e ^V3u.vT V,\o UNIVERSITY OF OREGON BULLETIN Yu> .5 _ ■ Cop, J NEW SERIES JANUARY, 1913 Vol. X, No. 5 CONCRETE ROADS VERSUS MACADAM e. h. McAlister Dean of the School of Engineering, University of Oregon Published monthly by the University of Oregon, and entered at the post-oflice in Eugene, Oregon, as second-class matter CONCRETE ROADS VERSUS MACADAM e. h. McAlister Dean of the School of Engineering, University of Oregon I CONCRETE ROADS VERSUS MACADAM. A NEW PROBLEM AT HAND. During 1 the past decade, the improvement of public highways has brought into existence a new class of traffic. Motor trucks and mechan- ical tractors drawing trains of loaded wagons have in many instances revolutionized the highway end of freight transportation. Mechanical haulage is able not only to effect a vast reduction in the cost of trans- portation per ton-mile on existing traffic, but also it makes possible the production of many articles which hitherto could not be produced with- out loss, on account of the expense and delay in getting them to market. This is notably true with reference to perishable products, but the production of such a staple article as wheat has been subject to consid- erable limitation. A few years ago, the general freight agent of one of the great railroads of the Northwest stated that the production of wheat was limited practically to a belt 12 or 15 miles wide on either side of 4 he railway line, and that the amount produced at greater distances did not figure in the sum total of production. Perhaps these limits have been slightly widened at the present time, but there is no doubt that mechan- ical haulage on good roads could very greatly increase the area of profit- able production for both staple and perishable products, and thereby in- crease the wealth of the state many fold, at the same time affording a comfortable living to increasing thousands on wide areas where now a bare existence would be difficult. Unquestionably this is one of the most important ways in which the wealth, population and prosperity of the state can be built up, and it is not surprising that the good roads move- ment has attracted the earnest attention of all classes, both in this country and in Europe. But mechanical haulage, wherever it has become considerable in mag- nitude, has brought new problems and difficulties. The good macadam roads, which made possible and brought forth the new traffic, have been quickly destroyed by the traffic — another case of the offspring devouring its parent. The destruction of the roads appears of course in the in- creased cost of maintenance and repairs. In Massachusetts, during the three years preceding 1910, the cost of upkeep for the state highways rose from a little over one cent to nearly 6 cents per square yard per annum. On many roads the cost became so great that water-bound ma- cadam was abandoned and some other type of surface was adopted. Other and even worse cases will be cited further on in this Bulletin. It is true that such serious conditions have not yet arrived in Oregon, and the cost of upkeep is still comparatively low. But mechanical haulage has just begun in this state, and it does not require the eye of a prophet to discern that in a very few years the same conditions will be upon us as are now found elsewhere. During the first nine months of 1912, the number of registered motor vehicles in Oregon showed an increase of 55 per cent over the total number registered in 1911. A little foresight may save considerable money to the taxpayers of Oregon, and give them much better roads in the meantime. [3 j SUITABLE TYPE OF ROAD SURFACE. Relative to the matters mentioned in the preceding paragraphs, the writer has recently perused several thousand pages of road literature, including the new books on highway construction, the reports of nu- merous state highway commissions, the proceedings of recent road con- gresses and conventions, including the International Road Congress at Brussels in 1910, and the more important articles in recent technical periodicals. The conclusion in all parts of the world is practically unan- imous that the standard water-bound macadam is not suited to the traf- fic of the future ; that mechanical haulage has come to stay, as an eco- nomical and social betterment. With scarcely an exception, the deter- mination is everywhere expressed that the roads must be made to serve. Commenting on a phase of this subject, a recent technical journal edi- torially remarks : 1 ‘ This is not a matter of pleasure vehicles and picnic parties. It is an issue of bread and butter. Many millions of dollars in mud taxes are coming either out of the people ’s mouths or out of their pocket books — depending upon whether they go without adequate food, or pay too much for it. Good roads and mechanical haulage will lift this mud tax, will lower the cost of living, and will make community existence possible under new conditions. Community existence will be possible without having people huddled together with an ever increasing tendency of the population toward the cities.” So far, there is practical unanimity, but when it comes to selecting a suitable type of surface to replace the water-bound macadam, there is every possible degree of divergence of opinion. However, for our present purpose, we may roughly group the opinions under three classes : First), there are those who advocate the use of some type of pavement found suitable in cities, such as standard asphalt, bitulithic, vitrified brick, etc. These are all admirable in most respects,’ but their high cost will doubtless prevent their adoption on country roads, except to a very limited extent. Secondly, we may class together the great number of schemes which are mere temporary palliatives of the difficulty, such as oiling the surface of existing roads, treatment with calcium chloride or various emulsions, or with tar. These have their merits in the way of cheapness and tem- porary relief, but do not seem desirable to adopt as a permanent policy of road improvement. Thirdly, there are the advocates of a concrete road, either with or without a bituminous wearing surface. The first cost of these roads will generally be somewhat greater than for macadam, but they are not torn to shreds by motor traffic, and their cost of upkeep is low enough to offset the additional first cost in a comparatively short time, wherever the traffic is severe, as it is likely to become on our more important roads within a very few years. In addition to their ultimate economy, con- crete roads are better than macadam all the time. The state Highway Commission of New York has concluded the expen- diture of $50,000,000, voted a few years ago for state roads, together with some $15,000,000 raised by counties and towns. The conclusions of the Commission, after spending $65,000,000 in constructing almost every kind of road known, are certainly of interest, and ought to carry weight with thinking people. In a report issued in 1912, discussing the change of policy of the Commission (from bituminous macadam to con- crete), the following statement occurs: “Having these factors in view, it seems wise, wherever possible, ,to construct a foundation of concrete and to cover this with a thin wearing course composed of bituminous material and screenings or sand, which is economical in first cost and easily and cheaply renewed. The foun- dation of a road, so constructed, is good for all time, and the wearing course serves the purpose of carrying all classes of traffic without rub- bing, raveling, or raising a dust.” The state of California recently voted $18,000,000 for the construction of trunk lines, and the Highway Commission has adopted the concrete base with thin bituminous surface for a number of roads connecting towns on the main line. Up to last October, more than a hundred miles of such roads had been laid out, with 78 miles under contract. The Highway Engineer of the California Commission is Mr. Austin B. Fletcher, a man of national reputation, formerly with the Massachusetts Highway Commission and special agent of the United States office of Public Roads. His judgment should carry great weight, especially when taken in connection with that of the New York commission. Between the extremes of California and New York, there is scarcely a state that has not some concrete roads, and the mileage is growing rapidly. Considering its many merits, and the fact that in the long run, it is relatively inexpensive, it seems probable that the concrete road, either with or without a bituminous wearing surface, will become the standard highway of the future. It is worthy of note that at the last meeting of the association for standardizing paving specifications, the committee on macadam declined to present any specification for water-bound ma- cadam, the members of the committee unanimously agreeing that it is not suited to modern conditions. Instead they presented specifications for bituminous macadam, which in Oregon would cost more than concrete. DECREASING THE COST OF CONCRETE. It has been known to scientific experimenters for some time that if certain volcanic materials be reground with Portland cement, the result- ing blend can be used to make concrete which in point of strength is equal or superior to concrete made in the usual manner. A material saving is effected in cost, especially where cement is high in price, and a further advantage is gained in greater resistance to certain destructive agencies which attack concrete in some locations. From the known chemical composition of certain volcanic materials abundant in Oregon, the writer inferred that possibly some of them might give good results when blended with cement, and accordingly un- dertook a series of tests at the civil engineering laboratory of the Uni- versity of Oregon. These tests have been carried on since last October and are still in progress. Already a few blends have been found which give results, both in tension and compression, superior to those obtained from the same cement without blending. Some of the results will be discussed on subsequent pages. Blends of this kind, after being prohibited for several years have re- cently received the approval of the German Government, following an elaborate investigation, and are now permitted on Government work. [5] M. Rene Feret, Chief of the Laboratory of Bridges and Roads of the French Government, after careful investigation, reported in 1908 that “The use of puzzolan in hydraulic mortars in combination with the cement increases the strength, and in a great many cases appreciably retards disintegration by sea-water.” Puzzolan is a volcanic material. Other European authorities have reached similar conclusions, includ- ing Dr. W. Michaelis, the great German chemist and cement expert, who has been called the “Dean of the German cement industry.” It is unfortunate that some have confused these puzzolan-Portland blends with the so-called “puzzolan cements.” There is no connection, other than the mere accident of name. The most extensive use of cement blends in this country has been on the 250-mile aqueduct which is in process of construction, to furnish water to the city of Los Angeles, California, and the surrounding country. The amount of cement required for this work is about 1,500,000 barrels, enough cement for 1200 miles of concrete road 15 feet wide, laid according to the California specifications. The aqueduct is large enough to carry a fair sized stream — equal to about half the low water flow of the Willamette river at Eugene. On this work blends, made by grinding together equal parts of Portland cement and volcanic rock, have been used with entire success during the past four years. The aqueduct crosses the Mojave Desert, and the concrete had to be laid, with a scant supply of water, in the heat and drouth of the desert. This furnishes the complete and final answer to those skeptics who de- clared that concrete made with these blends could not be successfully laid in dry weather. A lengthy account of tests made and processes used has been given by Mr. J. B. Lippincott, one of the engineers of the Aqueduct Commission, in a recent communication to the American Society of Civil Engineers. Samples, of the blends above referred to were sent to the United States Bureau of Standards for analysis and tests. The following is a quotation from the report of this Bureau : “You desired particularly to know whether there was any chemical reaction between the tufa and the cement. The enclosed report shows that such has undoubtedly been the case. The addition of tufa or puz- zolana to Portland cement undoubtedly does not reduce the strength of the latter when in the form of a mortar or concrete. ’ ’ The engineers of the United States Reclamation service have been carrying on an extensive series of tests along this line for over five years, using a great variety of materials in making the blends. As a result of these investigations, cement reground with local materials has been adopted for the large masonry dams and for the auxiliary works in connection with a large earthern dam, all under construction by the Reclamation service. Considering the very extensive tests referred to in preceding para- graphs, and the long practical experience with blends in Europe, ex- tending over a period of more than 30 years, and the more recent ex- perience of over four years on the Los Angeles aqueduct, and in vieflv of the still more recent adoption of blends by the United States Government for the construction of important dams, the only question that remained in the mind of the writer was whether the materials easily available in Oregon would prove satisfactory when reground with Portland cement. The tests made at the University of Oregon have shown that one material which is of wide distribution in the Willam- ette valley gives very excellent results, while other materials, available in certain localities, are also suitable. A brief account of these tests will now be given. The proposed stone was first crushed to fragments not exceeding one- twentieth inch in diameter. These fragments were then mixed with cement in definite proportions by weight. The mixture was then ground to a fineness somewhat exceeding that of the original cement so that 90 per cent would pass through a No. 200 sieve. The finely-ground mixture is for convenience called a “ blend,” and is subsequently used in the same manner as cement. For brevity, the blends are designated by letters, the letter “ A, ” however, designating pure Portland cement. Among the blends experimented with, the following may be noted: “A ”- — Pure Portland cement. “B” — 1 part cement to 1 part “Eugene puzzolan. ” “C” — 2 parts cement to 1 part “Eugene puzzolan.” “D” — 1 part cement to 1 part diatomaceous earth. “E” — 2 parts cement to 1 part diatomaceous earth. “F” — 1 part cement to 1 part “Grants Pass puzzolan. “H” — 2 parts cement to 1 part slag. “K” — 1 part cement to 1 part “Clackamas puzzolan. ” The stone designated above as “Eugene puzzolan’’ was obtained at Eugene, but is abundant in the Willamette Valley. It is needless to say that the term “puzzolan” does not adequately describe its mineralogical character; its chemical composition is similar to that of the tufa used on the Los Angeles aqueduct. The diatomaceous earth was taken from a deposit near Eugene; it is an excellent material for this purpose, but so far such deposits have been found only in a few places in Oregon, and it would not therefore be generally available. The Grants Pass and Clack- amas “puzzolans” are similar to the Eugene puzzolan, but not identical. The slag in blend “ H ” is a blast furnace slag from Oswego. It is some- times termed an artificial puzzolan, and similar slags are often used in Germany. Cement is almost never used pure in practical work; it is mixed with sand or stone screenings to form mortar or concrete. Accordingly, the practical question with reference to these blends is this: What is the relative strength of mortars made with the blends and of similar mortars made with the same cement not blended? To determine this question, both tensile and compressive tests have been made. The tension briquettes were composed of 1 part cement or blend to 3 parts sand by [7J weight. Some of the more important results are shown on the follow- ing diagram: The sand used was crushed quartz, screened to the same size as stand- ard sand. It is of very uniform character, but gives results in tension that are uniformly about 80 per cent of those obtained with standard Ottawa sand; accordingly all the tension curves, including curve “A,” have been reduced to the basis of standard sand, to facilitate comparison by those acquainted with cement testing. The compression tests given below have not been reduced to the basis of standard sand, because the relationship is not known for compression. It will be noted that all the blends shown surpassed the cement in tensile strength at the end of 28 days, except blend “K, M for which the 28-day tests had not been made at the time the figure was drawn. The tests made subsequently showed that blend “K” also surpassed the cement at 28 days. Some typical compression tests are as follows: Mortars similar in every resepect to those used in the tension briquettes, age 2 months : lbs. per sq. inch. With cement 3030 With blend “B” 4210 With blend 4210 Mortars similar to tension briquettes, except that river sand was used instead of quartz; age 2 months, 1 week: lbs. per sq. inch. With cement 4140 With blend “D” 3010 With blend “E” 5760 Mortars composed of 1 part cement or blend to 5.22 parts sand by weight ; the sand was coarser than the above ; age, 2 months : [ 8 ] lbs. per sq. inch. With cement 2526 With blend “B” 2125 With blend “C” 4240 From the results quoted above, and others not here quoted, it has been found that blends composed of equal parts by weight of cement and puz- zolanic material have somewhat less strength in compression than the original cement, but that blends composed of 2 parts cement to 1 part puzzolanie material by weight have far greater compressive strength than the original cement. However, it must be noted that all the puzzo- lanic materials experimented with are lighter than the cement, so that equal parts by weight give an excess of puzzolanie material by volume, and the results obtained indicate that 53 per cent by weight of cement to 47 per cent “ Eugene puzzolan” will give compressive strength equal to that of the original cement in either 1:3 or 1:5 mortars; while equal parts by volume of cement and diatomaceous earth will surpass the orig- inal cement. Blend (< E” is practically equal parts by volume, the diatomaceaus earth being only about half as heavy as cement. On the Los Angeles aqueduct, the blends chiefly used were composed of equal parts by volume of cement and tufa. In corroboration of his own experiments, the writer wishes to refer in this place to experiments made by Dr. W. Michaelis, of Chicago (the younger Michaelis), on blends made with cement and puzzolanie material obtained on the Columbia River. After describing his experiments, Dr. Michaelis says: “The experiments described in the foregoing show that the puzzolana used in these tests results in a commercial product) of excellent qualities, if reground with Portland cement in the proportions of 2 parts of Port- land cement to 1 part of puzzuolana, or even of equal parts of both in- gredients. The latter mixture has been found to be equal or even supe- rior to pure Portland cement, both in tension and compression, in all lean mixtures, which correspond with actual working conditions. Richer mixtures are only rarely used, and should they be desired, the puzzuo- lana-Portland cement mixture will prove to be of especial advantage on account of the greater saving involved. In rich mixtures, the pur§ Port- land cement has been found to be slightly superior in strength during the first months, but at the end of a year and at all subsequent dates, the puzzuolana-Portland cement mixtures mentioned are equal to pure Portland cement also in these rich mixtures. ’ ’ Up to the present time, the greater number of experiments conducted by the writer have been with the “Eugene puzzolan,” because of its known wide distribution. However, other materials, available in special localities, are under investigation, and some of them give promise of very excellent results. As to the expense of grinding these blends, the itemized costs of the three mills used on the Los Angeles aqueduct are available. In compar- ing these costs with the corresponding items in Oregon, the writer sees no reason why any of the items should be greater in Oregon than in California, with the possible exception of the cost of power.. The items are: Cost of power, cost of quarrying, cost of mill operations, and gen- eral expense, including labor, live stock, etc. The total of these items is 37 cents per barrel, of which the power item is 10.5 cents, and the other items 26.5 cents. In Eugene, the California power cost can be dupli- [ 9 ] cated, or practically so, and doubtless the same can be done in many other localities, but not in all. Probably a fair average cost in Oregon would be 40 cents per barrel. The California costs were obtained on relatively small mills, turning out an average of about 16 barrels per hour, or enough in 10 hours to lay about one-eighth mile of concrete road according to the California specifications. In any locality where material equal to the “Eugene puzzolan” is available, the cost of a suitable blend per barrel would be figured by taking 53 per cent of the cost of a barrel of cement and adding to thaft the cost of blending, say 38 to 40 cents. For example, the cost of cement in Eugene in carload lots is $2.38 per barrel, net. Allowing for a 3-mile haul from the railroad to the location of the plant, at 20 cents per ton- mile, the cost of hauling per barrel would be 10 cents, making a total of $2.48 per barrel for the cement. The cost of the blend would then be as follows : Cement, .53 barrel, at $2.48 $1,315 Total cost of blending, say 38 Cost of blend per barrel $1,695 At Portland the cost of such a blend should not exceed $1.44 per barrel, on the basis that the cost of cement at the grinding plant is $2.00 per barrel. It is proper to remark that the cost of the necessary grinding machin- ery is considerable, so that in order to keep the cost of grinding low it is necessary that the first cost of machinery be distributed over a large total output, extending over several years, or preferably over the eco- nomic life of the machinery. This w 7 ould be possible, for instance, in case any county should enter upon a systematic plan of building concrete highways for a series of years. It should also be stated that the economy of blends results from two sources : Nearly half of the cost of cement at the factory is saved, and nearly half of the freight bill is saved. The latter circumstance explains why it would not in general be economical to have the blends ground at the cement factory — a suggestion sometimes advanced. RELATIVE COST OF MACADAM AND CONCRETE ROADS. In considering the relative cost of macadam and concrete roads, it is sufficient to consider those items which are different in the tw T o cases. The cost of grading, drainage, shaping and rolling the roadbed, and in fact all costs up to the point of placing the road material, should be the same *for macadam as for concrete, and these costs will therefore not figure in the comparison. Cost of Macadam. — The cost of macadam varies of course within cer- tain limits, just as the cost of anything else varies, depending on local conditions. The very wide variations sometimes reported are generally due to two causes : The very low costs are for inferior construction which would not comply with standard specifications; the very high eosts are generally due to incompetent management, though of course unusual conditions sometime prevail. However, the cost of standard construction under normal conditions and competent management is fairly well known, and the following typical cases may be cited: [ioj In Massachusetts, in 1907, the average of 64 contracts figures out 54.7 cents per square yard of macadam, 6 inches thick, compacted in place. This was for roads on which local stone was used; where imported trap rock was used, the cost was from 10 to 15 cents greater per square yard. In New York, Mr. H. A. Van.Alstyne, who was state engineer, at the time New York began her recent good-roads campaign, estimated that 57 cents per square yard was a fair cost for macadam 6 inches thick and an average haul of two miles; this with crushed stone at 85 cents per cubic yard, common labor at $1.50 per 8-hour day, and team and driver at $4 to $4.50 per day. This figure (57 cents) does not include the con- tractor’s profit, which was estimated at 20 per cent. Where sand was used as a binder in the lower half, the cost was somewhat less. In New Jersey, in 1908, the average cost of 146 miles of macadam was 65 cents per square yard, including contractor’s profit, which would make 54.2 cents per square yard, if profit be figured at 20 per cent of cost. In Lane County, Oregon, last year, 5 miles of excellent macadam were laid at a cost of 51.3 cents per square yard for the macadam alone. This figure does not include interest and depreciation on machinery. H. P. Gillette (“Handbook of Cost Data”) works out an itemized cost of mac- adam in great detail for an ideal case. If freight charges on materials be omitted, then for a two-mile haul, Mr. Gillette’s estimate would be 55.5 cents per square yard for 6-inch thickness, with crushed stone at 75 cents per cubic yard, labor at $1.50 per 10-hour day, and team and driver at $3.50 per day. In many cases the cost of macadam has materially exceeded the figures quoted above, and there are few if any cases where the cost has been materially less, except where inferior construction was used, and inferior construction is generally not worth its cost, however small. On the whole, it seems fairly evident that standard macadam, 6 inches thick, cannot be laid under average conditions for much less than 54 cents per square yard. It must be remembered that this is for the macadam alone, not including grading, drainage, or any of the preliminary operations. As to the cost of upkeep of macadam roads, this varies widely of course Avith different conditions as to amount and character of traffic. The intensity of traffic is constantly increasing, and it will be well for the people of Oregon to have an eye to the future, bearing in mind what has already happened in other places. A correspondent of the magazine “Good Roads,” stated in the issue of Dec. 7, 1912, that macadam roads in Dallas County, Texas, cost $600 per mile per annum for maintenance. This would amount to 6.82 cents per square yard per annum for a 15- foot road. In speaking before the International Road Congress in 1910, L. Mazerolle, Engineer of Bridges and Roads, Paris, stated that the cost of maintaining macadam roads reaches 36 cents per square yard per year in certain instances. At the same meeting, Mr. Austin B. Fletcher, then Avith the Massachu- setts HighAvay Commission, now Highway Engineer of Calfornia, stated that the maintenance charge of state highways in Massachusetts has increased from 1.14 cents to 5.7 cents per square yard per annum in the last three years, that is from 1907 to 1910. During that time Massachu- setts’ improved highways were mainly macadam. W. P. Judson (“Roads and Pavements”) states that “in Paris the annual cost of maintenance of suburban macadamized streets having [ii] light traffic is about one-third the original cost of building them. In some cases of extra heavy city traffic, the annual care costs one-third more than the original building.” That is to say, under light suburban traffic, a macadam road has to be rebuilt every three years, and under heavy traffic every nine months. In examining the reports of numerous State Highway Commissions, the writer has found many instances where the cost of maintaining mac- adam roads is tabulated at $1000 per mile per- annum, or more, and $500 or over is a frequent figure. In an address delivered at Chicago last summer, Mr. Logan W. Page, Director of tlie United States Office of Public Roads, made the following statement: L ‘ Figures collected by the office of public roads relative to the cost of maintenance of plain macadam and bituminous macadam pavements under fairly heavy traffic conditions indicate that these pave- ments, when properly maintained, entail an annual absolute maintenance charge of approximately $450 per mile per annum for plain macadam, and possibly from $800 to $1000 per mile per annum for bituminous macadam, for 15-foot surfaces. These figures have led me to believe that we must seek a more permanent form of pavements for country road surfaces.” These figures, collected by the United States office of public roads make the cost of maintenance per square yard 5.11 cents per an- num for ordinary standard macadam, and about double this for bitumin- ous macadam. This is what Oregon is coming to in the next few years, in the matter of maintaining macadam roads, if the state makes the progress for which all loyal citizens are hoping. Maintenance costs published only a few years ago, now appear ludicrous. Five years ago the Massachusetts Highway Commission thought they would be able to maintain their roads indefinitely at a cost of 2.25 cents per square yard per annum, but with- in three years the cost had soared to 5.7 cents, and now the attempt to maintain macadam roads has been practically abandoned, and all sorts of surfaces, patented and otherwise, are being tried out in the hope of saving some of the many millions of dollars which the state has put into macadam roads. In New York, the experience has been the same, and so nearly every- where throughout this country and Europe. The cases cited are merely typical of what is happening generally, where the people are progressive enough to utilize the roads. But there is no thought anywhere that the right of the people to use the roads for which they have paid shall be curtailed. On the other hand, the people are calling for more and better roads, and are insisting that the roads must be made to serve. The state of New York, in a popular election last November, voted another $50,000,000 for roads, and the measure earned in every county. The mileage of improved roads in Oregon at present is relatively in- significant and almost wholly disconnected; so of course the traffic is small, since no large traffic is possible, and the present cost of main- tenance is low. But witih the expected development of the state, it is safe to say that the cost of maintenance of macadam roads within the next five or ten years will have reached the cost already passed in so many places, and the figures given by the office of public roads in the quotation above may be regarded as a conservative estimate of the cost which we must shortly expect, namely, 5.11 cents per square yard per annum. [ 12 ] Cost of Concrete Roads. — It is the present purpose to compare the cost of concrete roads with the cost of macadam roads subject to the traffic conditions just considered. In making such a comparison the first ques- tion that arises is this : What thickness of concrete may be fairly com- pared with the standard thickness of 6 inches for macadam? Plainly, it is not fair to compare 6 inches of concrete with 6 inches of macadam, because concrete is far stronger than macadam, and any comparison that pretends to be fair and rational must take this fact into account. No sane person would think of comparing the cost of a steel bridge, for in- stance, with the cost of a wooden bridge on the supposition that the vol- ume of steel should equal the volume of wood; on the contrary, the relative volumes would be made to correspond with the relative strengths of the two materials, and the same consideration should govern the com- parison of concrete and macadam. While it is not possible to state with precision just what thickness of concrete is equivalent to 6 inches of macadam, it is possible to state that 4 inches of good concrete will have greater supporting power than 6 inches of good macadam on the same subgrade, and conversely in any location where 4 inches of concrete would not be sufficient, neither would 6 inches of macadam be sufficient. Where a weak subgrade requires 6 inches of concrete, at least 9 or 10 inches of macadam would be required to give anything like the same supporting power; and where on a good subgrade 6 inches of concrete is required on account of very heavy individual loads, macadam should not even be considered. A thickness of 4 inches is considered the minimum permissible for a concrete base, and if we compare this thickness with 6 inches of macadam, the comparison will be more than fair to macadam; but to compare the cost of concrete and macadam inch for inch, as has sometimes been done, is an absurdity of which no engineer should be guilty. In cities the thickness of concrete base for pavements varies from 4 to 6 inches, the latter thickness being generally used for heavy traffic. For the large mileage of concrete roads now under way in California, a 4-inch thickness has been adopted, and the California Highway En- gineer is not a novice at the business. There are other competent men who believe that a thickness of 4 inches of concrete is sufficient. Mr. George C. Warren, president of Warren Brothers Company, in a paper read before the sixteenth annual convention of the American Society of Municipal Improvements, stated that in his judgment, based on over twenty-five years of experience, under ordinary conditions of well-rolled sub-soil, 4 inches of concrete was ample, provided the concrete was not disturbed until thoroughly set. He stated further that he had many cases of practical experience to back his judgment, and that with a 6- inch base, one-third of the concrete was a wanton waste, except over poorly-filled trenches, where settlement may occur. Even with the prospect of a greatly increased traffic, it hardly seems necessary that our country roads should be as heavy as the pavements in the congested districts of New York City, and it appears probable that 4 inches of good concrete will be sufficient, always with the proviso of proper drainage and compacting the subgrade, for either macadam or concrete. In special cases of extraordinary traffic or poor subsoil, a greater thickness will doubtless be necessary. Assuming, for the purpose of comparison, a thickness of 4 inches, we may arrive at a fair estimate of relative cost. The writer favors a [ 13 ] ( concrete base, with a thin bituminous wearing- surface, as adopted in New York and California, but will also give an estimate for a two- course type of all-concrete road. For the California type the estimate can best be arrived at by con- sidering the contract prices in California, and making the necessary changes to suit local conditions. The average of six contracts in Califor- nia, for which itemized bids are at hand, covering 56 miles of road, makes the cost of concrete in place $3.45 per cubic yard, exclusive of cost of cement, inclusive of cost of stone and sand, freight charges and hauling, mixing and placing, and protection from the. sun and sprinkling while hardening, and hauling the cement, which was furnished by the state. The concrete mixture is in the proportion 1 :2y 2 :5. This will require about 1.3 barrels of cement per cubic yard of concrete rammed in place. Taking conditions at Eugene as typical of local conditions, the cost of a suitable cement blend, equal to the cement of which it is made, as previously stated, will be $1.70 per barrel, making $2.21 for the cement in a cubic yard of concrete. There is no good reason why the cost of the remaining items enumerated above should exceed the California price quoted, and in fact with proper equipment the Eugene price would be less. Adding the items, we obtain $5.66 as the total cost per cubic yard of rammed concrete. Since one cubic yard will lay 9 square yards 4 inches thick, the cost of concrete per square yard, is about 63 cents. The cost of the bituminous surface or wearing coat in California is about 5 cents per square yard, the bituminous part being a heavy as- phaltic oil containing 90% asphalt. This price includes the cost of oil, freight, hauling, heating and spreading the oil, cost of stone screenings and sand, hauling and spreading same, and in fact the total cost of the surface coat. The only items which should be increased for Eugene conditions are the freight charges on the asphaltic oil from California to Eugene, and a possible increase in the cost of fuel used to heat the oil before placing. The freight charge in carload lots is $1.44 per barrel, and since one barrel will lay about 125 square yards, the extra freight charge will add 1.15 cents per square yard, and making a liberal allow- ance of 0.35 cent for the possible extra cost of fuel, the total cost in Eugene would be 6.5 cents per square yard. This added to the cost of concrete figured above makes the total cost of concrete and bituminous surface 69.5 cents. The length of life of this bituminous surface cannot be definitely stated, but experience with similar surfaces in the east indicate a prob- able life of five years. To. be on the safe side, we may cut this time in two, and assume that the life will be only two and one-half years, o«r four complete renewals required in a ten year period, or 2.6 cents per square yard per year, as an average cost of maintenance. On the basis of the preceding discussion, the relative cost of macad- am and of a concrete road of the California type for a ten year period, will be as follows: Macadam Concrete First cost $0,540 $0,695 Upkeep for 10 year period 0.511 0.260 Total per square yard $1,051 $0,955 [ 14 ] This indicates a saving of 9.6 cents per square yard in favor of the California type of concrete road over macadam, or about $845 per mile, for 15-foot road. This saving is effected by the use of a suitable cement blend. If cement were used without blending, the saving would be wiped out, but the concrete road would still compete with macadam on the basis of equality of costs for a ten-year period, greater supporting power, and a better road surface all the time. While it can hardly be denied that a comparison of costs of 4 inches of concrete and 6 inches of macadam is at least fair to macadam, con- sidering the relative supporting power of the two materials, yet the question ought to be raised whether or not 4 inches of concrete is a practicable thickness for Oregon conditions generally, outside the larger cities. If if is, then Mr. Warren is right in saying that any greater thickness would be a wanton waste. Believing that a practical test is of more value than any amount of opinion or argumentation, the writer intends, if possible, to construct a short stretch of concrete road during the coming spring, substantially as described above, with a 4-inch base, and bituminous surface. The effort will be made to find a location that is at once typical of general conditions, and at the same time subject to fairly heavy traffic. Without aiming to forestall the conclusions obtained from such a test, the following considerations may be adduced as pertinent to the question, and as a sufficient justification for making the test. In many eastern cities there is a large amount of concrete base 4 inches thick, laid years ago for sheet asphalt pavements, on streets where the traffic was then light; in later years, the traffic, has become heavier, and yet the concrete supports it safely, steam road rollers weighing 12 and 15 tons traversing the streets with impunity. In many cities, where a 6- inch base was considered to be required by the traffic, or other condi- tions, the concrete was made with natural cement, and a 6-inch thick- ness of such concrete is hardly equivalent to 4 inches of Portland ce- ment concrete. In the city of Eugene, among the many kinds of pave- ments in use, are some with a 4-inch base and a thin covering of bitu- men ; they bear all kinds of traffic with no evidence of injury to the base. Another type of road that' is worthy of consideration has a concrete base and also a concrete wearing surface composed of a richer mixture than the base. For the base course the mixture described for the pre- ceding type would be suitable, but for the wearing course the writer does not as yet recommend the use of cement blends. Their resistance to abrasion has not as yet been determined, and until he can do this, the writer recommends the use of Portland cement without blending for the wearing surface, iy 2 inches thick. The proportions would be 1:2:3 y 2 , which is the mixture adopted for one-course concrete roads in Illinois. The writer considers that the use of sand mortar for a wearing course is objectionable. For this mixture about one and two-thirds barrels of cement per cubic yard of concrete would be required. Taking the cost of cement at Eugene at $2.38 per barrel net, the total cost of the wearing surface would be $7.58 per cubic yard, or 31.6 cents per square yard. This, added to the 63 cents per square yard previously estimated for the 4-inch base, makes the total cost of concrete 94.6 cents per square yard. The wearing surface must of course be laid im- mediately upon the base course, so that the two courses solidify to- gether and form virtually one course. The cost of maintenance for a concrete wearing surface is not known, but is undoubtedly very low if [ 15 ] / the work is well done. If the cost should prove to be not more than one cent per square yard per year, on the average, this road could com- pete in cost with a 6-inch macadam road in a period of ten years, under the traffic conditions assumed; but any such comparison of costs would be wholly unfair to concrete, as above stated, since the 5y 2 inches of concrete would be equivalent to at least 8 or 9 inches of macadam in supporting power, assuming both on an equally satisfactory subgrade. On a considerable mileage of concrete roads built in New York last summer, the thickness was only 5 inches. In case roads are built from the proceeds of bonds, the interest on bonds of course is an item in the total cost. But since the chief justi- fication of bonds is that they enable a large mileage of roads to be built and put into service in a short .time, the interest account may fairly be balanced against the greater convenience and quicker use of the roads. If this does not offset the interest charge, then bonds would hardly be justified. For this reason no allowance for interest has been made in the preceding estimates. SUMMARY AND CONCLUSION. It is impossible to foretell how great the .traffic on Oregon roads may become during the next decade. But the experience of other states and our own ambitions toward rapid development all point to the conclusion that the further construction of macadam roads will be uneconomical and therefore unwise, at least on our more important highways. Concrete roads offer a far better solution, and by the use of methods which have found favor both in Europe and America, their first cost can be brought within reason, and their ultimate cost will be less than that of macadam. The numerous experiments that have been tried with bituminous ma- cadam have in the main proved costly, and the maintenance charge is found .to be so great that the Highway Commission of New York has declared that the results do not justify the cost, and the Commission has therefore adopted the concrete road as a standard for future construct- ion. Standard types of city pavements, such as bitulithic or vitrified brick, are too expensive for the great majority of country roads, and concrete is ‘the most promising material in sight that embodies at once a moderate first cost and reasonable maintenance charge. Macadam roads already built may be saved, at least for a time, by the use of various surface dressings, but such expedients do not embody a satis- factory policy for future construction. In view of the already rapid destruction of many well-built macadam roads in Oregon, and in view of the universal conclusion that macadam is not adapted to modern traffic, such as motor trucks, road tractors, and mechanical haulage generally, and in view of the fact that such traffic will in all probability be the prevailing traffic within a few years, the writer believes that any large expenditure in the future construction of macadam, whether by the state or by counties individually, would be unwise and result only in bitter disappointment. Hindsight is proverbially better than foresight, but it is vastly more expensive. If we are to vote millions of dollars in bonds to build good roads, we may well profit by experience elsewhere, and seek a type of construction that will stand modern traffic. “Seize time by the fore- lock;” quit, grabbing at the fetlock. [ 16 ] wim 3 0112 105762527