the inert mineral aggregate. The bond between the mineral aggregate and either the asphaltic cement or Portland cement is far weaker than the strength of the mineral aggregates. None of the aggregates used could be called soft or inferior, each type of aggregate having a toughness (impact test) over 8 and a French coefficient of wear (abrasion test) over 8. As our results with limestone have generally been as good as the results with tap rock, the only experiment was in the use of crushed gravel and even this has given good results. All of our results on bituminous macadam are based on maintenance work in the upkeep of water-bound macadam pavements according to the following method:

"The old macadam roadway was scarified, reshaped, and covered with 11⁄2-inch aggregate sufficient to make the total depth of 11⁄2-inch material 6 inches after rolling. When crushed stone was used a course of 4-inch stone was spread on top of the 12-inch stone. When crushed gravel was used no 4-inch material was used because the 12-inch gravel was sufficiently compact and close without the addition of intermediate material. The surface was sprinkled with sand and then rolled. The bituminous macadam binder previously described was then applied at the rate of 1% to 1% gallons per square yard. The surface was then covered with a mixture of material of the same nature corresponding in general grading requirements to brokenstone screenings, one cubic yard of material covering from 24 to 32 yards of roadway. No seal coat was applied. The bitumen is gradually drawn to the surface by the heat of the sun and capilarity." It must be noted, however, that our bituminous macadam with either trap rock, limestone or crushed gravel, as mineral aggregate, has not stood up under a procession of trucks loaded with building materials. The bituminous macadam pavements in the direct path of extensive building operations have been pretty well pounded to pieces. Under moderate conditions of traffic they have given good service and the lantern slides show a street in 1920 and in 1925 in which crushed gravel aggregates was used and on which there has not been one cent of maintenance expense in six years. (Photographs not furnished for reproduction here.)

Asphaltic Concrete. The same considerations with regard to the mineral aggregate seem to apply to asphaltic concrete. Owing to the difficulties inherent in the specifications for Topeka mix, we do not use Topeka specifications on new work any more, sheet asphalt top and binder course having replaced the Topeka mix entirely on new work. But we still have a large yardage of asphaltic concrete to maintain. The lantern slides show the condition in 1920 and 1925 of a street paved

with Topeka mix using crushed gravel from 12-inch downwards as mineral aggregate. There has been no maintenance cost on this street and the only replacements to the surface have been for cuts for house connections.

Key Block Pavement. This is a freak pavement but one of considerable merit. It consists of:

"Alternating courses of wood and granite block laid at right angles to the direction of traffic. The unique feature is an oak block of wedge or key-stone shape, 4"x5" on the wearing surface, tapering at the sides to a width which is 2" to 1" less at the bottom than that of the wearing surface; the depth is about 4". Courses of granite blocks are first laid with a space of 3% to 34 inches between the courses. The transverse spaces are then partly filled with sand and the keys are then firmly rammed in between the rows of granite block. The result is a series of alternating courses of granite with oak blocks tightly wedged between them. This type of construction has been used extensively for the past seven years in the franchise area of the New York Railways Company, particularly at street intersections which are subjected to unusually heavy traffic. It is essentially an economical method of repairing or repaving a worn granite-block pavement to meet conditions of excessively heavy traffic. As the oak blocks project slightly above the level of the granite the wood consequently bears the brunt of the traffic and the harder the oak blocks are pounded by traffic the tighter they should become wedged in, thereby Increasing the stability of the whole pavement.

In June, 1919, we replaced a badly cobbled Belgian block pavement on a macadam base with the key-block pavement. There was immediate relief from noise and slipperiness and although the pavement has been subjected to heavy cross-boro traffic for six years the expense of repairs has been about $100 on three city blocks of pavement on which the only drainage is surface drainage. Some of the oak blocks not coated with asphaltic grout have rotted in six years, blocks still protected by bituminous mastic are in good condition.

Paving with Rectangular Hard Wood (White Oak Blocks). Owing to the density and toughness of white oak and its low water absorption we have tried untreated oak blocks for repairs to creosoted pine pavements. For repairs they do not seem to offer any advantage. The Boro of Manhattan, however, has laid under contract a stretch on Second Avenue, in which rectangular white oak blocks were used with a seal coat of bituminous grout underneath the blocks and on the surface of the pavement, which is standing up well after three years of service.

It is gratifying to report that all of our experiments have

2 E. E. Butterfield, "Utility of Hardwoods, etc." Engineering NewsRecord, Vol. 85, p. 656.

given good service for from five to six years. The cinder and asphaltic oil surfacing of dirt roads has passed the experimental stage and has become a part of our regular maintenance program. The other experiments have all stood up as well as our regular contract work done under standard specifications. Resurfacing badly worn granite block by means of alternating courses of oak key blocks offers a method of reducing the noise and slipperiness of a pre-existing granite block pavement at a cost about the same as reclipping, renewing and replacing an entire granite block pavement.

THINNER BRICK PAVEMENTS IN STREET AND
HIGHWAY PAVING

By Webster L. Benham, President and Chief Engineer, Benham Engineering Co., Consulting Engineers, Kansas City, Mo.

Of the three outstanding points of interest in the consideration of vitrified brick pavements today, viz.: first, unusual and unique salvage in old brick surfaces; second, the wide and almost universal adoption of pure asphaltic cement for filler; and, third, the rapid increase in the use of brick of less than three inches in depth on light and medium traffic streets and highways, the latter is perhaps of the most interest from a standpoint of design.

It is the writer's opinion that for many years engineers have specified three, three and a half and four-inch brick on lighttraffic streets, where a thinner brick, of perhaps two and onehalf inches, would have served just as satisfactorily, effecting a considerable saving to the taxpayers. It must be borne in mind that the cost factor in the construction of highways and street pavements is becoming a more and more important item as time goes on and that henceforth the taxpayers will investigate costs of pavements with a view of obtaining something having qualities of durability and "wearability" at a reduced.

cost.

There was, of course, one very good reason for the heavy design of the paving wearing surface in years past. The pronounced abrasive wear due to steel-tire traffic and iron-shod horses made it necessary to construct pavement to withstand heavy impact and wear. The change in the character of traffic. during the last decade, however, has largely nullified this reason so far as present or probable future design of the brick wearing surface is concerned.

Although many of the brick pavements of the past have been constructed with almost total disregard of proper engineering principles, as we recognize them today, these pavements have demonstrated certain virtues in spite of this fact and have uniformly maintained a high standard of quality regardless of such neglect.

It is perhaps needless to point out here that the virtues inherent in our older brick pavements have been due to the toughness and wearing qualities of the bricks themselves rather than to any special virtues in the other component parts of which a pavement is constructed, such as the subgrade, foundation, cushion or filler. This important truth is borne out in instances too numerous to mention, and despite the lack of attention to the subgrade, the evenness of the surface of the foundation course, the thickness and evenness of the sand or mortar cushion, the type of filler, and the almost total lack of maintenance, brick pavement has, as a rule, demonstrated unusual merits.

It is universally conceded that early methods of construction were far from efficient. However, it is desired to emphasize that, in spite of the absence of any special study of the subgrade, adequate foundations or artificial base, in spite of poor drainage and lack of attention to the many important factors which tend to make a first-class pavement, in spite of the everincreasing volume of traffic, the brick wearing surface in scores of ancient brick pavements has demonstrated a strength and durability far beyond any reasonable expectation. Each individual brick in those anciently designed pavements, although poorly and irregularly supported, has rendered service which suggests that the brick themselves have a durability extending beyond the other component parts of the pavement structure, and that in many cases a thinner brick than that used would have served satisfactorily, and possibly more economically.

Sand filled brick pavements that have been down for as high as thirty-three years, on gravel base, with no artificial drainage, and subjected to frequent cutting for all sub-surface structures, have been taken up and relaid on renewed base courses, the brick seemingly good enough for another generation of use. These instances have been so numerous during the last few years that the procedure is no longer an experiment, and is accepted paving practice, constituting a unique economy.

It sounds paradoxical, but it is nevertheless a fact that with traffic becoming heavier year after year, more and more engi neers are adopting two and one-half-inch brick, and in some cases two and one-quarter-inch brick, for the wearing surfaces on light and medium traffic streets. At first glance this might seem incongruous, but a little consideration of the reasons why