if carefully surface-trued, can be made as even as desired; and if the grout is mixed slightly damp and the blocks laid in it immediately, it provides equally as good a compensation for minor inequalities in the height of the blocks as sand does. If tar can be used, a better method of accomplishing the same object is to mix the grout in the usual way and let it thoroughly set, after which a coating of tar should be applied and the blocks bedded in it. This is the method oftenest used in Europe. Sand makes a satisfactory cushion where the slope is negligible and the foundation is solid. It is sometimes preferred on the ground of greater elasticity and power of accommodation, and it has the merit of being cheaper than cement. On a gradient, however, if water, by any possibility, gets under the blocks it is likely to carry the sand to the bottom of the slope and seriously derange the pavement. On bridges, also, if there is much crown on the roadway the vibration of the structure is likely to shift the sand from the center of the crown toward the gutter. For bridges the usual practice is to lay the blocks directly on carefully creosoted planking.

The following publications on this subject have been issued by the Forest Service:

Bulletin 51. Report on the Condition of Treated Timbers Laid in Texas, February,

1902.

Circular 39. Experiments on the Strength of Treated Timbers.

Circular 80. Fractional Distillation of Coal-Tar Creosote.

Circular 98. Quantity and Character of Creosote in Well-Preserved Timbers.

Circular 101. The Open-Tank Method for the Treatment of Timber.

Circular 104. Brush and Tank Pole Treatments.

Circular 111. Prolonging the Life of Mine Timbers.

Circular 117. The Preservative Treatment of Fence Posts.

Circular 128. Preservation of Piling Against Marine Wood Borers.

Circular 134. Estimation of Moisture in Creosoted Wood.

Circular 139. Primer of Wood Preservation.

Extract from the Yearbook of the Department of Agriculture, 329. Life of Telephone Poles.

Prolonging the

BLOCKS.

The blocks should be rigidly inspected, especially as to imperfections of sawing, as to knot holes, decay, or defective corners or edges, as to squareness of the angles, and as to thoroughness of impregnation. Voids due to imperfections in any of these respects often can not be properly filled by the joint filler, and are very detrimental to the pavement. In European practice no variation greater than onesixteenth inch in any dimension of the blocks, and no measurable variation in the depth, are allowed. There the blocks are also required to be kept carefully protected from sun and weather after treatment and until they are laid. Deterioration from checking, which in America is often considerable, is thus prevented.

Exclusion of second growth material from wood-paving specifications is immaterial, except in so far as young trees contain a greater proportion of sapwood than older trees; and, since sapwood is separately provided for in most cases, the specification prohibiting secondgrowth timber could well be abandoned. Sapwood is entirely excluded by most wood-paving specifications. Under existing market conditions, however, it is quite impossible to obtain strictly allheart southern pine such as the specifications demand. Commercial longleaf southern pine lumber is also seldom free from an admixture of loblolly pine. The true longleaf pine has usually so narrow a sapwood that it could be neglected without danger to the life of the pavement. In loblolly the sapwood is often very wide; but loblolly pine is one of the species for which it has been proved that the sapwood under equal conditions of moisture content is as strong as the heartwood. Therefore, when effective seasoning of paving material can be assured before the creosote treatment, the prohibition of sapwood in southern pine material is needless, and should also be omitted from specifications.

A more pertinent specification would be one excluding fast-grown timber in pine paving-block stock-say all showing less than eight rings to the inch-since it is the porous wood resulting from fast growth, rather than the presence of sapwood, which unfits timber for this use.

ANGLE OF COURSES.

The angle at which the courses are laid in wood pavement is a matter of some importance. The most obvious angle is that of 90 degrees with the curb, which takes the courses straight across the street. Probably the greater part of the wood-block pavement in the United States is so laid. But this angle permits the calks of horses' shoes to strike in a direct line with the joints, and subjects the pavement to a wear and tear which may largely be avoided by laying the

I