THE CONSTRUCTION OF ROADS AND STREETS. THE -CONSTRUCTION OP ROADS AND STREETS IN TWO PARTS I THE ART OF CONSTRUCTING COMMON ROADS BY HENRY LAW, M.I.C.E. a * REVISED AND CONDENSED BY D. KINNEAR CLARK, M.I.C.E. II. RECENT PRACTICE IN THE CONSTRUCTION OF ROADS AND STREETS INCLUDING PAVEMENTS OF STONE, WOOD, AND ASPHALTE BY D. KINNEAR CLARK, M.I.C.E. AUTHOB OF "TRAMWAYS: THEIR CONSTRUCTION AND WORKING;" EDITOR OF " STEAM AND THE STEAM ENGINE," " CIVIL ENGINEERING," " LOCOMOTIVE ENGINES," " FUEL : ITS COMBUSTION AND ECONOMY," ETC. ETC. eMitlt IOURTH EDITION. LONDON CROSBY LOCKWOOD AND SON 7, STATIONERS' HALL COUET, LUDGATE HILL 1890 [All rights reserved] LONDON : ' PRINTED BY J. S. VIRTUE AND CO., LIMITED, CITY ROAD. PKEFACE. THE present work consists of two parts. The first com- prises " The Art of Constructing Common Koads," by Mr. Henry Law, revised and condensed ; the second consists of "The Eecent Practice in the Construction of Eoads and Streets/' by Mr. D. Kinnear Clark, C.E., in the investigation of which he has been indebted for much material to the excellent Eeports of Lieutenant-Colonel Haywood, Engineer and Surveyor to the Commissioners of Sewers of the City cf London. The whole is preceded by an historical sketch of the subject, also by Mr. Clark. The City of London is a microcosm of the best and most varied experience in carriage-way construction, under the superintendence of the Engineer who has lucidly described the various structures which have, from time to time, been laid down and tried, in a catholic spirit, and has recorded the results of his experience, in a series of Eeports ranging over a period of thirty years from 1848 to 1877. Mr. Clark has endeavoured impar- tially to set forth the merits and disadvantages of the systems of pavement which have come under his observa- tion, and he believes that the results of his investiga- tions will be useful to others. The varieties of wood pavement and of asphalte pavement which have been laid in the Metropolis more especially in the City have been fully described and, it is hoped, fairly criticised. Mr. Clark has also VI PREFACE. added a chapter on the Eesistance to Traction on Common Roads, in which he has endeavoured to educe the law of rolling resistance, and has contributed new formulas, with fresh data. Appended to the text will be found a portion of a paper by Sir John Burgoyne on Rolling New-made Roads ; some valuable extracts from Mr. Frederick A. Paget's Eeport on Eoad-rolling, containing several interesting historical facts ; and finally, a table showing the Condi- tion of Wood and Asphalte Carriage-way Pavements in the City of London, from a recent Report of Colonel Haywood. CONTENTS. HISTOEIOAL SKETCH. BY D. K. CLAHK. Country Roads. Barrelled Roads. Macadam's Roads. Tel- ford's Roads. Length of Metalled Roads, in 1868-69. Boulder Pavement. London Pavements. Wood Pavements in Russia. Wood Pavements in the United States. Stead's Wood Pavement. De Lisle's Wood Pavement. Carey's Wood Pavement. French iioeitia ..... PAET L CONSTKUCTION OF EOADS. BY HENRY LAW, C.E. CHAPTER I. EXPLORATION FOR ROADS : Principle of Selection of Route. Contour Lines. Taking Levels. Bench-marks. Sections. Laying out a Road 21 CHAPTER II. CONSTRUCTION OF ROADS : EARTHWORK AND DPATX^OE: Earthwork. Trial Pits. Working Plan. Onttinjra and Fillings. Side Slopes. Excavation in Rock. Slips. Drainage. Embankments. Catch-water Drains. Road on the Side of a Hill. Side-cuttings. Spoil-bank . 40 CHAPTER III. RESISTANCE TO TRACTION ON COMMON ROADS : M. Morin's Experiments. Sir John Macneil's Experiments. Resistance on Inclines. Table of Resistance on Inclined Roads. Professor Mahan's Deductions. Angle of Repose . 51 NOTE BY THE EDITOR: Sir John Macneil on Gradients. Professor Mahan on Gradients. M. Dumas on Gradients. M. Dupuit on Gradients 63 CHAPTER IV. ON THE SECTION OF ROADS : Gradients. Table of Gradients and Angles of Roads. Width and Transverse Section of Roads. Mr. Macadam's Views. Mr. Walker's Views. Proposed form of Cross Section. Professor Mahan's Views- Form of the Bed. Mr. Hughes' s Views. Drainage X CONTENTS. PAGE cal Sections and Plans of a Fifty-feet Street for the City. Southwark Street . i69 CHAPTER VIII. STONE PAVEMENTS OF LIVERPOOL : Mr. New- lands on the Length of Pavement in 1851. Tables of Cost for Construction of Set Pavements, Boulder Pavements, and Macadam ....... 193 CHAPTER IX. STONE PAVEMENTS or MANCHESTER: Early Boulder Pavements. Mr. H. Royle on Set Pavements. Use of Pitch Greeting. Cost. Macadam 198 CHAPTER X. WEAR OP GRANITE PAVEMENTS: In the City of London, with Tables. Data for Wear and Duration, with Table 202 CHAPTER XI. STONE TRAMWAYS IN STREETS: Mr. Walker's Tramways in the Commercial Road. Resistance on Stone Tramways. Granite Tramways in Northern Italy. Mr. P. Le Neve Foster, Jun.'s Data. Prices of Work at Milan . 208 CHAPTER XII. WOOD PAVEMENT: Dimensions of Blocks. Interspaces . . . . . . . . . .215 CHAPTER XIII. CAREY'S WOOD PAVEMENT : Carey's Pave- ment in the City of London, with Tables of Cost and Dura- tion. Carey's most recent Practice . . . . .217 CHAPTER XIV. IMPROVED WOOD PAVEMENT: First Laid in the City of London. Construction. Asphalte Grouting. Objections to the Flooring. Most recent Practice . . 223 CHAPTER XV. OTHER WOOD PAVEMENTS : Ligno-mineral Pavement. Asphaltic Wood or Copeland's Pavement. Harrison's Wood Pavement. Henson's Wood Pavement. Norton's Wood Pavement. Mowlem's Wood Pavement. Stone's Wood Pavement. Gabriel's Wood Pavement. Wilson's Wood Pavement. Table of Wood Pavements in the City of London 228 CHAPTER XVI. COST AND WEAR OF WOOD PAVEMENTS : Cost in the City of London. Mr. G. J. Crosbie-Dawson's Data. Mr. Ellice-Clark's Data 236 Wear in the City of London. Relation of Wear to Traffic. Table of Estimated Duration 238 CHAPTER XVII. ASPHALTE PAVEMENTS. First used in Paris. Mode of Construction in the City of London. Val de Travers Compressed Asphalte Pavement. Val de Travers Mastic Asphalte Pavement. Limmer Mastic Asphalte Pavement. Ba.rnett's Liquid Iron Asphalte Pavement. CONTENTS. XI PAQK Trinidad Asplialte Pavement. Patent British Asphalte Pavement. Montrotier Compound Asplialte Pavement. Societe Francaise des Asphaltes. Maestu Compound As- phalte. Stone's Slipless Asphalte. Bennett's Foothold Metallic Asphalte. Lillie's Composite Pavement. McDon- nell's Adamantean Concrete Pavement. Granite Pave- ments with Asphalte Joints. Table showing the Extent of Asphalte Pavements in the City of London, 18 73. -^Colonel Hay wood's Deductions from his Experience. Table show- ing Duration and Repair of Asphalte Pavements in the City of London, at March, 1873. Table showing the Wear of Asphalte Pavements in Proportion to Traffic. Colonel Hay wood's Conclusions as to the Durability of Asphalte Pavements. Cost and Terms of Contracts for Asphalte Pavements, with Table. Val de Travers Asphalte in Manchester 242 CHAPTER XVIII. OTHER PAVEMENTS: Metropolitan Com- pound Metallic Paving. Cast-iron Paving. Cellular-iron Pavement. Artificial Granite Pavement. Compound Wood and Stone Pavement. Concrete Pavement . . . .261 CHAPTER XIX. COMPARISON OF CARRIAGE-WAY PAVEMENTS : Comparative Costs of Pavements. Cost of Yorkshire Pav- ing-stones. Foot Pavements. Comparative Slipperiness, with Table. Comparative Convenience. Report of Com- mittee of the Society of Arts 264 CHAPTER XX. CLEANSING OP PAVEMENTS : Composition of Mud. Dr. Letheby's Analysis, with Table. Moisture in Mud. Cleansing by Machinery and by Manual Labour. Proportion of Granitic Detritus in Dust from a Granite Pavement. Comparative Cost of Cleansing Granite and Macadam. Watering with Jet and Hose, with Table Mr. J. Lovick's Experiments. Cleansing in Paris . . . 268 CHAPTER XXI. MOUNTAIN Ro ADS: Principle of Selection of Route Major James Browne's Data. Mr. Dobson on Road- making in New Zealand. Major Browne on the Cost of Roads in India, and Method of Construction . . .282 CHAPTER XXIL RESISTANCE TO TRACTION ON COMMON ROADS : Investigation of Rolling Resistance on Impressible Roads. Work in Compressing the Material. Resistance is In- versely Proportional to the Cube Root of the Diameter. M. Morin's Deductions. M. Dupuit's Deductions. M. De- bauve's Data, with Table. Resistance of M. Loubat's Om- nibus. Experiments by Messrs. Eastons and Anderson on Xll CONTENTS. ?AOI the Resistance of Agricultural Carts and Waggons. Sir John Macneil's Experiments on the Resistance of a Stage- coach, with Table. Formula for Resistance of a Stage- coach. M. Charie-Marsaines on the Performance of Flemish Horses, withTable. Mr. D. K. Clark's Data . . .290 APPENDICES. I. ON ROLLING NEW-MADE ROADS. By General Sir John F. Burgoyne, Bart .301 II. EXTRACTS FROM " REPORT ON THE ECONOMY OF ROAD MAINTENANCE AND HORSE-DRAFT THROUGH STEAM ROAD-ROLLING, WITH SPECIAL REFERENCE TO THE ME- TROPOLIS." By Frederick A. Paget, C.E. . . . 309 III. EXTRACT FROM THE REPORT OF COLONEL HAYWOOD, Engi- neer and Surveyor to the Commissioners of Sewers, City of London, ON THE CONDITION OF WOOD AND ASPHALTE CARRIAGE-WAY PAVEMENTS, ON THE IST FEBRUARY, 1877 324 INDEX ..,...,. -338 CONSTRUCTION OF EOADS AND STEEETS. HISTOKICAL SKETCH. BY D. K. CLARK, C.E IN the middle of last century, communication between towns was difficult. The roads were originally mere foot- paths, or horse-tracks, across the country, and the few wheeled carriages in use were of a rude and inefficient description, for which the roads were wholly unadapted. The roads were necessarily tortuous, every obstacle which the ground presented being sufficient to turn the traveller out of his natural direction. Many of these roads were carried over hills to avoid marshes, which were subse- quently drained off or dried up ; others deviated from their direct course in order to communicate with the fords of rivers now passable by bridges. The inland commerce of the country was chiefly carried on by transport on the backs of pack-horses, and the old-fashioned term load, commonly in use as a measure of weight, is a remnant of that custom meaning a horse-load. Gradually, the roads became practicable for the rude carriages of the times, and they were maintained, though in a very defective condition, by local taxes on the counties or parishes in which they were situated. So they remained until turn- pike-trusts were established by law, for levying tolls from B 2 HISTORICAL NOTICE. persons travelling upon the roads. Several of these trusts were established previous to 1765, and they subsequently became general, when the attention of all classes of the community was directed to the state of the highways. Bills for making turnpike-roads were passed, every year, to an extent which seems almost incredible ; and, in addition, every parish was compelled by the force of public opinion, supplemented by indictments and fines recoverable at common law against the trustees, when the roads were not maintained in proper repair. But the turnpikes formed a cumbrous system : they were trusts in short lengths about fifteen or eighteen miles and the surveyors em- ployed appear to have been ill -educated, and were appointed by favour of the trustees rather than for any professional knowledge. A long period elapsed before any good system of road- making was established. The old crooked horse-tracks were generally followed, with a few deviations to render them easy ; the deep ruts were filled with stones or gravel of large and unequal sizes, or with any other materials which could be obtained nearest at hand. The materials were thrown upon the roads in irregular masses, and roughly spread to make them passable. The best of those roads would, in our time, be declared intolerable. [Road- making, as a profession, was unknown, and scarcely dreamt of ; for the people employed to make the roads and keep them in repair, were ignorant and incompetent for their duties. Travelling was uncommon, and funds were scanty, and higher talent could not be commanded. Engi- neers, except in cases of special difficulty, such as the con- struction of a bridge over a deep and rapid river, cutting through a hill, or embanking across a valley, probably thought that road-making was beneath their considera- tion, and it was thought singular that Smeaton should have condescended to make a road across the valley of OLD COUNTRY ROAD. O the Trent, between Markham and Newark, in 1768. At the same time, civil engineers, according to Sir Henry Parnell, "had been too commonly deemed by turnpike-trustees as something rather to be avoided, than as useful and neces- sary to be called to their assistance." By-and-bye, as people became sensible of the value of time, easier and more rapid means of communication than the old roads were required : improved bridges were built with easier ascents ; and, in some cases, cuts were made to shorten the distances, though the general lines of the old roads were preserved. The roads, no doubt, were somewhat im- proved in this way, but there was no general system or concert between the district trustees. Mr. Arthur Young, in his " Six Months' Tour," pub- lished in 1770, writes of some of the roads in the north of England: " To Wigan. Turnpike. I know not, in the whole range of language, terms sufficiently expressive to describe this infernal road. Let me most seriously caution all travellers who may accidentally propose to travel this terrible country, to avoid it as they would the devil, for a thousand to one they break their necks or their limbs by overthrows or breakings down. They will here meet with ruts, which I actually measured four feet deep, and float- ing with mud only from a wet summer ; what therefore must it be after a winter ? The only mending it receives is tumbling some loose stones, which serve no other pur- pose than jolting a carriage in the most intolerable manner. These are not merely opinions, but facts ; for I actually passed three carts, broken down, in those eighteen miles of execrable memory." " To Newcastle. Turnpike. A more dreadful road cannot be imagined. I was obliged to hire two men at one place to support my chaise from overturning. Let me persuade all travellers to avoid this terrible country, which must either dislocate their bones with broken pavements, or bury them in muddy HISTORICAL NOTICE. sand." Even so much later as the year 1809, the roads answered to the description of Mr. Young. Mr. C. W. Ward, writing in that year,*' states that the convex sec- tion, as shown in Fig. 1, was the most prevalent in the Fig. 1. Common Convex Koad, in 1809. country. Under the impression that the higher the arch was made, the more easily the road would be drained, the materials were heaped up about the centre till the sides became dangerous, by their slope, for the passage of carriages. The carriages, therefore, ran entirely upon the middle till it was crushed and worn down, and then a fresh supply of materials was laid on, and the road was again restored to its dangerous shape. The sides of the road were but little used, except in summer, or until the heavy waggons had crushed the middle into a surface apparently compact and smooth. In some places, the rough materials were laid in a narrow line, not exceeding seven or eight feet in breadth, along the middle of the road, and the sludge collected from the scrapings of the roads or ditches was placed on each side, like banks, to prevent the stones from being scattered by the wheels. The high con- vex form was so exceedingly defective as to defeat the object for which it was constructed. Carriages were forced, for safety or for convenience, to keep to the middle, and ifc was speedily ploughed into deep ruts, which held the rain-water, even when the convexity approached to the form of a semicircle. The central elevation, therefore, was not kept dry ; and the central pressure of the traffic forced the material upon the sides, where they lay loose * Third Report from Parliamentary Committee on Turnpikes and Highways, 1809. OLD COUNTRY ROAD. 5 and unconnected, and obstructed the course of water from the middle. The condition of such a road, ploughed and disintegrated, is illustrated in section by Fig. 2, when it Fig. 2. -An Indicted Road. Its first state. Year 1809. was, probably, indicted. It was common for the parish- surveyor after harvest to make a contract with a stout labourer, who took job-work, for the reparation of the road, with a special injunction "to be sure that he threw up the road high enough, and made the stones of the old causeway, or foot pavement, go as far as they could." The diligent operator fell to work ; nor was he stopped by the equinoctial rains in September, for the work must be done, as contracted for, before the Michaelmas sessions. He accordingly produced something, Fig. 3. The clods Fig. 3. The Indie' ed Road thrown up, to tak^ wif the Indictment, under the direction of a Parish Survey jr. Its second state. and rushes were thrown into the bottom, and the soft soil which nourished the vegetation, and all other materials, hard or soft, were laid down, forming a convexity of con- siderable elevation, according to order: barrelling the road, as it was called. The whole was duly surmounted with the stones from the old broken footpath, with a little gravel raked over them, just to keep them together. Finished 6 HISTORICAL NOTICE. thus, say by Saturday night, then on the following Monday it was submitted for inspection to two magistrates, on their way to the quarter sessions. How could they possibly refuse to speak the truth ? they certified " that it was per- fectly smooth when they saw it, and that a vast deal had been done since the last time they were there." But besides tear and wear, decomposition immediately took place in the chaotic mass, and, in the second or third year after the repair, the road was reduced t ) the condition shown in Fig. 4, in its last and worst state. Although it appears that the practice of road-making, Fig. 4. The same Road, in its third year after repair, or its last and worst state. even at the commencement of the present century, was sadly deficient, it is, nevertheless, fair to add that persons of intelligence were aware of the first requisite for a good road. Mr. Foster, of Bedfordshire, in 1809, saw that it was desirable, " first, to lay a substantial founda- tion of the hardest stone or coarsest gravel that could be procured, and then to coat it with a finer and more level surface." It followed, from the imperfect condition of the roads, that the wheels of vehicles were required to be of great width, in proportion to the weight carried on each wheel. The following table shows the proportions and the distri- bution of weight on the wheels, according to the regula- tions of the Act which was in force in the early part of this century. The rolling widths are the slant widths of conical wheels : MACADAM S ROADS. TABLE No. 1. WEIGHT, HORSE-POWER, AND WHEELS OF VEHICLES ON COMMON ROADS. 1809. Weight Breadth of wheel. Gross weight. Number of horses. Draught of each horse. on the road at each Pressure per inch of width. wheel. inches. tons. cwt. Ib. cwt. Ibs. 16 . 8 10 16 40 280 9, rolling 16 6J 8 16 42 **i 404 9 6 8 15 30 373 6, rolling 1 1 H 6 18 37 27^ 513 6 . il 6 16 221 420 3 3 i 4 17 56 17j 653 2, Stage coach 4 4 20 20 1120 Here, it is apparent that the pressures per inch of width of tyres increased as the width diminished. In the opinion of the practical men of that day carriers and others the pressure should have been limited to about 4 cwt., or 448 Ibs., per inch wide ; and it was maintained that the minimum width of wheel for any vehicle should be 4tJ- inches. About the year 1816, Mr. James L. Macadam, who had for many years previously given his attention to the state of the roads, assumed the direction of the roads of Bristol, and he put in practice the leading principle of his system of road-making, namely, " to put broken stone upon a road, which shall unite by its own angles so as to form a solid, hard surface.' ' " It follows," he adds, "that when that material is laid upon the road, it must remain in the situation in which it is placed without ever being moved again ; and what I find fault with in putting quantities of gravel on the road is that, before it becomes useful, it must move its situation, and be in constant motion."* The principle was to substitute small angular stones, such * " Report of the Select Committee on the Highways of the King- dom, 1819," p. 22. 8 HISTORICAL NOTICE. as resulted from the breakage of larger stones, for rounded stones ; so as to form, a sort of mosaic or interlocking sys- tem. This is the distinctive novelty of the system of Macadam, and its value has been established by universal experience. Mr. Macadam also maintained that no greater convexity should be given to the surface of the road, in transverse sections, than was sufficient to cause rain-water to run readily into the side channels. The surface of the road was kept even and clean by the addition of proper fresh materials when necessary, distributed equally in thin layers immediately after rain, in order that the new mate- rials might bind and incorporate properly with the old. Macadam's system of construction consisted in simply laying a stratum of flints, or other hard materials, 10 or 1 1 inches thick, broken equally into small pieces about 2 inches in diameter, and spread equally over the intended road-surface. The broken " metal" became consolidated by carriages passing over it. Without any specialty of professional training, except the faculty of acute obser- vation, Macadam effected great improvement of the sur- face of the roads immediately under his charge ; and, by his business-like and extended views on road - adminis- tration, he established for himself a world -wide repu- tation. He professed to be a road-maker only, and he devoted his whole time and attention to the propagation of his system. He found the roads in the Bristol district loaded with two or three feet of materials, of large and irregular size, which had for years been accumulated on the surface. The heaps were utilised as quarries of stones partially broken on the spot ; the stones he excavated, separated from the mud, and reduced by breakage to a uni- form size, 6 ounces in weight. After having been so broken, the stones were relaid, and were carefully and regularly raked and levelled during the process of consolidation. In TELFORD'S ROADS. 9 this way, with the addition of effective drainage where necessary, he was enabled to make a good surface on roads which previously were almost impassable. As nearly every road had more metal upon it than was necessary, he, and the surveyors appointed by him, established economy in the construction and maintenance, as well as in the admi- nistration of the finances, and his system became generally adopted. Whilst Mr. Macadam deserved well as the pioneer of good road-construction, it may be observed that he had been anticipated in the promulgation of the system of a regularly broken -stone covering by Mr. Edgeworth, an Irish proprietor, whose treatise on roads, of which the second edition was published in 1817,* contains the results of his experiments on the construction of roads, with some useful rules. He advocated the breaking of the stones to a small size, and their equal distribution over the surface. He also recommended that the interstices should be filled up with small gravel or sharp sand a practice which, though it was condemned by Macadam, is now adopted by the best surveyors. Since Macadam's time, the practice of road-making has been greatly improved by the use of the roller for com- pressing and settling new materials, and of preparing at once a comparatively smooth and hard surface for traffic. Telford first directed his attention in 1803-4, to the construction of roads. He was employed chiefly in the construction of new roads hundreds of miles of roads in the Scottish Highlands ; also the high road from London to Holyhead and Liverpool, and the great north roads, formed in consequence of the increased communication with Ireland after the Union, and which were excellent models for roads throughout the kingdom. Telford set * u An Essay on the Construction of Roads and Carriages/' 2nd edition, 1817. B3 10 HISTORICAL NOTICE. out the roads according to the wants of the district through which they were made, as well as with a view to more distant communication ; and the acclivities were so laid out, that horses could work with the greatest effect for drawing carriages at rapid rates. As a notable instance of the wonderful improvements that were effected by Tel- ford's engineering skill applied to the laying out of new roads, an old road in Anglesea rose and fell between its extremities, 24 miles apart, through a total vertical height of 3,540 ft. ; whilst a new road, laid out by Mr. Telford between the same points, rose and fell only 2,257 ft., or 1,283 ft. less than the undulations of the old road, whilst the new road was more than 2 miles shorter. The road was formed by a substratum, or rough hand- set pavement, of large stones as a foundation, with suffi- cient interstices between the stones for drainage. The materials laid on this foundation were, like Macadam's materials, hard and angular, broken into small pieces, de- creasing in size towards the top, where they formed a fine hard surface, whereon the carriage wheels could run with but little resistance. Telford' s system was afterwards studied by his assistant, Mr. (afterwards Sir John) Macneil. The pressure of public opinion, acting through more than a century, has resulted in a network of fully 160,000 miles of good carriageable roads in the United King- dom, according to the following data supplied by Mr. Vignoles: * Length of Metalled Roads in 1868-69. United Kingdom France Prussia Spain The rolling of Macadam or broken-stone roads, though * Address of the President of the Institution of Civil Engineers, January llth, 1870. igth of Road. Area. Population. Miles. Square miles. Numbers. 160,000 122,519 30,621,431 100,048 210,460 38,192,064 55,818 139,675 23,970,641 10,886 198,061 15,673,481 BOTTLD'TCR FAVEMENT. 11 it seems to have been first applied in 1830, appears to have been but imperfectly appreciated in England until about the year 1843, when, according to Mr. F. A. Paget, the first published recommendation in the English lan- guage of horse road-rolling, as a measure of economy, was issued by Sir John Burgoyne^ Road-rolling is now very generally practised, by horse-power or by steam-power, f The first Act for paving and improving the City of London was passed in 1532. The streets were described, in this simply -worded statute, as ' i very foul, and full of pits and sloughs, so as to be mighty perillous and noyous, as well for all the king's subjects on horseback, as on foot with carriages " (litters). Previously to the introduction of the turnpike -road system, the streets of the metropolis and other large towns were paved with rounded boulders, or large irregular pebbles, imported from the sea-shore. They usually stood from 6 to 9 inches in depth for the carriage-way, and about 3 inches deep for the footpaths. Such a road could not be made with a very even surface ; the joints were neces- sarily very wide, and afforded receptacles for filth. The irregularity of the bases of the stones caused a difficulty in securing a solid and equal support; and, under the traffic, ruts and hollows were speedily formed. The boulder pavement was succeeded by a pavement composed of blocks of stone which, though ordinarily of tolerably good quality, and measuring 6 or 8 inches across the sur- face, were so irregular in shape that even their surfaces did not fit together. They formed a rubble causeway, * See a paper by Sir John Burgoyne "On Rolling new-made Roads," in the Appendix. t The history of Horse Road-Rolling and of Steam Road-Rolling, is given by Mr. Frederick A. Paget in his instructive "Report on the Economy of Road Maintenance and Horse-draught through Steam Road-rolling; with Special Reference to the Metropolis, 1870." Addressed to the Metropolitan Board of Works. 12 HISTORICAL NOTICE. in which the stones were but slightly hammer- dressed. Wide joints were made ; and far from being dressed square down from the surface, they most frequently only came into contact near the upper edges ; and, tapering downwards, their lower ends were narrow and irregular, leaving an insufficient area of flat base to support weight. With such irregular forms, considerable spaces were un- avoidably left between the stones, which were filled by the paviours with loose mould, sand, or other soft material, of which the bed or subsoil was composed. Another great deficiency in the construction of the pavement, was caused by inattention to the selection and arrangement of the stones according to size large and small stones were placed alongside of each other, and, as they acted un- equally in their resistance to pressure, they created a con- tinual jolting in wheel-carriages, and, adding percussive action to pressure, became powerful destructive agents. Again, the bed on which the stones were placed, being loose matter, for the most part, was easily converted into mud when water sank through between. It was unavoid- ably loosened by the paviour's tool, to suit the varying depths and narrow bottoms of the stones, and to fill up the chasms between the stones. The mud was worked up to the surface, and the stones were left unsupported. In consequence of these defects, the surface of the pavement soon became very uneven, and not unfrequently sunk so much as to form hollows, which rendered it not only incommodious but dangerous to horses and carriages. Such was the system of pavement met with in London fifty years ago. Mr. Telford, in 1824, clearly pointed out the deficiencies of the system ; and in his Eeport (referred to in the foot-note)* lie recommended first, a bottoming, * See Mr. Telford's " Report respecting the Street Pavements, &c., of the Parish of St. George's, Hanover Square." printed in Sir Henry Parnell's "Treatise on Roads," p. 348, 2nd Edition. LONDON PAVEMENTS. 13 or foundation, of broken stones, 12 inches deep ; second, rectangular paving-stones of granite, worked flat on the face, straight and square on all the sides, so as to joint close, with a base equal to the face, forming, in fact, an ashlar causeway. The dimensions of the stones were recommended to be as follows : Width. Depth. Length. Inches. Inches. Inches. For streets of the 1st class 6 to 7J 10 11 to 13 2nd 5 to 7 9 9 to 12 3rd 4jto6 7 to 8 7 to 11 Stones of such dimensions as those recommended by Telford, frequently having a depth of 12 inches, have been generally employed in street -paving. In some instances, they have been laid on concrete, with the joints grouted with lime and sand, to insure a great degree of stability. They have been proved to possess great dura- bility of which many instances will be adduced but they have been, for several reasons, generally abandoned in favour of narrower paving-stones, 3 or 4 inches in width, though many secondary streets in London and else- where, remain, at this day, paved with 6 inch stones. Macadam's system was introduced in some streets where the traffic was light, but it did not equal the granite paving. Pavements formed of blocks of wood appear to have been first employed in Russia, where, according to the testimony of Baron de Bode,* it has been, though rudely fashioned, used for some hundreds of years. After long and repeated trials of various modes of construction, wood pavement consisted, according to the approved method, of ; hexagonal blocks of fir wood, 6 inches across and 7 inches deep, planted, with the fibre vertical, close to each other, on a sound and level bottom ; a boiling mixture of * " Wood Pavement," by A. B. Blackie, 1842. 14: HISTORICAL NOTICE. pitch and tar was poured over them, and a small quantity of river sand was strewed over the tar. "The fabrication of these blocks," wrote Baron de Bode, "is extremely simple and expeditious. It is accomplished by fastening six strong blades into a strong bottom of cast-iron, and pressing the ready-cut pieces of wood through these six blades by means of a common or hydraulic press. The bottom of the press being open, these cut blocks drop on the floor, completely formed for immediate use. Eed fir is considered the best ; but none of it must be used when it has blue stripes on its edges, as that is a proof that it is in a state of decay. The blocks must be perfectly dried before they are used, and squeezed as close together as possible between the abutments, one on each side of the street or road, so as to keep the pavement from moving." In Norway, Sweden, Denmark, and Iceland, wood was, at the time of Mr. Blackie's writing, and it may be now, in general use for the pavement of streets and highways. In the United States, likewise, wood pavement was laid down experimentally in New York in 1835-6, and about the same time in Philadelphia. In New York, it was laid in three different forms. A hundred yards was laid in Broadway, consisting of hexagonal blocks of pitch-pine, 6 inches across, and 8 inches deep. No pitch or tar was applied to this pavement : it was simply strewed occasionally with gravel or sand for a month after it was laid. rig. 5. ^st^TTwood It na( l Iain f r two years, according to Paveme^Formof report? w i t hout having required any repair ; though it appears that very few carts passing over it carried more than half a ton of load, of which the widest wheel did not exceed three inches in width. An equal length of pavement was laid in William Street, a minor thoroughfare, in the end of 1836; the pavement consisted WOOD PAVEMENT. 15 of 6-inch square blocks of pine, 12 inches deep. The third specimen was laid in Mill Street, in the middle of 1837, consisting of the same size and kind of blocks as those laid in William Street, on a foundation of sand beat down very hard. It is stated in Mr. Blackie's pamphlet that the pavement of square blocks was laid on boards probably in William Street. Mr. David Stead was the first constructor of wood pavement in England. He patented his system in May, 1838: consisting of hexagonal blocks of Scotch fir or Norway fir, from 6 to 8 inches across, and from 3^- to 6 inches deep, according to the traffic of the thoroughfare in which they were to be laid. Each block was of the form shown in Fig. 5, chamfered at the upper edges. The ground having been well beaten and levelled, it was covered with three inches of gravel, upon which the blocks were placed, and which was designed to carry away the water which might penetrate below the surface. The pavement, when completed, looked substantial, and presented the appearance shown in Fig. 6. When the blocks were grooved across, they appeared together as in Fig. 7. Mr. Stead's pavement was, in several instances, laid on a bed of concrete. In Manchester, where it was thus laid, in front of the Eoyal Infirmary, the concrete bed was three inches deep, and was composed of three parts of small broken stones, j inch in diameter, Fig . e , flushed with Ardwick lime and stead's Wood Pavement, isss. Roman cement. The lime was mixed with sand in the proportion of one to two ; and the cement as one to twenty. The concrete was laid upon a hard, well- beaten clay substratum. Mr. Stead also laid pavements experimentally, consist- OF TUB 16 HISTORICAL NOTICE. ing of round blocks of wood sections of trees placed vertically, and laid together as in Fig. 8. The interspaces were filled with sifted gravel or sharp sand. The first example of wood-paving in London, was laid in the Old Bailey, in 1839, on Stead's system. It was laid haphazard on the bed of the roadway. The pavement did not wear well ; the blocks settled down irregularly in the unprepared stead's Wood lavement, 1838. foundation. At the end of three years and two months, in 1842, the pavement was lifted, and removed to pave the yard of the Sessions House ; there it decayed, and a large crop of fungi appeared in the places not touched by the traffic. Mr. Stead's system of wood paving was laid in several other localities in London about the same time as the piece which was laid in the Old Bailey, and also in Woolwich Dock- yard. It was laid also in Salford, Liverpool, and Leeds. Shortly after Mr. Stead's attempt, during the period from 1840 to 1843, seven other wood pavements, of various design, were laid in the City ; but they did not last, for the most part, more than three or four years. One of these was the invention of the Count de Lisle, patented in the name of Hodgson, in December, 1839 ; the in- vention was acquired by the Metro- politan Wood Pavement Company. The formation of the blocks was called the ' ! Stereotomy of the Cube." The upper and under surfaces of the blocks, Fig. 9, are cut diagonally to the direction of the grain, Stead's Wood Pavement. Round Blocks. Fig. 9. De Lisle' s Wood Pavement. Form of Blocks, 1839. WOOD PAVEMENT. 17 Fig. 10. De Lisle 's Wood Pavement, 1839. forming parallelepipeds, which are placed alternately in reversed positions, and when put together present a pave- ment having the appearance of Fig. 10. In each block, two holes are cut on each side to receive dowels or trenails, designed to lock the blocks together. Mr. Carey's wood pavement, patented in 1839, was one of the earliest pavements that were tried, and it proved to be the best at the time. It was first laid in the City, in the Poultry, in 1841, where it lasted six years ; and it was shortly afterwards laid in many other streets. It consisted of blocks of wood 6 or 7 inches wide, from 12 to 14 inches in length, and 8 inches deep, shown in side elevation, Fig. 11. The four-sided blocks of wood were of wedge-form, in and out, sidewise and endwise vertically, so as to form salient and re-entering angles, and to interlock on all the four sides, each block with its neighbour, when laid. It was anticipated that, by this arrangement, each block would receive support from its neighbours, and would be prevented from shifting or settling from its position, since the pressure of the load that was to come upon each block in succession would be distributed and dispersed over the neighbouring blocks. Later experience has demonstrated two things : that lateral sup- port of this kind was not re- Kg " l Ve quired ; and that, following the experience of stone-set . paving, the wood blocks of narrower dimensions answered better, and, with suitable interspaces, afforded the necessary foothold for horses. Asphalte, a natural, brittle compound of bitumen and 18 HISTORICAL NOTICE. limestone, found in volcanic districts, was introduced from France, for foot-pavements, in 1836. It has, since that time, been extensively employed in the City of London for the pavements of carriage-ways. In France, the art of the construction of roads, a hundred years ago, was far in advance of English practice. Pre- viously to 1775, the causeway was generally 18 feet wide, with a depth of 18 inches at the middle and 12 inches at the sides, according to the profile, Fig. 12. Stones were laid Fig. 12. Section of French Roads. Previous to 1775. flat, by hand, in two or more layers, on the bottom of the excavation ; on this foundation, a layer of small stones was placed and beaten down, and the surface of the road was formed and completed with a finishing coat of stones broken smaller than those immediately beneath. As the roads were, down to the year 1764, maintained by statute labour, with which the reparations could only be conducted in the spring and the autumn of each year, it was necessary to make the thickness of the roads as much as 18 inches, that they might endure during the intervals between repairs. With less depth, they would have been cut through and totally destroyed by the deep ruts which were formed in six months. The suppression of statute labour (la corvee), in 1764, was the occasion of a reformation in the design of cause- ways, whereby the depth was reduced to such dimensions as were simply strong enough for resisting the weight of the heaviest vehicles. The depth was reduced to a uniform dimension of 9 or 10 inches from side to side, and the cost was diminished more than one half. Writing in 1775, M. Tresaguet, engineer-in-chief of the generality of Li- moges, stated that roads constructed on the improved plan ROADS IN FRANCE. 19 lasted for ten years, under a S} 7 stem of constant maintenance, and that they were in as good condition as when first con- structed. The section of these roads, as elaborated by M. Tresaguet, is shown in Fig. 13. The form of the bottom is Fig. 13. Section of French Roads, elaborated by M. Tresaguet. 17T5. a parallel to the surface, at a depth of 10 inches below it. Large boulder stones are laid at each side. The first bed consisted of rubble stones laid compactly edgewise, and beaten to an even surface. A second bed, of smaller stones, was laid by hand upon the first bed. Finally, the finishing layer, of small broken stones, broken by hand to the size of walnuts, was spread with a shovel. Great care was taken in the selection of stone of the hardest quality for the upper surface. The rise of the causeway was 6 inches in the width of 18 feet, or 1 in 36. Tresaguet's method, here illustrated, was generally adopted by French engineers in the beginning of the present century ; although, on soft ground, they placed a layer of flat stones on their sides under the rubble work. In this case, the thickness was brought up to 20 inches. The rise of the causeway was as much as l-25th, and often equal to l-20th of the width. But, if the design was good, the maintenance was bad. Large and unbroken stones were thrown into the holes and ruts, and neither mud nor dust was removed. About the year 1820, the system of Mr. Macadam attracted some attention in France ; and the peculiar virtue of angular broken btone in closing and consolidating the surface was recognised. About the year 1830, it is said, the system of Macadam was officially adopted in France for the con- struction of roads; and M. Dumas, engineer-in-chief of 20 HISTORICAL NOTICE. the Fonts et Chaussees, writing in 1843,* stated that the system of Macadam was generally adopted in France, and that the roads were maintained, by continuous and watch- ful attention in cleansing the roads and with constant repair, in good condition realising his motto, " The maxi- mum of beauty." But the employment of rollers for the preliminary consolidation and finishing of the road, has been an essential feature in their construction and their maintenance ; for it has long been held in France that a road unrolled is only half finished. It appears, according to Mr. F. A. Faget, that the horse-roller was introduced in France in 1833. At al] events, in 1834, M. Folonceau, struck by the viciousness of the mode of aggregating or rolling the material of the road by the action of wheels, proposed, in the first place, to consolidate the bottom by a 6-ton roller, and to roll the material in successive layers consecutively, and thus to complete in a few hours what might, in the ordinary course of wheel -rolling, require many months to perform. "Annales des Fonts et Chaussees," 1843 ; tome 5, page 348. PAKT I. CONSTEUCTION OF EOADS. BY HENRY LAW. CHAPTEE I. EXPLORATION OF ROADS. THIS part of the work is confined to the art of constructing common roads, in situations where none previously existed, and to the repair of those already made. Before entering into the details of their construction, it is desirable to go into the subject of the exploration of roads, or the manner in which a person should proceed in exploring a tract of country, for the purpose of determining the best course for a road, and the principles which should guide him in his final selection of the same. Suppose that it is desired to form a road between two distant towns, A and B, Fig. 14, and for the present neglect D B Fig. 14. Laying out a Road. altogether the consideration of the physical features of the intervening country ; assuming that it is equally favour- 22 EXPLORATION OF ROADS. able, whatever line is selected. Now, at first sight, it would appear that, under such circumstances, a perfectly straight line drawn from one town to the other, would be the best that could be chosen. But on a more careful examination of the locality, it may be found that there is a third town, c, situated somewhat on one side of the straight line drawn from A to B ; and, although the primary object is to connect the two latter, it may, never- theless, be considerably better if the whole of the three towns were put into mutual connection with each other. Now this may be effected in three different ways ; any one of which might, under certain circumstances, be the best. In the first place, a straight road might, as originally sug- gested, be formed from A to B, and, in a similar manner, two other straight roads from A to c, and from B to c. This would be the most perfect way of effecting the object in view, the distance between any two of the towns being reduced to the least possible length. It would, however, be attended with considerable expense, and it would be requisite to construct a much greater length of road than according to the second plan, which would be to form, as before, a straight road from A to B, and from c to construct a road which should join the former at a point D, so as to be perpendicular to it ; the traffic between A or B and c, would proceed to the point D, and then turn off to c. By this arrangement, while the length of the roads would be very materially decreased, only a slight increase would be occasioned in the distance between c and the other two towns. The third method would be to form only the two roads A c and c B. In this case, the distance between A and B would be somewhat increased, while that between A and c, or B and c, would be diminished ; the total length of road to be constructed would also be lessened. Asa general rule, it may be taken that the last of these methods is the best, and most convenient for the public ; LAYING OUT A ROAD. 23 that is to say, if the pnysical character of the country does not determine the course of the road, it will generally be found best not to adopt a perfectly straight line, but to vary the line so as to pass through the principal towns near its general course. The public may thus be con- veyed from town to town with greater facility and less expense than if the straight line were adopted, and the towns were to communicate with it by means of branch roads. On the first system, vehicles established to convey passengers or goods between the two terminal towns, would pass through all those which were intermediate ; whilst, if the straight line and branch-road system were adopted, a system of branch coaches would be required for meeting the coaches on the main line. In laying out a road in an old country, in which the position of the several towns, or other centres of industry, requiring road accommodation, is already determined, there is less liberty for the selection of the line of road than in a new country, where the only object is to establish the easiest and best road between two distant stations. In the first case, the positions of the towns, and other in- habited districts situated near the intended road, are to be taken into consideration, and the course of the road may, to a certain extent, be controlled thereby ; whilst, in the second case, the physical character of the country would alone be investigated, and it alone would constitute the basis for the selection of a new route. Whichever of these two cases may be dealt with, in the selection and adoption of the line of road between two points, a careful examination of the physical character of the country should be made, and the line of the route determined in accordance with physical conditions. One of the first points which attract notice in making an examination of an ordinary tract of country, is the uneven- ness or undulation of its surface j but if the observation be 24 EXPLORATION OF ROADS. extended a little further, one general principle of con- formation is perceived even in the most irregular countries. The country is intersected in various directions by rivers, increasing in size as they approach their point of dis- charge ; towards these main rivers, lesser rivers approach on both sides, running right and left through the country ; and into these, again, enter still smaller streams and brooks. Furthermore, the ground falls in every direction towards the natural watercourses, forming ridges, more or less elevated, running between them, and separating from each other the districts drained by the streams. It is the first business of a person, engaged in laying out a line of road, to make himself thoroughly acquainted with the features of the country ; he should possess him- self of a plan or map, showing accurately the course of all the rivers and principal watercourses, and upon this he should further mark the lines of greatest elevation, or the ridges separating the several valleys through which they flow. It is also of peculiar service when the plan contains contour lines showing the comparative levels of any two points, and the rates of declivity of every portion of the country's surface. The system of showing upon plans the levels of the ground by means of contour-lines, is one of much utility, not only in the selection of roads and other lines of communication, but also for settling the lines of the drainage of towns, as well as of their water-supply, and of the drainage and irrigation of lands, and for many other purposes. A contour-plan of the City of London * (Fig. 15) illustrates the application of the system of con- tour levels. It will be observed that, upon this plan, there are a number of fine lines traversing its surface in various directions, and, where they approach the borders of the map, having figures written against them : these lines are * This plan is taken from a Report on the Health of Towns, and is made from levels 'aken from Mr. Butler Williams, Valley of the Fleet Ditch. Fig. 15. Contour Plan of London. 26 EXPLORATION OF ROADS. termed contour-lines, and they denote that the level of the ground is identical throughout the whole of their course : that is to say, that every part of the ground over which the line passes, is at a certain height above a known fixed point, the height being indicated by the figures written against the line. At the point A, for example, in Smithfield Market, a line with the figures 57 is attached, which indi- cates that the ground at that spot is 57 feet above some point to which all the levels are referred. If the course of the line be traced, it is found that it cuts Newgate Street at the point B, passes thence to the bottom of Paternoster Row at the point i, through St. Paul's Churchyard at c, to Cheapside at D. It then curves round towards the point from which it first started, and crosses Aldersgate Street twice, at E and F ; and, after intersecting Fore Street, Cripplegate, in the point G, it again meets the boundary of the City at H. It is thus shown that, tracing the course of this line, each of those points stands at the same height, namely, 57 feet above a certain fixed point, termed the datum. This point is, in the present instance, 10 feet below the top of the cap-stone at the foot of the step, on the east side of old Blackfriars Bridge. Each interval between the lines in Fig. 8, indicates a difference of level of 18 inches ; and by counting the number of these lines which intersect a street or road within any given distance, the rise or fall in the street is at once, ascertained by simple multiplication. Thus, looking at the line of Bishopsgate Street, near the north end, the contour-line 45 is seen, in- dicating that that point in the street is 45 feet above the datum, and nine lines are found intersecting the street between that point and the top of Cornhill. It is calcu- lated, therefore, that this point is (1-5 X 9 =) 13-5 feet above the other end of the street, or 58 -5 feet above the datum. The rate of inclination of the ground may also be estimated by the relative proximity or distance apart of LAYING OUT A ROAD. 27 these lines. Thus, on the northern side of the City, where the ground is comparatively level, the lines are far apart ; whereas, on the side next the Thames, and again on each side of the line of Farringdon Street, which marks the course of the valley of the old river Fleet, where the sur- face is very hilly, the contour lines lie close together. The plan, Fig. 16, shows an imaginary tract of country, to illustrate more clearly the mode of showing by means of contour-lines, the physical features of a country. The hatched line, E F G H i, is supposed to be an elevated ridge, encircling the valley shown in the plan; the fine black lines are contour-lines, indicating that the ground over which they pass is at the altitude above some known mark expressed by the figures written against them in the margin. It will be observed that these lines, by their greater or less distance, have the effect of shading, and make apparent to the eye, the undulations and irregu- larities in the surface of the country. In laying out a line of road, there are three cases which may have to be treated, and each of these is exemplified in the plan, Fig. 16. First, the two places to be connected, as the towns A and B on the plan, may be both situated in the same valley, and upon the same side of it ; that is, that they are not separated from each other by the main stream which drains the valley. This is the simplest case. Secondly, although both in the same valley, the two places may be on the opposite sides of the valley, as at A and c, being separated by the main river. Thirdly, they may be situ- ated in different valleys, separated by an intervening ridge of ground more or less elevated, as at A and D. In laying out an extensive line of road, it frequently happens that all these cases have to be dealt with: frequently, perhaps, during its course. The most perfect road is that of which the course is perfectly straight, and the surface perfectly level ; and, all c 2 9 12 15 15 15 18 21 24 27 42 45 48 51 54. 54 51 48 45 42 Tig, 16. Contour Plan of a Tract of Country. LAYING OUT A ROAD. 29 other things being the same, that is the best road which answers nearest to this description. Now, in the first case : That of the two towns situated on the same side of the main valley, there are two methods which may be pursued in forming a communication be- tween them. A road following the direct line between them, shown by the thick dotted line A B may be made ; or, a line may be adopted which should gradually and equally incline from one town to the other, supposing them to be at different levels, or which should keep, if they are on the same level, at that level throughout its entire course, fol- lowing all the sinuosities and curves which the irregular formation of the country might render necessary for the fulfilment of these conditions. According to the first method, a level or a uniformly-inclined road might be made from one to the other, forming embankments and cuttings where necessary ; or these expensive works might be avoided, and the surface of the road made to conform to that of the country. Now, of all these, the best is the straight and uniformly-inclined, or the level road, although at the same time it is the most expensive. If the importance of the traffic passing between the places is not suffi- cient to warrant so great an outlay, it will become a matter of consideration whether the course of the road should be kept straight, its surface being made to undulate with the natural face of the country ; or whether, a level or equally- inclined line being adopted, the course of the road should be made to deviate from the direct line, and follow the winding course which such a condition is supposed to necessitate. In the second case, that of two places situated on oppo- site sides of the same valley, there is, in like manner, the choice of a perfectly straight line to connect them, which would probably require a heavy embankment if the road were kept level ; or steep inclines, if it followed the surface 30 EXPLORATION OF ROADS. of the country ; or, by winding the road, it may be carried across the valley at a higher point, where, if the level road were taken, the embankment would not be so high, or, if kept on the surface, the inclination would be reduced. In the third case, there is, in like manner, the alterna- tive of carrying the road across the intervening ridge in a perfectly straight line, or of deviating it to the right or the left, and crossing at a point where the ridge is less elevated. The proper determination of the question, which of these courses is the best under certain circumstances, involves a consideration of the comparative advantages and disad- vantages of inclines and curves. What additional increase in the length of a road would be equivalent to a given in- clined plane upon it ; or, conversely, what inclination might be given to a road, as an equivalent to a given decrease in its length ? To satisfy this question, it is requisite to know the comparative force required to draw different vehicles with given loads upon level and upon variously-inclined roads : a subject which is treated in Chapter III. In laying out a new line of road, the first proceeding is usually, after a general examination of the country, to lay down one or more lines upon the best map which can be procured. On a contour-map of the district, this proceed- ing is greatly facilitated. The next step is to make an accurate survey of the lands through which the several lines sketched out pass, which should be plotted to such a scale as will admit of the smallest features being shown with sufficient accuracy and distinctness. A scale of ten chains to the inch, for the open country, with enlarged plans of towns and villages upon a scale of three chains to the inch, is generally found to be sufficient. Careful levels should also be taken along the course of each line ; and at suitable distances, depending upon the nature of the country, lines of levels should be taken at right angles to the original PLANS OF ROADS. 31 line. In taking these levels, the heights of all existing roads, rivers, streams, or canals should be noted; bench- marks should be left at least every half -mile, that is, marks made on any fixed object, such as a gate-post, or the side of a house or barn, the exact height of which is ascertained, and registered in the level-book. The bench-marks are useful in case of deviations being made in any portion of the lines, for the levels may be taken direct from the bench- marks, thus obviating the necessity of again levelling other parts of the line. A section should be formed from the levels, to the same horizontal scale as the general plan, with such a vertical scale as will show with distinct- ness the inequalities of the ground. If the horizontal scale is ten chains to the inch, the vertical scale may be 20 feet to the inch. ' A plan of this kind is exemplified in Fig. 1 7, plotted to a scale of ten chains to the inch, showing a district through which a road is to be constructed. One line is shown run- ning nearly straight across the plan, together with a devia- tion therefrom, which, although of greater length, would run on more favourable ground. The sections, Figs. 18 and 19, show the levels of the surface of the ground on the straight line, and on the deviation from it respectively. The required information is given on the plans, for enabling the engineer to lay down the course of the road, and to arrange the position and dimensions of the culverts, bridges, and other works necessary in its construction. It is shown in Fig. 17 that the straight line crosses a stream at B, and the river twice at o and D ; and also that it must pass from B to E, over a swamp or morass of such a nature that, if a solid embankment be formed, it is pro- bable that a very large quantity of earth would be absorbed beyond what is indicated in the section. It would, in addition, be necessary to form bridges with several capa- cious openings at the points where the intended road would EXPLORATION OF ROADS. SECTIONS OF ROADS. 33 cross the river, since the river would be liable to be flooded. Such disadvantages attending the more obvious route, would induce the engineer to sketch out some other line, by which they would be avoided. He would have the levels taken, with other needful information, to enable him to choose between the two routes. The manner in which the sections should be drawn, and the nature of the information to be given upon them, are exemplified in Figs. 18 and 19. In addition, data of the fol- lowing character should be obtained, and should be entered either in the survey field-book, or in the level-book. At the point B, fig. 17, the line crosses a stream 8 feet in width and 1 foot deep ; in flood, this stream hrings down a considerable quantity of water. At the point c on the section, the river is much narrower and is not so deep as at other places, in consequence of a great por- tion of its waters finding a passage through the marshy ground on either side. Its width is 16 feet, and its depth 2 feet; the velocity of the current is 95 feet per minute ; the height of its surface at the present time is 30-10 feet above the datum; and the angle of skew which the course of the stream makes with the line of the road is 62 degrees. At the point D the river is 27 feet wide, and 2J feet in depth ; its velocity 87 feet per minute ; the height of its surface above the datum 29-96 feet ; and the angle of skew 49 degrees. The ground from B to E is of a very soft, boggy nature, and full of water. The height to which the river has risen during the highest flood known, at the bridge at F on the plan, is 35 feet above the datum ; the water-way at that time was 90 feet, and the sec<- tional area of the opening through which the water then flowed was 550 square feet. The same flood at the lower bridge, at G on the plan, was 35-3 feet above the datum; the water-way was 102 feet, and the sectional area nearly 600 square feet. The deviation-line only crosses one stream, at M, on the plan and the section. The width of this stream at present is 15 feet, and its depth 1 8 inches ; but in times of flood it rises to the same height as the river, and brings down a large body of water. The height of its surface at present above the datum is 31*25 feet, and the angle which its course makes with the line of road is 85 C 3 EXPLORATION OF ROADS. Junction with) Existing Road J 37-26 River. Stream. River. Stream. Junction with') Existing Road. ) fj " iKs Jl Fig. 18 -Laying out a new Road. Section. SECTIONS OF ROADS 7-25 Junction with ) Existing Road, j 87.25 Junction wi^ KxuUag Road, j H 41 2f> . 19. Laying out a new .Road. Section- 36 .EXPLORATION OF ROADS. The information relative to the rivers crossed, such as is given above, should always be obtained, in order that the bridges constructed over them may be adequate for the passage of the water brought down in time of floods. A cross section should be taken of each of the existing roads, near their junctions with the intended road ; to show to what extent, if any, the levels of the existing roads might be altered to suit the levels of the proposed new road. Laying out a Road. On the sections Figs. 18 and 19 the line of the road is to be laid down ; in other words, the levels at which it shall be formed are to be determined. As the road should always be dry, it should be placed at least a foot above the level of the flood ; and if it be placed at 37-25 feet above the datum, which is the height of the existing road at I, this object will be effected. Drawing a line at this level upon the section, it appears that an embankment will have to be formed across the valley from the road at i, to the point where the line meets the ground at K; and that the remainder of the road from K to H will be in a cutting. Now, the obvious principle in arranging the levels of a road, would be so to adjust the cuttings and embankments that the ground taken from one should form the other. In the present instance, this is impossible, because the level of the road is determined by other circumstances, and necessitates the formation of a very long embankment with but very little cutting. It therefore becomes necessary that ground for the formation of the embankments should be obtained from some other source. But, in order to produce as much cutting as possible, the line should be kept at the same level as before until it becomes necessary to raise it so as to attain the level of the existing road at H. If an inclination of 1 in 50 be given to this last part of the road, the distance at which the rise will commence will be 200 feet from H, the LEVELS OF ROADS. 37 difference of level being 4 feet. There is therefore to be added to the other disadvantages already mentioned, as belonging to the straight line of road, that of the formation of a large embankment, with the necessity for making an excavation in some other place, to supply the earth for that purpose. Examine the section of the deviation-line, and see what improvement can be thereby effected. The level of the lowest portion of the road must, as before, be placed 37'25 feet above the datum ; and if a line be drawn at that level on the section, Fig. 19, it will be found that the quantity of embankment is very much reduced, compared with what would be required for the straight course, and that there is now no difficulty in adjusting the cutting between H and L, so as exactly to afford the amount of filling required. A few trials will show that, if the line be kept at the same level until within sixteen chains of the point H, and then carried up at a regular inclination, this object will be effected, and that the quantities of cutting and embank- ment will be very nearly equal. The deviation-line is, therefore, the line which the engineer would select as the better of the two. Having made his selection, he would proceed to mark the course of the road on the ground, by driving stakes into the ground, on its centre line, at inter- vals of one chain-length, or 66 feet. In the next place, he would take very careful levels of the ground at every one of these points, and at any intermediate point, where an undulation or change of level occurred ; and wherever the level of the ground varied to any extent in a direction at right angles with the course of the road, he would take levels from which he would make transverse or cross sections of the ground. From these levels a working section should be made, to a horizontal scale of not less than five chains to the inch, and a vertical scale of 20 feet to the inch. A portion of 38 EXPLORATION OF ROADS. the section plotted to these scales is shown in Fig. 20 ; the level of the surface of the ground above the datum, at every chain -length, at the points where stakes have been driven into the ground, should be figured-in on the section, as shown in the column A, and the depth of cutting or height of embankment, at the same points, should be given in another column, B. The entries in this last column are obtained by taking the difference between the level of the surface of the ground and the level of the road. It will be observed that, upon the section, there are two parallel lines drawn as representing the line of road ; the upper line is intended to represent the upper surface of the road when finished, while the lower thick line represents what is termed the formation-surface, or the level to which the surface of the ground is to be formed, to receive the foun- dation of the road. In the section, the formation-surface is shown 15 inches below the finished surface of the road ; the difference of level is therefore the thickness of the road itself. All the dimensions on the section are understood to refer to the formation-level ; and the height of the latter above the datum should be figured-in wherever a change in its rate of inclination takes place, and should be marked by a stronger vertical line ; as shown at a i r s> w 40-00 d. 1 40*33 0-58 I 40-45 1-2U I 40*56 1*56 M I 40*66 i-yi M> I 49-54 2*04 9 8 1 40*12 1*87 5 o I 39-90 1*90 M-M ^~ 1 39'82 2-07 5" 1 39'67 2*17 IP g r 39-60 2*35 I 1 39-30 2*30 p 1 39 '00 38*75 2-25 1 38*30 2-50 I 38 33 2-05 3600 38-52 38-20 2*62 r 37-60 2-20 j 36*75 1-60 I 35*45 0*75 !\ CO.Qrt 0*5i * 1 09.40 2-20 \ 1 52 3-62 -J 8 32*00 4*00 3 /Stre 1 N 3*79 \ s- 33*40 2*60 \i M i 5 34*75 1-25 M . I 0-30 5 _ 1 36*00 Parnell on Roads, p. 73. f Ibid., p. 464, SIR JOHN MACNEII/S EXPERIMENTS. 53 nature of the surface over which the carriage is drawn, and the value of which for several different kinds of road is as follows : On a timber surface . . . . . c = 2 On a paved road . . . . . . 2 On a well-made broken stone road, in a dry clean state . 5 On a well-made broken stone road, covered with, dust . 8 On a well-made broken stone road, wet and muddy . 10 On a gravel or flint road, in a dry clean state . . 13 On a gravel or flint road, in a wet and muddy state . 32 Stage wagon, *=^f + ^ + c v . . . (1.) Stagecoach, * = ^^ 4- + cv . . . (2.) ETJLE 1. Divide the gross weight of the carriage when loaded, in pounds, by 93 if a wagon, or by 100 if a coach, and to the quotient add one-fortieth of the weight of the load only ; to the sum, add the product of the velocity in feet per second, by the proper constant for the particular kind of road. The sum is the force in pounds required to draw the carriage at the given velocity upon that descrip- tion of road. For example : What force would be requisite to move a stage-coach weighing 2,060 Ibs., and having a load of l,100lbs., at a velocity of 9 ft. per second, along a broken- stone road covered with dust ? By the rule, MM +_UOO + 1100 + (8 x 9)=131>1 ^ the force required. To consider, next, the additional resistance which is occasioned when the road, instead of being level, is inclined against the load, in a greater or less degree. In order to simplify the question, suppose the whole weight to be supported on one pair of wheels, and that the tractive force is applied in a direction parallel to the surface of the road. Let A B, Fig. 21, represent a portion of an inclined 54 RESISTANCE TO TRACTION ON COMMON ROADS. road, c being a carriage just sustained in its position by a force acting in the direction c D. The carriage is kept in position by three forces, namely, by its own weight w, acting in the vertical direction c F, by the force F, applied in the direction c D pa- rallel to the surface of the road, and by the pressure p, which is exerted by the carriage against G the Surface of the road acting in Fig. 21. Gravity on an inclined the direction c E, perpendicular to the surface. To determine the relative magnitude of these three forces, draw the horizontal line A G, and the vertical line B G ; then, since the two lines c F and B G are parallel, and are both cut by the line A B, they must make the two angles c F B and A B G equal ; also the two angles c E F and A G B are equal, being both right angles ; therefore the remaining angles FOE and BAG are equal, and the two triangles c F E and A B G are similar. And as the three sides of the triangle c F E are proportional to the three forces by which the carriage is sustained, so also are the three sides of the triangle A B G ; that is to say, A B, or the length of the road is proportional to w, or the weight of the carriage ; B G, or the vertical rise is pro- portional to F, or the force required to sustain the carriage on the incline ; and A G, or the horizontal distance for the rise is proportional to p, or the force with which the car- riage presses upon the surface of the road. Therefore, w : A B ; : P : G B, and w I A B * ' p t AG, And if A G be made of such a length that the vertical rise, B G, of the road, is exactly one foot, then, y - JL = W = w . sin ft . . . (3.) AB v/AG 2 + 1 RULES FOE RESISTANCE. 55 =w.cos/? . . . (4.) - A B V /A G 2 in which, ft is the angle BAG. These formulae reduced to verbal rules are as follows : RULE 2. To find the force requisite to sustain a carriage upon an inclined road (the effects of friction leing neglected), divide the weight of the carriage, including its load, by the inclined length of the road, the vertical rise of which is one foot, and the quotient is the force required. RULE 3. To find the pressure of a carriage against the sur- face of an inclined road, multiply the weight of the loaded carriage by the horizontal length of the road, and divide the product by the inclined length of the same ; the quo tient is the pressure required. Example. What is the force required to sustain a car- riage weighing 3,270 Ibs. upon a road, the inclination of which is one in thirty, and what is the pressure of the carriage upon the surface of the road ? Here the horizontal length of the road, A o, being equal to 30, for a rise of 1 foot, the inclined length, A B VA a 2 + 1 = 30-017, and by the first rule, 3,270 -f- 30'017 rz 108-93 Ibs. for the force required to sustain the carriage on the road. By the second rule, 3,270 x 30 -H 30-017 = 3,269-9 Ibs., the pressure of the carriage upon the surface of the road. Since the pressure of a carriage on a sloping road is found by multiplying its weight by the horizontal length of the road and dividing by the inclined length, and as the former is always less than the latter, it follows that the force with which a carriage bears upon an inclined road is less than its actual weight. In the foregoing example, it is about two pounds less ; but, unless the inclination is very steep, it is not necessary to distinguish the difference of pressure, as the pressure may be assumed to be equal to the weight of the carriage. 5() RESISTANCE TO TRACTION ON COMMON ROADS. If the resistance which, is to be overcome in moving a carriage, at a given rate, upon a horizontal road, be ex- pressed by R, then R -f F is the resistance in ascending a hill, and R F descending a hill, with the same velocity ; neglecting the decrease in the weight of the carriage pro- duced by the inclination of the road. Taking, however, this decrease into consideration, the following modification in the formulae (1.) and (2.) will be requisite to adapt them to an inclined road : ~*l& ' cos ^ "*" W + ' sn ^ + c v ' 5 ') in the case of a common stage wagon ; and in that of a stage coach, E = (iSr +S) ' cos P + (w + w) ' sin ft + c v (6>)> the upper sign being taken when the vehicle is drawn down the incline, and the lower when it is drawn up the same. To ascertain the resistance in passing up or down a hill, therefore, the resistance on a level road is first to be calcu- lated, by Rule 1, page 53. To this is to be added the force necessary to sustain the carriage on the incline, in ascending, calculated by Rule 2, page 55 ; or, in descending, the same force is to be subtracted from the resistance on a level. As an example, take, as before, the case of a stage coach weighing 2,060 Ibs., besides a load of 1,100 Ibs., at a velo- city of 9 ft. per second, up a broken stone road of which the surface is covered with dust, and which is inclined at the rate of one in thirty. The force to sustain the coach on this slope is, by Eule 2, 21? = 105-3 Ibs, Adding this force to the force already found at page 53, requisite to move the same coach on a level road, the sum is (105-3 + 131-1 ==) 236-4 Ibs., for the force required to RULES FOR RESISTANCE. 57 move the coach, with, a velocity of 9 ft. per second up the inclined road of one in thirty. To draw the coach down the same incline, at the same velocity, the resulting force re- quired is the difference of the two forces already found, or it is (131-1 105-3 = ) 25-8 Ib, The same example worked by formula (6) will give ( 2 - - ~- - ) -9995 + (2060 + 1100) -0333 +(8x9) = 236-3 Ibs, when the carriage is drawn up the incline ; and * 9995 - ( 206 + 110 ) * 0333 + ( 8 x 9 ) = 25-84 Ibs., when the carriage is drawn down the incline, the result being the same as that given by the rule. The following table has been calculated in order to show, with sufficient exactness for most practical purposes, the force required to draw carriages over inclined roads, and the comparative advantage of such roads and those which are perfectly level. The first column expresses the rate of inclination, and the second the equivalent angle ; the two next columns contain the force requisite to draw a common stage wagon weighing with its load 6 tons, at a velocity of 4*4 ft. per second (or 3 miles per hour) along a macadamized road in its usual state, both when ascending and descending the hill ; the fifth and sixth columns con- tain the length of level road which would be equivalent to a mile in length of the inclined road, that is, the length of level road which would require the same mechanical work to be expended in drawing the wagon over it, as would be necessary to draw the wagon over a mile of the in- clined road. The next four columns contain the same information as the four just described, with reference to a stage coach supposed to weigh with its load 3 tons, and to travel at the rate of 8-8 ft. per second, or 6 miles per hour. D 3 58 RESISTANCE TO TRACTION ON COMMON ROADS. TABLE No. 4. EESISTANCE TO TRACTION ON INCLINED ROADS. i i 1 FOB A STAGE WAQOK. FOR A STAGE COACH, i 6 tons gross. 3 tons gross. 1 w o a o a "o d *-d lo* o,a A ft r3 03 1 1 its' |li III Its *J sM fjj MI 1 s ! 2"*^ 5 a ||| 111 CJ 1 1 & |5*i PR 'p s r 1*3 Jr^ g ro s IP 8 cji ^ 8 o , // Ibs. Ibs. Miles. Miles. Ibs. Ibs. Miles. Miles. 1 in 600 5 44 286 241 1-085 9150 373 350 1-030 9690 575 5 59 287 240 1-088 9116 373 350 1-032 9676 550 6 15 288 239 1-093 9074 374 349 1-033 9662 525 6 33 289 238 1-097 9029 374 349 1-035 9646 500 6 53 291 237 1-102 8979 375 348 1-037 9629 ,, 475 7 14 292 235 1-107 8926 376 347 1-039 9605 450 7 38 294 334 1-113 8869 377 347 1-041 9588 425 085 295 232 1-120 8801 377 346 1-043 9563 400 8 36 297 230 1-128 8725 378 345 1-046 9535 375 9 10 300 228 1-136 8642 380 344 1-049 9505 350 9 49 302 225 1-146 8543 381 342 1-053 9469 325 10 35 305 222 1-157 8433 382 341 1-056 9430 300 11 28 309 219 1-170 8301 384 339 1-061 9381 290 11 51 310 217 1-176 8245 385 338 1-064 9358 280 12 17 312 216 1-182 8179 386 338 1-066 9336 , 270 12 44 314 214 1-189 8111 386 337 1-068 9314 , 260 13 13 315 212 1-196 8039 387 336 1-071 9286 , 250 13 45 317 210 1-204 7^63 388 335 1-074 9259 , 240 14 19 320 208 1-212 7876 390 334 1-077 9226 , 230 14 57 322 205 1-222 7785 391 332 1-080 9192 , 220 15 37 325 203 1-232 7683 392 331 1-084 9156 , 210 16 22 328 200 1-243 7573 394 330 1-088 9115 200 17 11 331 197 1-255 7451 !| 395 328 1-092 9071 , 190 18 6 334 193 1-268 7319 1 397 326 1-097 9024 , 180 19 6 338 189 1-283 7171 l 399 324 1-103 8968 , 170 20 13 343 185 1-300 7004 i 401 322 1-109 8908 . 160 21 29 848 180 1-319 6814 404 320 1-116 8839 150 22 55 353 174 1-341 6587 ; 406 317 1-123 8761 , 140 24 33 360 168 1-364 6359 ' 410 314 1-132 8673 130 26 27 367 160 1-392 6079 413 310 1-142 8573 120 28 39 376 152 1-425' 5752 .' 418 306 1-154 8451 , 110 31 15 386 142 1-451 5491 , 423 300 1-169 8308 . 100 34 23 398 129 1-510 4903 , 429 294 1-185 8142 ', 95 86 11 405 122 1-537 4634 ,] 432 291 1-195 8045 , 90 38 12 413 114 1-566 4338 ;' 436 287 1-206 7937 , 85 40 27 422 106 1-600 4004 , 441 282 1-219 7801 , 80 42 58 432 96 1-637 3629 446 278 1-232 7677 i 1 i TABLE. 59 TABLE No. 4. (Continued.) g FOB A STACK WAGON. FOB A STAGE COACH. i i i 6 tons gross. 3 tons gross. & g $ fl H 'Sod o . HI III < || d I'SS tii flj o H w ta I* OC_g Ill ill ~S p s> III If] til ii "a | S> OJ fn .~ ||| H M p 9 fl s"* 3 - 2 l ff 111 I| l-M *lls W-8 * 111 / // Ibs. Ibs. Miles. Miles. Ibs. lb s. Miles. Miles. 1 in 75 45 51 443 85 1-680 3204 451 272 1-247 7522 70 49 7 456 72 1-728 2719 457 265 1-265 7-345 65 52 54 470 57 1-784 2161 465 258 1-285 7143 60 57 18 488 40 1-850 1505 474 250 1-309 6903 55 2 30 508 19 1-926 0736 484 239 1-337 6620 50 8 6 533 2-019 __ 496 227 1-371 6283 45 16 24 562 2-133 511 212 1-412 5871 40 25 57 600 2-274 530 194 1-464 5354 35 38 14 648 2-456 554 170 1-530 4690 34 41 8 65.) 2-499 559 164 1-546 4535 33 44 12 671 2-544 565 158 1-562 4370 32 47 27 684 2-593 572 152 1-580 419.3 31 1 50 55 697 2-644 578 145 1-599 4007 30 1 54 37 712 2-699 586 138 1-619 3805 29 1 58 34 727 2-758 593 130 1-640 3592 28 225 744 2-820 602 122 1-663 3363 27 272 762 2-888 610 113 1-688 3119 26 2 12 2 781 2-960 620 103 1-714 2854 25 2 17 26 801 3-038 630 93 1-743 2566 24 2 23 10 823 3-120 641 82 1-774 2257 23 2 29 22 847 3-213 653 69 1-808 1919 22 2 36 10 874 3-313 666 56 1-844 1554 21 2 43 35 903 3423 681 42 1-884 1150 20 2 51 21 933 3-538 696 26 1-926 0730 19 3 46 970 3-677 714 8 1-977 0221 18 3 10 47 1009 _ 3-826 734 2-032 ^^ 17 3 21 59 1053 3-991 756 2-092 16 3 34 35 1102 4-178 780 2-160 15 3 48 51 1157 4-388 807 2-234 14 4 5 14 1221 4-629 _ 839 2-322 13 4 23 56 1294 4-906 875 2-423 12 4 45 49 1379 5-229 918 2-540 11 5 11 40 1480 5-611 968 2-679 10 5 42 58 1600 6-067 1028 2-846 9 6 20 2n 1747 6-623 1101 3-048 8 7 7 30 1929 7-315 1192 3-300 ,, 7 8 7 48 2162 8-199 1308 3-621 60 RESISTANCE TO TRACTION ON COMMON ROADS. The foregoing table may be considered as affording a view of the comparative disadvantage of hilly roads with light and heavy traffic ; the stage wagon weighing 6 tons and travelling at the speed of 3 miles per hour, may be taken as a fair average for goods traffic, and the stage coach, weighing 3 tons and running 6 miles an hour, for passenger traffic. It is shown that the resistance on hills is much more unfavourable to the wagon than to the coach. The force which would be requisite to move the wagon on a level road would be 264 Ibs., and that to move the coach 362 Ibs., being an excess of 98 Ibs. for the traction of the coach. But, with a road inclined at the rate of 1 in 600, this excess is only (373 286 =) 87 Ibs. ; arid when the inclination of the road amounts to about 1 in 70, the forces required to draw them become equal. As the inclination of the road increases beyond this, the excess of force requisite to draw the waggon over that necessary to move the coach, increases rapidly until, at an inclination of 1 in 7, it amounts to (2162 1308 =) 854 Ibs. Comparing the forces required to draw either the wagon or the coach up and down any given incline, the former is as much greater than the force required on a level road as the latter is less. It might thence be concluded that, when a vehicle passes alternately each way along the road, no real loss is occasioned by the inclination of the road, since as much power is gained in the descent of the hill as is lost in its ascent. Such is not, however, practically the fact, for whilst it is necessary in the ascending journey to have either a greater number of horses, or more powerful horses, than would be requisite if the road were entirely level, no corresponding reduction can be made in the descending journey. There must be horses sufficient to draw the vehicle along the level portions of the road ; nor, generally speak- ing, have the horses less to do in descending the hill. ANGLE OF REPOSE. 61 since they frequently are required to push back, to prevent the speed of the coach from being accelerated to a rate beyond the limits of safety. In a practical sense, therefore, it may be considered that the fifth and ninth columns in the foregoing table express the length of level road which would be equivalent to a mile of road with the stated inclination, the fifth giving the result for heavy traffic, and the ninth for passenger traffic. For instance, against the incline 1 in 75, there is a length of 1'247 miles, or about a mile and a quarter, in the ninth column, given as the equivalent length of level road for 1 mile of ascent on the incline, in the sense that the same quantity of work of traction would be requisite to move a coach of 3 tons, at a velocity of 6 miles per hour, along one as along the other. But, in other respects, the incline might be more advantageous than the level; for instance, the shorter road would cost less for repair, and would be passed over in less time. The table, therefore, merely expresses the equivalent length as far as the mechanical work required for the traction is concerned. From the results of Sir John Macneil's experiments on fractional resistance, page 52 ante, Professor Mahan de- daces " that the angle of repose in the first case is repre- sented by -K^-, or 1 in 71-34 nearly; and that the slope of the road should therefore not be greater than one per- pendicular to 71*34 in length; or that the height to be ascended must not be greater than one seventy-first part of the distance between the two points measured along the road, in order that the force of friction may counteract that of gravity in the descent of the road. " A similar calculation will show that the angle of repose in the other cases will be as follows : No. 2, . . . 1 to 50-9 nearly. 3, . . .1 to 36-1 ,,4, . . . itoie 62 RESISTANCE TO TRACTION ON COMMON ROADS. " These numbers, which give the angle of repose between 1 in 36*1 and 1 in 50*9 for the kinds of road-covering, Nos. 3 and 2, in most ordinary use, and corresponding to a road-surface in good order, may be somewhat increased, to from 1 in 28 to 1 in 33, for the ordinary state of the surface of a well-kept road, without there being any neces- sity for applying a brake to the wheels in descending, or going out of a trot in ascending. The steepest gradient that can be allowed on roads with a broken-stone covering is about 1 in 20, as this, from experience, is found to be about the angle of repose upon roads of this character in the state in which they are usually kept. Upon a road with this inclination, a horse can draw, at a walk, his usual load for a level without requiring the assistance of an extra horse ; and experience has further shown that a horse at the usual walking pace will attain, with less apparent fatigue, the summit of a gradient of 1 in 20 in nearly the same time that he would require to reach the same point on a trot over a gradient of 1 in 33. "A road on a dead level, or one with a continued and uniform ascent between the points of arrival and departure, where they lie upon different levels, is not the most favour- able to the draft of the horse. Each of these seems to fatigue him more than a line of alternate ascents and descents of slight gradients ; as, for example, gradients of 1 in 100, upon which a horse will draw as heavy a load with the same speed as upon a horizontal road. "The gradients should in all cases be reduced as far as practicable, as the extra exertion that a horse must put forth in overcoming heavy gradients is very considerable ; they should, as a general rule, therefore, be kept as low at least as 1 in 33, wherever the ground will admit of it. This can generally be effected, even in ascending steep hill-sides, by giving the axis of the road a zig-zag direc- tion, connecting the straight portions of the zig-zags by MAXIMUM GRADIENTS. 63 circular arcs. The gradients of the curved portions of the zig-zags should be reduced, and the roadway also at these points should be widened, for the safety of vehicles descend- ing rapidly. The width of the road may be increased about one-fourth, when the angle between the straight portions of the ziz-zags is from*120 to 90 ; and the increase should be nearly one-half where the angle is from 90 to 60." * NOTE BY THE EDITOR. Sir John Macneil, in 1836, maintained that no road was perfect unless its gradients were equal to or less than 1 in 40. In thus limiting the ruling gradient to 1 in 40, he justifies the assertion by the much greater outlay for repair on roads of steeper gradients. For instance, he adduces as a fact not generally known, that if a road has no greater inclinations than 1 in 40, there is 20 per cent, less cost for maintenance than for a road having an inclination of 1 in 20. The additional cost is due not only to the greater injury by the action of horses' feet on the steeper incline, which has already been noticed, but also to the greater wear of the road by the more frequent necessity for sledging or braking the wheels of vehicles in descending the steeper portions. Professor Mahan, it has been seen, page 62, recom- mends, as a general rule, that the gradients should be kept as low as 1 in 33 ; whilst M. Dumas, engineer-in - chief of the French Fonts et Chaussees, writing in 1843,f recommended, as a maximum rate of inclination, 1 in 50 ; for, he says, "not only are the surfaces of steeply-inclined roads subjected to abrasion by the feet of horses clambering up the hill, but, in the intervals of rest, loose stones are placed as props behind the wheels of vehicles, which are usually allowed to remain where they have been tempo- rarily placed, and may be the causes of serious accidents." * "A Treatise on Civil Engineering," by D. H. Mahan, 2nd edition* page 407. t " Annales des Ponts et Chaussees," 2nd series, 1 Semestre, 1843, page 343. 64 RESISTANCE TO TRACTION ON COMMON ROADS. Besides, lie states as the result of experience, tliat on broken-stone roads, in perfect condition, the resistance to traction is l-50th of the gross weight, or 45 Ibs. per ton, for which the angle of repose is 1 in 50 ; and he adds, with scientific acuteness, " that for the ascent of an incline of 1 in 50, the traction force required is just double that which is required on the level. " " Evidently, ' ' he continues, " there is no danger, under such conditions, in making the descent, since it requires but the slightest effort to check the vehicle ; whilst, in ascending, the horses can, without trouble, exert double the customary force for a short time.' 7 In fact, horses can easily enough surmount gradients of more than 3 per cent., or 1 in 33, at a trot, on roads in mediocre condition. M. Dupuit recommends for the maximum gradients of roads For metalled roads . . 3 per cent, or 1 in 33 For pavements . . .2 ,, 1 in 50 It can but be observed, upon the foregoing evidence, that Sir John Macneil's proportion of 1 in 40 for the maximum slopes of roads, is most nearly an average of the deductions which have been cited. But there is another condition the minimum longi- tudinal slope of a road. It should not be quite level, for provision must be made, by inclining the road, for running off surface-water. The minimum slope is fixed by one authority at 1 in 80 ; by another, at half a degree, or 1 in 115 ; and by the Corps des Fonts et Chaussees, at 1 in 125. In the Second Part of this work, by the Editor, he has given an analysis of the Boiling or Circumferential Ke- sistance of Wheels. CHAPTEB IV. ON THE SECTION OF ROADS. WHERE hills or gradients are unavoidable, they should be made as easy as possible ; and, although a certain amount of additional power must be required to draw a carriage up a hill, compared with the resistance on a level, yet so o 3 a) Description of the road. 3J3 6cg .9 fl 2 ^ s s | o fl !& PH S - i t Well -laid pavement Broken stone surface on a "bottom of) rough pavement or concrete . . / Broken stone surface laid on an old flint \ road j 31-4 44 62 48 1 35 1 in 71 1 in 51 1 in 36 Gravel road .... 140 3 35 1 in 16 The following table of gradients is of considerable value in laying out and arranging roads. The first column con- tains the gradient, expressed in the ratio of the height to the length ; the second and third columns contain the ver- tical rise in a mile and a chain respectively ; the fourth column, the angle of inclination with the horizon ; and the last column, the sine of the same angle, which is inserted for facilitating the calculation of the resistances occasioned by the gradient. GRADIENTS. 67 TABLE No. 5. GRADIENTS AND ANGLES OF INCLINATION OF ROADS. , oS 43 1 1o 11 cJ a || Angle (0) which gradient makes with the horizon. oi .2 1 02 Gradient. 1 cJ .3 grf 11 oj .a P Angle (0) which gvjulient makes with the horizon. Sine of angle 0. 1 in 10 528-0 6.60 5 42 58 09960 1 in 60 88-0 1-10 57 18 01667 11 480-06-00 5 11 40 -09054 99 60 81-2 1-02 52 54 01539 12 440-0 5-50 4 45 59/0830C 99 TO 75-4 94 49 7 01429 13S 406-1 5-08 4 23 56! -07570 ,, 7o 70-4 88 45 51 01334 14 377-14-71 4 5 14 -07128 99 80 66-0 82 42 58 01250 15 352-04-40 3 48 51 -06652 99 85 62-1 78 40 27 01177 16 330-0 4-12 3 34 35 -06238 ,, 90 58-7 73 38 12 01111 IT 310-6 3-88 3 21 59 -05872 99 95 55-6 69 36 11 01053 99 18 293-3 3 67 3 10 47 -05547 99 100 52-8 66 34 23 01000 19 277-9 ;3-47 3 46 05256 99 HO 48-0 60 31 15 00909 20 264-0 '3.30 2 51 21 04982 99 120 44-0 55 28 39 00833 21 251-4 :3-14 2 43 35 04757 99 130 40-6 51 26 27 00769 22 240-0 J3-00 2 36 10 04541 99 140 37-7 47 24 33 00714 23 229*6 2*87 2 29 22 04344 99 l^O 35-2 44 22 55 00666 99 24 220-0 275 2 23 10 04163 160 33-0 41 21 29 00625 25' 211-2 12-64 2 17 26 03997 99 1TO 31-1 39 20 13 00588 26 203-1 2-54 2 12 2 03840 99 180 29-3 37 19 6 00556 27 195-5 2-42 272 03694 190 27-8 35 18 6 00527 28 188-5 2-36 225 03551 200 26-4 33 17 11 00500 29 182-1 2-28 58 341-03448 210 25-1 31 16 22 00476 30 176-0 2-20 54 37 03333 220 24-0 30 15 37 00454 31 170-3 2-13 50 55 03226 230 23-0 29 14 57 00435 32 165-0 2-06 47 27 03125 240 22-0 27 14 19 00417 33 160-0 2-00 44 12 03031 250 2M 26 13 45 00400 34 1553 1-94 41 8 02941 260 20-3 25 13 13 00385 35 150-9 1-88 38 14 02857 99 270 19-6 24 12 44 00370 99 36 146-7 1-86 35 28 02777 280 18-9 24 12 17 00357 37 142-7 1-78 32 53 02702 290 18-2 23 11 51 00345 38 138-9 1-74 30 27 02631 300 17-6 22 11 28 00334 99 39 135-4 1-69 28 8 02563 325 16-2 20 10 35 00308 40 132-0 1-65 25 57 02500 350 15-1 19 9 49 00286 99 41 128-8 1-61 1 23 50 02438 i 375 14-0 18 9 10 00267 9, 42 125-7 1-57 1 21 50 02380 99 400 13-2 17 8 36 00250 43 122-8 1-53 1 19 56 02325 99 425 12-4 16 085 00235 44 120-0 1-50 1 18 7 02272 99 450 11-7 15 7 38 00222 99 45 117-3 1-47 1 16 24 02222 99 475 11-5 14 7 14 00210 46 114-8 1-44 1 14 43 02173 500 10-6 13 6 53 00200 47 112-3 1-40 1 13 8 02127 525 10-1 12 6 33 00191 48 110-0 1-37 1 11 37 02083 550 9-6 12 6 15 00182 49 107-7 1-35 1 10 9 02040 99 575 9-2 11 5 59 00174 50 105-6 1-32 1 8 6 01981 600 8-8 11 5 44 00167 55 96-0 1-20 1 2 30 01818 68 ON THE SECTION OF ROADS. Width and Transverse Section of Roads. It is recom- mended that roads should be wide. It is an error to suppose that the cost of repairing a road depends entirely upon the extent of its surface, and increases with its width. The cost per mile of road depends more upon the extent and the nature of the traffic; and it may be asserted, generally, that the same quantity of material is necessary for the repair of a road, whether wide or narrow, subjected to the same amount of traffic. On the narrow road, the traffic, being confined very much to one track, the road would be worn more severely than when the traffic is spread over a larger surface. The expense of spreading the material over the wider road would be somewhat greater, but the cost for material might be taken as the same. One of the advantages of a wide road is, that the air and the sun exercise more influence in keeping its surface dry. The first cost of a wide road is certainly greater than that of a narrow road, nearly in the ratio of the widths. For roads situated between towns of importance, and exposed to much traffic, the width should not be less than 30 ft., which would admit of four vehicles abreast ; besides a footpath of 6 ft. In the immediate vicinity of large towns and cities, the width should be greater. The form of the cross section of a road is a subject of much importance, and it is one upon which much difference of opinion exists. Some persons advocate a considerable degree of curvature in the upper surface of the road, with the view of facilitating the drainage of its surface ; whilst others are averse to a road being much curved. It is the practice of others, again, to form the road on a flat surface transversely; whilst others give a dip to the formation- surface each way from the centre, supposing that the drainage of the road is thereby facilitated. The only advantage resulting from the curving of the MACADAM'S VIEWS. 69 transverse section of the road is, that the water, which would otherwise collect upon its surface, is allowed to drain freely off into the side ditches. It has been urged that, in laying on fresh material upon a road, it is necessary to keep the centre much higher than the sides ; because, in consequence of the greater number of carriages using the middle of the road, that portion wears more quickly than the sides, and that, unless it is made originally much higher, when so worn it necessarily forms a hollow or depression, from which water cannot drain. Now it is entirely overlooked by those who advance this argument, that the cause of carriages using the middle in preference to the sides of a road, is its rounding form, since it is only in that situation that a carriage stands upright. If the road were comparatively flat, every portion would be equally used ; but on very convex roads, the middle is the only portion of the road on which it is safe to travel. On this subject, Mr. Macadam remarks, in his evidence before a committee of the House of Commons,* "I consider a road should be as flat as possible with regard to allowing the water to run off it at all, because a carriage ought to stand upright in travelling as much as possible. I have generally made roads 3 in. higher in the centre than I have at the sides, when they are 18 ft. wide; if the road be smooth and well made, the water will run off very easily in such a slope." And, in answer to the question, " Do you consider a road so made will not be likely to wear hollow in the middle, so as to allow the water to stand, after it has been used for some time ?" he replies, " No ; when a road is made flat, people will not follow the middle of it as they do when it is made extremely convex. Gentlemen will have observed that in roads very convex, travellers generally follow the track in the middle, which * Parliamentary Report on the Highways of the Kingdom, 1819, page 22. 70 ON THE SECTION OF ROADS. is the only place where a carriage can run upright, by which means three furrows are made by the horses and the wheels, and water continually stands there ; and I think that more water actually stands upon a very convex road than on one which is reasonably flat." On the same subject, Mr. Walker remarks,* " A road much rounded is dangerous, particularly if the cross section approaches towards the segment of a circle, the slope in that case not being uniform, but increasing rapidly from the nature of the curve, as we depart from the middle or vertical line. The over-rounding of roads is also injurious to then^ by either confining the heavy carriages to one track in the crown of the road, or, if they go upon the sides, by the greater wear they produce, from their constant tendency to move down the inclined plane, owing to the angle which the surface of the road and the line of gravity of the load form with each other; and, as this tendency is perpendicular to the line of draught, the labour of the horse and the wear of the carriage wheels are both much increased by it." f The drainage of the surface of the road is then the only useful purpose answered by making it convex. But the surface of a road is much more efficiently drained by a small inclination in the direction of its length, than by a much greater transverse slope. On this subject, Mr. Walker has very justly remarked, J " Clearing the road of water is best secured by selecting a course for the road which is not horizontally level, so that the surface of the road may, in its longitudinal section, form, in some degree, an inclined plane ; and when this cannot be obtained, owing to the extreme flatness of the country, an artificial * Parliamentary Report, 1819, page 49. f Remarks on the evils of " barreled roads," as they were called, have been made in the Historical chapter, page 4. EDITOR. J Parliamentary Report, 1819, page 48. TRANSVERSE GRADIENTS. 71 inclination may generally be made. When a road is so formed, every wheel-track that is made, being in the line of inclination, becomes a channel for carrying off the water much more effectually than can be done by a curvature in the cross section or rise in the middle of the road, without the danger or other disadvantages which necessarily attend the rounding a road much in the middle. I consider a fall of about li inches in 10 feet to be a minimum in this case, if it is attainable without a great deal of extra expense." Whilst, then, the advantages attending the extreme con- vexity of roads is so small, the disadvantages are consider- able. On roads so constructed, vehicles must either keep to the crown of the road, and so occasion an excessive and unequal wear of its surface, or use the sides, with the liability of being overturned. The evidence of coach- masters and others, taken before the committee of the House of Commons, and appended to the report already quoted from, fully bears out the view here taken, and shows that many accidents have arisen from the practice of forming roads with an excessive amount of convexity. With reference to the above remarks, it is only intended to express disapproval of the practice of forming roads with cross sections rounding in an extreme degree and not to advocate a perfectly, or nearly, flat road, as many, who have fallen into the opposite error, have done. It is recommended, as the best form which could be given to a road, that its cross section should be formed of two straight lines inclined at the rate of about 1 in 30, and connected at the middle or crown of the road by a segment of a circle, having a radius of about 90 feet. This form of section is shown in Fig. 22, and the rate of inclination there given is quite sufficient to keep the surface of a road drained, provided it is maintained in good order, free from ruts. If the maintenance is neglected, no degree of convexity which can be given to the road will 72 ON THE SECTION OF ROADS. be of any avail, as the water will remain in the hollows or furrows. The form of cross section here suggested is equally adapted to all widths of road, as the straight lines have merely to be extended at the same rate of inclination, until they meet the sides of the road. Professor Mahan is of the same opinion with respect to the proper section of a road namely, that it should be formed of two straight sides, connected at the middle by a flat circular arc. The slope which he recommends is 1 in 48, or 1 inch in 4 feet. With regard to the form which should be given to the bed upon which the road is to be formed, a similar dif- ference of opinion exists as to whether it should be flat or rounding. Except where the surface upon which the road is to be formed is a strong clay, or other soil imper- vious to water, no benefit results as far as drainage is con- cerned, in making the formation-surface or bed of the road convex. It should be borne in mind that, after the road materials are laid upon the formation-surface, and have been for some time subjected to the pressure of heavy vehicles passing over them, they become, to a certain extent, intermixed : the road materials are forced down into the soil, and the soil works up amongst the stones, and the original line of separation becomes entirely lost. If the surface upon which the road materials are laid were to remain a distinct flat surface, perfectly even and regular, into which the road materials could not be forced, then it would be useful to give such an inclination to it as would allow any water which might find its way through the crust or covering of the road, to run off to the sides. Even so, it would have to force a passage between the road materials and the surface on which they rest. Such is, however, far from being the case ; and, therefore, unless under peculiar circumstances, no ON THE SECTION OF ROADS. 7& water which, finds its way through the hard compact surface of the road itself is arrested by the comparatively soft sur- face of its bed, and carried off into the side ditches, what- ever the slope which might be given to the bed. While, however, it is believed, that, as far as drainage is con- cerned, it is useless to form the bed or formation surface of the road with a transverse slope, it should, nevertheless, be formed to the same outline as that recommended for the outer surface ; making the two surfaces parallel, and thus bestowing an equal depth of road material over every portion of the road. Nevertheless, some road-makers not only recommend a less depth of road materials to be put on the sides than on the middle of the road, but they further advise that an inferior description of material should be employed at the sides. On this subject the following remarks of Mr. Hughes are very much to the purpose:* "A very common opinion is, that the depth of material in the middle of the road should be greater than at the sides, but, for my part, I have never been able to discover why the sides of the road should be at all inferior to the middle in hardness and solidity. On the contrary, it would be a great improvement in general travelling, if carriages could be made to adhere more strictly to the rule of keeping the proper side of the road ; and the reasonable inducement to this practice is, obvi- ously, to make the sides equally hard and solid with the middle. In many roads, even where considerable traffic exists, the only good part of the road consists of about 8 or 10 feet in the middle, the sides being formed with small gravel quite unfit to carry heavy traffic ; and the consequence is, that the whole crowd of vehicles is forced into the centre track of the road ; thus at least doubling or trebling the wear and tear which would take place if * "The Practice of Making and Repairing Roads," by Thomas Hug-hes, 1838, page 12. 1 /4 ON THE SECTION OF ROADS. the sides were, as they ought to be, equally good with the centre. Another mischievous consequence is, that when it becomes necessary to repair the centre of the road, the carriages are driven off the only good part on to the sides, which consist of weak material, and are often even dan- gerous for the passage of heavily-laden stage coaches. On the other hand, if equal labour and materials be expended on the whole breadth of the road, it is evident that the wear and tear will be far more uniform ; and when any one part requires repair, the traffic may with safety be turned on to another part. Hence, I should always lay on the same depth of material all over the road : and this alone will of course render it necessary to curve the bed of the road." Great attention should be paid to the drainage of roads, with respect to their upper surface as well as to the sur- face of the ground on which they rest. To promote the surface-drainage, the road should be formed with the transverse section shown in Fig. 22, and on each side of the road a ditch should be formed of sufficient capacity to receive all the water which can fall upon the road, and it should be of such a depth and with such a declivity as to conduct the water freely away. When footpaths are to be constructed on the sides of the road, a channel or water- course should be formed between the footpaths and the road, and small drains, formed of tiles or earthern tubes, such as are used for underdraining lands, should be laid under the footpath, at such a level as to take off all the water which may collect in this channel, and convey it into the ditch. In the best constructed roads, these side channels are paved with flints or pebbles. The drains under the footpath should be introduced about every 60 feet, and should have the same inclination namely, 1 in 30, as is recommended for the sides of the road, as shown in Fig. 22. A greater inclination would be objec- ON THE SECTION OF ROADS. 75 tionable. It is a very frequent mistake to give too great a fall to small drains, for such, a current through, them is produced as may wash away or undermine the ground around them, and ultimately cause their destruction. When a drain is once closed by any obstruc- tion, no amount of fall which could be given to it would suffice again to clear the passage ; whilst a drain having a considerable current through it, would be much more likely to be stopped by foreign matter carried into it, than a drain with a less rapid stream. When the surface of a road, constructed of suitable materials, compactly laid, is drained in the manner which has just been described, very little water finds its way to the sub- stratum. For some descriptions of soil, how- ever, it is desirable to adopt additional means for maintaining the foundation of a road in a dry state; as, for instance, when the surface is a strong clay through which no water can percolate, or when the ground beneath the road is naturally of a soft, wet, or peaty nature Under such circumstances a species of under- drainage should be provided. When the sur- face of the ground is formed to the level intended for the reception of the road materials, trenches should be cut across the road from, a foot to eighteen inches in depth, and about a foot wide at the bottom, the sides being sloped as shown in Fig. 23. The distances at which these drains should be formed depends in a great measure on the nature of the soil ; in the case of a strong clay soil, or a soil which is naturally very wet, there should be a cross E2 .3 Fig. 22 76 ON THE SECTION OF ROADS. drain at intervals of 20 ft. As the ground becomes firmer and drier, the interval may be increased in length. A drain not less than 4 inches square internally should be formed in the trench, of old bricks, drain-tiles, or flat stones, as shown in Fig. 23 ; or the drain may be formed by any other mode used for under-drains ; and the remainder of the trench should be filled with coarse stones free from all clay or Formation Surface. Fig. 23. dirt, in the manner shown in the figure. These drains must have a fall from the centre of the road into the ditches on either side : an inclination of 1 in 30 will be sufficient. When the road is level in the direction of its length, the drains should run square across ; but, on inclined portions of the road, the drains should be formed, as shown on the ROAD-WAY Fig. 24. plan, Fig. 24, somewhat in the form of a very flat Y, the apex being in the middle of the road, and the drains mak- ing an obtuse angle with the line of the road, in the direc- tion in which it falls. The amount of this angle should not be greater than is shown in Fig. 24. When a road with footpaths is underdrained in the manner just described, it is not necessary to form drains from ON THE SECTION OF ROADS. 77 the side channel under the footpath into the ditch, as shown in Fig. 22. It is sufficient to carry up a little shaft, constructed in the same way as the drain, from the drain to the chan- nel, covering the shaft with a small grating to prevent leaves or other substances, which might choke the drain, from being carried into it. This method of forming the drains is shown at A in Fig. 25. The footpath should be not more than 9 inches higher than the bottom of the side chan- nel. The surface of the footpath should have a pitch of 2 inches, towards the side channels, to convey its surface-water into them. When the natural soil is firm and sandy, or gravelly, its surface will serve for the footpath ; but in other cases the natural soil must be thrown out to a depth of 6 in., and the excavation filled with fine clean gravel. To prevent the foot- path from being damaged by the current of water in the side channel, its side slope, next to the side channel, must be protected by a facing of good sods, or of dry stone. Independently of the drainage for marshy soils, they will require, when the subsoil is of a spongy elastic nature, an artificial bed for the road-covering. This bed may, in some cases, be formed by simply removing the, upper stratum to a depth of several feet, and supplying its place with well-packed gravel, or any soil of a firm character. In other cases, when the subsoil yields readily to the ordinary pressure that the road-surface must bear, a bed of brushwood from 9 to 18 in. in thickness, must be formed to receive the Fig. 25. 78 ON THE SECTION OF ROADS. soil on which the road-covering is to rest. The brush- wood should be carefully selected from the long straight slender shoots of the branches or undergrowth, and be tied up in bundles, termed fascines, from 9 to 12 in. in diameter, and from 10 to 20 ft. long. The fascines are laid in alternate layers crosswise and lengthwise, and the layers are either connected by pickets, or else the withes, with which the fascines are bound, are cut to allow the brushwood to form a uniform and compact bed. This method of securing a good bed for structures on a weak wet soil has been long practised in Holland, and experience has fully tested its excellence. CHAPTEE V. g CONSTRUCTION OF ROADS. IT has been maintained, by Mr. Macadam and others, that a soft and yielding foundation for a road is better than one which is firm and unyielding. Mr. Macadam has stated that he ' * should rather prefer a soft one to a hard one ;" and even a bog, " if it was not such a bog as would not allow a man to walk over it." * The principles upon which this opinion was founded were, that the road on the soft foundation being more yielding or elastic, the materials of which the covering of the road was formed would be less likely to be crushed and worn away by the passage of a heavy traffic over them than when placed on a hard solid. The contrary opinion is, however, maintained by the largest number of advocates ; and it is certain that there is no more general cause of bad roads than soft foundations. A firm, solid, and dry substratum is necessary for the road mate- rials to rest upon. However good the materials themselves may be, and however much care may be bestowed upon the distribution of them, the material and labour are of no avail unless a good foundation has been prepared. The outer surface of the road should be regarded merely as a covering to protect the actual working road beneath, which should be sufficiently firm and substantial to support the whole of the traffic to be carried over it. The proper func- tion of the road materials is to protect the actual road from * Parliamentary Report, 1819, p. 23, 80 CONSTRUCTION OF ROADS. being worn and injured by horses' feet and wheels, or by the action of the weather. This lower, or sub-road, as it may be called, being once properly constructed, would last for ever, provided that the outer case or covering is re- newed from time to time, so as to maintain a sufficient depth for the protection of the sub-road. Roads may be classed as follows, according to the manner in which the foundation is formed : 1st. Roads having no artificial foundation, but in which the covering materials are laid on the surface of the ground. 2nd. Eoads having a foundation of concrete. 3rd. Eoads having a paved foundation. And each of these might be sub-divided according to the kind of material employed as a covering. The first class of roads comprises by far the largest pro- portion of the roads in this country. But it should only be employed in cases where the road is not sufficiently important to warrant any large expenditure, and where the anticipated amount of traffic is small. Every care should be taken to make the road as solid as possible. If the ground is at all of a soft or wet nature, deep ditches should be cut on each side of the line of the road, and cross under -drains should be formed in the manner already described at page 76. And where the ground is very soft, a layer of faggots or brushwood, from 4 to 6 inches in depth, should be laid over the surface of the ground to receive the road materials. For embankments, or in other situations where the ground has been recently deposited, the surface should be either rolled or punned, that is, beaten with heavy beetles, so as to confer upon it as great a degree of solidity as possible. The same mode of proceeding should be followed, even where it is intended to form either a paved or a concrete foundation. The employment of concrete composed of gravel and CONSTRUCTION OF ROADS. 81 lime was first proposed by Mr. Thomas Hughes, and the following remarks upon its use are quoted from his work on roads : * "The use of lime concrete, although an introduction of modern times, and certainly one of rather a novel charac- ter, derives its real origin from a very remote period. Wei have indisputable evidence that the Romans, in construct- ing their military ways, particularly in France, adopted the practice of forming a concrete foundation composed of gravel and lime, on which also they placed large stones as a pavement. The consequence of a construction so solid has been, that, in many parts of Europe, the original bed or crust of the Roman roads is not at the present day en- tirely worn down, even after a lapse of fifteen centuries. ' ' With the view of affording a modern example in which lime concrete has been used, I would refer to the Brixton Road, where a concrete composed of gravel and lime has been recently applied by Mr. Charles Penfold, surveyor to the trust. In this case the proportion of gravel to lime is that of four to one. The lime is obtained from Merstham or Dorking, and before being used it is thoroughly ground to powder. The concrete is made on the surface of the road, and great care is taken, when the water is added, that every particle of the lime is properly slaked and saturated. The bed of concrete having been spread to the depth of 6 inches over the half breadth of the road, the surface is then covered over with 6 inches of good hard gravel or broken stone, and this depth is laid on in two courses, of 3 inches at a time, the first course being frequently laid on a few hours after the concrete has been placed on the road. The carriages, however, are not on any account allowed to pass over it until the concrete has become sufficiently hard and solid to carry the traffic without suffering the road material to sink and be pressed into the body of concrete. On the * "The Practice of Making and Repairing Roads," p. 44. E 3 82 CONSTRUCTION OF ROADS. other hand, the covering of gravel is always laid on before the concrete has become quite hard, in order to admit of a more perfect binding and junction between the two beds than would take place if the concrete were suffered to be- come hard before laying on the first covering. The bene- ficial effect arising from the practice of laying on the gravel exactly at the proper time is, that the lower stones, pressed by their own weight, and by those above them, sink par- tially into the concrete, and thus remain fixed in a matrix, from which they could not easily be dislodged. The lower pebbles being thus fixed, and their rolling motion conse- quently prevented, an immediate tendency to bind is com- municated to the rest of the material a fact which must be evident, if we consider that the state called binding, or rather that produced by the bidding, is nothing more than the solidity arising from the complete fixing and wedging of every part of the covering, so that the pebbles no longer possess the power of moving about and rubbing against . each other. It is found that, in a very few days after the first layer has been run upon, the other, or top covering, may be applied ; and, shortly afterwards, the concrete, and the whole body of road material, becomes perfectly solid from top to bottom. The contrast thus presented to the length of time and trouble required to effect the binding of road materials where the whole mass is laid on loose is alone a very strong recommendation in favour of the concrete. " The experiment of using concrete on the Brixton Eoad, although not at present on a very extensive scale, has been tried under circumstances very far from being favourable, and on a part of the road which had hitherto baffled every attempt to make it solid. Since the concrete has been laid down, however, there is not a firmer piece of road in the whole trust." Mr. Penfold gives the result of an experiment made by CONSTRUCTION OF ROADS. 83 him upon the Walworth Eoad. "It was raised by nine inches of concrete, and six of granite and Kentish rag- stone mixed ; and in some parts it was covered by rag and flints. The improvement is so great, with respect to the draught, and so desirable with respect to the saving in the annual repair, that the trust have directed it to be applied to upwards of two miles of road upon which the greatest traffic exists." * One of the principal advantages attending the employ- ment of concrete as a foundation for roads is, that a good and solid road may be made with materials, such as round pebbly gravel, which, on any other mode of application, would be ill-suited to the purpose, and would form a very imperfect road. This description of gravel is that which is by far the most frequently met with. The gravel selected for this purpose should be free from any kind of dirt, clay, or other impurity, and should consist of stones and sand, mixed in about such proportions that the latter would just fill the interstices of the former. The gravel should then be mixed with the proper quantity of ground unslaked lime in ordinary cases five or six parts of gravel and one of lime will be found to answer ; after which, sufficient water being added to effect the slaking of the lime, the whole should be quickly, but thoroughly, mixed up, and then immediately thrown into place, and trimmed off at once to the proper form intended to be given to its upper surface ; the first layer of broken stones or screened gravel should then, as Mr. Hughes directs, be placed just as the concrete is about to set. The third mode of forming an artificial foundation was introduced by Mr. Telford. It consists in laying a rough pavement on the top of the formation-surface, which is afterwards covered by the road materials. The following * " Practical Treatise on the best Mode of Making and Repairing Roads," by Charles Penfold, p. 31. 84 CONSTRUCTION OF ROADS. is an extract from one of Mr. Telford's specifications for a portion of the Holyhead Eoad : " Upon the level bed pre- pared for the road materials, a bottom course or layer of stones is to be set by hand, in form of a close, firm pave- ment ; the stones set in the middle of the road are to be 7 inches in depth : at 9 feet from the centre, 5 inches ; at 12 from the centre, 4 inches; and at 15 feet, 3 inches. They are to be set on their broadest edges lengthwise across the road, and the breadth of the upper edge is not to exceed 4 inches, in any case. All the irregularities of the upper part of the said pavement are to be broken off by the hammer, and all the interstices to be filled with stone chips, firmly wedged or packed by hand, with a light hammer ; so that when the whole pavement is finished, there shall be a convexity of 4 inches in the breadth of 15 feet from the centre." * The stone which Telford employed for this purpose, was generally such as would have been totally unfit for most other purposes, whether on account of its inferior quality, or the smallness of its dimensions. In comparing the two methods of forming artificial foundations of roads, regard must be had to the nature of the materials found in the locality in which the road is to be formed. Where stone is plentiful, and easily procured, the paved foundation would be the best ; whilst, where stone is scarce, and gravel and lime abundant, the preference must be given to the concrete foundation. The foundation of the road having been prepared, the next proceeding is, to lay a firm and compact covering upon the foundation, to form a smooth surface for carriages to travel upon. The materials of which the covering is composed should possess the property of becoming quickly united into one solid mass, of which the surface should be smooth and hard, and at the same time not liable to * Sir H. Parnell on Eoads, p. 132. CONSTRUCTION OF ROADS. 85 be broken to pieces, or ground into dust, by wheels or horses' feet. The materials which have been employed for this purpose are of two kinds ; angular fragments of broken stone of different sorts, and gravelly pebbles. It is essential to the formation of a good road that the distinc- tion here pointed out be always kept clearly in view, because a totally different mode of proceeding must be adopted to form a perfect road with these two classes of material. The want of attention to the distinction here pointed out has led to much discussion and misapprehen- sion on the employment of clay, chalk, or other material, as a binding upon roads. If the materials of which the covering is to be formed are in angular masses, no binding of any description is requisite ; as they quickly become united by dovetailing, as it were, amongst each other much more firmly than they would by the use of any kind of artificial cement. When the stones, instead of being angular, are round and pebbly, like gravel stones, it is necessary to mix with them just sufficient foreign matter, of a binding nature, as will serve to fill up the interstices between the stones, for otherwise these would roll about, and would prevent the road from becoming solid. There are, then, two methods of cementing or solidifying the surface of a road : one, by the mechanical form of the materials themselves forming a species of bond ; the other, by the use of some cementing or binding matter. And in comparing the relative merits of the two, the preference must certainly be given to the former, that in which the stones are united in virtue of their angular form, without the use of any cementing material. The principal reason for this preference is, that roads formed with stones so united, are not materially affected by wet or frosty weather ; whereas, roads whose surfaces are composed of pebbly stones united by cementing material, become loose 86 CONSTRUCTION OF ROADS. and rotten under such circumstances : the cementing mate- rial becoming softened by the wet, and reduced to a loose pulverulent condition by subsequent frost. The first method, that of forming the road-covering entirely with angular pieces of stone, without any other material, was first strongly recommended by Mr. Macadam, and subsequent experience has shown its superiority over every other which has been employed. The most im- portant quality in stone for road-making is toughness : mere hardness without toughness is of no use, as such stone be- comes rapidly reduced to powder by the action of wheels. Those stones which have been found to answer this pur- pose best are, the whinstones, basalts, granites, and beach pebbles. The softer descriptions of stone, such as the sandstones, are not fitted for this purpose, being far too weak to resist the crushing action of wheels. The harder and more compact limestones may be employed ; but, generally speaking, the limestones are to be avoided, in consequence of their great affinity for water, by which in frosty weather, which has been preceded by wet, they are split up into powder, when the solidity of the road is de- stroyed. Next in importance to the quality of the stone is its preparation. This consists in reducing it to angular frag- ments of such a size that they will pass freely through a ring of 2| inches in diameter, in every direction ; that is, that their largest dimensions shall not exceed that mea- sure. The stone thus prepared, should be evenly spread over the surface prepared for the foundation of the road, to a depth of about 6 inches ; and the road should then be opened for traffic. In Mr. Telford's specifications, he usually directed that, on the top of this coating of broken stone, a layer of good clean gravel, about 1J inch in depth, should be spread before throwing the road open for CONSTRUCTION OF ROADS. 87 use. The reason for this practice was, to lessen the ex- treme unevenness of the surface, and to render the road more pleasant to pass over when first opened. It would be better, however, for the public to put up with the tem- porary inconvenience of a rough road, because the gravel does a permanent injury to the road, and reduces in a considerable degree the facility with which the stones unite into a compact, solid mass. Broken stone, being superior to gravel for the purpose of road-making, should always be employed where it can be easily obtained. The quality of gravel varies so consider- ably, that while some kinds may, when properly prepared, form a very excellent road, others may be entirely worth- less ; such as those kinds of gravel the stones composing which are of the sandstones and flints, for even flints, although hard, are so excessively brittle as to be immedi- ately crushed by the passing of the wheels over them. The gravel when taken from the pit should be passed over a screen which will allow all stones less than } of an inch to pass through it, and the fine stuff, or hoggin, as it is technically termed, thus obtained, should be reserved for forming the footpaths ; the remainder, which has not passed through the screen, should have all the stones whose greatest dimension is more than 2 inches removed and broken, and it would be desirable that these broken stones should be reserved for the upper layer. In screen- ing the gravel, especially as it comes from the pit, a cer- tain portion of loam is generally found adhering to the stones, and this should by no means be separated from them, for, although angular broken stones require no extraneous substance to cause them to bind, the pebbles of which most gravel is composed, require a cer- tain amount of loam, clay, or chalk, to fill up the inter- stices between the stones, and to prevent them from being rolled about, as they otherwise would be. The gravel 88 CONSTRUCTION OF ROADS. thus prepared by screening should be laid on and spread to a uniform depth of not more than 6 inches over the whole surface. On this subject, Mr. Hughes says :* " In laying on this upper covering many surveyors commit a great error in not making a distinct difference between angular or broken stones and those rounded smooth pebbles of which gravel is usually composed. The former cannot be too well cleaned before being laid on the road, because, even when entirely divested of all earthy matter, they soon become wedged and bound closely together when the pres- sure of carriages comes upon them. But the case is dif- ferent with the smooth, round surfaces of gravel ; for if this material be entirely cleaned by means of washing and repeated siftings, the pebbles will never bind, until in a great measure they become ground and worn down by the constant pressure and rubbing against each other. Before this takes place, the surface of the road must be considerably weakened, and will, in fact, be incapable of supporting the pressure of heavy wheels, which conse- quently sink into it, and meet with considerable resistance to their progress. Under these circumstances, it seems that the practice of too scrupulously cleaning the rounded pebbles of gravel must be decidedly condemned ; and the question then arises, to what extent should the cleaning process be dispensed with ? or, in other words, what pro- portion of the binding material found in the rough gravel, as taken out of the pit, should be allowed to remain in the mass intended to be placed on the road ? * * * A long course of experience, accompanied by attentive observa- tions on these details in the practice of road-making, has convinced me that it is much better and safer, as a general rule, to leave too much of the binding material in the * u The Practice of Making and Repairing Roads," p. 16. CONSTRUCTION OF ROADS. 89 gravel than to divest it too completely of this substance- When the gravel is placed on a road without being suffi- ciently cleaned, the constant wear and tear, aided by the occurrence of wet weather, causes the harder material or actual gravel to be pressed close together ; and the surplus of soft binding material remaining after the interstices be- tween the pebbles are filled up, being then forced to the top, and usually mixed with water, becomes mud, and according to the usual practice should be scraped to the sides of the road. When this has been done, the surface is usually firm and solid ; because the hard gravel below the mud has become perfectly bound, without, at the same time, being broken or ground to pieces. Suppose, next, a road covered with gravel, too much cleaned, where it is evident that the destruction of the gravel will continue until it becomes broken into angular pieces, and a suffi- cient quantity of pulverised material has been formed to hold the stones in their places and thus to effect the bind- ing of the mass. I need hardly say, that the deterioration thus occasioned to the road is an evil of much more im- portance, and one much more to be avoided, than that occasioned by employing stones not sufficiently cleaned. Begardless of all this, however, it is the practice of many road-surveyors to insist that all gravel, of whatever quality, shall be rendered perfectly clean by repeated siftings, and even by washing, until it becomes entirely divested of all that may properly be considered the binding part of the material." Particular care and attention is required to be given to new roads when opened for traffic ; a sufficient number of men should be employed to keep every rut raked in the moment it appears, and guards or fenders should be placed on the road, in order that vehicles may be caused to pass over every part of the surface in turn. If these precautions are not taken, years may elapse before the road attains a 90 CONSTRUCTION OF ROADS. firm condition; and many roads have been permanently ruined through want of sufficient attention when first used. When ruts are once formed, vehicles using the road keep in the same track, deepening and increasing the rut. In wet weather, the rut is filled with water, which, having no other means of escape, slowly penetrates the sides and bottom of the rut, rendering them so soft as to be more susceptible to the action of every succeeding carriage. Huts thus formed, involve a much larger outlay to repair the injury than that which would have been sufficient to prevent it; besides the inconvenience, danger, and ex- pense to the public, in travelling on a road in such a condition. Amongst the substances which were mentioned as bind- ing material to be mixed with clean gravel, was chalk. Many roads have been ruined by its improper use. There are two modes in which chalk may be advantageously employed in the construction of roads. It may be laid in the very bottom of the road, to form the foundation ; but io must be at such, a depth as to be entirely beyond the influence of frost, otherwise it will quickly destroy the road, for chalk has a very powerful affinity for water ; or rather, to speak more correctly, capillary attraction for it, in consequence of which it readily absorbs all the moisture which finds its way through the road covering. Herein consists the value of chalk if judiciously applied, for the water thus absorbed would otherwise penetrate to the foundation of the road and render it soft. If, however, the chalk be placed within the reach of frost, the water, which is only mechanically held by the chalk, will, in the act of congealing, expand, and by so doing rend the chalk into fragments, and reduce it, in fact, to a pulverulent state, in which condition it is changed by the succeeding thaw into a soft paste or mud. The other purpose for which chalk may be employed is, as before mentioned, to be CONSTRUCTION OF ROADS. 91 mixed with gravel in order to make it bind ; though, in using chalk for this purpose, it is only required when the gravel is perfectly clean and free from other binding matter. The mixing of chalk with gravel already contain- ing sufficient clay or loam is not only useless, but is posi- tively injurious ; and, even when the gravel is of such a nature as to require the intermixture of chalk, care should be taken not to add too much chalk, for it is not with chalk as with the loam or clay, with which gravel is naturally combined. Clay, generally speaking, possesses little power of absorbing water, but the superabundant chalk would soon be reduced to the state of a soft paste by the action of the weather, in the manner which has been described. Chalk, therefore, if used as a binding material with gravel, on the surface of roads, should be reduced to a state of powder, and should be perfectly and thoroughly mixed with the gravel before the latter is spread on the road. Although the use of bushes or bundles of faggots has been recommended for the foundation of roads over very soft or boggy ground, they should only be employed in such, a situation, and they should be placed at such a depth below the surface, as will ensure their always being damp ; for when they are placed where they would be alternately wet and dry, they quickly become rotten, and form a soft stratum beneath the road. NOTE BY THE EDITOR. Mr. Walker,* in 1819, speaking of the great advantage of filling up or grouting the joints of granite pavement with lime-water, which finds its way into the gravel between and under the stones, and forms the whole into a concrete mass, made a suggestion of value, in recommending, for the same purpose, a mixture of a * Evidence before the Select Committee on the Highways of the Kingdom, 1819. 92 CONSTRUCTION OF ROADS. little of the borings or clippings of iron, or small scraps of hoop-iron, with the gravel used in filling up the joints of the paving. The water would, he said, very soon create an oxide of iron, and form the gravel into a species of rock. " I have seen," he adds, "apiece of rusty hoop taken from under water, to which the gravel had so connected itself, for 4 or 5 inches round the hoop, as not to be separated without a small blow of a hammer. And the cast-iron pipes which are laid in moist gravel, soon exhibit the same tendency." CHAPTER VI. ON REPAIRING AND IMPROVING ROADS. THE improvement of existing roads may be divided into two distinct branches ; namely, the improvement of their general course and levels, and the improvement of the materials of the road. The first of these consists in the application of the principles which have been laid down for the construction of new roads, and consists generally in straightening their course by extinguishing unnecessary curves and bends ; improving their levels by either avoid- ing or cutting down hills, and embanking valleys ; increas- ing their width, where requisite, and rendering it uniform throughout. With regard to the improvement of the surface, the operations may consist in reducing the transverse section to the form shown in Fig. 22, page 75, filling- up all ruts, cleansing and deepening, if necessary, the side ditches, cutting down trees or hedges by the side of the road, removing mud-banks which but too often exist on the road-sides, and placing proper materials on its surface. Of all these, the most important to explain, because the most difficult to effect, and the least generally understood, is the method by which the condition of the surface of the road may be improved. In the practice of many surveyors, the remedy for a bad road is to heap on fresh material ; whereas, as Mr. Macadam has very justly observed,* " Generally, the roads of the kingdom contain a supply * Parliamentary Report, 1819, p. 21. 94 ON REPAIRING AND IMPROVING ROADS. of materials sufficient for their use for several years, if they were properly lifted and applied." Generally speak- ing, the cause of inferiority is the imperfect transverse form of the roads, and the improper manner in which the road-materials are used. The remedial measures to be adopted must, in a great degree, depend upon the nature of the materials composing the upper surface of the road ; but, whatever these may be, the road must be brought to the proper form of section before much improvement can be expected. This should be done by cutting down those parts which are too high, and raising the depressed parts. But where the surface of the road is so rotten or brittle that the materials lifted are not fit to be again used, the renewal may be done gradually, and rather by the addition of fresh material to the lowest parts. Unless the materials of which the road has been formed are found to be brittle or rotten, or to be already very thin, the course to be pur- sued is that which is technically termed " lifting the road," which consists in loosening and turning the surface of the road to a depth of about 4 in., and carefully removing such portions of the materials as may be found in an improper state ; such as large stones, which should be broken into pieces of the proper dimensions, and then restored to the road. Where, however, the materials of the road are of such a nature that in lifting they would crumble or fall to powder, a different mode of proceeding must be adopted : the surface of the road must be carefully cleared of mud and dirt, and fresh material, prepared as described in the preceding chapter, should be laid on in a very thin coat, never exceeding at one time 3 inches, and, under ordinary circumstances, not more than 2 inches in thickness. Where the surface of a road, although hard, is found to be very thin, it is necessary, instead of lifting the old materials, to add a fresh coat; and, preparatory to doing this, it is to loosen the surface of the road with a pick, so that ON REPAIRING AND IMPROVING ROADS. 95 the new material may become more rapidly incorporated with the old material. Autumn is the best season of the year for repairing roads, when they are in a wet state, for the depressed and soft parts of the road are then not only the most readily detected, but the surface of the road is softer, and the new materials are more easily worked into it. As was observed, the quantity laid on at one time should never exceed 3 inches in depth ; and, generally speaking, a half, or even a third, of this thickness would suffice, if judiciously em- ployed. It is certain that roads are more frequently spoilt by having too much material put upon them than by having too little. On this subject Mr. Penfold, whose experience in road-making cannot be questioned, remarks: "It is one of the greatest mistakes in road-making that can be com- mitted, to lay on thick coats of materials, and when under- stood, it will no longer be resorted to. If there be substance enough already in the road, which, indeed, should always be carefully kept up, it will never be right to put on more than a stone's thickness at a time. A cubic yard, nicely prepared and broken, as before described, to a rod super- ficial, will be quite enough for a coat, and will be found to last as long as double the quantity put on unprepared and in thick layers. There is no grinding to pieces when so applied ; the angles are preserved, and the material is out of sight and incorporated in a very little time. Each stone becomes fixed directly, and keeps its place : thereby escap- ing the wear and fretting which occur in the other case."* Although autumn is the best time of the year for repair- ing roads, it is not to be assumed that it is the only season for executing repairs. Roads should, so to speak, be always under repair : every road should be divided into lengths, on each of which an intelligent labourer, who * " A Practical Treatise on tlie best Mode of Making and Repairing Roads," p. 15. 96 ON REPAIRING AND IMPROVING ROADS. thoroughly understands his business, should be placed, to attend constantly and at all times to the state of the road, for which he should be held responsible. His duty should consist in keeping the road always scraped clean and free from mud, in filling any ruts or hollow places, the moment they make their appearance, with broken stones. A supply of broken stones should be kept in depots or recesses formed at the sides of the road at intervals of a quarter of a mile. The depots should be capable of containing about 30 cubic yards of material, and should be formed with walls, so that the quantity of material in them can be easily mea- sured. Each man should be provided with a wheelbarrow, a shovel, a pickaxe, and a scraper. As autumn approaches, additional labourers should be engaged; but the constant labourers should alone be responsible for the good order of the road. Not only should the mud formed in wet weather be carefully scraped off from the surface to the sides, and re- moved altogether as soon as it becomes sufficiently solid, but in dry weather the roads should be constantly and regularly watered. After a long season of drought, the surface of a road becomes, as it were, baked : and in this state, being brittle, it is quickly injured and worn to dust by constant traffic. But a regular and moderate supply of water entirely obviates this undue wear, and preserves the road in a proper state. Care should be taken that the water is properly applied, as much injury may be done by the water being discharged in too great quantity, or un- evenly distributed. The manner in which the water should be poured upon the road should resemble, as nearly as possible, a gentle shower of rain. The system of water- ing roads in particular conditions, even in winter, has been practised with advantage, as is shown by the following ex- tract from the evidence of Mr. Benjamin Farey, surveyor of the Whitechapel Eoad, before the Committee of tho ON REPAIRING AND IMPROVING ROADS. 97 House of Commons : " The wheels stick to the materials, in certain states of the road, in spring and autumn, when it is between wet and dry, particularly in heavy foggy weather, and after a frost ; by which sticking of the wheels the Whitechapel Eoad is often, in a short time, dreadfully torn up and loosened ; and it is for remedying this evil that I have, for more than eight years past, occasionally watered the road in winter. As soon as the sticking and tearing-up of the materials is observed to have com- menced, several water-carts are employed upon these parts of the road, to wet the loamy and glutinous matters so much that they will no longer adhere to the tire of the wheels, and to allow the wheels and the feet of the horses to force down and again fasten the gravel-stones ; the traffic, in the course of from four to twenty hours after watering, forms such a sludge on the surface as can be easily raked off by wooden scrapers, which is performed as quickly as possible ; after which, the road is hard and smooth. The advantages of this practice of occasional winter-watering have been great ; and it might, I am of opinion, be adopted with like advantages on the other entrances into London, or wherever else the traffic is great, and the gravel-stones are observed to be torn up by stick- ing to the wheels."* Proceeding with a description of the tools or implements employed in the construction and repair of roads, the most important of these is the level or plummet-rule used for forming the true transverse section of the road. It is shown in Fig. 26, and consists of a horizontal straightedge or bar A c, having in the centre of its length a plummet D, for ascertaining when the straightedge is horizontal. Thus far it exactly resembles an ordinary bricklayer's level. A line is drawn near the end A of the bar, and at every 4 feet from this line a gauge (a, b, c, d,) is fixed in a * Parliamentary Report, 1819, p. 40. F 98 ON REPAIRING AND IMPROVING ROADS. Length of Gauge, 13 ins. Slit lor screw-pin, 7 ius. Length of Gauge, 11 ins. Slit for screw-pin, 4 ins. Length of Gauge, 7 ins. Slit for screw-pin, 2 ins, Middle of Road dovetailed groove, in such a way as to be ca- pable of being moved up or down, so as to adjust the depths of its lower end below the horizontal line of the bottom of the straight edge ; and there are thumb screws (one of which is shown on an enlarged scale in Fig. 27) passing through each gauge, by tight en- A ing which the gauge 3 can be fixed when so P3 I adjusted. When the bottoms of the gauges a, b, c, and d, have been adjusted as shown in Fig. 26, they will co- incide with the surface recommended to be given to a road 30 feet in width, and such as is shown in Fig. 22 ; and, in order to ascertain whether the surface of any existing road is con- structed to the proper inclination and form, it is only requisite to apply the level, which, when placed perfectly horizontal, by means of INSTRUMENTS. 99 the plummet D, should rest upon the road at the lower extremity of each of the gauges, a, b, c> and d. For forming the sides of roads of greater width than 30 feet, it would be convenient to have a level constructed in the manner shown in Fig. 28, in which a straightedge about 15 feet long, has a plummet at the centre of its length, so adjusted that when hanging truly in its place, the lower Inches. 10 9 8 7 6 5 4 321 Fig 1 . 27. Gauge. side of the straightedge should be inclined from a hori- zontal line at the rate of 1 in 30. The pick used for lifting the surface of roads is shown in Fig. 29. The bent iron head (a I) should weigh about fcen pounds, having a large eye in the centre (0), in which is fitted the handle, which should be of ash, rather more than 2 feet in length ; one extremity (0) should be formed Tig. 28. Plummet Rule. Inclination 1 in 30. like the end of a chisel, while the other (3) should terminate in a blunt point. Both ends should be tipped with steel. The most useful form of shovel for road purposes is shown in Fig. 30. The blade should be somewhat pointed, and the handle bent, so as to enable the person using it to bring the blade flat upon the surface of the road without excessive stooping. The ordinary ivheellarrows are of ash or elm, with cast- F2 100 OS KEF AIRING AND IMPROVING KOADS. ,-L Fig. 29. Pick. iron wheels ; but they may be made of wrought iron, which would combine strength and durability with light- ness. Of whatever material they are con- structed, they should not exceed 9 inches in depth, and their sides should be splayed with a slope of 2 to 1. It is also desirable to have hooks placed on their sides to re- ceive a shovel and a pick. The screens, or sieves, employed for sepa- rating coarse gravel from hoggin or small gravel, consist of iron wires or slender rods, placed at equal distances apart, and fixed in a frame of wood, the sides of which are raised about 5 inches above the plane of the wires. In the screens the frames are rectangular, about 5 feet 6 inches in height and 3 feet wide, and the wires are stretched in the direction of its length at distances varying from i inch to 1J inch, according to the size of the stone required ; and these wires are kept in place by others crossing them at intervals of 5 or 6 inches. When used, they are placed so that the plane of the wires is inclined about 30 from the upright, and the gravel to be screened being dashed or thrown forcibly against them, the finer particles pass through and fall on the further side of the screen, while the large stones roll down its surface and fall on the nearest side. The sieves are somewhat different in form : the frame is circular, forming a cylinder about 6 inches in depth and 20 inches in diameter, and the wires placed either as already de- scribed or equally close in both directions, forming a kind of bottom to the cylinder. The sieve is held hori- Fig. 30.- Shovel. INSTRUMENTS. 101 zontally by one man, while the other throws into it a Iron. 21bs. Length, 6| ins. 1 inch diameter by f . Total Length. 1$ foot. Tiff. 31. Hammer. shovelful of gravel ; upon shaking the sieve, the fine hoggin falls through, leaving the stones in the sieve, which are then thrown by the man into anything which may be placed to receive them. This is gene- rally the best and cheapest mode of screening gravel. The hammers generally employed for breaking stones are of two sizes, and are shown in Figs. 31 and 32. The handles should be of straight-grained ash, and the iron heads of the weight and form shown in the drawings ; the faces should be spherical, and case-hardened or steeled. Fig. 33 represents the ring to be used for testing the size of the broken stones. Its internal diameter is 2 inches, and the largest stones should in all positions be passable through the ring. Fig. 34 represents a pronged fork, to be used instead of a shovel for taking up the stones to throw upon the road. The ad- vantages attending its use are, that a man can take up the stones much more quickly and more easily than with a shovel, free from dirt and extraneous matter. Flg< 82 '~~ Hammer - It is advantageous to roll the surface of new roads, in 102 ON REPAIRING AND IMPROVING ROADS. order to consolidate the material ; and for this purpose a cast-iron roller is usually employed, about 5 feet wide, 4 feet in diameter, and weighing about 4 tons. The rakes, which should be employed in filling in ruts and hollow places in the surface of roads, should be formed with prongs between 2 and 3 inches in length, fixed at the distance of three- quarters of an inch apart, into a wooden head about 11 inches in length. The handles should be formed of ash, and should be about Fig. c3. . _ _ Ring-gauge. 6 feet in length. Scrapers are indispensable for preserving roads in a proper state and free from mud. They are usually con- structed of wood shod with wrought iron ; but it is better inches. Fig. 34. Pronged Fork. to make them entirely of iron. They should be 6 inches in depth, and about 1 8 inches in length, and slightly curved at each extremity to prevent the escape of mud at each side. INSTRUMENTS. 103 Scraping machines have been invented, and are very generally employed, by means of which the surface of a road may be scraped much more regularly and quickly than with the old scrapers. They consist of a number of iron scrapers, attached to a frame mounted on wheels, which are so placed that, when the body of the machine is raised somewhat, the wheels are lifted from the ground, and the- whole weight of the machine is thrown upon the scrapers, which, upon the machine being drawn across the road, scrape all the mud from its surface, and carry it to the sides. A machine has also been invented by Mr. [now Sir Joseph] Whitworth, of Manchester, which has been extensively employed, both there and in London, for sweeping up the mud from the roads and carrying it away at once. It consists of a species of endless broom, passing round rollers attached to a mud cart, and so connected by cogged wheels with the wheels of the cart that, when the latter is drawn forwards, the broom is caused to revolve, and sweeps the mud from the surface of the road, up an inclined plane, into the cart. The machine is drawn by one horse ; and, by its aid, the roads are swept much more rapidly and better than by the old system of scraping, and with less injury to the surface of the road, and less annoyance to the passengers. CHAPTEE VII. OK HEDGES AND FENCES. IN moot situations, fences are required to mark the boun- daries of roads, and separate them from the adjoining lands. They should be dispensed with wherever it is possible, for the reason that fences, of whatever kind, deprive, to a greater or less extent, the surface of the road of the benefit of free exposure to the action of wind and sunshine, both of which are essential for maintaining it in a dry state. Pew persons are aware of the extent to which a road may be injured by high hedges, or lines of trees. Trees are worse than hedges, because they not only deprive the road of the action of the air and sun, but they further injure it by the dripping of rain from their leaves, as a consequence of which the road is kept in a wet state long after it would otherwise have become dry. When fences are indispensable, they should be placed as far as may be from the sides of the road, and should be kept as low as possible. When there is a deep ditch on either side of the road, it becomes necessary, to prevent accident, that the fence should be placed between the road and the ditch; but, in other situations, the fence should be placed on the field- side of the ditch. In so doing, the surface-draining of the road into the side ditches is less interfered with, and the action of air and sunshine is less obstructed by the fence. The different descriptions of fence which may be em- HEDGES AND FENCES. 105 ployed are various. In districts where stone is plentiful, and especially in the immediate neighbourhood of Quarries, where stone rubble can be obtained at a trifling cost, dry rubble walls, without any mortar, are very good and cheap, and require little or no repair. For the road itself, an open post-and-rail fence is the best which can be employed, because it scarcely impedes the action of the wind and the sun upon the surface of the road ; but the great practical objection to timber fences is, their liability to decay, which occasions frequent and constant expense for renewal. The most common, and, all things considered, the most useful, fence is the quickset hedge. If properly planted, and carefully attended to for the first few years, a natural fence may thus be obtained, sufficiently strong to resist the efforts of cattle to break through, and very economical in cost for maintenance. A bank or mound of earth, at least two feet in depth, should be prepared for the recep- tion of the quicks, which should be three-years plants which have been transplanted two years. The best kind of soil is one of a light sandy nature, admitting sufficient moisture to nourish the plants, and retaining moisture in dry seasons. Heavy clay soils are not sufficiently pervious to water, and plants placed in such soils are never found to thrive. A mixture of peat or of rotten leaves is of great use, and causes the plants to grow with much vigour. The quicks are most commonly planted in a single row, at distances of about 4 inches apart. But a much better hedge is formed by planting them 6 inches apart, in a double row, as shown in Fig. 35, with a space of 6 inches be* tween the rows, o o o o o c o o o o o Fig 35. Planting Hedges. and so arranged that the plants in one row are opposite F 3 106 HEDGES AND FENCES. the spaces in the other. By this arrangement, although the plants are really not so crowded, and have more space round their roots from which to derive nourishment than in the single row, they form a thicker hedge. The proper time for planting quicks is during the autumn or the spring, and, in fine seasons, the operation may be continued during the whole winter. A temporary fence should be put up to protect the young plants from injury ; and the fence should be retained until the hedge has attained sufficient strength to require its protection no longer : at the end of a period, under favourable circumstances, of three or four years after the quicks are planted. That the plants may thrive, they must be very carefully attended to at first, and it is essen- tial that they should be properly cleaned and weeded at least twice every year. Once every year, towards the end of the summer, the hedge should be judiciously trimmed, not to such an extent as to produce stunted plants, but by merely cutting off the upper and more straggling shoots, so as to bring it to a level and even surface. By proceed- ing in this manner, a neat, strong, and compact hedge of healthy plants will be obtained in about three years after planting. When the hedge or fence is placed between the road and the side ditches, it is essential that small d?ains be formed at least every fifteen yards, to convey the water from the side tables or gutters, through the fence, into the ditches. Professor Mahan insists that fences and hedges should not be higher than 5 feet ; and that no trees should be suffered to stand on the road-side of the side drains, for, independently of shading the roadway, their roots would in time throw up the road-covering. NOTE BY THE EDITOR. Sir John Macneil proved experi- mentally, on the Holyhead Eoad, the unfavourable in- fluence of close and high hedging in interfering with HEDGES A1X 7 D FENCES. 107 what maybe called the ventilation of a road : keeping it moist, and incurring excessive draught. By means of his experiments on tractional resistance, the trustees and the surveyors of the roads "have," he said, "perceived the defective parts of the road ; and within three months after the [Report of the Parliamentary Commissioners became public, there was not a hedge on that part of the road where the draught was shown to be excessive, that was not cut down, and improved on the surface."* Mr. Walker had previously, in 1819, remarked that nothing is more injurious to roads than the permitting of high hedges and plantations near them, f * Report of the Select Committee on Steam Carriages, 1831 ; p. 103. f Report of the Select Committee on the Highways of the Kingdom, 1819. ORAPTEB Yin. PAVED ROADS AND STREETS. roads or streets, through towns or cities where the traffic is considerable, a paved surface is preferable to a macadamised surface. Macadamised roads in such situa- tions are exposed to incessant and heavy traffic, their sur- face is rapidly worn, and it requires constant repair, and much attention in scraping or sweeping, or in raking-in ruts. The cost of maintenance is very heavy, whilst there are frequent interruptions of, or interference with, traffic. In dry weather, the macadamised road is dusty, and in wet weather it is covered with mud. The only advantage which such a road really possesses over a pavement con- sists in the less noise produced by carriages in passing over it. Several different methods have been employed of forming the foundations of pavements, such as concrete and broken stone ; but where it can be done, it is perhaps best to lay the new pavement on the old surface of the road, whether paved or macadamised, taking care, of course, that its sur- face has been brought first to an even state, and of the required form of cross section. The practice of laying the new pavement on the top of the old has been a great deal used in Paris, and has there been found to answer extremely well. It is usual to take up and relay the old pavement, in order that its surface may be even and true ; after which it is covered with gravel, on which the new stones are bedded. Mr. Telford PAVED ROADS AND STREETS. 109 strongly recommended the surface upon which the pave- ment was intended to be laid to be prepared as though intended for a macadamised road, and that it should be used, in that state, by carriages until it had become thoroughly consolidated; when the pavement should be laid on the top of the hard road so formed, the stones being properly bedded in a kind of coarse mortar. Mr. Edgeworth, in his work on roads, states that this method of forming paved roads had been extensively employed in Dublin, and was attended with considerable success. In constructing a paved road or street, the following method should be employed for forming the foundation. The loose ground at the surface should first be entirely removed. The depth to which it may be requisite to do this depends upon the nature of the grjund; and, unless the ground be very solid, it should be removed to such a depth as will allow of 18 inches of concrete beneath the pavement. In some situations so great a thickness as this may not be requisite; but it is better to err rather in forming too strong a foundation, than in forming one the failure of which necessitates the taking up and relaying of the pavement. When the loose ground has been re- moved, a layer of concrete, prepared in the manner already described at page 83, should be evenly spread over the whole area of the intended road. The depth of concrete should never be less than 1 2 inches ; and, under ordinary circumstances, it should not be less than 18 inches. Its upper surface should have the true form of cross section intended to be given to the road. The stone sets should be well bedded upon the concrete in a kind of coarse mortar, which should also be well filled in between their joints. For the sets, several of the harder kinds of stone are used, such as granite, whinstone, and the very hardest varieties of limestone and freestone. Of all materials, granite is the best for sets, and more 1JO PAVED ROADS AND STREETS. particularly those kinds, such as the Guernsey and Aber- deen, which do not wear smooth and acquire, a polished surface. Smooth wear is a great practical objection attend- ing the employment of excessively hard stone. With regard to the form of the stones, experience has shown that the best form is that of rectangular blocks, from 7 to 9 inches in depth, depending on the amount and nature of the traffic, and not more than 3 or 4 inches in width. Until lately, it was considered better to have stones of much greater width, under the supposition that, having a larger base, they would be better able to support the superincumbent weight, but experience has shown that the narrow stones are much the better. The stones should be sorted according to their depths and widths : for, if the stones are of unequal depth, and the surface of the concrete has been made, as it should be, even and parallel to the intended surface of the road, any stones less than the general depth would require more mortar to be placed under them, and would consequently . - ^_^_^ settle down more than the others, I I I ' i-* I and form hollows on the surface of | | I I I the road. They should also be sorted I I I I I according to their width, so that they -" ' ' may run entirely across the street Fig. 36.- Plan of Pavement, ^ paralM courseSj as shown in Eig. 36, and the stones in each course should be so selected as to break joint with those in each adjoining course, as there shown. A firm and substantial curb should be laid on each side of the road, to act as an abutment ; and, in laying the pave- ment, the courses should be commenced at each side, and worked towards the middle ; the joints between the stones of each course should be as thin as possible, and the last stone should fit tightly, so as to form a kind of key to the course. After the stones have been set they should be well rammed down with a heavy punner, and stones which sink PAVED ROADS AND STREETS. Ill below the general level should be taken up, and re-packed underneath. It is not usual to incur the expense of bed- ding the stones in mortar, in the manner here recom- mended. The ordinary practice is to pour a thin grouting of sand and lime over the surface, after the pavement has been laid, which finds its way, although very imperfectly, into the interstices between the stones. But, this is mis- taken economy, for a pavement laid as here described, upon a firm concrete foundation, and having the joints perfectly formed with good mortar, would last almost for ever, since nothing else than the positive wearing away and destruction of the stone would render its renewal necessary. Paved roads should be watched for a few months after having been opened for traffic, in order to prevent irre- gular settlement in the stones, and consequently an uneven and irregular surface. When any portion of the road is found to settle below the general level, it should be taken up, and a sufficient quantity of fine concrete put under- neath it to bring it slightly above the level of the sur- rounding surface. Great attention should be given to the manner in which the pavement is re-laid after having been disturbed for the repair of sewers, water-pipes, or gas- pipes : the excavated ground, when thrown back, should be well punned, or beaten down in layers of not more than a foot in thickness, and at least 18 inches of concrete should be laid on the top, under the pavement. The surface of the fresh concrete should be about an inch above the general level of the concrete, and the stones should be properly bedded in the manner already described, care being taken that the stones correspond in depth and width with those already in place on either side, and, further- more, that the last stone, in making good each course, fits tightly into its place. In laying pavement in streets having a considerable 112 PAVED ROADS AND STREETS. inclination, two methods have been employed to afford a more sure and perfect hold for the horses' feet than the ordinary pavement. The first method is shown in Fig. 37, and consists in laying between the rows of paving-stones a course ^ S l a t6, rather less than an thickness, and about an inch than the stones. By this Fig. 37. r . , , Pavement on inclines, means a series oi small channels or grooves, about an inch in width and depth, are formed Between the rows of stones, which affords sufficient stay for the horses' feet. The other method is somewhat simpler, and consists in merely placing the ordinary paving-stones somewhat canted on their beds, as shown in Pavement on inclines. Fig. 38, so as to form a series of ledges or steps, against which the horses' feet being planted, a secure footing is obtained. Roadways and streets in cities must be accompanied by sidewalks, and crossing-places, for foot-passengers. The sidewalks are made of large flat flag-stones, 3 inches thick, laid on the ordinary ground, or on gravel, well rammed and settled. The proper width of the sidewalks depends on the amount of the traffic. It would, in all cases, be well to have them at least 12 feet wide; they are laid at a slope, or pitch, of 1 inch to 10 feet, towards the pavement, to convey the surface-water to the side channels. The pavement is separated from the sidewalk by a row of long slabs, termed curb-stones, which confine both the flagging and the paving-stones. They are usually about 12 inches by from 6 to 10 inches in section, and they are laid, in some cases, on their side, in other cases on edge. They form the sides of the side channels, and they should for this purpose project a few inches above the outside paving-stones, and be sunk at least 4 inches below their upper surface. The curb-stones are flush with the upper surface of PAVED ROADS AND STREETS. 113 the sidewalks, to allow water to run over into the side channels. The crossings should be from 4 to 6 feet wide, and slightly raised above the general surface of the pavement, to keep them free from mud- CHAPTER IX. ON TAKING OUT QUANTITIES FOR ESTIMATES. THE process of making out an estimate for any description of engineering work may be divided into two distinct parts ; namely, in the first place, calculating the actual quantity of each description of work to be executed ; and, in the second place, affixing to these quantities just and reasonable prices, such as the work might really be executed for. In the construction of roads, the principal item of ex- pense is the earthwork, or the cost of forming the cuttings and embankments to obtain the required levels for the formation service, in excavating the ditches and forming the banks, and in laying on the metalling or ballast to form the road. Of these different descriptions of work, the first, namely, the cuttings and embankments, are the only ones the estimation of the quantity of which is attended with any difficulty. The others, being generally constant, are readily obtained by ascertaining the quantity in a given length, as a yard, and then multiplying that quan- tity by the total length of the road. In the calculations of solid contents required in balancing the excavations and embankments, the most accurate method consists in subdividing the different solids into others of the most simple geometrical forms, as prisms, prismoids, wedges, and pyramids, whose solidities are readily determined by the ordinary rules for the mensu- ration of solids. As this process, however, is frequently long and tedious, other methods requiring less time, but ESTIMATES. 115 not so accurate, are generally preferred, as their results give an approximation sufficiently near the true for most practical purposes. They consist in taking a number of equidistant profiles, and calculating the solid contents be- tween each pair, either by multiplying the half sum of their areas by the distance between them, or else by taking the profile at the middle point between each pair, and multiplying its area by the same length as before. The latter method is the more expeditious ; it gives less than the true solid contents, but a nearer approximation than the former, which gives more than the true solid contents, whatever may be the form of the ground between each pair of cross profiles. In calculating the solid contents, allowance must be made for the difference in bulk between the different kinds of earth when occupying their natural bed and when made into embankment. From some careful experiments on this point made by Mr. Elwood Morris, published in the Journal of the Franklin Institute, it appears that light sandy earth occupies the same space both in excavation and embankment ; clayey earth about one-tenth less in embankment than in its natural bed ; gravelly earth about one-twelfth less ; rock in large fragments about five- twelfths more, and in small fragments about six-tenths more. The calculation may often be simplified by considering the slopes apart from the trunk or main body of the cut- ting. For example, let Fig. 39 be the section of a cutting, A D being the natural surface of the ground ; then if B c is the width of the formation surface, B c F E will be the trunk or central portion of the cutting, and A E B, F c D, will be the side slopes. Now, the width of the first is constant, being the formation width, while its depth varies as the depth of the cutting ; and therefore its cubic content for a given length forward is directly proportional to the depth 116 TAKING OUT QUANTITIES FOR ESTIMATES. of the cutting. If, for instance, the line G H were the natural surface of the ground instead of A D, the cutting being now twice as deep as before, the cubic content of the trunk i B c K would be twice as great as that of E B c F. 03 C Pig. 39. Section of a Cutting. But the cubic content of the slopes increases in the pro- portion of the square of the depth of the cutting, as is evi- dent from the figure, in which i B being twice as great as E B, the volume of the slope G i B is four times as great as A E B, the triangles a, b, c, and d being evidently equal. In consequence of these two quantities varying in a differ- ent proportion, it is convenient to estimate them separately. In order to facilitate the calculation of earthwork, several tables have been published, the most complete and elabo- rate of which are those by Sir John Macneil. The following table, No. 6, will be found very useful in estimating the content of cuttings or of embankments of moderate depth, and will give the result very nearly true, except in cases in which the two ends of the cutting are of very different depths. The first, fourth, and seventh columns contain the depth of the cutting or height of embankment, in feet, for every tenth of afoot, from O'l to 25 -2 feet; the second, fifth, and eighth columns express the content, in cubic yards, of one foot in width, and one chain in length, of a portion of the trunk or central part of a cutting, whose mean depth is shown in the preceding column. The quantities taken ESTIMATES. 117 from these columns must be multiplied by the forma- tion width. The third, sixth, and ninth columns express the content in cubic yards of a length of one chain of both slopes, when the slopes are formed at 1 to 1. With any other ratio the quantities derived from these columns must be increased in the same proportion. Thus, if the slopes are 3 to 1, the quantity obtained from the table must be multiplied by 3. TABLE No. 6. CONTENTS or CUTTINGS OR OP EMBANKMENTS. Content for a Content for a Content for a Length of one chain. Length of one chain length of one chain. i &* g PI .3^ 1 f s-g 1 ^ I| & 3 CD 11 1 || 1 o 49 'bn H -t- 3 IB S tt-t +* K Ho &*% II Is 1 o' 3^ W * & s9 1 W S'S ss Feet. Cub. yds. Cub. yds. Feet. Cub. yds. Cub. yds. Feet. Cub. yds. Cub. yds. o-i 24 02 4-3 10-51 45-20 8-5 20-78 176.6 1 0-2 49 in 4-4 10-76 47-32 8-6 21-02 1808 0-3 73 22 4-5 11-00 49-50 8-7 21-27 185-0 0-4 99 39 4-6 11-24 51-72 8-8 21-50 189-3 0-5 1-22 61 4-7 11-49 54-10 8-9 21-76 193-6 0-6 1-47 88 4-8 11-73 56-42 9-0 22-00 198-0 0-7 1-71 1-20 4-9 11-98 58-79 9-1 22-24 202-4 0-8 1-96 1-56 5-0 12-22 61-11 9-2 22-49 206-9 0-9 2-20 1-98 51 12-47 63-58 9-3 22-73 211-4 1-0 2-44 2-44 5-2 12-71 66-10 9-4 22-98 216-0 1-1 2-69 2-96 5-3 12-96 68-66 9-5 23-22 220-6 1-2 2-93 3-52 5-4 13-20 71-28 9-6 23-47 225'-3 1-3 3-18 4-13 5-5 13-44 73-94 9-7 23-71 230-0 1-4 3-42 4-79 5-6 13-69 76-66 9-8 23-96 234-8 1-5 3-67 5-50 5-7 13-93 79-42 9-9 24-20 239-6 1-6 3-91 6-26 5-8 14-18 82-23 10-0 24-44 244-4 1-7 4-16 7-06 5-9 14-42 85-09 10-1 24-69 249-4 1-8 4-40 7-92 6-0 14-67 88-00 10-2 24-93 254-3 1-9 4-64 8-82 6-1 14-91 90-96 10-3 25-18 259-3 2-0 4-89 9-78 6-2 15-16 93-96 10-4 25-42 264-4 2-1 5-13 10-78 6-3 15-40 97-02 10-5 25-67 269-5 2-2 5-38 11-83 6-4 15-64 100-1 10-6 25-91 274-6 2-3 5-62 12-93 6-5 15-89 103-3 10-7 26-16 279-9 2-4 5-87 14-08 6-6 16-13 106-5 10-8 26-40 285-1 2-5 6-11 15-28 6-7 16-38 109-7 10-9 26-64 290-4 2-6 6-36 16-52 6-8 16-62 113-0 11-0 26-89 295-8 2-7 6-60 17-82 6-9 16-87 116-4 11-1 27-13 301-2 2-8 6-84 19-16 7-0 17-11 1198 11-2 27-38 306-6 2-9 7-09 20-56 7-1 17-36 123-2 11-3 27-62 312-1 3-0 7-33 22-00 7-2 17-60 12G-7 11-4 27-87 317-7 3-1 7-58 23-49 7-3 17-84 130-3 11-5 28-11 323-3 3-2 7-82 25-03 7-4 18-09 733-8 11-6 28-36 328-9 3-3 8-07 26-62 7-5 18-33 137-5 11-7 28-60 334-6 3-4 8-31 28-26 7-6 18 58 141-2 11-8 28-84 340-4 3-5 8-06 29-94 7-7 18-82 144-9 11-9 29-09 346-2 3-6 8-80 31 68 7-8 19-07 148-7 12-0 29-33 352-0 3-7 9-04 33-46 7-9 19-30 152-6 12-1 29-58 357-9 3-8 9-29 35-30 8-0 19-56 156-4 12-2 29-82 363-8 3-9 9-53 37-18 8-1 19 80 160-4 12-3 30-07 369-8 4-0 9-78 39-11 8-2 20-04 164-4 12-4 30-31 375-9 4-1 10-02 41-09 8 3 20-29 168-4 12-5 30-56 381-9 4-2 10-27 43-12 8-4 20-53 172-5 12-6 30-80 388-1 Dontentfor alength of one chain. Content for a length of one chain. Content for alength of one chain. i II g 1 i *a jl a "P, 1 || g o ^-"SfH o Sr^ fcD || 12 1 || fto *2 1 ' i ! fS 1 o'i W o' S-8 1 si o9 Feet. Cub. yds. Cub. yds- Feet. Cub. yds. Cub. yds. Feet. Cub. yds. Cub. yds. 12-7 31-04 394-3 16-9 41-31 698-2 21-1 51-58 1088 12-8 31-29 400-5 17-0 41-56 706-4 21-2 51-82 1099 12-9 31-53 406-8 17-1 41-80 714-8 21-3 52-07 1109 13-0 31-78 413-1 17-2 42-04 723-2 21-4 52-31 1119 13-1 32-02 419-5 17-3 42-29 731-6 21-5 52-56 1130 13-2 32-27 425-9 17-4 42-53 740-1 21-6 52-80 1140 13-3 32-51 432-4 17-5 42-78 748-6 21-7 53-04 1151 13-4 32-76 438-9 17-6 43-02 7572 21-8 53-29 1162 13-5 33-00 445-5 17-7 43-27 765-8 21-9 53-53 1172 13-6 33-24 452-1 17-8 '43-51 774-5 22-0 53-78 1183 13-7 33-49 458-8 17-9 43-76 783-2 22-1 54-02 1194 13-8 33-73 465-5 18-0 45-00 792-0 22-2 54-27 1205 13-9 33-98 472-3 18-1 44-24 800-8 22-3 54-51 1216 14-0 34-22 479-1 18-2 44-49 809-7 22-4 54-76 1227 14-1 34-47 486-0 18-3 44-73 818-6 22-5 55-00 1238 14-2 34-71 492-9 18-4 44-98 827-6 22-6 55-24 1249 14-3 34-96 499-9 18-5 45-22 836-6 22-7 55-49 1260 14-4 35-20 506-9 18-6 45-47 845-7 22-8 55-73 1271 14-5 35-44 513-9 18-7 45-71 854-8 22-9 55-98 1282 14-6 35-69 521-1 18-8 45-96 864-0 23-0 56-22 1293 14-7 35-93 528-2 18-9 46-20 873-2 23-1 56-47 1305 14-8 36-18 535-4 19-0 46-44 882-4 23-2 5671 1316 14-9 36-42 542-7 19-1 46-69 891-8 23-3 56-96 1327 15-0 36-67 550-0 19-2 46-93 901-1 23-4 57-20 1339 15-1 36-91 557-4 19-3 47-18 910-5 23-5 57-44 1350 15-2 37-16 564-8 19-4 47-42 920-0 23-6 57-69 1362 15-3 37-40 572-2 19-5 47-67 929-5 23-7 57-93 1373 15-4 37-64 579-7 196 47-91 9391 23-8 58-18 1385 15-5 37-89 587-3 19-7 48-16 948-7 23-9 58-42 1397 15-6 38-13 594-9 19-8 48-40 958-3 24-0 58-67 1408 15-7 38-38 602-5 19-9 48-64 968-0 24-1 58-91 1420 15-8 38-62 610-2 20-0 48-89 977-8 24-2 59-16 1432 15-9 38-87 618-0 20-1 49-13 987-6 24-3 59-40 1443 16-0 39-11 625-8 20-2 49-38 997-4 24-4 59-64 1455 16-1 39-36 633-6 20-3 49-62 1007 24-5 59-89 1467 16-2 39-60 641-5 204 49-87 1017 246 60-13 1479 16-3 39-84 649-5 20-5 50-11 1027 24-7 60-38 1491 16-4 40-09 657-5 20-6 50-36 1037 24-8 60-62 1503 16-4 40-33 665-5 20-7 50-60 1047 24-9 60-87 1516 16-6 40-58 673-6 20-8 50-84 1058 25-0 61-11 1527 16-7 40-82 681-7 20-9 51-09 1068 25-1 61-36 1540 16-8 41-07 689-9 21-0 51-33 1078 25-2 61-60 i 1552 1 I 120 TAKING OUT QUANTITIES FOR ESTIMATES. As an example of the use of this table, we may estimate the quantities in the cutting and embankment shown on the working section, Fig. 20, page 39. In the following table, the first column contains the number of the peg, the second the depth of cutting or height of embankment, the third the cubic content of the corresponding portion of the trunk, and the fourth column the content of the slopes : CUTTING No. 1. bb ft ^ &b Trunk. Slopes. *o 5.2 1 fi o 1 37 11 2 6 1.47 88 3 9 2-20 1-93 ' 4 1-2 2-93 3-52 5 1-6 3-91 6-26 6 1-9 4-64 8-S2 7 2-0 4-89 9-78 8 1-9 4-64 8-82 9 1-9 4-64 8-82 10 2-1 5-13 10-7S 11 2-2 5-38 U-83 12 2-4 5-87 14-08 13 2-3 5-62 12-93 14 2-3 5-62 12-93 15 2-5 6-11 15-28 16 2-1 5-13 10-78 17 2-3 5-62 12-93 18 2-5 6-11 15-28 19 2-2 5-33 11-83 20 1-6 3-91 6-26 21 8 1-96 1-56 91-53 185-46 40 3661.20 185-46 3846-66 cub. yds. EMBANKMENT No. 1. "3 & o S Trunk. Slopes. <*H O S j3 fcc 5 6 'S 22 6 1-47 88 23 2-2 5-38 11.83 24 3-5 8-56 29-94 25 4-0 9-78 39-11 26 3-8 9-29 35-30 27 2-6 6-36 16-52 28 1-3 3-18 4.13 29 3 73 22 44-75 137-93 40 1790-00 137-93 6896 1996-89 cub. yds. ESTIMATES. 12J It will be remarked that, at peg I, where the cutting has no depth, we have yet inserted quantities in the third and fourth columns. The manner in which these were derived is as follows. Although at the peg itself there is no cutting, at the next peg the depth is 0'6, and therefore the mean depth is 0*3, which, in the table, gives 0*73 and 0-22. As, however, we have here only half a chain (as it is the commencement of the section), we insert half these quantities, or 0'37 and (HI. The sum of all the separate portions of the trunk, being obtained by addition, is then multiplied by 40, the width of the formation, to which the sum of the slopes being added, gives 3846-66 cubic yards as the content of the cutting. In the case of the embank- ment, the slopes being lj to 1, we add one and a half times the sum of the slopes, and thus obtain 1996*89 cubic yards as its content. By reference to the working section, Fig. 20, it will be seen that the quantities there given agree with the above PAET II. EECENT PEACTICE IN THE CONSTEUC- TION OF EOADS AND STEEETS. BY D. K. CLAKK, C.B. CHAPTEE I. MATERIALS EMPLOYED IN THE CONSTRUCTION OF ROADS AND STREETS. For carriage-ways: Stones. The hardest and toughest Mnds of stones are those which are employed in the construction of paved and macadamised carriage-ways. First in geolo- gical order, as well as in the order of usefulness, is granite. Granite is an unstratified or igneous rock, generally found inferior to or associated with the oldest of the unstratified rocks, and sometimes penetrating them in the form of dykes and veins. It is a crystalline compound of three simple minerals felspar, silica or quartz, and mica ; in which the proportion of silica varies from 65 to 80 per cent. Granite derives its name from its coarse granular structure granum, Latin for grain. The silica generally occurs in an amorphous condition, enclosing and cementing together the felspar and the mica in the state of crystals. Some granites are ternary compounds ; others are quaternary, or even quinary compounds, consisting of silica, two varieties of felspar, or two varieties of mica. MATERIALS EMPLOYED. 123 Two felspars are present in some of the granites of Gal- way, Donegal, and Aberdeenshire. Two micas occur in the Wicklow granite, imparting black and grey hues. Instead of mica, another substance, hornblende, is found in some granites. Such granites, of which the component parts are quartz, felspar, and hornblende, are distinguished as syenite , or syenitic granites ; so called, because it was first found in the island of Syene, in Egypt. Of the constituents of granite, quartz, glassy in appear- ance, is a compound of the metallic base silicium and oxygen ; felspar, opaque, is of a yellowish or a pink colour, composed of silicious and aluminous matter, with a small proportion of lime and potash ; mica consists principally of clay and flint, with a little magnesia and oxide of iron ; hornblende is a dark green crystalline substance, composed of flint, alumina, magnesia, with a considerable propor- tion of black oxide of iron. The granite districts of the United Kingdom, whence most of the material for road-making is obtained, are in Devon, Cornwall, the Channel Islands of Guernsey and Herrn ; Mount Sorrel, in Leicestershire ; Clee Hills, in Shropshire ; Port-Nant and Portmadoc, in North Wales ; Aberdeen, in Scotland ; and Newry, in Ireland. The syenitic granites of Guernsey, Herm, Mount Sorrel, and North Wales, are celebrated for their hardness and dura- bility. The granite of Devon and Cornwall is of several varieties, and it has a prevailing greyish hue. Aberdeen granite is of a bluish - grey tint ; so also is the Newry granite. Mount Sorrel granite is of a rich pink colour. Granites vary much in hardness and resistance to crushing force. In some places, by the decomposition of the felspar or the mica, they are sufficiently soft to admit of their being dug out with a spade. The specific gravity of granite varies from 2*60 to 3*00, and the volume of 1 ton is from 12 to 14 cubic feet. Conversely, the weight G 2 124 MATERIALS USED IN RECENT CONSTRUCTION. of a cubic yard of solid granite varies from 1'93 to 2'25 tons, averaging, say, 2 tons ; whilst the average weight of a solid cubic foot is 1 J cwt. Granite is capable of absorb- ing, on an average, a gallon, or 10 Ibs., of water per cubic yard, or 1 -450th of its weight of water. Mr. Mallet found that the maximum resistance of 1-inch cubes of various granites to crushing force, ranged from 1 to 6 tons. The annexed table, No. 7, gives the results of experiments to determine the resistance of granites : * TABLE No. 7. CRUSHING RESISTANCE OP GRANITES. Surface ex- Pressure per square inch. Locality of Stone. posed to pressure. To fracture. To crush. Herm Square in. 16 Square in. 4.77 Square in. 6-64 Aberdeen (Blue) l 4-13 4-64 Heytor . 16 3 94 6 19 Dartmoor 16 3-52 5-48 Peterhead (Red) 18 2-88 4-88 Peterhead (Bluish grey) 18 2-86 436 Penrhyn 16 2-58 3-45 Killiney (Grey felspathic) 9 4-81 Ballyknocken (Coarse grey) Ballybeg, Carlow (Grey fels 9 1-43 pathic) 9 3-17 Some results of experiments made by Sir William Fairbairnf on the crushing strength of building stones are added, in Table No. 8. The specimens were, for the most part, 2-inch cubes : * The data for this table are derived from Sir John Bnrgoyne's " Rudimentary Treatise on Blasting and Quarrying," 1862, pai?e 94; except for the last three stones, for which the data are borrowed from Professor Hull. f " Memoirs of the Literary and Philosophical Society of Man- chester," 1857 ; vol. xiv. page 31. STONES. 125 TABLE No. 8. CRUSHING STRENGTH AND ABSORBENT POWER OF VARIOUS STONES. (Deduced from Sir W. Fairbairn's data.) Weight Crushing of water Stone. Locality of Quarry. Specific gravity. Cubic feet in one ton. resist- ance per square absorbed after 48 hours' inch. immer- sion. Cub. ft. Tons. 1 part in Porphyry France . . 18-04 Greywacke . Penmaenmaur 2-75 13-04 7-54 1641 Do. Ingleton . . 2-79 1287 1963 Granite . . Aberdeen . . 5-16 Do. Mount Sorrel . 2-67 13-45 5-74 490 Do. JBonaw, Inverary 4-87 Sandstone Yorkshire . . . 241 14-85 4-38 47-5 Do. Runcorn . . 98 Limestone . 3-80 Do.Magnesian Auston, Worksop 2-26 Brick, hard . 84 Do. red . 36 The Trap Rocks are a large group of igneous rocks allied to granite, composed of felspar, augite, and hornblende. The various proportions and states of aggregation of these simple minerals, and their differences in external forms, give rise to varieties known as Basalt, Dolorite, Green- stone, Whinstone, Greywacke, and others. The term trap is derived from trappa, a Swedish word for stair, because the rocks of this class sometimes occur in large tabular masses, rising one above another like steps. Basalt is one of the most common varieties ; it is a dark green or black stone, composed of augite and felspar, very compact in texture, of considerable hardness. It often contains iron ; whence the name basalt, an Ethiopian word for iron. Dolorite is likewise composed of augite and felspar. Green- stone is composed of hornblende, which is dark green, and felspar. Greywacke is a very hard rock, as heavy as granite, and much harder. The rock is very often of a grey colour ; whence the name. Wacke, is a provincial miner's term in Germany. 126 MATERIALS USED IN RECENT CONSTRUCTION. The principal uses of the trap rocks are for paving and macadamising. The most important quarries are those of Barden Hill, in Leicestershire, and of Penmaenmaur, in North Wales. The greywacke of Penmaenmaur is manu- factured into squared sets, which are sent by ship and by rail in enormous quantities to Manchester, Liverpool, and other large towns in the north of England and in Scot- land, and to Ireland. Comparative Wear of Stones. Mr. Walker tested the com- parative durability of various granites and one whinstone. In his first experiments, he submitted square specimens of Guernsey, Aberdeen, and Peterhead granites, having equal weights and rubbing surfaces, to frictional motion on a large block of stone, with sand and water between them. The Guernsey granite showed the smallest amount of wear. He made further tests for the durability of granites and whinstone, under working conditions, by laying down, in 1830, two parallel lines of rectangular tram-stones, 18 inches wide and 12 inches deep, in the gateway of the Limehouse turnpike, so as to be exposed to all the heavy traffic from the East and West India Docks. The experi- mental stones were all new. After an exposure of 17 months to the wear and tear of the traffic, the stones were taken up, and the loss by wear was found to be in order as follows : TABLE No. 9. RELATIVE WEAR OF GRANITES, ETC. (MR. WALKER.) Name of Stone. Super- ficial area. Loss of weight per sq. foot. Vertical wear. Sq. foot. pounds. inch. relative. Guernsey granite Herm granite . 473 5-25 95 I'Oo 060 075 1-000 1-190 Budle whinstone 6-34 1-22 082 1-316 Peterhead blue granite 3-48 1-80 131 2-080 Heytor granite Aberdeen red granite . 4-31 5-38 1-92 2-14 141 159 2-238 2- 24 Dartmoor granite 4-50 2-78 207 3-285 Aberdeen blue granite 4-82 3-06 225 3-571 STONES. 127 From the table, it is apparent that Guernsey granite is many times tougher and more durable than Aberdeen granite. Mr. Walker made a still further test of the relative wear of these two stones on old Blackfriars Bridge, which was paved, in 1840, with these granites, in sets 3 inches wide by 9 inches deep, on a bed of concrete 12 inches thick. When the pavement had been down 13 years, it was found that the Aberdeen stone was worn down to the extent of 1 inch of vertical wear, whilst the Guernsey stone had only worn down J inch showing that the Guernsey had worn only one-sixth as much as the Aberdeen granite.* The durability of the Welsh stones the greywacke of Penmaenmaur, and the syenites of Portmadoc, &c., is almost beyond estimation. After 20 years of work in the streets of Manchester, Penmaenmaur stones, the resistance of which to crushing is considerably greater than that of granite, have suffered little more wear than a rounding of the upper surface. But the blue variety is extremely slippery, and it makes a dangerous pavement. The grey variety, the Irish, the syenitic, and the blue Aberdeen granites retain rough -wearing surfaces; but, in con- sequence of their roughness, they are not so durable as the Guernsey and the Penmaenmaur stones. The rough- wearing sets are excellent stones, and they are specially suitable for the paving of streets on steep inclines. For macadam, the hardest stones as Guernsey granite, and Penmaenmaur greywacke are the most suitable. Their slippery qualities are of no moment in macadam, whilst their hardness and toughness are valuable qualities. In India, the stones used for macadam are granite, trap, and the hard limestones and sandstones. Laterite, which is a hard sandstone, is very much used on the Madras roads but it is comparatively soft, and does not bear much traffic. * Proceedings of the Institution of Civil Engineers, 1853-54; vol. xiii. page 237. 128 MATERIALS USED IN RECENT CONSTRUCTION. Kunkur is the material chiefly used in Hindostan : it is a peculiar formation of oolitic limestone, found generally in nodules, sometimes in masses a little below the surface of the earth. It makes an excellent road ; but it requires constant repair if the traffic is heavy. For Footpaths. The flagstones usually employed for foot- paths are of sandstone. The Old Eed Sandstone of Scot- land produces the well-known sandstones and flags of Arbroath and Dundee, which are largely used in Glasgow, Edinburgh, London, and other large towns. The Arbroath pavement is of a light greenish, grey colour, and of dense structure; it resists the weather successfully. The flag- stones of Caithness, Cromarty, and Nairn are also used for paving. The carboniferous sandstones are largely used for paving in the north of England and in Scotland ; they are generally hard and durable, of yellowish or greyish tints, and of various degrees of coarseness. They are derived from the millstone-grit and the coal measures. Excellent flagstones, which are sent to all parts of Eng- land, are obtained from the lower part of the coal for- mation in Cheshire, near Macclesfield; Lancashire, near Wigan, Burnley, and St. Helen's ; and from Elland, in Yorkshire. The sandstones of the coal measures for in- stance, those of Heddon and Kenton, near Newcastle are seldom durable, and they generally become iron-stained on exposure, owing to their containing considerable quan- tities of iron and alumina.* The following particulars of the composition of sand- stones are derived from the Commissioners' Report of 1839: * See "Quarries and Building Stones," "by Professor Hull; " British Manufacturing Industries." STONES. 129 TABLE No. 10. COMPOSITION, SPECIFIC GRAVITY, AND STRENGTH OF SANDSTONES. Craigleith. Barley Dale. Heddon. Kenton. Mansfield. i Composition. Silica per cent. 98-3 per cent. 96-4 per cent. 95-1 per cent. 93- 1 per cent. 49-4 Carbonate of lime . 1-1 0-3 0-8 2-0 26-6 Carbonate of mag- nesia o-o o-o o-o o-o 16-1 Iron, alumina 0-6 1-3 2-3 4-4 3-2 Water and loss o-o 1-9 1-8 0-5 4-8 ', 100-0 100-0 100-0 100-0 100-0 I SPECIFIC GRAVITIES. Of dry masses 2-232 2-628 2-229 2-247 2-338 Of particles 2-646 2-993 2-643 2-625 2-756 COHESIVE POWERS. In tons per square inch 3-52 3-17 1-78 2-22 2-28 A few particulars of sandstones are given in Table No. 8, page 125. All sandstones are porous, and they absorb water. The Yorkshire sandstones are capable of absorbing 1 part by weight of water in 48 parts of stone. Sandstone flags are in this respect objectionable, for they retain wet impurities, and remain damp some time after rain. An instructive test of the comparative durability of various kinds of flagstones was conducted in Liverpool, by Mr. Newlands, in 1856. The stones were found to range, in the order of durability, as follows Kilrush (Irish), most durable, Caithness, Knowsley, Port Eheuddin, Llangollen, Yorkshire, least durable. G3 130 MATERIALS USED IN RECENT CONSTRUCTION. Of these, Port Bheuddin stone, one of the softest, had the best appearance. Next to this was the Caithness stone. The Kilrush flags had a rough or wrinkled surface ; but in the course of time they wore smooth, and their appearance was good. The Llangollen- slate flag looked well, but it was not durable. Caithness flags were extensively substituted for Yorkshire paving ; Welsh and Irish flags were also laid to a considerable extent. In substituting granite curbstones for sandstone curbs, Mr. Newlands formed the wearing surface of the new curbs to the section produced by abrasion upon the softer stone : thus presenting a larger wearing surface to the strokes of heavy wheels, and minimising the superficial lateral wear. Asphalte. Asphalte, the mineral, is a bituminous lime- stone, or pure carbonate of lime naturally impregnated with bitumen. At the same time, asphalte is, in a scientific sense, synonymous with bitumen. Bituminous limestone, or asphalte, is a rock consisting of from 90 to 94 per cent, of carbonate of lime, and from 6 to 10 per cent, of bitumen. The rock is of a liver-brown, or deep chocolate colour ; it takes an irregular fracture, with definite cleavage, and the texture and the grain vary with the layers. The specific gravity is from 2*20 to 2*30. Exposed to the atmosphere, asphalte gradually assumes a light grey tint, caused by the evaporation of bitumen from the surface. The asphalte used for the construction of carriage-way pavements, is brought chiefly from the Val de Tr avers, a few miles from the town of Neuchatel, in Switzerland. Limmer asphalte is brought from a mine situated at Limmer, near the city of Hanover, and from another mine at Yorwohle, near Alfeld, in Brunswick. The Seyssel asphalte is obtained from mines at Pyrimont and Garde Bois, at Seyssel, in France. ASPHALTE. WOOD. 131 Asphalte, as excavated, must be pulverised before being used, by decrepitation at a temperature of from 212 to 300 Fahr., or by mechanical means. It is applied in two modes for the formation of pavements : 1st, as compressed asphalte, which is obtained by heating the powder to from 212 to 250 Fahr., and causing the particles to cohere by the application of pressure. 2nd, as liquid asphalte, or asphaltic mastic, for which a manufacturing process is neces- sary. The powder is heated with from 5 to 8 per cent, of free bitumen ; the bitumen acts as a flux, the mixture melts, and is run into moulds for use, forming cheeses 12 inches in diameter and 4 inches thick. Like rock asphalte, mastic, heated alone, does not melt ; it only be- comes soft. To be re-melted, a fresh quantity of bitumen must be added. For paving, a large proportion of sand or small gravel is mixed with it, even to the extent of 60 per cent. No chemical union takes place between the mastic and the sand, but the cohesion is so complete that the fracture of sanded mastic shows the simultaneous fracture of the grit ; and the power of resistance to the heat of the sun is increased by the admixture. The Yal de Travers and the Seysell rocks are the only two natural asphaltes used in laying compressed-asphalfce roadways. The former has not quite so close a grain as the latter, and it is softer, and is more regularly impreg- nated with bitumen. Artificial Asphalte. By mixing heated limestone and gas-tar, a material possessing some of the properties of asphaltic mastic has been obtained. But gas-tar is a very inferior form of bitumen for paving purposes : it passes from the dry to the liquid state, and conversely, according to the season, and it is brittle. Wood. Both soft and hard woods are employed for paving : closed-grained wood of the pine family, as Cana- dian and Baltic yellow pines ; and hard woods, as oak, 132 MATERIALS USED IN RECENT CONSTRUCTION. elm, ash, and beech.. Spruce and white deals are not very suitable for paving purposes. The following Table,* No. 11, shows the average crushing resistance of cubes of timber, of from 1 inch to 4 inches, lineal dimension, in the direction of the fibre : TABLE No. 11. CRUSHING RESISTANCE OF TIMBER. Cubes, 1 inch ; 2 inch ; 3 inch ; and 4 inch. Name of Timber. Crushing resistance per sq. inch. tons. English oak, unseasoned 2-19 Do. seasoned 3-34 Dantzic oak . 3-34 American white oak 271 Moulmein teak 2-56 Iron wood 5-21 Greenheart . 6-44 Honduras mahogany 2-85 English elm . 2.58 Canadian elm 2-45 Dantzic fir . 3-10 Biga fir 2-34 Spruce fir 217 Larch 2-60 Red pine 2-54 Yellow pine . 1-88 Pitch pine 2-88 Mr. D. T. Hope,f i n 1844, recorded the results of experi- ments made by him on the comparative wear and endur- ance of wood-paving, laid in the same line of carriage-way, resting on the same concrete foundation, and exposed to the same kind and amount of traffic. The blocks of wood were constructed with the fibre placed vertically, placed horizontally, and also placed at various inclinations to the * Derived from " A Manual of Rules, Tables, and Data," by D. K. Clark, 1877, page 541. t "Transactions of the Royal Scottish Society of Arts," 1844, page 231. WOOD. 133 vertical. Continuous observations were made for a period of 18 months. It was found that the wear or reduction of height of the wood blocks was greater during the first three months than during the following quarters. The more rapid wear at the commencement of the trial was ascribed by Mr. Hope to initial compression of the fibres : the blocks with vertical fibres presented a more compact surface than when they were originally laid, and the surface was so impregnated with sand that it had more of the appearance of sand than of wood. The following shows the total amount of wear at the end of 18 months : Wear at the end of 18 months, "Wood, vertical fibre . . '125 inch, or | inch fibre inclined at 75 147 , 4-0 bare 60 -;" 182 , , ^g- > 91 45 241 > 4 , 30 . 312 , A , 15 379 , , f , full , horizontal 480 , i , bare It is clear that the resistance of the wood was most effective when the fibres were in a vertical position, and least effective when they were horizontal. The horizontal fibres presented, when worn, the appearance of heaps of broken strings. It is evident, further, that the more nearly the fibres were presented in an upright position, the less was the wear, and, of coarse, the greater was the durability. CHAPTEE II. CONSTRUCTION OF MODERN MACADAM ROADS. IN addition to the detailed account of the construction of roads, given by Mr. Law, in the first part of this work, a few words may be added as to the usual modes of construc- tion now adopted. The levels, gauges, plummet-rules, and prong-shovels, formerly employed in the setting out and construction of roads, described at pages 84 to 88, are not now used. "Boning rods" are used for fixing the inclination of the surface, longitudinally and trans- versely, by the eye. First-class Metropolitan Roads. When the ground has been excavated and levelled, it should be rolled, when it consists of clay. A bottoming, or bed, 12 inches thick, of "hard- core," is laid on the ground; it may consist of brick- rubbish, clinker, old broken concrete, broken stone or shivers, or any other hard material, in pieces. The bed should be rolled down to a thickness of 9 inches, and any loose or hollow places made up to the level. Next comes a layer of Thames ballast, 5 inches thick, rolled solidly to a thickness of 3 inches. The ballast serves to fill up the vacancies in the bottoming, and, being less costly, saves so much of the cost for broken granite. Broken granite, or macadam, is laid upon the prepared surface of the ballast, in two successive layers, 3 inches thick, rolled successively, to a combined thickness oi 4 inches ; a layer of sharp sand, J inch or J inch thick should be scattered over the second layer, and rolled into METROPOLITAN ROADS. 135 it with plenty of water. But it is better to add the sand and the water as the second layer of granite is laid, and to roll them well together. Broken Guernsey granite, which is hard and enduring, is the best material for the coating ; Aberdeen granite is softer, and is not so good for the purpose. Flints are, in some instances, laid instead of granite ; but they are worst of all for the coat, for they are brittle, and are soon reduced, by the traffic, to small fragments. In transverse section, the contour of the road is a segment of a circle. Indeed, all roads and streets are, or should be, circular in section. The general practice is, in this respect, at variance with the practice recommended by Mr. Law, page 75, where he recommends that the section of the surface should be formed of two inclined straight lines, joined by a flat curve at the middle, form- ing a species of ridgo. The advantage of the circular section consists in the fulness given to the " shoulders" of the road, which lie in the lines of the traffic on each side of the centre line of the road, and thus present a full surface for wear. Second-class Metropolitan Roads are usually constructed of a hard core of brick rubbish, or other material, about 9 inches thick, and a layer of broken granite, or of flints, about four inches thick. The material is not, in general, submitted to rolling, except by the action of the traffic. Country Roads. The regular width of country roads is 18 feet; but some of them are only 10 feet wide, and others may be more than 18 feet. The bottom is excavated with a flat floor, not being rounded to the contour of the surface. The bottom layer, or hard core, consists of any hard and dry rubbish: broken bricks, lumps of chalk, or hard flints. The material most usually employed con- sists of flint-stones, which are got out of the fields. The largest stones, which may be as much as 6 or 8 inches 136 MODERN MACADAM ROADS. across, are laid at the bottom. Chalk flints answer best for the bottoming. They are rough, and they bind firmly together, and are open, affording free space for drainage. If the bottom is of solid material, as chalk or rock-stuff, the bed is laid in to a thickness of 12 inches. If it be soft, as of clay or of sand, it is made up to a thickness of 18 inches. The finishing layer, or coating, consists of flints of a smaller size ; they are broken up, if necessary, to the usual size of macadam, and are laid 6 inches thick. No binding of any kind is employed, nor is it the practice to roll the roads. Jig. 40. Country Road, with open ditches. The rise of a road, at the middle, is 3 inches for a width of 18 feet. The rule for the rise is to allow 1 inch for every 6 feet of width: giving a slope of 1 in 36, which is sufficient for running off rain-water, and leaving the roads dry. For roads 10 feet wide, a greater degree of rise is given, where ditches cannot be got in. Openings are made through the fences at each side to let off surface- water into the adjoining fields. Mr. J. GK Crosbie Dawson states that he has made many gravel roads, from 12 to COUNTRY ROADS. 137 18 inches thick, at a cost of from 2s. 6d. to 3s. 6d. per square yard. The general conformation of a country road, with hedges, and with an open ditch at one or both sides, for collecting surface and drainage water, is shown by Fig. 40. Side ditches are a frequent cause of overturns and other accidents, The improved country road, proposed by the Committee of the Society of Arts * is represented by Fig. 41, Fig. 41. Proposed Country Road, with wheel- tracks and subsoil drains. in which the side drains are suppressed, and replaced by covered drains, which act also as subsoil land-drains. Wheel-tracks of asphalte, horse-tracks of concrete, and a foot-path of asphalte are also proposed. * Report of the Committee of the Society of Arts on Traction on Xoads, 1875. CHAPTEE m. MACADAMIZED ROADS. WEAB, IT was maintained by Macadam (see page 90) that for ease of draft, a yielding, elastic road was preferable to a road constructed according to the prescription of Mr. Telford, on a rigid or paved foundation. The dis- tinction rests entirely upon the precise signification attached to the word "elastic." If the elasticity be perfect, like that of an indiarubber wheel -tire, where the reaction or uplifting force is exactly equal to the resistance to depression under a load, it may be logically maintained that the elastic road is better for draft. But this condition is impossible in practice ; and Mr. Provis exactly defined an elastic road as " that which will give way by pressure, and rise again to its original position when the pressure is removed" tacitly assuming that time is consumed in the reaction, after the pressure is removed. Moreover, as Sir John Macneil justly observed, " where there is a yielding and elasticity in the road, there must be motion among the particles with which the road is formed, and this motion produces wear." * And, again: "If the road be weak or elastic, and bend or yield under the pressure of the wheels the particles of which it is composed will move and rub against each other, or perhaps break by the action of heavy wheels over them." f Herein lies the gist of the * "Roads," by Sir Henry Parnell, page 412. f Report of the Sele Committee on Steam Carriages, 1831, page 97. WEAR. 139 argument against the admission of any perceptible range of elastic action ; and Sir John Macneil touched the essen- tial element of legitimate elastic action, when he located it " on the surface and at the point of contact." This is pre- cisely the mode of elastic action of a wood-pavement : a material which, with its vertical fibre, acts as a coat of velvet might be supposed to do, simply to absorb and neutralise minute vibration ; and it is the only kind of action that is admissible in the work of a road. Sir John Macneil adduced the results of experiments on London streets, in which he found that, when broken granite was laid down, there was little crushing on the surface, but a good deal below. He found "that a great portion of the wear took place near the bottom; the stones there got round after a very short time. They were all jumbled together ; the lower part of the stratum was on clay. Before it became a solid mass at all, the wheels worked through, and the stones were kept in motion, and rubbing against each other, from top to bottom." Then he tried the experiment of laying " a portion on a very solid foun- dation, and the same quantity was put over it. I took up a portion of both roads," he adds, " and we found, where the road was about 15 or 16 inches thick [with a 9-inch pavement], and the stones 6 inches thick over the broken pavement, they were quite square, and as perfect as when put on. In the other case [simple macadam], it was not BO." " There are two things to be considered, when a road is newly made : there is very great wear indeed, in the first instance, if there be not a pitched foundation ; that is different from the wear that takes place when it is nearly consolidated. To bring it into a solid state, the wear is great. If there are 4 inches of broken stone on the top of a pitched foundation, you may get that road into a perfectly hard and consolidated state, by the ordinary travelling over a turnpike road, injjjuoi^ t^ree months; 140 MACADAM ROADS. but if you put on the same thickness of stone, without a foundation, you will not get it into a perfect state in three or four times as long." * The bearing of Sir John Macneil's exhaustive evidence is very direct upon the comparative wear of "weak or elastic roads, " and roads with a rigid pavement-bottom. He adduces an instance in which an experiment was accu- rately made to test the relative wear of roads so con- structed : "The wear was found to be 4 inches of hard stone, when it was placed on a wet clay bottom ; while it was not more than half an inch on a solid dry foundation, or with a pavement bottom, on a part of the same road, where it was subject to the same traffic." f Estimates have been formed of the relative tear and wear of roads due to the action of the horse-shoes and the action of wheels. Mr. John Farey stated that a cart- horse, walking at a speed of 2^ miles per hour, could draw with a force of traction of 100 Ibs., on an average, for 8 hours a day ; but that a stage-coach horse, running at a speed of 10 miles an hour, could not exert more than 28 Ibs. force of traction, on an average, for ! hours a day. The comparative efficiency of the animal is readily worked out. The work done per hour, on duty, and per day, are as follows : Work done in pounds through one mile. Speed in Traction per hour. per day. miles per hour. ' mile-pounds, mile-pounds. Cart-horse . 2J 100 Ibs. 250 2000 Stagecoach-horse 10 28 280 420 The comparison clearly indicates that, whilst the work done by a horse per hour at work, is a little greater at 10 miles per hour than at a fourth of the speed, the day's work, on the contrary, is very much less, little more than * " Roads," by Sir Henry Parnell, pages 412, 413. t Report of the Select Committee on Steam Carriages, 1831, page 97. WEAU. 141 a fifth. It is probable that the injury done to the road is in the same proportion as the day's work done. " In heavy wagons," says Mr. Farey, " drawn slowly by horses, the horses do far more injury, by digging and scraping with their feet, than is done by the horses in coaches and vans travelling quickly ; because the wagon-horses, having a heavy pull to make, must choose places in the road where they can place their feet in depressions in order to get hold; hence, on a smooth, good road, they slip, and scrape up the surface." Again : " The horses, by treading with their feet, excavate and scrape out depressions in the surface of the road ; that is particularly the case before the road-materials are consolidated into a solid mass. In this manner, the horses, after injuring the road them- selves, prepare the way to further injury to the road by the wheels of carriages." * With respect to the comparative action on a road of the feet of horses and the wheels of vehicles, Mr. Telford con- sidered that the tearing up of a well-made road by horses 7 feet was much more injurious to the road than the pressure of the wheels. Sir John Macneil stated that a four-horse stage-coach weighed from 15 J cwt. to 18 cwt., and frequently carried upwards of 2 tons of passengers and goods ; and the total average weight amounted to from 2J to 2^- tons gross. The tires of the wheels were, in general, 2 inches wide ; in some instances rather less than 2 inches. Some tires were rounded, forming in cross section a segment of a circle 1J inch in diameter. The action of these rounded tires was found to be extremely injurious to the road ; coach- men complained that carriages fitted with such rounded wheel-tires "ran wild" in descending hills in summer; and that they ran " heavy" in winter, when the roads were * Report of Select Committee on Steam Carriages, 1831, page 57. 142 MACADAM ROADS. soft and muddy. Mail-coaches weighed very nearly 20 ewt., and frequently carried a ton of letters and parcels, in addi- tion to their complement of passengers. The tires of their wheels were 2| inches wide. Again, the four-horse vans, which travelled at a speed of 6 miles per hour, weighed, on an average, 4 tons gross, with 2J-inch tires. There were four descriptions of wagon in general use 8-horse, 6-horse, 4-horse, and farm-wagons of 2 or 4 horses. Sir John Mac- neil maintained that, on all gravel roads, however made, without foundation or bottoming, the load on the road for a wheel 4 inches wide at the tire should not exceed 15 cwt. ; and for wheels of less width, 10 cwt. He gave a table showing the weight actually carried per inch of width of tire- bearing, and what he conceived might be the width of tires if they were made cylindrical, with an even bearing, allow- ing 1 inch of total width of four wheels for each ton of gross weight ; or 1 inch of width in each tire per 5 cwt. of gross weight. TABLE No. 12. WEIGHT OF VEHICLES AND WIDTH OP TIRES ON COMMON ROADS. (Sm JOHN MACNEIL.) Description of Vehicle. Speed in miles per hour. Average gross weight. Width of wheels and tires. Pres- sure of each wheel. Pres- sure per inch of width. Width of wheel, ! allowing 5 cwt. per inch of tire. Mail coach miles. 9 to 11 tons. 2 inches. 2i cwts. 10 cwts. 4-40 inches. 2 Stage coach 8 to 11 2i 2 12-5 6-25 2J Van 6 to 7 ** 2* 21-25 829 4 i Wagon . 2jto 3 6 9 25 277 6 Do. 2 to 3 4 i 6 22-5 3 75 4J Do. 2i to 3 &i 4 17-5 4-37 3J From the table it appears that, in the four-horse van, the greatest load per inch of tire upon the road amounted to 8-3 cwt. i and though Sir John Macneil found that good WEAR. 143 hard roads may support concentrated loads of 10 cwt. per inch of width of tire, he considered that, for " the generality of roads," the concentrated load per inch of width of tire ought not to exceed a maximum of 5 cwt. He judged that the wear of roads caused by horses' feet increased with the speed of transit ; and adduced in support of this conclusion the facts that, whilst a coach-horse, going at 10 miles per hour, ran 270 miles per month, and wore away 4 pounds of iron in shoes ; a wagon-horse going at 3 miles per hour, and walking 26 miles per day, or 416 miles per month, wore away 4' 8 pounds of iron in shoes. From these data, it appears that the coach-horse ran 67*5 miles for each pound of iron worn off, and that the wagon- horse walked 86*5 miles for each pound of iron worn off. These data, though they show a greater rate of wear at the higher speed, scarcely suffice to confirm the conclusion drawn by Sir John Macneil ; for the wear of shoes is due entirely to abrasion, whilst that of the surface of the road is due to abrasion combined with excavation. According to another of his estimates, it appears that, in running 63,000 miles, 1,000 pounds of shoe-iron were worn away; showing that 63 miles were run by a horse for each pound of shoe-iron worn off. It also appeared that for an equal mileage of mail-coaches and stage-coaches, of which the hind wheels were 4 feet 8 inches high, and the fore wheels 3 feet high, the tires were worn down from the original section, 2 inches wide by J inch thick to a thickness of i inch, losing 327 pounds; showing that 193 miles were run by a coach for each pound of tire-iron worn off, or three times the mileage run per pound of shoe-iron. From the evidence of the comparative wear of shoe-iron and tire-iron, and other considerations, Sir John Macneil framed an estimate of the relative proportions in which a road was deteriorated by the action of atmospheric changes, 144 MACADAM KOADS. wheels, and horses' feet. For " the generality of roads" the proportions were as follows : Cause of Wear or Injury. For Fast Coaches. For Wagons, per cent. per cent. Atmospheric changes 20 20 Wheels .... 20 35*5 Horses' feet that draw the vehicles 60 44-5 100 100 Mr. James Macadam, in 1831, considered that a vehicle of any description, required to carry a great load, say, 5 or 8 tons, ought to have wheels with tires at least 4% inches wide, flat or cylindrical. He preferred that width to any other width, being of opinion that it was the maximum width of tire that could at the same instant touch the sur- face of a well-formed road.* For a gross weight of 8 tons, or 2 tons per wheel, it is seen that the weight amounted to 9 cwt. per inch of width considerably in excess of the working limit assigned by Sir John Macneil. M. Dupuit tested the width of surface in active contact with tires of various widths, on roads slightly wetted; and he found that the marks of contact with the road traced on the middle of the tires were practically of equal width for various widths of tire, amounting to about 3f inches on a tire 6| inches wide. The pressure appeared not to have been uniformly distributed over the width of surface in contact ; and near the edges of the tires the pressure was nearly nothing. On the tires of a stage-coach, 5 inches wide, the width of contact was limited to 2^ inches. M. Dupuit ascribes the limitation of width of contact to the natural rounding of the tires by wear and stress, towards the edges, by which the circumference of the wooden fellies themselves are likewise rounded ; and he condemns, as illusory, the use of wide tires for the better distributing of excessive loads. He maintains that wide tires are only * Report of the Select Committee on Steam Carriages, 1831, page 91. WEAR. 145 useful when they take a bearing for the whole width, which can only happen when the road is soft, loose, or in bad condition ; or when the ruts are filled up and require to be rolled. As a matter of fact, a 6f -inch tire practically becomes, after a few days, a 5 \ -inch tire ; and, after a few months of wear, a 4J-inch tire. He adds that 6J-inch tires wear out after running 10,000 miles, losing 530 pounds of iron ; showing that for each pound of iron worn off, 19 miles were traversed: less than a fourth of the mileage per pound deduced from Sir John MacneiTs data for wagons, and countenancing, certainly, M. Dupuit's opinion that great width of tire is illusory, and adds weight without efficiency. From these and other considerations, it is obvious that the wear of macadam roads must necessarily be much greater than that of paved roads or streets. Mr. Mitchell's remarkable analysis of the material of a macadam road, about to be noticed, places this conclusion in a clear light. A cubic yard of broken stone metal, of an ordinary size 2 inches or 2J inches cube when screened and beaten down in regular layers 6 inches thick, contains, according to Mr. Mitchell, 1 1 cubic feet of interspaces, as tested by filling up the metal with a liquid. Herr E. Bokeberg, of Hanover, who, says Mr. Paget, made many very careful experiments on the proportion of vacuity to solid material, found that, in a loosely-heaped cubic yard of broken stones, void space amounted to one -half of the total volume. As the stones became rounded at the corners by wear, the vacuities were reduced to 37 per cent, of the gross bulk, or to 10 cubic feet These results indicate that the interspaces of the new stones were reduced by compression to nearly the same volume as was attained after the corners had been rounded without compression. The operation of consolidation necessarily reduces the thickness 146 MACADAM ROADS. of the coating. A large portion of the vacant space be- comes filled with mud, which forms the cementing matter, ground from the metal. The primitive mass of 16 solid cubic feet of broken stone is, in fact, crushed into every variety of form down to the finest sand, a small proportion only of the stones remaining of their original dimensions. Mr. Mitchell* gives the results of an analysis of a portion of the crust 2 J cubic feet of the macadamised road in the Mall, in St. James's Park, which was taken up for examina- tion. The component parts of the sample were carefully separated and classed, when it was found that one cubic yard contained Cubic Per feet. cent. Mud ....... 11-00 or 41 Sand, with pebbles, not exceeding 3-16 inch thick 2-40 9 Stones, from 3-16 inch to J inch . . . 6-56 24 Stones, from J inch to 1 inch . . . 4-48 16J Stones, from 1 inch to 2J inch . . . 2-56 9J Total vo"!ume, 1 cubic yard, or 27 '00 100 From this analysis, it appears that less than 9 per cent. say, one-tenth of the original stone escaped unground ; whilst 40 per cent, of it was reduced to the state of mud. These proportions, taken as they stand, are too favourable for the duration of the stone in that instance, for, no doubt, the sample was a sample of the remains of stone, much of which must have been swept or washed off out of sight. Mr. Bart, therefore, cannot be much amiss in his estimate that one-third of the loose road-material used in London is literally wasted by being ground up under the traffic before the consolidation of the surface is effected. The logical inference is that a macadamised road is not properly fit for traffic unless it is condensed, consolidated, and reduced to a hard and regular surface by suitable * "A New Mode of Constructing the Surface of the Streets and Thoroughfares of London and other Great Cities," 1870. WEAR. Id7 appliances. The road-roller has long been successfully employed in France and Germany for this purpose. Mr. Paget* states that a well-rolled road-covering contains at least from 70 to 80 per cent, of mass of stone, leaving only from 20 to 30 per cent, of intermediate space, most of which is filled, especially at the top, with clean sand. In a cubic yard, the spaces would amount to from 5 to 8 cubic feet, which would prove that the spaces are reduced by rolling to nearly a half of the amount when the new metal is not rolled. It is scarcely necessary to insist on the in- crease of durability, and the clear gain in economy of maintenance, of the road, by efficient rolling. The reduc- tion of gross bulk by the process of rolling, is measured by the difference of the volumes of interspace, 13 cubic feet, or half a cubic yard, and, say, 6J cubic feet, the mean of J- and 8 cubic feet ; that is, the reduction is (13 j- 6J =) 6 J cubic feet, which is one-fourth of the primitive bulk, or one-half of the primitive volume of the interspaces. It was said by Macadam that the annual wear of metalled roads amounted to from 1 inch to 4 inches of depth. The rate of vertical wear is readily calculated when the area of surface covered and the quantity of metal deposited are known. Since a cubic yard of loose broken stone contains only one-half of its volume, or 13J cubic feet, of solid stone, its weight, allowing 12 cubic feet of solid granite to 1 ton, is 13-5 approximately, 1 X - = 1J ton. Again, one cubic yard is equivalent to 36 square yards 1 inch deep ; and 1 ton of metal laid without compression to a depth of 1 inch covers an area of (36 X =) 32 *8 square yards. When the metal is laid and rolled, the * " Report on Steam Road- Rolling," p. 10. H 2 H8 MACADAM ROADS. primitive volume is reduced, as was found, by one-fourth, and 1 ton of rolled metal laid to a depth, of 1 inch covers an area one-fourth less than 32, or (32 x f = ) 24 square yards. On these data, the following rules are framed : RULE 1. To find the average vertical depth of wear per year of a macadamised road, when the area of surface, and the weight of metal laid per year, are given. Multiply the quantity of broken granite laid per year, in tons, by 32 when not rolled ; or by 24 when rolled ; and divide the product by the area covered, in square yards. The quotient is the average vertical depth of wear per year. For example, Mr. Redman stated, in 1854, that on the Commercial Road, in the east end of London, 10 yards wide and 2 miles long, the quantity of coating of broken granite required was from 1 inch to 1^ inches deep per year, or a total of 1,200 tons of metal. The heavy traffic of the road was then, as it is now, conveyed on two lines of granite tramway, consisting of four tracks 16 inches in width, occupying together a width of 64 inches, or T78 yards. The net width for macadam was, therefore, (10 1-78 ==) 8-22 yards. The length was (2 x 1,760 =) 3,520 yards; and the area of macadam was (3,520 X 8'22 r=) 28,934 square yards. By the rule, for unrolled metal, the average vertical depth of material supplied, and of 1,200 X 32 course worn off, was = 1-33 inches ; which is ^Jo,yo4 nearly the mean of 1 and H inches given by Mr. Redman. The converse rule follows directly : RULE 2. To find the area of surface that can he covered ly one ton of 'broken granite, when the thickness of the layer is given. Divide 32 by the thickness of the layer, in inches, unrolled ; or, divide 24 by the thickness of the layer, in inches, when rolled. The quotient is the area in square yards. RULE FOR WEAR. 149 When the quantity is given in bulk, the rule is as follows : B-TJLE 3. To -find the area of surface that can be covered by one cubic yard of broken granite, when the thickness of the layer is given. When the metal is not rolled, divide 36 by the thickness in inches ; the quotient is the number of square yards that can be covered. When the metal is rolled, divide 27 by the final thickness in inches, to give the required quotient. The carriage-way of old Westminster Bridge, 30 feet wide, was worn away, according to Mr. Browse, at the rate of 5^ inches annually ; the carriage-way of Bridge Street, adjoining the bridge, wears off 7 inches per year. The quantity of broken metal worn away annually from the surface of roads, is an exceedingly variable quantity, dependent not only upon the nature of the traffic, but also upon the mode of maintaining the road. A simple method of approximating to the annual rate of wear, when direct evidence is wanting, is suggested by data given by Mr. Paget.* He takes the average price of the five-ounce Guernsey granite metalling, used in London in 1869-70, at 16s. per ton delivered on the roads, and he shows that the cost of the material formed a large proportion of the total cost for maintenance. For instance : Total cost for Cost of maintenance, material. Bond Street . -. 245 216 or 88 per cent. 1384 1359 98 341 305 89 528 473 90 869 744 86 ,. 1588 1414 90 Piccadilly Bridge Street, "Westminster Parliament Street Victoria Street Kensington Road . Totals . 4955 4511 91 showing that about nine-tenths of the whole cost is * " Report on Steam Road-Rolling," page 32. 150 MACADAM ROADS. expended on the purchase of metal. When the total cost is known, therefore, together with the superficial area of carriage-way maintained, the cost may be converted into tons by dividing it by 16s., and the quantity and depth per square yard, approximately, may be deduced. CHAPTEE IV. MACADAMISED ROADS: COST. LONDON. THE price of Guernsey metalling, delivered in London, is now (1877) about 17 s. 6d. per ton. An ordinary mac- adamised road, constructed in the best manner, with, metal laid 9 inches deep, costs, at London prices, 6*. 3d. per square yard. It appears, from a table elaborated by Mr. Paget,* that in 1869 there were 1,127 miles of macadamised roads in the metropolis, maintained by 39 parishes and district boards of works, comprising an area of upwards of 21 millions of square yards. The expenditure for maintenance, exclusive of cost for cleansing and watering, averaged from four- teen of the returns, amounted to 250 per mile of road per year, equivalent to 2s. Wd. per lineal yard. If the average width of roadway be taken at 10 yards, the cost would be at the rate of 3*4 pence per square yard. The ascertained costs of maintenance appear to have varied from If pence to 2s. per square yard per year for broken granite covering ; and for granite, and flint and gravel coverings, taken to- gether, from If pence to 14^ pence. The following table, No. 13, is abstracted from the larger table of Mr. Paget : * " Eeport on Steam Road- Rolling." 152 MACADAM ROADS. TABLE No. 13. MACADAMISED EOADS IN LONDON: ANNUAL COST OP MAINTENANCE OF SOME OF THESE ROADS IN 1868-69. (Exclusive of Cleansing and Watering.) Laid with broken granite. Laid with broken flints or gravel. Total of mac- adamised road. *3 >> . it . |"4 Parish or District 5 5 a 5 Board of Works. ^ TJ! *o M j3 !S M OJ . "S *o * 1 1ft f | 1ft 1 6 o ft 3 1 11 r 1 1? 3 1 "3 * * * fl rt ^ II mis. yds feet. d. m. y. feet. d. m. y. feet. d. St. Mary, Lambeth 15 5J St. George, Han- over Square . . 14 40 6J 21 40 1-31 St. Leonard, Shore- ditch .... St. Mary, Newing- ton 5 13* St. Giles, Camber- well .... 7 43 50 4|* St. James and St. John, Clerken- well .... 2 9 5 St. George the Mar- tyr, Southwark . 15 33 2 St. Martin' s-in-the- Fields .... 1 413 30 10 f Mile End Old Town 21 3J* St. John, Hamp- stead .... 33 31 Westminster Dis- trict .... 14|.+ Wandsworth Bat- tersea .... 2f Wandsworth-Toot- ing, Streatham . 3J Hackney District . 7 8J* Strand District 170 24 St. Saviour's . . 8 if Plumstead . . . 27 20 13 40 if * Including Cleansing and Watering. i Principal thoroughfare of St. Martin's- in-the-Fields cost 2s. per square yard per year. * This cost, of 14|d., is the cost for maintenance of Parliament Street, White- hall, Great George Street, Broad Sanctuary, Victoria Street, Bridge Street, Great Smith Street, James Street, Kensington Road, Exhibition Eoad, Princes Street, and Rochester Row. Total area 72,329 square yards. COST. 153 The actual extent of the variation of the cost of mainte- nance of macadamised roads in London, is not revealed by these tabulated statements. The costs for individual roads or streets, principal thoroughfares, exclusive of cleans- ing, have been stated, by various competent authorities, to be as follows : Annual cost. Year. Per sq. yd. Parliament Street , . . 1856 2s. 4d. Do. ... 1869 3s. 3d. Bridge Street, Westminster . . 1856, 1869 3s. 6jd. Great George Street, Westminster . 1856 6Jd. Westminster Bridge (Old) . . 1854 2s. Od. Piccadilly .... 183463 2s. 5d. Do. .... 1870 3s. 6d. to 4s- Regent Street . . . . 1876 3s. 7d. Cranbourne Street and North Side of Leicester Square . . . 1870 2s. Od. Average annual cost . . 2s. 7jd. If the average amount of vertical wear be estimated from these rates of cost, in the manner before indicated, taking 90 per cent, as the cost for metal, the vertical wear would be found to vary from 1 inch to 7 inches per year. But, it must be admitted, this mode of estimation, though suggestive, is inexact. Suburban Highways. The Author is indebted to Mr. George Pinchbeck for statistics of the cost for maintaining the suburban highways of the Metropolis, Middlesex side, for one year, in or about the year 1855. The highways were comprised in the following Metropolitan Trusts : Kensington, Uxbridge, Harrow, Kilburn, Highgate, Hamp- stead and Islington, Stamford Hill, and Hackney Lea Bridge. Total length of roads for this account, 123J miles. H 3 154 MACADAM ROADS. Total, Per mile. Day labour, including the pumping of water . . 9,500 77*03 Digging gravel and preparing materials, per contract 1,830 14*84 Tram labour, including watering . . : 13,500 109*50 Flints, per contract ....... 1,951 15-82 Gravel do 2,460 1994 Granite and hard stone 29,430 238-60 Wheel wright work, per contract , . . 640 5*19 Implements 255 2*07 Smiths 158 1-28 Carpenters . . ... . . . . 251 2*04 Bricklayers 328 2*66 Stand-posts, water-plugs, pumps : Repairs . 365 2*96 Pavior'sbill 165 1*34 Water for Roads, from Companies .... 2,000 16-22 Material and Labour . . < . . 62,833 or 509*5 Rent of Premises 270 2-19 Salaries of Eight Surveyors, at 170 . , . 1,360 11*03 Do. Engineer-in-Chief 700 5*68 Do. Solicitor . . . . . . 400 3-25 Do. Accountant ... 350 2-85 General Expenses ..... 3,080 or 25-0 Total Cost 65,913 or 534*5 If the width be taken at an average of 30 feet, or 10 yards, the area maintained per mile would be 17,600 yards; and the total cost would amount to 7*29 pence, say 7%d. f per square yard for the year, made up as follows : Per square yard. Material and Labour . . t , 6*95 pence General Expenses . . ., , *34 7*29 pence. There were required, in addition, the services of 42 Com- missioners, which were rendered gratuitously. COST. 155 The cost for material that is, gravel, flints, and granite and other hard stone amounted to 35,671 per year, or 289-20 per mile, or about 4 pence per square yard ; being 54 per cent, of the total expenses. For the purpose of forming an estimate of the quantities of granite and flints consumed by wear, Mr. Pinchbeck calculates them from the total costs for the materials, allowing 15 shillings per cubic yard for granite and hard stone, and 4 shillings for flints. He thence deduces that the total quantities consumed were 4,000 cubic yards of granite laid over 100 miles of road, and 10,000 cubic yards of flint laid over, say, 25 miles of road ; being at the rate of about 7i inches deep of granite and of flints for the year. Local Roads. The minor roads of the parish of Islington, 36 miles in length, were maintained by the authorities themselves. The average cost during four years was 8,424 per year, or 234 per mile; which, taking an average width of 30 feet, would be at- the rate of 3-13 pence per square yard. BIRMINGHAM. According to the results of the experience of mac- adamised roads in Birmingham, in 1853,* it was found that the vertical wear in Bull Street was less than 6 inches per year, where 2,484 vehicles, comprising very heavy traffic, passed in ten hours in one day. With this exception, the greatest amount of wear was 4 inches per year, and, taking the average wear for the whole borough at 2 inches per year, which was a very high rate, the total cost for maintenance was estimated at 4d. per square yar J. per year. Adding 2d. for the cost of watering and cleansing, * " On Macadamised Roads for the Streets of Towns," "by Mr. J. P. Smith, in the Proceedings of the Institution of Civil Engineers, vol. *in. page 221. 156 MACADAM ROADS. the total annual cost was estimated at 6d. per square yard. But the actual cost for the year 1853, when the actual average wear was 1 J inch per year, amounted to a little over 3%d. per square yard. The items of cost are given as follows : Length. Area. Material used. Cost Ox Miles. Sq. yds. Tons. material. Urban streets . 34 359,040 20,112 rag stone, &c., at 5s. lid. 5,949186 Suburban streets and roads . 116 1,429,120 36,107 pebbles and gravel, at 2s. 6d. 4,513 76 150 1,788,160 56,219 10,463 6 Mr. Smith takes the volume of materials at the rate of a cubic yard per ton, whence the calculated rates of vertical wear : Vertical Cost of Material. wear material Cub. yds. per yd. persq.yd. inches, per year. Urban streets , , . 20,112 2 4d. Suburban streets and roads . . 36,107 0-91 fd. Total . 56,219 1J- 1*40 The cost for labour, haulage, management, &c., inclusive of cleansing and watering, amounted to 16,131 5s. 6d.; and the items of cost may be presented concisely thus : Birmingham. 1853. Per sq. yd. Pe^lmeal Macadam . . . 10,463 6 l'40d. 70* Labour, management, &c. 16,131 5 6 2-16d. 108 26,594 11 6 3-56d. 178 showing a total cost of 3'56 pence per square yard per year, as before stated; or 178 per lineal mile of street and road. The statement in detail, supplied by Mr. Smith, shows the respective amounts for the various items of cost, as here given. The proportional values of these amounts are added as percentages of the total cost : COST. 157 Birmingham. 1. Staff of officers and men 2. Keep of 63 horses, shoeing, and vete- rinary expenses 3. Ragstone and Hartshill stone 4. Pebbles and gravel 5. Water .... 6. Cleansing machines and brooms 9. Team hire .... 10. Rent ..... 11. Gearing . . . 12. Wheelwright .... 18. Stationery , , 22. Sundry accounts Total cost for maintenance, cleansing, and watering . 1853. 6,979 10 7 Per cent 26-3 3,276 12-3 5,949 18 6 22-4 4,513 7 6 170 771 3 8 2-9 965 6 8 3-6 631 7 2 2-4 12 0-0005 114 6 4 0-4 423 7 3 1-6 117 7 9 0-4 2840 6 1 10-7 26,594 11 6 100*0 Here it is shown that the cost for macadamising materials for maintenance amounted to about 40 per cent, of the total cost. The cost of coarse-grained schist, with 23 per cent, of lime, was 4s. 9d. per cubic yard. For heavy thoroughfares, it was broken up into cubes of about 2 inches ; but, in some cases, where the traffic was more than ordinarily severe, the pieces were 4 inches in thick- ness. Generally, for streets with less traffic, the thickness was from 1 inch to H inches. The size of the stones was proportioned to the wear and tear of the streets, and was suited to the season of the year. In the management of repairs by Mr. Smith, they were executed during the winter months November, December, January, and February when the roads were in their wettest condition ; and during the other months of the year, the horses were employed in bringing, and the men in pre- paring, the broken stone and gravel. The new material, laid in the wet season, bonded, and became more rapidly consolidated, than if it had been laid down during drier weather, and there was less need for watering it. The consequence was that, during summer, loose stones were scarcely to be seen upon the roads. 158 MACADAM ROADS. Whilst the average cost for maintaining the whole of the macadamised streets and roads of Birmingham has been shown to be about 3 %d. per square yard per year, the following estimates were framed by Mr. W. Taylor, of the first cost, and the cost for maintenance and cleansing the streets of the greatest traffic in Birmingham, when mac- adamised, and when paved with granite sets ; taking a period of fifteen years for the duration of the sets, and taking a vertical wear of 4 inches of macadam annually : Macadam Birmingham. Per square Per year yard. d. Macadam 6 inches thick, first cost . . Is. 6d. or 1-2 Coating laid twice a year, 4 inches thick, \ , A n -, , . annually, 14 years at Is. / 14s ' Ud> 11>2 15s. 6d. 12-4 Cleansing, 15 years at 4d. ... 5s. Od. 4-0 Total cost for macadam, in 15 years . 20s. 6d. 16'4 Paving, first cost Repair, 15 years Deduct, value of old stone Cleansing, 15 years at Id. Paving. Per square yard. 7s. 6d. or Is. 6d. Per year, d. 6-0 1-2 9s. Od. 2s. 6d. 7'2 2-0 6s. 6d. . Is. 3d. 5-2 1-0 Total cost for paving, in 15 years 7s. 9d. 6-2 DERBY. The streets of Derby are, or at least they have been until recently, macadamised, with one exception, which is paved. The following instructive table gives particulars of cost for maintenance and cleansing of twelve principal streets in Derby, and the proportional traffic.* * Slightly adapted from a Table given in a Report of Mr. E. B. Ellice-Clark, Surveyor to the Borough of Derby, May, 1876. COST. 159 TABLE No. 14. DERBY. MACADAMISED STREETS, COSTS FOB MAINTB- NANCE AND CLEANSING. Total Annual Annual annual cost for Number ist lor cog ^. f or mainte - of vehi- Name of Street. Area. Material. mainte- c j eari g_ nance cles m 3 iner Der and 12 per square yard. square yard. cleans- ing per sq, yd. hours. sq. yd. s. d. d. s. d. vehicles Bridge Gate 2931 Broken boulders 1 6 6J 2 J 1268 St. Helen's Street 1203 Do. granite . 1 2J & 1 7} 585 Wardwick 984 Do. do. 1 2-J- 8* 1 11 1258 lion Gate . 1562 Do. do. 7- 11^ 1506 Corn Market 1695 Do. do. rl 3- 10} 2031 St. Peter's Street 3230 Do. do. i 2! 3 1 5} 2762 London Road 4859 Do. boulders } and granite J 7J 2 1272 London Street . 2847 Do. do. 10} 3 1 1} 1200 Tenant Street 1245 Do. do. 1 2J 3f 1 6J 1250 St. James's Street 995 Do. do. 6f 41. 10 711 Siddals Roud 4728 Do. do. n 3 10} 1931 Averages for ) macadam j 11 j 1 31 143i Derwent Street, \ paving down 3483 Sets 3 " x 5 " nil 2 4 2 i 1658 7 years ) It is to be remarked, in this table, that the boulder macadam of Bridge Gate incurs a higher cost for mainte- nance than any of the granites. Wardwick, upon which the greatest cost for cleansing is expended, is cleansed daily. The average annual cost for maintenance for eleven macadamised streets is 11^. per square -yard, and for cleansing 4%d. ; whilst, for the seven years during which the paving has been down, it has not cost anything for main- tenance, but it has cost for cleansing 2%d. Mr. Ellice- Clark has formed estimates of the cost for construction and maintenance of granite -set paving, in substitution for macadam, for five streets in Derby ; from which the following table, No. 15, is compiled : 160 MACADAM ROADS. TABLE No. 15. DERBY: ESTIMATED COST FOB PAVING AND ING FIVE STREETS. Granite Sets 3 inches wide by 5 inches deep. Name of Street. Dura- tion of Pave- First cost per square yard. Cost for maintenance per square yard. Total cost per square yard. J"S ibr construc- itenance, and g per square per year. ment. t*> ao.S.g'd Total. Per year. Total. Per year. Total. Per year. ii P Bridge Gate 24 s. d. 13 d. 6-5 s d. 7 6 d. 3-8 s. d. 2J 6 d. 101 d. '4 s. d. St. Helen's Street 29 13 5-4 7 6 3-1 20 6 8} 2 10* London Road 32 12 6 4-7 6 6 2-44 19 7* 1 8^ Tenant Street . 28 13 5-5 7 2 3-1 20 2 81 U- 10 Sidda's Eoad 32 12 6 4-7 6 6 2-44 19 H 8f Averages 29 5-36 3-0 8-36 1-6 10 ii The first cost includes the cost for a foundation of con- crete. No deduction has been -made by Mr. Ellice-Clark for the value of the old materials ; this is taken as equi- valent to the cost of removing the existing macadam. In the estimates for maintenance, provision is made for taking up, re-dressing, and re-laying the stones once during the period. From the foregoing data, a comparison is readily framed between the costs in perpetuity for macadam already laid and new granite-paving, including a foundation of con- crete, for the five streets named in the last table: COST. 161 TABLE No. 16.- DERBY: COMPARATIVE COSTS FOR GRANITE PAVE MENT AND MACADAM. Pavement of Granite Sets, 3 inches by 5 inches deep. Total cost for pave- ment per sq. yd., Total cost for macadam, already Name of Street. Dura- tion of including founda- tion of conciete. laid, per square yard. Number of vehicles Pave- in 12 ment. For the For the hours. whole Per year. whole Per year. period. period. Bridge Gate Years. 24 s. d. 25 8. d. i i s. d. 49 s. d. 2 J Vehicles. 1.68 St. Helen's Street 29 25 4 10} 47 9 1 7| 585 London Road 32 21 8 8} 24 8 9* 1272 Tenant Street 28 23 8 10 42 7 1 6J 1250 Siddals Road 32 22 4 8* 18 6 10} 1931 Averages 29 24 2 10 39 11 i H 1261 SUNDERLAND. Mr. D. Balfour* reported, in 1876, on the cost for main- tenance of the highways of the Sunderland and Houghton- le-Spring District, before and after the passing of the Highway and Turnpike Act of 1869. For the maintenance of 77^- miles of highway, the following were the respective costs. The last two lines are added by the writer : AVERAGE ANNUAL EXPENDITURE. Materials and Labour . General Charges. During last 3 years, Old Act. 1,903 Total Cost . Do. do. per mile . Do. do. per square yard \ allowing width of 30 feet j 1,903 24-5 3-34d. During last 6 years, New Act-. 1,735 284 2,019 26 3'55d. During last 3 years, New Act. 1,509 277 1,786 23 3'14d. * " Road Legislation and Surveying." By D. Balfour, A.I.C.E., Surveyor and Engineer to the Sunderland and Houghton-le-Spring District Highway Board, 1876. 162 MACADAM ROADS, DISTRICTS NEAR EDINBURGH, GLASGOW, AND CARLISLE. Mr. James H. Cunningham, in his paper already noted, gives the following particulars of cost of construction of macadam roads in the south of Scotland : Country Eoads (1861 1874) cost from Is. 2d. to Is. 6d. per square yard. Town .Roads, founded on shivers, or on hard set stone : 15 inches of metal (1871) cost Is. 9d. per square yard. 16 (1874) 2s. 4d. 18 (1873) 2s. lljd. Of whin metal, from 2 inches to 2J- inches thick, the average price (1861 1874) was 4s. Wd. per cubic yard. In Edinbugh (1874) the price was 5s. 9d. to 7s. 3d. per ton, averaging 6s. Sd. per ton, and as, according to Mr. Cunningham, one cubic yard of whin metal, containing 40 per cent, of vacant space, weighs 27 cwt., the average price is at the rate of 4s. Wd. per cubic yard, the same as the average already given. GHAPTEE V. CONCRETE KOADS. MR. JOSEPH MITCHELL designed a concrete macadam sur- face, in which the spaces, otherwise vacant, and ultimately filled with the muddy cementing matter of worn macadam, are, in the construction of the concrete road, filled with an admixture of Portland cement, or other hydraulic cement grout. The concrete thus formed rapidly becomes a uniform and impervious mass, which is wholly unaffected by heat or moisture. It is mixed in these proportions : Broken stones , , .4 measures. Clean sharp sand . , 1 J to 1J- Portland cement . . 1 So, for a cubic yard, or 27 cubic feet, of broken metal, 6| cubic feet, or 1J barrels (of 4 cubic feet), of Portland cement, are required. The broken metal should be of the hardest quality, of uniform size, thoroughly screened ; and it should, when in the screen, be dipped up and down in a large tub of water, and then thrown on the platform on which the concrete is to be made. Cement of the best quality must be employed, and the sand should be sharp, clean, and gritty. The surface of the ground is brought to form, and rolled several times. The concrete is then laid on the surface in a layer 3 or 4 inches, and is left for three days to harden. The second layer of 3 or 4 inches is next laid on the first, and immediately rolled to form with a heavy iron roller, as heavy as two or three men can 164 CONCRETE ROADS. draw. The cement should be left for three weeks, to allow it to become quite hard, before the road is opened for traffic, although a week has been found to be a sufficient interval. Mr. Mitchell states that a concrete road, 7 inches deep at the middle and 5 inches at the sides, is sufficient for ordinary traffic. For heavy traffic, a depth of 8 inches is recommended. The first piece of concrete road was laid, in 1865, in Inverness, and consisted of 45 lineal yards of the approach to the goods station of the railway. In 1870, after the road had been under traffic for 4J- years, it was reported that the wear of the surface was scarcely appreciable, whilst the adjoining macadamised road had been coated frequently every year. Another specimen, 50 yards long and 15 yards wide, was laid in 1866, on George IV. Bridge, Edinburgh, where the traffic is heavy and continuous. At the end of 3J years, under traffic, the surface was perfectly sound and immovable. The amount of vertical wear, during the periods above- named, appears not to have exceeded J inch. But Mr. J. H. Cunningham, writing in January, 1875,*' stated that it was then much worn at the surface, in consequence, he thought, of its great hardness and rigidity. A concrete road, of 6 inches average depth, would cost in London, by Mr. Mitchell's estimate, 65. Wd. per square yard, against 6s. 3d. per square yard for an ordinary macadamised road, constructed in the best manner, with 9 inches of metal. * See Mr. Cunningham's paper on " Streets," in the Proceedings of 1he Edinburgh and Leith Engineers' Society > 1874-75, page 18. CHAPTEE VI. MACADAMISED ROADWAYS IN FRANCE. M. DUMAS, in the article already referred to (page 20), maintained that, if a macadamised road be properly and actively watched, levelled, and cleaned, a foundation, as such, is of no special utility, and may be suppressed ; and that the road should be simply a uniform bed of small materials. The width of the old roads, he says, was excessive, amounting sometimes to nearly 80 feet, with a pavement in the middle, from 16 to 20 feet wide. They are now made of widths of from 8 to 14 metres, or 26 to 46 feet. Type-sections of French and Belgian roads and streets are shown in Figs. 42 to 44. There is no need, says M. Dumas, for great thickness of Fig. 42. Section of Hue de Eivoli, Paris. road, provided that a compact and impermeable mass be formed, completely shielding the ground from moisture. The least thickness that may be given depends on the size of the materials. If the pieces are 2 inches or 2J inches diameter, mixed with pieces of smaller size, a solid road- 166 MACADAMISED ROADWAYS IN FRANCE way may be formed, of a thickness not exceeding 4 inches. Inferior quality of material, bad ground, or heavy traffic, does not imperatively demand greater thickness, if the road be well and continuously maintained. Four inches Fig. 43. Type-Section, Bas-Rhin. of thickness may be regarded as sufficient : it may be pru- dent to make a thickness of 6 inches ; and any excess above 8 inches is completely useless. Fig. 44. Type-Section, Belgium. The curvature of the old roads was excessive. The rise was from 1 in 14 to 1 in 12. The most convenient rise is 1 in 33, which is much more than sufficient for a well-kept road. The minimum appears to be 1 in 50 ; which is the CONSTRUCTION. 167 slope upon which the weight is in equilibrium with the frictional resistance on roads in good condition. When newly made, the rise may be 1 in 25, to allow for the greater wear at the centre than at the sides. The form of the base, or surface of the ground, may follow the curva- ture of the surface ; but M. Dumas prefers a horizontal line, as in all respects preferable. A road 10 metres, or 33 feet wide, with a rise of 1 in 33, and an average thick- ness of 4 inches, on a horizontal base, would have, accord- ing to M. Dumas, a thickness of 8 inches at the middle or crown, and 2 inches at the sides. But these measure- ments are not consistent with each other. The material for the road must have the qualities of hardness, and facility for binding. The largest pieces must pass through a 2-inch ring; the smaller pieces occupy the interspaces. The 2^-inch gauge is perhaps too large for very hard material, but it is more than sufficient for the softer materials. The smallest debris of broken material should be preserved, and spread on the surface. For gravel, the lower limit of size is fixed by the use of a sieve having J-inch meshes ; otherwise there may be no- thing but sand for binding. The smallest pieces should be on the upper surface ; if not, the upper pieces are broken. It is well, finally, to break the pieces on the made surface : the debris binds. For the consolidation of the road, it should be rolled and watered. The cylinder of the roller should be from 6 feet to GJ feet in diameter, and 5 feet wide ; weighing, empty, 3 tons, and full, 6 tons. The maximum weight, when loaded, should be from 8 to 10 tons. These weights give a pressure varying from 112 pounds to 370 pounds per inch of width; the sufficiency of which has been proved by experience. The empty roller is first used; then the full roller ; and, lastly, the weighted roller. Sand or other binding, with water, are thrown on the 168 MACADAMISED RO \DWAYS IN FRANCE. surface at intervals ; the water helps to bind the new material. The material binds most speedily when the thickness is 4 or 5 inches. For this thickness 8 or 1C traverses are sufficient : two traverses dry, with the empty roller ; two, with a layer of detritus and the empty roller ; two with the full roller, two with the loaded roller, and one at an interval of from 8 to 15 days after the road is opened for traffic. The same number of horses per- form the work of rolling from first to last. But the sur- face of inclined portions may be injured by the feet of the horses ; and the injury is partially prevented by increas- ing the number of horses. It is nearly impossible to roll over inclines steeper than 1 in 20. If the metal is more than 4 or 5 inches thick, it should be deposited in two successive layers, to be rolled succes- sively ; although the first layer should not be completely rolled ; and the second layer should be laid after the sur- face of the first has been roughed by the traffic. The surface is kept up by the use of a stamper or rammer, weighing from 15 to 20 pounds, which is 8 inches in diameter at the base, and is shod with iron. In some cases, stamping may be applied with advantage, instead of rolling. A road constructed on the system above described, 4 inches thick, is superior to a road 8 or 10 inches in thickness, consolidated by wheels, and with successive additions of material. The flanks of the road or the bermes (accotements), con- sisting of the natural ground, are fit for traffic during the greater part of the year. " All these proceedings," says M. Dumas, " have, for their basis, the principle of the maximum of leauty" CHAPTEE YII. STONE PAVEMENTS CITY OF LONDON. IT lias been stated that the streets of London were, pre- ! viously to the introduction of granite sets, paved with pebbles or boulder-stones, which were bedded in sand, and constituted pebble-paving. This pavement was gradually superseded by a pavement of irregular blocks of stone, and ultimately by rectangular granite sets, or " cubed granite," as they are sometimes designated ; until, in 1848, there re- mained only one mile of pebble-paving in a total length of 50 miles, in the City alone. Granite sets of comparatively large dimensions were at first employed. They were from 6 to 8 inches in width on the surface, by from 10 to 20 inches in length, with a depth of 9 inches. As origi- nally laid, they were merely laid in rows on the subsoil, and, after the usual process of grouting and ramming, the street was thrown open for the traffic which was ex- pected to perform the last duty of the pavior, and to settle each stone upon its bed. The large wooden rammer, of 84 pounds weight, was obviously insufficient for the pur- pose of enabling the pavement to resist, without further movement, the percussion of heavily weighted wheels. In 1850, however, and probably for some time previously, it had become the general practice to make a good sub- stratum of hard core," consisting of shivers, broken stone, brick rubbish, clinker, or other hard material, not necessarily of the size of macadam, nor, on the contrary, i 170 STONE PAVEMENTS CITY OF LONDON. laid on a pavement on the Telford system of road-making, but a medium between these extremes of practice, such as Telford himself recommended in 1824. The substratum of hard core has been usually laid to a depth of from 9 inches to 12 inches ; but a thickness of 15 inches has been laid down in the principal streets of the City. Upon Fig. 45. Stone Pavement of King William Street, City of London. the hard core was laid a stratum of sand, into which the stone sets were bedded. Fig. 45 shows a section of King William Street, as originally paved. The following table, No. 17, gives the dates of the for- mation of some of the earliest carriage-way granite pave- ments laid in the leading thoroughfares in the City of London, with their superficial dimensions, and the lengths of time that they remained before being removed to other places.* * This table is compiled from the Reports of Lieutenant- Colonel, Haywood, Engineer and Surveyor to the Commissioners of Sewera of the City of London. The writer is indebted to this valuable series of Reports for the greater proportion of the materials employed in the preparation of what follows with respect to the streets of the City of London. DURATION OF EARLY PAVEMENTS. 171 TABLE No. 17. CITY OP LONDON : EARLIEST CARRIAGE-WAY GRANITE PAVEMENTS IN LEADING THOROUGHFARES. Aberdeen Granite Sets, C inches wide by 9 inches deep. Name of Thorough- fare. When laid. i When taken up. Time down. Remarks. Year. j Year. Years. Bridge Street 1828 1818 20 Ludgate Hill 18i4 16 Ludgate Street 1847 19 Bishopsgate St. } 1853 25 Without . J M Poultry 1829 1841 & '43 11&13 Removed for wood paving Cheapside 1843 & '44 14&15 / Do do. ( 1843 to 1846-47. Farringdon Street 1830 Relaid in 1858 Skinner Street 1845 15 Newgate Street . 1842 12 Removed for wood paving Aldersgate Street . 1831 1857 26 St. Paul's Church- } yard . j n 1847 16 Holborn n 1849 18 Threadneedle St. . J? 1862 31 Fleet Street . 1832 1846 14 Fenchurch Street . 1852 20 Cannon Street n 1855 24 Walbrook Little Moorfields 1 (sets 5 in. by 7) j 1833 Still down in 1853 King William St. . 1834 1849 15 Princes Street 1835 1850 15 Mark Lane . 1837 1856 19 Houndsditch . 1t 1853 16 Fore Street . 1838 1856 18 Nearly the whole of the stones which formed these early pavements was removed to and laid in places of secondary traffic within the City, where their duration has been as great as, and in some instances greater than, it had already been in the original thoroughfares. The cost for reparation, up to the year 1840, of 36 streets laid with granite sets 6 inches wide and 9 inches deep, in- cluding nearly all the streets named in the preceding table, was calculated by Mr. Kelsey, formerly the City Surveyor, and the results are given in the following table, No. 18.. 12 172 STONE PAVEMENTS CITY OF LOXDOX. TABLE No. 18. COST FOR REPARATION OF CARR T AGE-WAY PAVEMENTS, Laid with 6-inch Granite Sets, in Ihe City of London, up to the year 1840. Name of Street. Average cost of repa- ration per square yard per year. Name of Stieet. Average cost of rep -i- ration per square yard per year. Aldgate High Street pence, over J Holborn . pence, over 1 Aldgate Street, and King Street, Snow Aldgate over J Hill over | Aldersgate Street . over J King Street, Cheap- Arthur Street West f. side not J Bishopsgate Street Leadenhall Street . nearly ' Without over J Long Lane not i Bishopsgate Street Lombard Street nearly 1| Within over J Ludgate Hill and . Budge Row and Street . not I Watling Street . 1J. Newgate Street 3 Cannon Street . nearly 1 New Bridge Street Cheapside and Po 1- (part of) not | try . . . 1J Old Bailey (part of) 4 Coleman Street not f Pavement, The not g Cornhill . over Queen Street . not | Dowgate Hill . Farringdon Street . nearly J not J Shoe Lane (part of) . St. Paul's Church- not | Fenchurch Street . over f yard . over f Fleet Street . nearly If Threadneedle Street not 1 Fish Street Hill not J Walbrook 2-| Gracechurch Street . not 1 Wood Street . over 1 Holborn Bridge and Skinner Street HolbornHill . not -J if Average cost for re- pair, say 1 penny The average cost for repair, according to the table, may be taken at Id. per square yard per year. The first cost was considerably affected by the nature of the pavement laid, the size of the stones, the nature of the granite, and the market price of the material. Colonel Haywood esti- mated that, for stones 9 inches in depth, it ranged from ^11$. to 17 s. per square yard. The long continuance of the system of paving with large blocks resulted from the experience of their great dura- bility and economy in first cost. But they did not afford INTRODUCTION OF THREE-INCH SETS. 173 sufficient foothold for horse-traffic. Granite sets of less width were subsequently laid: they were 5 inches and 4 inches in width; and, finally, sets of only 3 inches in width were laid. The 3 -inch sets, with a depth of 9 inches, al- though they were considered to be the least durable, and the greatest in first cost, proved to be by far the safest for paving, and they gave more satisfaction than the wider sets. The merit of their introduction is due to Mr. Walker, under whose direction Blackfriars Bridge was, in 1840, paved with 3-inch granite sets. Mr. Walker attached great importance to the obtainment of a solid substratum. Blackfriars Bridge, to the paving of which allusion has already been made, was closed for some weeks, in order that the concrete foundation might have time to set and harden, before the pavement was laid down. The narrow granite sets were laid with a great degree of accuracy, and the whole mass was bedded as if it was composed of bricks : the stones were bedded in good mortar, and the joints were well filled with it, and in consequence of the careful work- manship, with the use of narrow stones, the work remained good for 13 years, at the end of which time it was lifted. From that time 1840 the use of 3-inch sets was ex- tended, until, in 1848, there were, in the City of London alone, three miles of carriage-way paved with them. The following analysis, prepared by Colonel Haywood, shows the lengths of the various kinds of pavement in the of London, in November, 1848 : CITY OF LOXDOX PAVEMENTS, IN NOVEMBER, 1848. Miles. Carriage-way pavement of pebbles or boulders . . about 1 Carriage-way pavement of 6-inch, 5-inch, and 4-inch granite sets (nearly the whole being 6-inch sets) . ., 28 J- Carriage-way pavement of 3-inch sets . . . 3 Wood pavement, various kinds . . . ,, j Macadamised road, in Finsbury Circus . . Flagstone paving, courts, alleys, &c. 16 J Totc^l . 5CT 174 STONE PAVEMENTS CITY OF LONDON. In 1851, there were 51 miles of public ways in the City containing 441,250 square yards of carriage-way, averaging for, say, 34 J miles. about 22 feet wide. 328,907 square yards of footway. 770,157 In 1866, Colonel Hay wood reported that there were then 50 miles of public way : Miles. Main thoroughfares . . . .7 Collateral thoroughfares . . . . .28 Minor streets, courts, alleys, passages, &c. .15 50 Area of carriage-way, about . . 390,260 square yards. Do. footway . . 309,018 699,278 Regarding the question in all its bearings, Colonel Haywood concluded that the 3-inch granite sets, being safest, as giving the best foothold, were the best for large towns with great traffic. On such a pavement, horses were less strained, there was less wear and tear of vehicles, and a greater degree of quietness and general comfort. They made the most even pavement, and they retained an even surface, longer than any other stone pavement that had been tried. In adopting, in 1848, the conclusion that 3-inch sets that is, granite sets 3 inches in thickness, with a depth of 9 inches were the best for large towns, Colonel Haywood had before him the experience of a trial piece of what was called the Euston pavement in Mount Sorrel granite, which had been laid in 1845, in Watiing Street, and relaid in 1848. This pavement derived its name from the circum- THE EUSTON PAVEMENT. 175 stance of its having been laid about the year 1843, at the departure side of the Euston Station of the London and North Western Kailway. The same sort of pavement had previously been tried experimentally, and for the first time, in 1838, at Birmingham, at the crossing of a street, where heavy loaded wagons were constantly passing over it, and, in 1850, it was stated by Mr. "William Taylor* to be in as perfect a condition as when it was first laid. Mr. Taylor's motives for advocating the Euston pavement, were based upon the principle advocated by Macadam, to provide a foundation possessing a certain amount of elasticity, but of sufficient strength to support the surface material ; but he substituted "one stratum of solid granite/' in the form of paving-stones 4 inches deep, for an equal depth of mac- adam or broken ring-stone: thus endeavouring to com- bine the elasticity, so-called, of a macadam road, with the durability of a paved road. As laid at Euston, the ground was first removed to the depth of 16 inches below the in- tended level of the pavement, the bottom being formed to the convexity of the intended surface of the street. A layer of coarse gravel, 4 inches thick, was spread upon the bottom ; and compressed by being rammed equally through- out. Upon this layer was placed another layer, 4 inches thick, of gravel mixed with a small proportion of chalk, or of hoggin, and rammed likewise. A third layer of the same materials as the second, but of a finer quality, was distributed, and the whole was well rammed together, to form a solid and level surface ready to receive the pave- ment. Upon a 1-inch bed of sand, stones of Mount Sorrel granite were laid, measuring 3 inches wide, averaging * Proceedings of the Institution of Civil Engineers, 1849-50; vol. ix., p. 214 : " Observations on the Street Paving of the Metro- polis ; with an Account of a peculiar system adopted at the London and North- Western Railway Station. Euston Square," by William Taylor. 176 STONE PAVEMENTS CITY OF LONDON. 4 inches in length, and from 3 to 4 inches deep, neatly dressed and squared ; and carefully and closely jointed, in order to prevent any single stone from rocking in its bed. A rammer of 55 pounds weight was then applied over the whole surface, after which the pavement was covered with a sprinkling of screened gravel. The maximum cost of the Euston pavement, the foundation included, was said to amount to 12s. per square yard. Mr. Dock- ray, resident engineer of the railway, in testifying to the excellency of the Euston paving, stated that the car- riage-way had previously been substantially executed, according to the old system, with stones 8 inches deep, laid upon a substratum of concrete ; but that the stones became much rounded on the upper surface, and were required to be removed. In the Euston pavement he stated that, by reason of the smallness of the stones, they never became rounded on the upper surface by wear, as was the case in ordinary pavements of large stones. The Euston pavement was subsequently laid in the large esplanade in front of the booking offices at Euston station. The carriage-way traffic at Euston sta- tion has of course been of a light order, limited to carriages, cabs, and omnibuses. Eeverting to Colonel Hay wood's piece of Euston pave- ment, in Watling Street, he had experienced a difficulty in obtaining a satisfactory test of the durabilities and cost of various pavings ; and, with a view towards the settling of the question of the most useful dimensions of granite sets, for the carriage-ways of the City, he recommended that different sorts of stone-paving should be laid in Moorgate Street. The experimental pieces were laid, all at the same time, in October, 1848. The situation was well chosen, for the different pieces of paving were as nearly as possible sub- jected to the same amount and kind of traffic, with the usual disturbing and destructive influences. The dimer EXPERIMENTAL PAVING. 177 sions and the first costs of the experimental paving are exhibited in Table No. 19. TABLE No. 19. CITY or LONDON : EXPERIMENTAL MOOKGATE STREET, 1848 56. Dimensions and Firat Cost. PAVING Dimensions of Sets. First cost Name of Granite. Depth* Width or thickness at the sur- T ength at the surface. per square yard com- plete. face. inches. inches. inches. s. d. 1. Blue Aberdeen 9 4 8 to 12 13 8 2. Mount Sorrel 9 4 8 to 12 14 9 3. Blue Aberdeen 9 3 8 to 12 16 3 4. Mount Sorrel 9 3 8 to 12 17 5. Blue Aberdeen 5 3 5 10 9 6. Mount Sorrel 5 3 5 12 NOTE TO TABLE. The cost of the substratum is not included in the above prices, but the whole of the specimens were laid upon a substratum of equal depth and compactness, and the cost for a founda- tion may be taken for all practical considerations, as being the same for the whole. The prices given, therefore, are for the new granite surface only, laid and grouted with lime and sand, and left complete. The pavings were duly and carefully maintained during the eight years from October, 1848, to October, 1856, under Colonel Haywood's personal superintendence. The accounts of charges for maintenance were accurately kept, and tho results, with additional particulars, are given in Tables Nos. 20 and 21, 1 3 178 STONE PAVEMENTS CITY OF LONDON. TABLE No. 20. CITY OF LONDON : EXPERIMENTAL PAVING IN MOORGATE STREET, 184856. Cost for Reparations. Cost for Cost for Annual repara- Tepara- interest tion and tion in at 5 per interest Date of first reparation after laying Name of Granite. 8 years, cent, on to- down the sever 1 pavements, in per sq. first gether, October, 1848. yrd. per cost, per per sqr year. sq. yrd. yard per year. d. d. d. 1. Blue Aberdeen . 2-076 8-20 10-27 In quarter ending M'mas, 1852 2. Mount Sorrel . 96 8-70 9-66 In half year 1852 3. Blue Aberdeen . 4-62 9-74 14-36 In 1851 4. Mount Sorrel . 1 872 10.20 12-07 In 1856 5. Blue Aberdeen . 6-684 6-46 13-14 In quarter ending Mar. ?5, 1 851 6. Mount Sorrel . 5-844 7-20 13-04 In half year ending M'mas, 1852 Averages . . . 3-672 8-41 12-08 TABLE No. 21. CITY OP LONDON: EXPERIMENTAL PAVING IN MOORGATE STREET, 1848 56. Data showing Comparative Values. Name of Granite. Number of stones in a square yard. Area of surface of one stone. Cubic contents of one stone. Area relaid in 8 years. Area of new stones laid in 8 years. square cubic 1. Blue Aberdeen . stones. 29 inches. 44-7 inches. 402-2 per cent. 1198 per cent. 1-63 2. Mount Sorrel . 25 51-8 466-6 41-8 75 3. Blue Aberdeen . 35 37-0 333-3 243-6 3- 80-01 75-83 50-12 52-70 64-28 50-33 63.20 Labour ) 01* 6s. Sd. Gs. 3 Jd 4s. 8d. 4s.4fd. 5s. 4jd. 4s.2|d 5s. 3Jd. 1 i STONE PAVEMENTS OF LIVERPOOL. 195 TABLE No. 26. LIVERPOOL STREETS : COST FOR CONSTRUCTION PEK SQUARE YARD OF THE CARRIAGE-TV AYS OF Six STREETS IN DIF- FERENT LOCALITIES. Boulder Pavement. Name of Street. Average First Costs. Spencer Hygeia. Sheri- dan. Jubilee. Hope. Hope Place. MATERIALS. Boulders . Channels . Crossings . Gravel and Sand Materials Cartage Total Cost for 1 Materials j pence. 14-61 4-08 27 pence. 17'52 3-96 1-88 pence. 12.28 9-82 13-60 22 pence. 13-06 6-39 8-51 61 pence. 9-76 4-51 3-35 41 pence. 11-25 6-80 1-66 69 pence. 13-08 5-93 4-88 32 18-96 283 23-36 3-76 35-92 8-34 28.57 5-49 18-03 3-06 20-40 1-21 24-21 4-11 21-79 37-12 44-26 ^4-06 21-09 21-61 28-32 LABOUR. Paviors Labourers . Labour . Use of Tools . Total Cost for 1 Labour j 3-51 5-27 3-24 4-86 2-83 11-74 4-17 4-98 2-88 3-30 3-80 4-53 3-40 5-78 8-78 51 8-10 48 14-57 98 9-15 37 6-18 62 8-33 46 9-18 57 9-29 5-58 15-55 9-52 6-80 8-79 9-75 Total Cost for ] Materials & V Labour, J or 31-08 35-70 2/11} 59-81 4/11} 43-58 3s.7Jd. 27-87 2s. 4d. 30-40 2s.6jd. 38-07 3s. 2d. These tables afford good selections of instances to show how much the cost is influenced by the circumstances under which the work is done. The cost of the materials used in set-paving does not vary so much as the cost of cartage and of labour, which depend upon the distance of K2 196 STONE PAVEMENTS OF LIVERPOOL. a street from the docks whence the material is conveyed ; and upon the length of the day and the state of the weather, and other circumstances affecting the conve- nience for operations. From these tables it appears that the average costs for the construction of the carriage-ways of Penmaenmaur sets and of boulders, were composed as follows : LIVERPOOL. Penmaenmaur sets. Boulders. s. d. s. d. Materials . 4 1J or 78 per cent. 2 4j or 74 per cent. Labour . . 1 2 or 22 9J or 26 Total cost. 5 3J 100 3 2 100 In introducing a comparative statement of the annual cost for maintenance of set-paving, boulder -paving, and macadam, Mr. Newlands fixed the first cost LIVERPOOL. First cost. For square-set paving, at 4s. 6d. per square yard. boulder paving Is. 8d. macadam Is. 8d. These amounts were probably based on wider averages than those which have been deduced from the foregoing tables. However that may have been, Mr. Newlands, in 1854, stated that, at Liverpool, the first cost for a set-paved road was 6s. per square yard, and that of a macadam road was 2s. per square yard.* The average cost for maintenance for the three years 1848 50, is given for 27 streets, of which nine were paved with squared sets, nine with boulders, and nine were laid with macadam. Summaries of the costs are given in the table, No. 27. Mr. Newlands explained that the averages here derived were not to be taken as for long periods, especially for the boulder paving, much of which was required shortly to be * Proceedings of the Institution of Civil Engineers, vol. xiii., 1853-54, age 240, STOXE PAVEMENTS OF LIVERPOOL. 197 TABLE No. 27. LIVERPOOL STREETS: ANNUAL COST FOR MAIN- TENANCE OF 27 STREETS, DEDUCED FBOM THE AVERAGED CHARGES FOR THREE YEARS, 1848-50. Penmaenmaur set paving. Boulder paving. Macadam. Annual Annual Annual Street cost per Street. cost per Street. cost per square square square yard. yard. yard. pence. pence. pence. St. James' 08 Blundell .00 Church 8-60 Park Lane 29 Mersey . 79 Berry 8-33 Whitechapel 11 Flint . 73 Renshaw . 18-31 Dale . 38 Simpson 1-95 Gt. George 3-81 Castle . North John 17 26 Sparling Limekiln Lane 37 62 Park Lane Myrtle 16-29 5-82 Mar yb one 09 Burlington . M4 Hanover . 14-61 Water . 19 Soho . 18 Gt. Howard 12-54 Leeds . 06 Fox 1*01 St. Anne . 3-80 Averages 18 . 82 . . . 10-23 renewed. He deduced from the tables that, when the mac- adam was subjected to heavy loads, the annual cost of maintaining it was nearly equal to the cost of renewing it, and that the vertical wear amounted to 12 inches per year. Where the loads were light, although the traffic was great, the wear of the macadam was only about half as great as under heavy loads ; and that, under the same circumstances of traffic, the expense was reduced when the street was wide and open, and the moisture was rapidly evaporated by the action of the sun and the air. He concluded from the experience of Castle Street, Dale Street, and Marybone, that set-paving would last 20 years before requiring to be renewed. He decided, finally, that the total annual costs of maintenance were respectively, LIVERPOOL. For square-set paving Boulder paving . Macadam Maintenance annually. 28 pence per square yard. . 6-80 16-00 CHAPTEE IX. STONE PAVEMENTS OF MANCHESTER. THE early pavements for the streets of Manchester con- sisted of boulder stones brought from the sea-coasts of Wales, Westmoreland, and Cumberland, like those of Liverpool, which have already been described. But the construction of boulder pavement has long since been dis- continued ; the importation of boulders into Manchester ceased about the year 1840. There yet remains a con- siderable area of such pavement, particularly in the older streets of small traffic, in the outskirts. They are in process of renewal with granite sets. When finely broken, the boulder stones make an excellent "racking" or packing for paving sets. The principal thoroughfares, according to Mr. H. Hoyle, District Surveyor,* are paved chiefly with syenitic granite, or with trap -rock sets. The most usual dimensions of the sets are 5 inches, 6 inches, and 7 inches in depth ; from 3 to 3 1 inches in width, and from 5 to 7 inches long. The largest and deepest sets are laid in streets having the heaviest traffic. Cubes of 4 inches were tried, some years ago ; but it appeared that they were unfitted for resisting the lateral stress of the traffic, particularly on streets of considerable inclination. Stones are obtained from, 1st. The Welsh Granite Company, and the Portnant Granite * See a paper " On Street Pavements in Manchester," by Mr. H. Royle, District Surveyor; read July 9, 1876, before the Association ot Municipal and Sanitary Engineers and Surveyors. STONE PAVEMENTS OF MANCHESTER. 199 Company, near Carnarvon. 2nd. The Syenite Granite Company, Portmadoc. 3rd. From greywacke quarries at Penmaenmaur, North "Wales. 4th. From quarries at Clee Hills, Shropshire. 5th. From Newry, Ireland. It is said that the prices of sets from these quarries are nearly the same. The secondary streets, of little traffic, are paved with sets of millstone-grit, brought from the mountainous dis- tricts of Lancashire and Derbyshire. Macadam, which is regarded in Manchester as an expen- sive nuisance, is being gradually replaced with granite pavement. The paved carriage-ways are constructed in the follow- ing manner: A foundation, not exceeding 15 inches in thickness, is laid, of cinders and other hard material, including 3 inches of gravel as a bedding for the sets. The traffic is turned over this foundation until it becomes solid ; and the temporary gravel surface is renewed from time to time. When the surface has become sufficiently solid, the sets are bedded upon it, and well beaten, and they are racked with clean small broken stones, or with washed gravel. The joints of the stones are then filled up with an asphaltic fluid mixture composed of coal-tar, pitch made from coal-tar, gas-tar, and creosote oil ; in the pro- portion of 1 cwt. of pitch to 4 gallons of tar, and 1 gallon of creosote, proportions which are varied somewhat accord- ing to the quality of the pitch employed. The mixture is melted and boiled for from 1 to 2 hours, in a boiler adapted for the purpose, before it is poured into the joints. The use of the pitch-composition for closing the joints of the stone -sets was originally suggested by Mr. Eon- chetti, a chemist, of Manchester ; and the good reputation of Manchester pavement is, in a great measure, due to the employment of this compound for jointing, together with the solid foundation prepared for the pavement. The 200 STONE PAVEMENTS OF MANCHESTER. jointing is impervious to moisture, and possesses a degree of elasticity sufficient to prevent it from cracking. It adapts itself to all temperatures. Sometimes, as Mr. Koyle observes, in the heat of the summer, the asphalte, as he calls it, rises out of the joints, and slowly flows to the channels ; but this, it is said, rarely happens, and it hap- pens only when the asphalte has not been properly pre- pared. It follows that the foundation is always dry, and as no material rises from below, the formation of mud is pre- vented, and the cleansing of the surface is easily done. The foundations of hard core, prepared in the manner described, have given entire satisfaction ; and there has not been any need for the employment of concrete in foundations. The pitch obtained for the jointing of the pavement has, during the last 10 or 12 years, deteriorated in value for the purpose, since the demand for gas-tar for the manu- facture of aniline and other dyes, which has greatly in- creased, by the removal of that ingredient in virtue of which the asphalte retained its elasticity and adhesive- ness. The deficiency is partially supplied by the addition of the gas-tar, which is added to the pitch whilst it boils. Still, it is said, the asphalte laid in the old pavements was much better than that which is now made ; and the carriage-way pavements more recently constructed are not likely to last so long in good condition as the older asphalted pavements. The cost for repaving an old street with sets 6 inches deep, exclusive of the cost for foundation, and without taking credit for old material, is as follows : For stone sets, 6 inches deep . . 8s. 4d. per square yard. Labour, carting, gravel, asphalte . 2s. 6d. to 2s. 9d. Total 10s. lOd. to 11s. Id. The amount of the second item for labour, &c., is a mini- STONE PAVEMENTS OF MANCHESTER. 201 murn. In summer, when the men work for 54J hours per week; and is a maximum in winter, when the hours of labour are fewer. The average cost for the year is 11*. per square yard. Mr. Royle states that the cost for maintenance of asphalte-granite pavement is in simple proportion to the density of the traffic ; at the centre of the City, the cost is greatest. He adds that the average duration of the pavement, as first laid, is 14 years; at the end of which time it requires to be entirely relaid ; and that a sinking fund of 3d. per square yard per year would provide labour and material for maintenance in perpetuity. Macadam in Manchester. The cost of macadamising Albert Place, Manchester, in 1852, exclusive of founda- tion, comprising an area of 1,373 square yards, was as follows : Per sq. yd. s. d. 606 tons 9 cwt. of macadam at 10s. . 303 46 45 Labour , . , . . 11 18 8 2 Carting 4 . . . 24 15 4J 4J Total , , .339 18 6J 4 11J Taking the weight of one cubic yard of macadam at 1-J- tons, the bulk of the macadam used was (606-5 x f =) 539 cubic yards, which was equivalent to a layer 9 inches in thickness. The cost of maintenance, including scav- enging, was Is. l%d. per square yard per year. CHAPTEE X. OF GRANITE PAVEMENTS. IT has been stated that the greywacke sets from Pen- maemnaur are practically limitless in their resistance to wear ; and, as a corollary, they make dangerously slippery pavements. Next to these, the sets of Guernsey granite may be ranked. Mount Sorrel granite comes after ; and Aberdeen granite sets, the most generally popular, are the softest and the least durable ; that is, they wear the most rapidly. The order of resistance to wear of other stones, the results of Mr. Walker's experiments, has been given, page 126. The data for the vertical wear of granite pavement, forming the basis of the following table, No. 28, are collected from various sources*: TABLE No. 28. WEAR OP CARRIAGE-WAY PAVEMENTS IN THE CITY. Aberdeen-granite stones. Num- Amount of vertical wear. Num- ber of LOCALITY. Year of obser- Width of ber of years years to wear vation. stones. down. Total per Year. away 1 inch. Inches. Years. Inches. Inch. Years. Watling Street 1844 6 17 1* 065, or A 15-5 Great Tower Street . ? 6 9 1?, 139, or Y nearly 7-2 Bishopsgate Street Without 1850 6 20 2^ 105, or yV fully 9-6 St. Paul's Churchyard 1847 6 16 2 125, or | 8 Fleet Street . 1846 6 14 2 143, or ^ 7 Poultry . 1852 3 6 IJtolJ 230, or f 4-4 London Bridge 1851 3 9 2 222, or f 4-5 Blackf riars Bdg. (old) 1853 3 13 115, or i 8-7 Do. (Guernsey stones) 1853 3 13 i 019, orVfr 52 * Chiefly from the Proceedings of the Institution of Civil Engineers, Yol. ix., page 214 ; and vol. xiii., page 221. WEAR OF GRANITE PAVEMENTS. 203 It may be expected that there is a correspondence be- tween the intensity of the traffic and the rate of vertical wear of the several granite pavements. But, in forming such a comparison, there is another factor to be introduced, the width of the pavement ; for the vertical wear may be expected to be directly as the quantity of daily traffic, and inversely as the width of the street ; and the quotient of the traffic by the width, or the number of vehicles per foot or per yard of width, measures the intensity of the traffic. The following table, No. 29, has been constructed to show to what extent the relation holds good for such of the thoroughfares as have had their traffic observed. The traffic, showing the number of vehicles that traversed each locality in ten hours in one day, in 1850, already given in a previous table, is calculated for one foot of width of each locality ; and the last column contains the products of the relative intensities of traffic, in column 4, by the periods of year required to wear the pavement through equal vertical depths of one inch. These products represent proportion- ally the relative amounts of traffic which caused equal amounts of vertical wear : TABLE No. 29. WEAR OF CARRIAGE-WAY RELATIVE TO INTENSITY OF TRAFFIC. Aberdeen Granite Stones. LOCALITY. Approxi- ' mate average width of pavement. Vehicles in one day in 1850. Number of years to wear away one inch. Relative amounts of traffic for equal amounts of vertical wear. Total. Per foot of width. Sets 6 inches wide. Feet. Vehicles. Vehicles. Years. 1 Col. 4 X col. 5 j Gt. Tower Street 16 2,8*0 181 7'2 1,303 Bishopsgate St. \ Without i 22 (narrowest 4,110 190 9-6 1,824 St.Paul's Church- part.) yard 33 6,829 207 8 1,656 Fleet Street 30 7,741 258 7 1,806 204 WEAR OF GRANITE PAVEMENTS- TABLE No. 29 continued. Relative Approxi- Vehicles in one day in 1850. Number of amounts of traffic for LOCALITY, mate average width of pavement. years to wear away one inch. equal amounts of Tertical wear. Total. Per foot of width. Sets 3 inches wide. Feet. Vehicles. Vehicles. Years. Col. 4 x col. 5 Poultry 22 10,274 467 4-4 2,055 London Bridge . 35 13,099 374 4-5 1,683 Blackfriars Bdg. (old) . . 28 5,262 188 8-7 1,636 . 1 2 3 4 5 6 Supposing that the intensities of traffic be truly measured by the quantities in the 4th column of the table, and that the wear had been exactly in the inverse ratio of the in- tensity of traffic, the relative products in the last column should have been equal to each other. But the quantities in the 4th column can be but roughly approximate to each other, for they are based on but one day's observations ; besides, the traffic, though it was distinguished according to the number of horses to a vehicle, was not classified so as to distinguish goods traffic from passenger traffic. There is, nevertheless, a remarkable degree of correspon- dence in the products or measures of performance. The Poultry exhibits the highest product ; and that is con- sistent with the fact that the traffic consisted of a less pro- portion of waggon traffic than that of London Bridge or Fleet Street, and that the traffic of the Poultry was limited to a walking pace. Again, though there is a much larger proportion of heavy traffic over London Bridge, yet the heaviest traffic moves at a walking pace, and thus the pavement holds a high position in the order of relative duration. It is remarkable, too, that Blackfriars Bridge, with an intensity of traffic exactly one half of WEAR OF GRANITE PAVEMENTS. 205 that of London Bridge, shows, in column 5, about twice the actual duration ; and that, consequently, the products in the last column are substantially identical. The per- formance of Bishopsgate pavement is high ; and the highness of the rate may be partially attributable to the fact of the irregularity of width of the street, for here and there it expands into large bays which dilute the traffic to some extent at certain localities. The evidence in the last table supports generally Colonel Hay wood's conclusion, drawn from the results of his observations, that the duration of a pavement is " almost exactly " in the inverse ratio of the amount of the traffic over it. From the foregoing materials, some useful general data may be adduced for the wear and duration of granite pave- ments. For the seven pavements named in Table No. 29, the traffic and the wear were as follows : ABERDEEN GRANITE PAVEMENTS. Vehicles per Vertical wear foot of width. per year. Great Tower Street . . . 181 -139 inch. Bishopsgate Street Without , . 190 -105 St. Paul's Churchyard . . .207 -125 Fleet Street .... 258 -143 Poultry . 467 -230 London Bridge . , .374 -222 Blackfriars Bridge (old) , . 188 -115 Totals .... 1,865 vehicles. 1-079 inches. Averages . . , 266 -154 From these averages, it appears that the wear per 100 vehicles per day per foot wide, was equal to ( 154 X -HM =) 058 inch, or about -FT inch per year. To apply this rate of wear to the representative streets in Table No. 24, page 188, the daily traffics in 12 hours are here taken, for the sake of uniformity of comparison :* * These traffics are taken from a table cf Asphalte Pavements, in a following page. 206 WEAR OF GRANITE PAVEMENTS. Cheapside . Poultry Old Broad Street Moorgate Street Lombard Street Average 298 vehicles per foot of width. 363 103 188 147 220 The average duration of the granite pavements in these streets is given in Table No. 24 as 15-6 years. Take it at 15 years, as a round number. Then, the total vertical wear of the pavements when removed is (A X 15 X -f-S =) 2-06 inches, say 2 inches. Again, the stones, after having been redressed, last 20 years longer, in minor streets, for which the traffic averages 150 vehicles in 12 hours, per foot of width. Then the total vertical wear in the minor streets is (re X 20 X Jg-g- =) 1-875 inches, or, as a round number, 2 inches. TABLE No. 30. CITY OF LONDON : RECAPITULATION OF DATA ON THE WEAR AND DURATION OF ABERDEEN GRANITE PAVEMENTS. Sets, 3 inches wide ; 9 inches deep. Aberdeen Granite Pavements. Vertical wear. Duration. Inches. Years. Vertical wear per 100 vehicles in 12 hours per foot of width, per year . A 1 Total vertical wear in principal streets 2 15 Ditto additional ditto in minor ditto . 2 20 Total vertical wear when laid aside 4 35 Remaining depth when laid aside 5 Depth of new sets .... 9 Here it is seen that 3 -inch sets may be reduced by wear to 5 inches of depth, when they are treated as worn out, and are laid aside. The depth of a stone-set should be proportioned to the width at the surface. For insuring steadiness and re- sistance to the vertical and oblique forces applied to it by WEAR OF GRANITE PAVEMENTS. 207 the traffic, the depth of a set cannot be too great ; but it may be too small, for the resistance to displacement be- comes less as the depth is reduced in proportion to the width. A flagstone laid in a foot pavement is an extreme case in point : it can, if at all loose, be more easily tilted on its bed, by forces acting on it, which may act with leverage, than if it were one of a row of flags on edge placed side by side. For reasons of this nature, old paving stones, from 6 to 8 inches wide, are to be met with in London, having a depth of 11 or 12 inches. Modern granite sets, 3 inches wide, are made 7 or 9 inches deep ; and if 2 -inch sets were used, a depth of 6 inches would be sufficient for stability. Thence there is a saving in first cost by laying narrow sets. CHAPTER XI. STONE TEAMWAYS IN STREETS. MR. WALKER laid down two granite tramways on the Commercial Road, in the east of London ; they consisted of two uniform lines of " tram-stones," each 16 inches wide, and 12 inches deep, and 5 or 6 feet in length, placed at a suitable distance apart, to carry the wheels of vehicles. The interspaces are paved. Such tracks are laid in several narrow streets in the City of London, at Holyhead, and in other places. When they were first laid down in England, each wheel -track had kerbs at the outer sides; but in recent constructions these have been dispensed with, and vehicles may move freely from one part of the road to another. From the experiments of Sir John Macneil on the resistance of granite trackways, the resistance amounted to only 12 Ibs.. or 13 Ibs., per ton of gross weight : -less than half the resistance of granite paving. On an incline of 1 in 20, it is reported, the resistance was 132lbs. per ton, against a resistance of 295 Ibs. per ton before the tracks were laid. In Glasgow, a tramway consisting of cast-iron plates, 2 inches thick, 8 inches wide, and cast in lengths of 3 feet, is laid in Buchanan Street, on a gradient of 1 in 20, where it has been down for 40 years. Granite tramways* are in general use in Northern * For the particulars of Italian tramways, the writer is indebted to the valuable "Report on Stone Tramways in Italy," by Mr. P. Le Neve Foster, Jun., C.E., appended to the Report of the Committee of the Society of Arts on "Traction on Roads," June 25th, 1875. STONE TRAMWAYS. 209 T Italy, not only in the streets of the principal cities as Turin, Milan, and Verona but also in smaller towns, as Chi- vasso, Mcrtara, and Vigevano. They consist of two parallel lines of granite blocks, usually 24 inches wide, 8 inches deep, and 5 feet in length, bedded in a layer of sand. See Fig. 51. The lines are 28 inches apart, and the interspace, or footway for horses, as well as the other portions of the road- 1 way, are paved with cobbles obtained from the Po, or from other rivers. These stones should be egg-shaped, with a maximum diameter of from 3 to 4J inches, and a depth of from 4J to 5J inches. The roadway is usually formed with 1 a slight inclination downwards ~>p towards the centre. By this arrangement, the space between the trams serves as a channel to receive the surface water, and is provided with stone grat- ings, placed at suitable inter- vals, as shown in Fig. 52, by which the water escapes into the sewers. The surfaces of the trams are slightly in- clined towards each other, the inner edges being f inch lower than the outer edges; whilst the interspace is concave, having a versed sine or 210 STONE TRAMWAYS. depression of .1 J inches. The foundation of the roadway lip-. 52. 1 Ian of Tramway, Northern Italy, showing stone gulley. Scale l-100th. Tie 53 Tramways in Northern Italy. Intersection of Tramways at Eight Angles. STONE TRAMWAYS. 211 consists of a layer of screened gravel, about 6 inches deep, placed on the surface of the ground, and well rammed and 212 STONE TRAMWAYS. watered, so as to form a compact mass. Two inches of sand is laid on the gravel, as a bed for the paving- stones. The upper surfaces of the trams are dressed flat, and the ends square, to form close joints. The stone gratings for the gullies are 32 inches long, formed with three slots, 12 inches long and IJ inch wide. After the trams are placed, the other portions of the pave- ment are completed. After the surface has been well rammed with a wooden ram- mer, it is watered and covered with a bedding of sand J inch deep, so as to fill the joints by de- grees. On steep gra- dients, the surfaces of the trams are grooved diagonally. STONE TRAMWAYS. 213 SCHEDULE OF PRICES IN MILAN. Granite "blocks for trams, from the quarries of St. Fedelino, including labour for setting, per lineal yard ...... 10s. 3|cL Granite blocks from Mont Orfano per lineal yard 8s. Do. do. do. | not including setting J " " 6s ' 10d ' Paving with cobbles . . per square yard lOjd. Pit sand . . . per cubic yard Is. 6Jd. Eiver sand, from the Ticino or the Adda 3s. OJd. Screened pit gravel . . . ,, Is. IJd. Do. do. from the Adda . 2s. lid. Cobbles for paving, from the Adda or 1 theBrembo j Do. do. from the Ticino . 4s. 3d. Do. do. from the Lambro | or the Senese } 4s ' 10 ? d ' Stone gratings for gullies . . . each 13s. 9d. Wages : Stone mason . . per day 2s. to 2s. 9d. Labourer ... lOjd. to Is. 7d. Pavior ... lOjd. to Is. 7d. Labour : Dressing ends of granite ) blocks to form joints ) *> er S 1 Uare * ar per square yard 8d. apart, on trams . / From these prices, it is calculated that the cost of a line of tramway, comprising two lines of trams and the inter- mediate paving, amounts to 1,778 per mile. On wide thoroughfares, as at Milan, there are three or four lines of granite tramway. But, on some lines, the interspace is paved with Dutch clinker bricks. 214 STONE TRAMWAYS. A Member of the Committee of the Society of Arts pro- posed a new form of tramway and road, in which a line of asphalte trams is laid on a road formed of hard concrete. See Fig. 55. The design for a concrete road is based upon the prin- Fig. 56. Proposed Eoadway with Wheel- tracks of Asphalte. ciple of Mr. Mitchell's concrete macadam, already noticed. Another design by the same member is illustrated by Fig. 56. CHAPTEB WOOD PAVEMENT. "W GOD-PAVING, originally, was, like granite-paving, laid in large blocks, and without any width of joint or inter- space, like the hexagonal paving introduced from Eussia, measuring 8 or 9 inches across ; and like Carey's original paving, in blocks 6 inches wide, and 12 inches long, a sample of which remains to this day in Bartholomew Lane. But, like granite-paving, wood-paving best answers its purpose when it is restricted, in the width of sets or blocks, to 3 inches, 3J inches, or 4 inches, with inter- spaces, to afford the necessary stability for a minimum depth, and the necessary foothold for horses. Wood-paving is, by almost universal practice, laid with the fibre in a vertical position, on end. This is the best position for wear, for endurance, and for safety. According to the best experience of wood-paving, it should consist of plain rectangular blocks, solidly set upon a foundation of concrete, with water-tight joints. A wood pavement so constructed as to fulfil these conditions, gives satisfaction on the five points of convenience, cleans- ing, maintenance, safety, and durability. Unless the foundation be rigid, it is impossible to preserve a sound pavement ; and as in macadam, so in wood-pavement, the dogma of elastic action has been exploded by experience, for it was found that such a degree of elasticity as is afforded by the reaction of vertical wood-fibre, against a vertical pressure, is quite sufficient to absorb the shock 216 ^OOD PAVEMENT. of a horse's hoof, and to soften the strokes of loaded wheels. As with granite sets, so with wood blocks, the gauge of a horse's hoof is the measure of the proper maximum width. The most common width of blocks is 3 inches ; but they are sometimes made 3j- inches or 4 inches wide. The advantage of the narrower width consists in this, that, besides affording a more ready foothold, narrower blocks have more stability than wider blocks of the same depth ; and that, in short, the narrower the block, the shallower it may be made, so economising material. Mr. Carey's practice affords an illustration in point. His original blocks, which were 6 or 7 inches wide, were made 8 inches and 9 inches deep ; according to his latest practice, his blocks are only 4 inches wide, and they are only 5 inches or 6 inches in depth. The course of the practice of wood pavement is strictly analogous to that of granite-set pavement. It may be added, that the length of a block should be suitably proportioned to the width. A length of 12 inches for a block 3 inches wide, has been tried and found to be too much : the blocks were subject to splitting across. The normal dimensions of wood blocks in current practice are : width, 3 inches ; depth, 6 inches ; length, 9 inches. These are in the proportions of 1, 2, and 3. The next essential provision is an interspace of suit- able width between the courses of blocks, to make a good joint, affording, at the same time, foothold where the joints supply the only fulcrum. The blocks are set end to end in each course ; and the width of interspace between the courses varies from about J inch to J inch. Experience has decided that the narrower the interspaces the better for the durability of the pavement. The streets of the City of London afford the best and most exhaustive available experience of wood-pavements. GHAPTEE XIII. CAREY'S WOOD PAVEMENT. CAREY'S pavement, already described, page 17, was tlie first durable pavement that was laid in the City. In July, 1841, this pavement, of pine blocks, was laid in Mincing Lane, and in October and November, 1842, it was laid in Gracechurch Street, covering an area of 1,623 square yards. The blocks were from 6 to 7j- inches wide, 13 to 15 inches long, 9 inches deep for Mincing Lane, and 8 inches deep for Gracechurch Street. The foundation consisted of a layer of Thames ballast. The cost for the wood pavement of Gracechurch Street, including a foundation of ballast, but not excavation, was 14s. 2d. per square yard. In 1844, two years after the pavement was laid, it was "much worn," and there were " several short holes in various places." It had been for a considerable time in a bad condition, when it was relaid in September, 1847, five years after it was laid ; and a portion was relaid a second time in 1850, whilst ordinary repairs were made when required. In 1853, the paving was worn out, having been down 1 1 years, and it was re- placed with new paving of the same kind, at a cost of 12s. Sd. per square yard. The second pavement was relaid in 1857, four years after it was laid, and again in 1861 ; and in June, 1865, it was removed, having been down for 11 years and 7 months. The costs of laying and main- taining the two successive pavements were as follows L 218 CAKEY S WOOD PAVEMENT. TABLE No. 31. CITY OF LONDON: COST OF CAPEY'S WOOD PAVE- MENT IN GRACECHURCH STREET, 1842 to 1865. Costa. 1st Pavement ; duration 11 years. 2nd Pavement ; duration 11'6 years. BothPave-j ments ; duration 22 '6 years. per square yard. per square yardper year. per square yard. per square yardper year. per square yard per year. First cost, including') a foundation of J> ballast . . J Relays and repairs Total expenditure Deduct value as old } material . . . ) Net total cost Add, cost for gravelling Total cost of the pavem< s. d. 14 2 4 s. d. 1 31 4J s. d. 12 8 17 l\ s. d. 1 1 1 5| s, d. 1 2J 11J 18 2 1 1 8 1 29 9J 1 2 6| 1 2 1J 1 17 2 1 7 28 9J 2 5| 2 OJ 1J 3nt for 22-6 years, 1842-65 . .22 The foregoing statement exhibits the cost for two com- pletely new wood pavements laid and worn out in Gracechurch Street. They were succeeded by a third new pavement, on the same system, costing Us. 6d. per square yard, laid in 1865. This was taken up, half worn, in 1871, having been laid down 6 years ; and replaced by asphalte. The cost for maintenance during that time was 6s. lid. per square yard, or Is. l%d. per square yard per year. The cost for maintenance appears to have been a very variable item of expenditure. The costs for the three successive pavements are here brought together for comparison: CAREY'S WOOD PAVEMENT. Gracechurch Street. 1st pavement , 2nd ditto . 3rd ditto . Totals 28-6 Duration. Cost for main- years. tenance per squan yard per year. s. d. 11 4 11-6 1 5f 6 1 1! CAREY S WOOD PAVEMENT. 219 The fluctuations of cost may not be readily accounted for ; but it is clear that the generally increased traffic was one cause of the greater expenditure on the second and third pavements. The general total cost for maintenance may be taken at Is. per square yard per year. Carey's pavement was laid in Mincing Lane in July, 1841, and was taken up and replaced by pavement of the same kind, 9 inches deep, in August, 1860, having been down 19 years and 1 month. This was the oldest wood pavement then in existence in the Metropolis. During that period it was turned and relaid, and again relaid, the tops of the blocks having been cut off; and it was at other times exten- sively repaired. There is no other instance in this country of so long a duration in a wood pavement in a public thoroughfare. The second pavement, though not worn out, was taken up in August, 1873, to be replaced by asphalte. The following table, No. 32, shows the first costs and the costs for maintenance of Carey's wood pavement in Mincing Lane : TABLE No. 32. CITY OF LONDON : CAREY'S WOOD PAVEMENT IN MINCING LANE, 18411873. Costs. 1st Pavement : duration 19 '1 years. 2nd Pavement : duration 13 years. BothPave- ments ; duration 32'1 years. per square yard. per square yard per year. per square yard. per square yard per year. per square yard per year. First cost Relays and repairs Total expenditure Deduct value as ) old material . ( Net total cost B. d. 14 4 13 4 s. d. 9 8-4 s. d. 9 2 22 6| 8. d 8J 1 8f s. d. 8f 1 1J 27 8 1 1 5-4 0-6 31 8| 1 2 5J 1 1 10i OJ 26 8 1 4J 30 8} 2 4J 1 9J 220 CAUET'S WOOD PAVEMENT. In Mincing Lane, as in Gracechurch Street, the cost for maintenance of the second pavement was much greater than that of the first pavement, and for the same reason ; thus : CAREY'S WOOD PAVEMENT. Mincing Lane. 1st pavement 2nd ditto . Totals . Duration. Years. 19-1 13 32-1 Cost for main- tenance per square yard per year. s. d. 8-4 H The two series of Carey-pavements, in Gracechurch Street and Mincing Lane, have supplied valuable expe- rience of wood paving, and they have, therefore, been treated here somewhat in detail. The following table, No. 33, is given by Colonel Haywood, comprising the general results of the duration and cost of Carey's pave- ments in the City : TABLE No. 33. CAREY'S WOOD PAVEMENTS IN THE CITY OF LONDON, CONSISTING or LARGE BLOCKS : DURATION AND COST. Total First cost Cost for average STREET. Date when laid new. Duration. per square yard. repairs per square cost rer square yard. yard per year. LARGE TRAFFIC. Yrs. Mon. s. d. s. d. s. d. Cornhill . . { May, 1855 July, 1865 10 2 6 8 12 2 11 6 17 4J 8 9| 2 11 3 OJ Gracechurch \ Nov. 1853 11 7 12 8 17 1J 2 6| Street . J June, 1865 6 11 6 6 11 3 Of Lombard \ May, 1851 9 4 9 6 6 1 7i Street . . ) Sop, i860 10 7 9 2 20 2 2 9 CAREY'S WOOD PAVEMENT. TABLE No. 33 continued. 221 Cost for Total STRKETS. Date when laid new. Duration. First cost per square yerd. repairs per square yard. average cost per square yard per year. SMALL TRAFFIC. Yrs. Mon. s. d. s. d. s. d. Lothbury . j May, 1854 Aug. 1866 12 3 6 1 12 6 12 6 28 4| 3 6} 3 4 2 7i Mincing Lane \ July, 1841 Aug. 1860 19 1 13 14 4 9 2 13 4 22 6} 1 5J 2 5J Bartholomew 1 Lane . . J May, 1854 Aug. 1866 12 3 5 5 12 6 12 6 17 5| 3 Hi 2 5J 3 Oi Cost for Foundations included, but no Excavation, is made for the value of old material. No deduction All the second wood pavements referred to in this table were removed before they were worn out, and replaced with asphalte, except the last. Nearly all of them would, by relay, with the insertion of some new wood, have en- dured a few years longer. Taking all the pavements together, the following averages are deduced : Averages for Carey's Pavement. Streets of large traffic . ^treets of small traffic Total averages . Duration of Pavements.> Years. 9-05 . 11-33 10-19 Total cost per square yard per year. 2s. 7id. 2s. 4jd. 2s. 6d. It may be added that Carey's paving in Bartholomew Lane has recently (February, 1877) been extensively re- newed. It was newly laid in August, 1866, and the original paving, in its entirety, has thus lasted 10 J years. As just now mentioned, Carey's recent wood pavement consists of wood blocks, 4 inches wide, and 5 inches or 6 inches deep, according to the traffic of J&e street, and 222 CAREY'S WOOD PAVEMENT. 9 inches long. The ends only of the blocks are formed on Mr. Carey's original model, Fig. 11, with double- bevelled surfaces, salient and re-entering to the extent of 5-16 or -f inch, which come together for the purpose of preventing the shifting of the blocks, and of distribut- ing the pressure on one block over the contiguous blocks. The paving is placed on a bed of ballast or sand, 2 inches deep, laid on the old bed of the street ; and the joints, f inch wide, are grouted with lime and sand. It is needless to remark that this pavement can only endure when it is laid on a previously existing foundation ; other- wise, if laid with 2 inches of sand on a loose excavated bottom, it cannot be durable. CHAPTEE XIV. IMPROVED WOOD PAVEMENT. THIS pavement was introduced from the United States, where it has been known as the Nicolson pavement. It has already been noticed, page 14. The first piece was laid, in 1871, in the City of London; but the practice of the proprietors of this pavement has, since that time, been considerably modified. Their practice has been, after excavating to the profile and section of the street, to lay a bed of sand, fine mortar stuff, or dry earth, 4 inches deep, on the bottom. On this bed, two layers of 1-inch deal boards, previously dipped in boiling tar, were close-laid transversely and longitudinally. On these boards, wood blocks, 3 inches wide, 6 inches deep, and 9 inches long. also dipped in tar, were placed in transverse courses, the heading joints abutting, whilst the courses were inter- spaced by f-inch fillets of wood nailed to the floor, and through the blocks. The f-inch joints were made tight by a boiling mixture of tar and pitch poured into the spaces to a depth of an inch or two. The spaces were then filled up with dry ballast, which was rammed down by means of a flat caulking iron made for the purpose ; and filled up with the tar mixture. The surface was then strewed lightly with small gravel, over which boil- ing tar was scattered from a pail fitted with a flat- nozzled spout. Finally a sprinkling of sand was thrown over the surface. The gravel and tar coating was worked into the surface by the wheels of traffic. Spaces, J inch 224 IMPROVED WOOD PAVEMENT. wide, were left between the heads of the blocks in each course, and grouted. The tar and pitch mixture, in the proportion of 30 gallons of tar and 4 cwt. of pitch, were boiled together in an iron kettle, for at least four hours, until they ac- quired such a consistency that, when cold, the compound should become elastic and tough. When not sufficiently boiled, the mixture became brittle when it cooled, and was broken up by wheels and feet. The object of the elastic binding was to form a surface which should con- tinue water-tight, as had long been effected by the same means in the granite pavements of Manchester. The principal advantage claimed for this system was, that the flooring of planks formed an elastic foundation, and distributed the weight applied on one block over a large surface, whilst the additional elastic action mini- mised the wear of the blocks. There was an inconsis- tency in the principle of the construction, for it was presumed that the bed of sand would solidly support the flooring, at the same time that the flooring was expected to yield elastically under pressure. Again, so soon as leak- age, even to the slightest extent, commenced, by which surface-water was allowed to penetrate downwards be- tween the paving blocks, there was nothing to prevent its reaching to and saturating the substratum of sand; since the boards, although close-laid, were not tongued, and the water could pass through them with comparative freedom. The saturated substratum became mobile, and subject to movement under variations of pressure. Consequently, when a load passed over the surface, the boards, opposing an inconsiderable resistance to deflection, were pressed downwards by the load, into the quicksand, and they re- covered their normal position when the load passed away. In this manner, a pumping action was set up, and the sand and water, mixed with other loose matter at the bottom. IMPROVED WOOD PAVEMENT. 225 was pumped up to the surface in the form of mud and slime. Thus the pavement became gradually undermined, and the undermining process was accelerated by the form of the pavement itself, which presented a continuous dia- phragm under which the exhausting process was extended, as by a diaphragm-pump. It is scarcely necessary to add, that the wetter the weather, the greater was the action of undermining. But, in addition to the general liability to leakage of water through the pavement, there is a special difficulty in keeping it watertight at the kerbs, where it is compara- tively overhung and unsupported, and where there is, at the same time, a constant supply of water for penetration, so long as there is any water in the channels. A serious consequence of the flexibility of the pavement is the numerous breakages of the blocks by splitting, caused by the unequal stress and leverage of the load on blocks which are supported by a floor partly non-resisting and partly resisting. A small sample of the Improved Wood Pavement, 468 square yards, was laid in Bartholomew Lane, in the City of London, in December, 1871. This is a short street of very little and light traffic. This pavement, which was the first of its kind laid in London, continues now (1877) in fair condition, but many of the joints are open. The enduring power of the pavement has also been tested in thoroughfares of heavy traffic. In August, 1872, and January, 1873, an area of 6,066 square yards was laid, with 4-inch blocks, in King "William Street and Adelaide Place, on the approach to London Bridge, from the Statue. In the course of the following three years, the foundation was rather extensively undermined, by the causes which have already been described, particularly near the kerb on one side ; so much so as to cause the pavement to be deflected in many places to the extent of an inch under L 3 226 IMPROVED WOOD PAVEMENT. the wheels of passing vehicles, whilst the mud was seen bubbling up from below. The pavement was lifted about the month of January, 1876; and it was found that the substratum of sand had been partially washed away, and that, in the lines of regular traffic, the blocks had lost about -J- inch in height, on an average, by vertical wear. The pavement was renewed on the same system. The Improved Wood Pavement which was laid, in November, 1874, in Bishopsgate Street Without, in the City of London, is a more recent instance of the undermining influences set up under the action of heavy traffic, brought to a head during the past winter (1876-77), which has been a season of unusually heavy and continuous rain. The pavement, though it had not (in February, 1877) been down 2j years, had already been extensively undermined, and it was observed to be subject to deflection to the extent of half an inch, at some places, under passing wheels. At the same time, the collateral symptom of mud working up through the joints, became manifest. The obvious remedy for the pumping action through the pavement is to lay a rigid foundation under the pavement. The flooring, to be stiff enough, would require to be con- structed of two layers of 3-inch battens. But a less costly method of providing rigidity has been adopted in the con- struction of the new pavement now being laid to replace the defective pavement above referred to, which has been taken up. There is a thin bed of concrete, upon which a single layer of 1-inch boards is laid, and the blocks, 3^ inches wide, are laid, as before, upon the board- ing. The joints between the courses of the blocks are only -TQ inch wide ; they are filled for half the depth with an artificial asphalte, consisting of chalk, pitch, and coal- tar, boiled together, and are filled to the top with a grout- ing of gravel and blue lias lime. It may be added that, for width of 1 8 inches at each side of the roadway, at the IMPROVED WOOD PAVEMENT 227 channels, the boards are not laid, but their place is occu- pied by an additional depth of concrete, to afford a rigid support, so as to resist deflection, and the entrance of water at the edges. The carriage-way of Ludgate Hill was paved on the original system of the Improved Wood Company in November, 1873. The pavement has been worn into bumps and hollows, and is about to be entirely renewed. It has been proposed to renew the pavement by laying the wood blocks on a foundation of concrete 6 inches thick, with a 2-inch bed of sand intervening. CHAPTEE XV. OTHER WOOD PAVEMENTS. Ligno-Mineral Pavement. This pavement, known also as Trenaunay's system, was imported from France. It was the first system of wood pavement that was provided with a hard concrete foundation, and moulded exactly to the required curve of the road. The concrete is laid with cement or blue lias lime, and besides supplying a firm foundation, it closes the surface against underground exha- lations. Blocks of hard wood, oak, elm, beech, or ash, are laid on the concrete ; their upper and lower surfaces are cut at an oblique angle, of about 60, to the grain, in order that the fibre may be presented obliquely to the wearing surface, and that each block may be partially supported by the next block in the course upon which it leans. The courses are laid transversely, and the inclina- tion of the blocks is alternately to the right and to the left in the successive courses. The upper edges of the blocks are chamfered. A groove is cut horizontally along the sides of the blocks, near to the base, which is filled with the asphaltic mastic used for jointing so forming an addi- tional tie. The joints are only partially filled with mastic, and they are filled up with a grouting of lime and sand. The blocks are previously subjected to a process called mineralisation : they are placed in drying chambers heated with steam-coils, and exposed to a continuous current of air. When dried, the blocks are withdrawn, and plunged into tanks containing the mineral oils, consisting of hydro- OTHER WOOD PAVEMENTS. 229 carburets ; they quickly absorb the oils, with which, it is said, the wood becomes saturated, and are so rendered tougher and more durable. Though the employment of hard woods is a specialty of this process, mineralised firs also are used, and they are placed upright, with vertical fibres. But it is claimed that, in hard wood, a less depth of block suffices than in soft wood, and that thus the costs for the two classes of wood are equalised. The first piece of this pavement laid in the City of London, was laid in Gracechurch Street, near Talbot Court, in August, 1872. On a bed of concrete, 4 inches thick, mineralised elm blocks were laid, bedded in Portland cement. The blocks were 3 inches wide, 7j- inches long, and only 4 inches deep. The joints were partially filled with heated mastic, and filled up with a grouting of lime and sand. The pavement was relaid in 1875, after having lain three years, and a portion of it was replaced with new blocks. The next piece of paving was laid, in 1874, in Fore Street, which is entirely paved on this system. On a con- crete bed of blue lias lime, 6 inches thick, hard-wood blocks were placed, 3|- inches wide, 7J- inches long, and 4J inches deep. The wood used for nearly the whole of the surface was beech ; supplemented by blocks of oak and of elm. The lines of channel are formed throughout of elm blocks. Lastly, Coleman Street was, in 1875, paved with vertical blocks of mineralised fir, 3 inches wide, 9 inches long, and 6 inches deep, with f -inch joints ; on a concrete founda- tion 6 inches deep. Asphaltic Wood Pavement. On this system, originally patented by Copland, a solid concrete foundation, 6 inches thick, is laid to the curvature of the road ; It is composed of blue-lias lime and ballast, in the proper- 230 OTHEft WOOD PAVEMENTS. tion of 1 to 5 or 6. Upon this foundation, a coat of mastic asphalte, f-inch or f- -inch thick, is laid, as a bedding for the wood blocks. The blocks are 3 inches wide, 6 inches deep, and 9 inches long, of Baltic fir, laid in transverse courses, butt-jointed, with -r^-inch interspaces. The inter- spaces or joints are run up with melted asphalte to a depth of about 1^-inches. The correct interspacing of the courses is ensured by placing long strips of wood A-inch. thick, against each course as it is laid. The asphalte, when poured, remains for two or three minutes in a fluid con- dition, and it partially re-melts the coating on which it rests, and unites with it, whilst it adheres firmly to the blocks. The whole structure thus becomes solidified as one mass, and the joints are filled up with a grouting of sand and hydraulic lime. The grouting serves as a non- conductor of heat, and fixes the grit and gravel which it is usual to strew over a newly -laid wood paving, for the pur- pose of indurating the surface of the wood. The first piece of Copland's pavement in London was laid at the east end of Cannon Street, in 1874, on a concrete foundation, 9 inches thick, made with Portland cement ; a layer of mastic asphalte, f-inch thick, was spread over the concrete, and carried the wooden blocks, which were of soft wood, and had been steeped in tar. The blocks were laid in courses across the street, with f-inch joints between the courses. Each block had two holes ii -inch in diameter, and f-inch deep, bored in each side ; and when the joints were filled to half their depth with " liquid asphalte," the asphalte of course occupied the holes. The upper part of the joint was then rammed with screened gravel, and grouted with tar and asphalte, and the pavement covered with screened gravel and sand. The lateral holes have not, in later pavements on this system, been made in the blocks. It was found, no doubt, that they were an entirely useless refinement. OTHER WOOD PAVEMENTS. 231 The asplialtic wood pavement appears to have answered well. In Bristol, it has been laid for upwards of two years, in Broad Street and at the Exchange, and Mr. F. Ashniead, the Engineer, reports that it is wearing well. He maintains that the principal advantage of the pave- ment is "its imperviousness to water." It is stated that, on lifting a portion of this pavement after having been down two years, the foundation was found to be as dry as when the pavement was laid. Harrison's Wood Pavement. This pavement is analogous to the Asphaltic Wood Pavement. Upon a concrete founda- tion, it is proposed to place strips of wood, 2 inches wide by i-inch in thickness. Upon these, 3-inch wood blocks are to be placed, and heated asphalte is to be poured into the joints, penetrating under and adhering firmly to the blocks. On this system, an under coating of asphalte would be formed in sections by the successive pourings into the joints, instead of in large sheets as formed by the Asphaltic Wood Company. This system does not recom- mend itself. The multiplication of attachments by pieces of wood is objectionable. Hemorfs Wood Pavement. On a solid substratum of blue lias lime concrete, 6 inches thick, covered by a 2-inch layer of cement-concrete of a finer quality, a coat of ordinary roofing felt is spread ; the felt having been previously saturated with a hot asphaltic composition of distilled tar and mineral pitch. On this felt, as on a carpet, cushiony and impervious to moisture, blocks of Swedish yellow deal, containing resin sufficient for preservation, are laid. The blocks are 3 inches wide, 6 inches deep, and 9 inches long, are placed with the grain upright, and laid closely to- gether end to end, in rows across the street. The rows are also driven together and close-jointed with a strip of saturated felt in each joint. The width of the interspaces is thus reduced to the simple thickness of the felting, and 232 OTHER WOOD PAVEMENTS. does not exceed, if it even amounts to, a quarter of an inch. At intervals of every three or four rows, a row of blocks grooved along the middle is laid, to aid in giving foot- hold. The surface is dressed with a hot bituminous com- pound and fine clean grit. This pavement was laid in Oxford Street in two lengths, between Princes Street and Marylebone Lane, and be- tween Hereford Gardens and Edgware Eoad; and was opened for traffic in December, 1875. For the 18 months (to June, 1877) during which this pavement has been open, it has, according to the report of Mr. H. T. Tomkins, the Surveyor, given very great satisfaction, and it continues in excellent condition. A piece of Henson's wood pave- ment was laid in Leadenhall Street, in the City of London, in August, 1876. Norton's Wood Pavement. This pavement was evidently designed as a compromise, to do away with a foundation of concrete, whilst securing the advantage of a firm base. A thin bed of ballast is laid on the subsoil, and brought to the profile of the street, to receive the pavement. The pave- ment consists of slabs or structures, 7 feet long, and 3 feet wide, composed of blocks of yellow Baltic timber, 3 inches wide, 6 inches deep, and 7 inches long, fixed to a backing of 2-inch boards, with a strong bituminous cement. The joints are run with the same material. Between the slabs, the joints, J-inch wide, are filled with powdered rock asphalte, in a heated state, well rammed and levelled to the surface of the blocks. A small piece of paving on this system was laid at the east end of Cannon Street, in 1874. It has been occasionally under repair, and is now (Feb- ruary 1877) in bad condition: bumpy, and depressed at the edges of the slabs. Mowlem^s Wood Pavement. This pavement reverts to the simple type of wood pavement, formed of fir blocks, 3 inches wide, 6 or 7 inches deep, and 9 inches long, with OTHER WOOD PAVEMENTS. joints, transversely, J-inch wide, filled with a grouting of blue lias lime and sand. It is laid on a foundation of concrete made with Portland cement, 6 inches thick. The earliest piece of Mowlem's paving was laid in Duke Street, Smithfield, in June, 1873. A small piece was laid at the east end of Cannon Street, in September, 1873. Portions of the blocks were creosoted. Stone's Wood Pavement. A concrete foundation is formed, compressed by machinery. Grooves, f-inch deep, and 3 inches apart, are cut by machinery in the surface of the concrete. Wood blocks, 3 inches wide, 5 inches deep, and 6 inches long, are placed on the concrete, being shaped to fit into the grooves. The blocks are in transverse courses, with 1 J-inch joints, which are filled with gravel, and run with heated tar. A specimen was laid in King William Street, near to Clement's Lane, in July, 1873. Having been completely worn out by the traffic, it was taken up and replaced by the Improved Wood Pavement, in October, 1876. This new pavement was constructed with one layer of boarding on concrete and sand. GalrieVs Wood Pavement. Wood blocks, 3 inches wide, 6 inches deep, and from 7 to 11 inches long, are placed, with the intervention of a little sand, on a foundation of concrete, composed of Thames ballast and Portland cement or ground lias lime, of from 6 to 9 inches in thickness. Two small fillets are secured to one side of each block, so as to keep the blocks steady during the process of grout- ing. The joints are filled with a grout of lime and sand, and the whole surface of the pavement is covered with a layer of hoggin and sand. The eastern and the greater part of the south side of St. Paul's Churchyard were laid with this pavement in 1875. Wilsorfs Wood Pavement. This pavement is formed of blocks 3 inches wide, 6 inches deep, 8 inches long, laid on a foundation of blue lias concrete 6 inches deep, on 234 OTHER WOOD PAVEMENTS. which a |-inch layer of sand is spread, to bed the blocks. Two small fillets are secured by brads on the side of each block, so as to keep the blocks steady until the process of grouting is completed. The grouting for the joints con- sists of blue lias lime and sand. A piece of this pave- ment, 419 square yards in extent; was laid in Fetter Lane, in October, 1876. TABLE No. 34. CITY OF LONDON : WOOD PAVEMENTS. Date when the Situation. Name of Wood Length. Area. paving was Pavement. completed. Yards, Sq. yds. Bartholomew Lane . Carey's Wood 40 468 Jan. 1872 Birchin Lane . >j 28 77 June, 1866 Jewry Street . 42 253 Feb. 1872 Little George Street 24 148 Feb. 1872 Cannon Street ft> 9,051 Sept. 1874 Houndsditch,N.W. ) half . . . j 1,826 May, 1874 Barbican, a third part ~ 800 Oct. 1875 Bartholomew Lane . Improved Wood 48 392 Dec. 1871 Great Tower Street \ and Seething Lane 1 > 76 448 Aug. 1873 King William St., "I between Grace- church Street and > M 311 6,066 Jan. 1876 Cannon Street, and Adelaide Place . j LudgateHill . n 266 2,639 Nov. 1873 Aldersgate Street, &c. If 6,884 Sept. 1874 Barbican, a third part 606 Nov. 1875 St. Mary Axe, &c. . 1,662 July, 1875 Bishopsgate Street 1 Without . / Feb. 1877 GracechurchStreet, } near Talbot Court j Ligno-Mineral 27 440 Aug. 1872 Fore Street . j> 3,822 Dec. 1874 Coleman Street . N 2,291 June, 1875 OTHER WOOD PAVEMENTS. 235 TABLE No. 34 continued. Situation. Name of Wood Pavement. Length. Area. Date when the paving was completed. Duke St., Smithfield Houndsditch, S.E. \ half , . ) Wormwood St., &c. Mowlem's Wood * Yards. 134 Sq. yds. 676 1,832 689 June, 1873 May, 1874 June, 1875 Cannon Street, sam- \ pie at east end . j Barbican, a third part St. Bride Street Queen Street . Asphaltic Wood y> n 299 785 1,774 2,169 July, 1874 Oct. 1875 Nov. 1876 July, 1876 Cannon Street, sam- | pie at east end . j St. Paul's Church- \ yard, east end and > part of south side j Norton's Wood Gabriel's Wood 4,907 1874 Jan. 1876 In the Appendix is given a table by Colonel Hay wood, showing the condition of the wood pavements in the City of London, as existing on the 1st February, 1877. CHAPTEE XYI. COST AND WEAK OF WOOD PAVEMENTS. Cost. The first cost of wood pavements in the City cf London, including foundation, but not excavation, has been as follows : IMPROVED WOOD. Laid. First cost. King William Street . . 1873 18s. Od. per square yard Ludgate Hill . . . . 1873 18s. Od. Great Tower Street and Seething Lane . . 1873 16s. Od. Bartholomew Lane . , 1871 16s. Od. CAREY'S WOOD. Bartholomew Lane . . 1872 12s. 6d. Houndsditch . 1874 13s. 6d. ,. MOWLEM'S WOOD. Duke Street .... 1873 15s. 3d. Houndsditch .... 1874 17s. Od. The Asphaltic Wood Pavement has been laid in Bristol for 14s. 6d. per square yard. With regard to the cost for maintenance of wood paving, the Improved Wood Company contracted to maintain the pavements of King William Street and of Ludgate Hill, 1 year free, and 15 years at Is. 6d. per square yard per year; and the pavements of Great Tower Street and Seething Lane, 1 year free, and 15 years at Is. 3d. per square yard. These terms are equivalent to a total cost per square yard per year for 16 years, of 2s. 6^d. for the first and second COST. 237 streets, and 2s. 2d. for the third. But in King William Street, the pavement was renewed three years after it was first laid ; so that the actual total cost must be something considerably greater than the contract cost. The Asphaltic Wood Paving Company maintain the Bristol pavement for Is. per square yard per year for 1 years ; equivalent to a total cost for this period of 2s. 5d. per square yard per year. Messrs. Mowlem and Co. maintain the pavement of Duke Street for 2 years free, and 3 years at Is. per square yard per year. Also, the Houndsditch pavement, for 2 years free, and 5 years at 9d. per square yard per year. Mr. Carey maintains the Houndsditch pavement, for 2 years free, and 5 years at Is. per square yard per year. Mr. GL J. Crosbie Dawson* gives the cost of the Ligno- Mineral Pavement as follows : Per square yard. Hard wood, including concrete foundation . .14s. 6d. Maintenance for 10 years, after the first 2 years . Os. 8d. Yellow deal, including concrete foundation . . 11s. 6d. Maintenance for 10 years, after the first 2 years . Os. lOd. Footways, including foundations . . . .5s. Od. The price of Mr. Henson's pavement, as laid in Oxford Street, is 12s. per square yard, if only 2 inches of Port- land cement concrete is required; and 145. if with 6 inches of the same concrete. Mr. Ellice-Clark gave, in 1876, the cost for various pavements, if laid in Derby, from which the following particulars are deduced : Total in First cost Period of Cost annual cost ASPHALTES. DERBY. per sq. yard, maintenance, per year. per year. Limmer . . . 16s. Od. 13 yrs. Os. 7d. Is. 9|d. Val de Travers, with iron studs . .17s. 6d. * " Street Pavements," in the Journal of the Liverpool Polytechnic Society, 1876. 238 COST AND WEAR OF WOOD PAVEMENTS. WOOD. Carey . . 14s. 9d. 10 yrs. Is. 5d. 2s. lOJd. Asphaltic Wood . . Us. 3d. 13 Os. 7d. Is. 8Jd. Ligno-Mineral : Yellow deal , . 11s. 6d. 17 yrs. Is. l-'d. Is. 9|d. Hard wood . . 14s. 6d. 17 Os. 5Jd. Is. 3|d. Improved wood . . 15s. Od. 16 Os. 9jd. Is. 8d. GRANITE . . 13s. Od. 28 Os. 3d. Os. 8|d. In these estimates for Derby, except those for Carey's Wood, and the Improved Wood, the cost for a foundation of concrete is included, but no excavation. That wood pavement, properly seasoned and properly laid, with watertight joints, whether lime -grouted or asphalted, may last for many years without suffering decomposition, has been amply demonstrated by the expe- rience of Carey's pavements in the City of London, which were laid in 1841. In New York, it appears, the average duration of wood pavement is only four years, for "it is estimated that the wood pavements become worn out and useless at the rate of 25 per cent, per annum." The failure arises, not from simple wear, but from decay ; and ' ' the process of disintegration is much accelerated by the con- stant wear and pressure from the wheels of vehicles, and by the sudden changes of temperature which characterise our climate." * But the information is not of that pre- cise character which would make it of use for comparison in England. It has, indeed, been averred that the wood was unsound when it was laid. Wear. In the City of London, the abrasive wear of wood pavement in the lines of traffic has been recently ascer- tained, in a few instances, by direct measurement, by the writer and others, to be approximately as follows : * Communications from the municipal authorities of New York and Paris on the "Results of their Experience of Wood Pavements in those Cities." 1876. WEAR. 239 Time Vertical Wear. WOOD PAVEMENT. down. Total. Per year. 1. Adelaide Place, London Bridge Improved Wood Co. 3 yrs. f inch -29 inch 2. Bishopsgate St. Without . do. do. 2 J f 30 3. Tower Hill . Carey's , . . 5 3J 70 4. Bartholomew Lane . . do. , , . 5 f -15 5. Cannon St. . do. . 2J J -30 The amounts of wear in the first, second, and fifth in- stances are practically the same, namely, *30 inch per year. The widths and the daily traffics at the places where the measurements were made, may be taken, for the first in- stance, at 41 feet and 16,000 vehicles in 12 hours ; and for the second instance, 22 feet and 8,000 vehicles. For the fifth instance, in Cannon Street, the measurement was made just opposite Mansion-House Station, where there is much cross,traffic. The normal traffic is upwards of 6, 000 vehicles, and it may be taken as double, or 12,000 vehicles, at this spot, with a width of 36 feet. The comparison, then, stands thus Width of street. Traffic per day. 1. Adelaide Place . 41 ft. 16,000 vehicles, or 390 vehicles per ft. wide 2. Bishopsgate St. Without . . 22 ft, 8,000 or 364 5. Cannon Street . 36 ft. 12,000 or 333 362 From these data, it appears that, for a traffic of 362 vehicles in 12 hours per foot of width, the vertical wear amounts to '300 inch per year. The wear for 100 vehicles per day per foot of width, is, therefore, equal to (300 X i-J-g- = ) -083 inch, or iV inch per year. Experience is wanting, at least in England, for deter- mining the minimum depth to which wood-block paving should be worn, betore being removed. It is said that, in 240 COST AND WEAR OF WOOD PAVEMENTS. some instances, ligno-mineral and other wood blocks have been worn down to a depth of 2J inches when they were removed. Sufficient depth of block is required for stability, in wood-paving as in stone paving ; and, reasoning by comparison, it would appear that there may be some mistake in the first design of wood-paving, by limiting the depth, when new, to 6 inches. The motive for this limitation of depth, it can hardly be questioned, is one of economy in first cost; but when it is observed that Carey's earlier wood-pavements in the City of London, which had 8 inches and 9 inches of depth, have far surpassed in duration more recent pavements having only 6 inches of depth origi- nally, it would be well for wood-paviors to try back, and revise their designs on the basis of a 9-inch depth of block for heavy traffic. In forecasting the amount of vertical wear to which such blocks may judiciously be subjected, it may be premised that, for equal original dimensions, the amount of vertical wear may be at least as much for wood, as it may probably be, and as has been indicated, for stone-sets ; for the geometrical form of the wood-block admits of a more secure and more permanent joint being made than for the roughly squared stone-set. Empirically, then, the total vertical wear may be taken at 1 inch on a depth of 6 inches, and 4 inches on a depth of 9 inches, leaving stumps of 5 inches to be re- moved. But, it is scarcely necessary to add, that in order that a wood-pavement should be capable of undergoing such amounts of wear, it must be solidly and rigidly con- structed with thoroughly seasoned close-grained wood and watertight joints, and thoroughly drained. Moreover, time, the destroyer, is an element, and, in making esti- mates, it is necessary to impose a limit to the duration of wood, independent of the resistance to wear. In the following table, the limit of endurance is taken at 20 yeais : WEAR. 241 TABLE No. 35. WOOD PAVEMENT: ESTIMATED DURATION. Vertical wear, tV inch per year per 100 vehicles per day per foot of width. Height of blocks. 6 inches. | 7 inches, j 8 inches. 9 inches. Trnffic per foot of width per year. Maximum amount of vertical wear. 1 inch. 2 inches. 3 inches. 4 inches. Vehicles. Years. Years. Years. Years. 100 13 20 200 6 12 18 24 300 4 8 12 16 400 3 6 9 12 500 *i 5 7 9* CHAPTEE XVIL ASPHALTE PAVEMENTS. THE employment of asphalte for the carriage-way of streets was commenced in Paris, where, in 1854, the Eue Bergere was laid with Yal de Travers asphalte. In 1858, three sides of the Palais Eoyal were laid with the material, which was brought to the ground in the state of rock crushed into small pieces, and was heated and powdered by decrepitators. The foundation of concrete was about 6 inches thick ; the asphalte was 2-4 inches thick in the Eue St. Honore and the Eue Eichelieu, and 2 inches in the Eue de Yalois. The price was 20 francs per square metre, or I3s. 4d. per square yard. The success of these pave- ments was, it is said, complete, in spite of an unpropitious season ; and, in 1859, the Eue des Petits Champs was laid. But, the conditions were unfortunate. The street had been occupied by drainage works, and had been filled with loose soil to a depth of from 10 to 13 feet. On a 4-inch bed of concrete, 2 inches of asphalte was laid ; the work was carried on in rain and snow, and in January the street was opened for traffic. Cracks in the roadway immediately appeared, owing to the constant wet during the process of laying, and the settling of the soil. Each damaged part was repaired in succession. The price paid for the work was 15 francs per square metre, or 10s. \d. per square yard, including the concrete at 2 francs per metre, or Is. 4d. per yard. This price was to serve as a basis for future pro- posals for re-paving streets with compound asphalto :* to ASPHALTE PAVEMENTS. the standard of 4 inch.es of concrete, and 2 inches of asphalte* The conversion of the street-pavements into asphalte work, on a large scale, was commenced in 1867. Seyssel-rock asphalte was laid in the Hue Richelieu and elsewhere, and, according to report, answered perfectly well. ASPHALTE PAVEMENTS IN THE CITY or LONDON. The first pavement in asphalte, laid in the City of London, was laid of compressed Yal de Travers asphalte, in Threadneedle Street, near Finch Lane, in May, 1869. The next pavement was laid, in October, 1870, in Cheap- side and the Poultry. Asphaltic pavement was rapidly extended to other streets in the City, until, at the end of 1873, the total area of carriage-way covered by asphalte amounted to about 61,000 square yards, extending over 7,484 lineal yards, or 4 miles, of thoroughfare. From these data, it appears that the average width of car- riage-way so laid was 24 feet 4 inches. The details are given in a following table. Meantime, the various kinds of asphaltic pavements that have been laid and tried in the City of London, are now to be described. The par- ticulars for description have been gleaned, for the most part, from Colonel Hay wood's reports. Val de Travers Compressed Asphalte Pavement. Carriage- ways are constructed with a foundation of concrete, of from 6 to 9 inches in thickness, according to the traffic in the street. The rock in its natural state is broken up and reduced to powder by exposure to heat in revolving ovens. It is then lodged in iron carts with close-fitting covers, and brought upon the ground, taken out, laid over the surface, and, whilst hot, compressed with heated irons into a homogeneous mass without joints. The finished thick- ness is from 2 to 2 inches, according to the traffic, and the material is further compressed and consolidated by the M 2 244 ASPHALTE PAVEMENTS. action of the traffic : by as much as 20 or 25 pel- cent., according to the statements of the Company. The pavement laid in Threadneedle Street, already noticed, was laid on a foundation of concrete 8 inches thick ; a coating of mastic, -ineh thick, was laid over the concrete, and over this, 2 inches of Val de Travers com- pressed asphalte. The mastic was applied to expedite the work ; but it has not been used in any other pavement formed in the City. The next pavement that was laid, in Cheapside and the Poultry, was 2 inches thick, with a foundation of concrete, 9 inches thick. The cost was Is. 9d. per square yard for concrete, and 16s. 3d. for asphalte: together, 18s. per square yard. Val de Travers Mastic Asplialte Pavement. In laying this pavement, the rock is ground to a fine powder, and placed in caldrons on the ground, with an addition of from 5 to 7 per cent, of bitumen. When heated, it becomes semi- fluid ; then, 60 per cent, of grit or dry shingle is added to the melted mass, and after having been thoroughly mixed, the compound is spread over the bed of concrete in one layer. In March, 1871, George Yard, Lombard Street, was paved with this mastic asphalte, in a layer 1J inches thick, on a bed of concrete 6 inches thick. The price, including concrete, was 12s. per square yard. Limmer Mastic Asphalte Pavement. The rock is broken up and mixed with clean grit or sand, of different sizes, according to the place in which the pavement is to be laid. A small quantity of bitumen is added to the materials. The mixture is heated and made liquid in caldrons on the spot, and the compound is run over the surface of the foundation of concrete, and smoothed with irons to the required profile. It is run in two layers, the lower of which is made with grit of a larger size than that of the upper layer. The total thickness of asphalte, when finished, ASPHALTE PAVEMENTS. 245 is from 1J to 2 inches, according to the traffic. This asphalte was first laid in Lombard Street, and the pavement was finished in May, 1871. The concrete was 9 inches, and the asphalte 2 inches thick. The price of the concrete was 2s. 8d., and of the asphalte 13s. 4d.; together, 16s. per square yard. Barnettfs Liquid Iron Asphalte Pavement. This is made either of natural or artificial asphalte, mixed with pulverised iron ore or sesqui-oxide of iron, and a small proportion of mineral tar. The materials are heated in a caldron, and melted, then run over the surface, and smoothed like other liquid asphaltes. The usual thickness is 2 inches. It was first laid in Moorgate Street, in October, 1871. The price was 1 3s. 6d. per square yard, exclusive of foundation. Trinidad Asphalte Pavement. This is a compound of Trinidad pitch, broken stone, chalk, and other ingredients comminuted and mixed together. It is laid as a heated powder. A piece of this pavement was laid in Princes Street, in June, 1872, for which the price was 10s. Sd. per square yard. This pavement showed signs of wear directly it was opened for traffic. It was continually under repair, but it got cut to pieces, and was taken up in the end of October, and replaced by Yal de Travers Asphalte. Patent British Asphalte Pavement. This pavement does not contain any natural asphalte. It is composed of cer- tain oils, caustic lime, pitch, sawdust, and iron slag or grit ground to a powder. The mixture is heated in a boiler, and, in a semi-liquid state, run over the surface of a bed of concrete. This pavement was laid next the Trinidad asphalte, in Princes Street, in July, 1872, at the price of 12s. per square yard. By the month of December, it had suffered severely from the wear of the traffic, by which it was pulverised ; and, after having been repaired, ineffectually, it was taken up in the same month and re- placed by the Val de Travers Asphalte, 246 ASPHALTE PAVEMENTS. Montr otier Compressed Asphalte Pavement. This is a natural asphalte from the mines of Montrotier, in the department of Haute Savoie, France. It was laid as heated powder, in Princes Street, at the north side, and compressed with rammers in the same manner as the Val de Travers asphalte. The work was completed in August, 1872. Price, 15s. per square yard. The compressed asphalte of the Societe Frangaise des Asphalte is brought from the mines of Garde Bois in Seyssel. It was laid 2f inches in thickness, in Princes Street, in July, 1872, on 9 inches of concrete. The price was 14s. 9^. per square yard. Maestu Compound Asphalte. This asphalte is brought from mines at Maestu, in Spain. The pavement is formed of blocks of asphalte compressed into the shape of bricks, about 61 inches wide, 2f inches thick, and 13 inches long, laid on a bed of concrete, and the joints grouted with bitumen. This pavement was laid in Threadneedle Street, in December, 1871, at the price of 12s. per square yard. It failed at places, owing to defective concrete ; and, after having been repaired, it was taken up in January, 1872, li month after it was laid. Stones Slipless Asphalte. A foundation, 9 inches thick, is composed of a peculiar cement, compressed while in a fluid state by machinery, under a pressure of 112lbs. per square inch. A layer of asphalte, 2J inches thick, in a heated state, was laid on this foundation; it was com- posed of tar, cement, sand, and lead- ore, subjected to com- pression as the concrete was. A specimen of the pave- ment was laid in King William Street, near Nicholas Lane, in July, 1873 ; but it failed to sustain the traffic, and was taken up in September, after having been down only 5 or 6 weeks. Bennett's Foothold Metallic Asphalte. This pavement con- sisted of blocks about 2 feet square and 4 inches thick. ASPHALTE PAVEMENTS. 247 The lower portion 3 inches thick of the WOOES is com- posed of British manufactured asphalte and burnt ballast. The superstratum of the blocks, 1 inch thick, is composed of foreign asphalte, bone-dust, sulphate of lime, and cer- tain metallic substances, including lead. The strata were fused together. Portions of the superstratum, 6 inches in diameter, at 8 J-inch centres, were raised above the general level of the surface, and made hard by some process, with the object of increasing the wearing powers of the pave- ment, and to preserve its convexity. A specimen of the pavement was laid in King William Street, near Nicholas Lane, in October, 1873 ; but, failing to sustain the traffic, it was taken up in January, 1874, after having been down three months. LUlie's Composite Pavement. This pavement was com- posed of asphalte, wood, and broken granite. A specimen for trial was laid in Lombard Street. HcDonneWs Adamantean Concrete Pavement. This material was composed of blocks made of broken stone, chalk, lime, and clay, mixed with vegetable or mineral pitch or tar. A portion of the carriage-way of Carter Lane was, in April, 1869, paved with this material. The blocks were 18 inches long, 12 inches wide, and 6 inches deep; they were laid with f-inch joints, upon a solid foundation, and run up with an asphaltic composition. The price was 20s. per square yard. The pavement showed a serious degree of wear at the end of the first year after it had been laid ; and at the end of 18 months it had to be extensively repaired by the contractor, according to the terms of his contract. In 1872, it was in such a bad con- dition, that, not having been repaired by the contractor, it was taken up in September of the same year, after having been down 3 years and 5 months. The traffic of the Lane did not exceed 700 vehicles in 24 hours. Granite Pavements with Asphalte Joints, Trials were made 248 ASPHALTE PAVEMENTS. of asphalte jointing for granite sets, as in the pavements of Manchester and other northern towns. A portion of Duke Street, Smithfield, was paved with Carnarvon granite sets, 3 inches wide, 6 inches deep, and 6 inches long. They were laid with 1 -inch joints, which were filled with small clean pebbles, and run up with a composition of pitch and oil boiled together. The work was completed in May, 1868. A piece of Aberdeen granite pavement was laid for comparison at the same time, contiguously, in the same street. The sets were 3 inches wide, and 7 inches deep, and were jointed with lime-grout, in the usual manner. In 1871, the two pavements were in nearly the same condition ; but the lime-grouted pavement showed, if anything, the greater amount of wear : the difference having been ascribed by Colonel Haywood, no doubt correctly, to the comparative softness of the Aberdeen sets. Both pavements were removed in May, 1873, after a five-years' trial, and replaced by wood. Another piece of this kind of paving, formed with Aberdeen granite sets, 3 inches wide and 9 inches deep, was laid in St. Paul's Churchyard, at the east end, in April, 1870, and removed in December, 1875, after having been down 5 years and 8 months. During this time, no . repairs were done to the pavement. For the first three years, the surface continued in very good condition ; after- wards, it gradually fell into disrepair, and, in the last year, it was in a very bad state. Small repairs were not done to it, in the way customary with lime-grouted sets, by reason of the difficulty and expense of taking up and relaying small detached portions : which required a boiler and special apparatus to be brought to the ground. Colonel Haywood reported that all the asphalte-jointed granite pavements were noisier and less pleasant to travel over than the pavements grouted with lime. ASPHALTE PAVEMENTS. 249 The following table, No. 36, gives the length and area of carriage-ways paved with asphalte, at December, 1873. Since this date, several additional pavements in asphalte have been laid : TABLE No. 36. CITY OF LONDON : EXTENT OF ASPHALTE CARRIAGE- WAY PAVEMENTS, AT DECEMBER 31, 1873. (COLONEL HAYWOOD.) Number Pavement. of streets, Length. Area. &c., paved. Yards. Square yds. Val de Travers, compressed 20 4,185 34,876 Val de Travers, mastic 1 69 232 Limmer, mastic . 8 1,446 8,477 Barnett's, mastic . 6 1,705 16,544 Societe FranQaise, compressed \ (Seyssel) ... / 1 39 327 Montrotier, compressed 1 40 346 Total number of pavements . 37 7,484 60,802 NOTE TO TABLE. The Val de Travers compressed pavements and the Limmer-mastic pavements for carriage-ways, comprised in this table, were, with three exceptions, made with 2 inches of asphalte ; they were laid on concrete of from 6 to 9 inches in thickness. The exceptional thicknesses were 2J inches for Cheapside and the Poultry, as before noted; and 2|- inches for Gracechurch Street and Queen Street. The pavements of Barnett's Iron Asphalte were laid on 9 inches of concrete, and were 2J inches thick. The pavement of the Societe Franchise was 2f inches thick, on 9 inches of concrete ; and that of Montrotier was 2 inches thick. From the experimental results just recorded, it is ap- parent that no preparation of. asphalte compounds can compete successfully with simple asphalte as a material for roadways. Experience has amply confirmed Colonel Hay- wood's belief, expressed in his report of July, 1871 : "It has been seen," he says, referring to the Yal de M 3 250 ASPHALTE PAVEMENTS. Travers compressed asphalte and liquid asphalte, the Limmer mastic, and Barnett's iron-asphalte mastic, "that the modes of compounding and forming the asphalte pave- ments differ materially, the one being a mineral unmixed and laid in a state of dry heated powder, the other three being composed of asphalte largely mixed with grit, sand, and other ingredients, and laid in a heated liquid state. There is consequently a marked difference between their structures, and I incline to the belief that the asphaltes which will form the most durable pavements for carriage- ways are those capable of being laid and compressed in the shape of heated powder." In a report by Colonel Haywood, dated April 18, 1873, he described the condition of the pavements laid in the City of London, at the 1st March of that year, from which the following table, No. 37, has been compiled. Portions of some of the pavements were cut out under his direction, in order to ascertain whether they had lost materially in thickness. "There is no doubt, " he says, "that the asphaltes have somewhat diminished in thickness under the wear of the traffic, but owing to inequalities when they were first laid, and to the compression which takes place in all those laid in the shape of heated powder, it would be difficult to ascertain the exact wear without making a very large number of openings." By "holes" or "short holes," in the table, it was not meant that the asphaltes were worn down to the concrete foundation. The larger holes upon compressed asphalte are in most cases depressions caused by the traffic, and do not necessarily indicate surface wear. In the mastic as- phaltes, on the contrary, they are for the most part the result of disintegration and wear of the surface. Minute holes are noticeable in compressed asphaltes shortly after they are laid, which seem, after a time, to close up or disappear, whilst others open. The cause of this ASPHALTE PAVEMENTS. 251 " flow of solids," which Colonel Hay wood thinks may be due to the presence of foreign substances, or of moisture from the foundation, has not yet been satisfactorily ex- plained. The lengths of time during which these asphalt es had been down, at the date of the inspection, were 3 years and 9 months for one, 2 years and 2 months for two, less than 2 years for eighteen, less than 1 year for four, and less than 6 months for five. 252 ASPHALTE PAVEMENTS. TABLE No. 37. CITY OF LONDON : DURATION VAL DE TRAVEKS LOCALITY. Approximate average width of pavement. Vehicles in 12 hours in 1872-73 Time down at March 1, 1673. Total. Per foot of width Vehi- Vehi- Feet. cles. cles. Yrs. Mos. Poultry : next kerb, general surface 22 7,997 363 2 2| Cheapside : next kerb, general 30 8,949 298 2 2J surface. Old Broad Street .... 24 2,473 103 1 iu New Broad Street .... 24 1,515 63 Throgmorton Street 14 661 47 1 11 Milk Street 14i 516 36 1 lit 7 97 14 1 II 1 15 2,292 153 1 11 4 1 10 Old Bailey 22 2,903 132 1 10 Gracechurch Street . . 21 4,730 225 1 7i Finsbury Pavement and Moorgate 42 5,361 128 1 6J Street. Moorgate Street .... 32 6,000 188 1 6 Wood Street Hi 980 68 1 5| London Wall .... 24 2,500 104 1 4 J Threadneedle Street, near Finch 26 3,696 142 3 9 Lane. Ditto, east end .... 14 264 1 2} 264 1 lj Mansell Street .... 498 4 1 4 Mansion House Street . . . 46 13,767 300 8J Princes Street . , 24 5,628 234 3i Ditto ..... 234 2 1 George Yard (I 58! VAL DE 1 11 I ASPHALTE PAVEMENTS. 253 AND KEPAIR OF ASPHALTE PAVEMENTS, 1873. ASPHALTE (COMPRESSED). Repairs and Condition on March 1, 1873. Continually repaired at night. Many holes in the entire surface. Channels much worn. Extensive repairs required this year. Continually repaired at night. Many small holes in the surface ; mostly appeared within the last 6 months. Very extensive repairs required this year. No repairs done. A few small holes ; otherwise in good condition. No repairs done. Channels somewhat worn. One or two repairs done. One or two holes, channel somewhat worn. Otherwise in good condition. One or two repairs made. In good condition. Trifling repairs. In fair condition. No repairs. In good condition. Frequent repairs at south end, where there are many small holes. Many depressions towards north end. Generally 'in good con- dition; but not so good as others of this Company. Trifling repairs. Small holes over the surface. Generally in good condition. A few small holes. In very good condition. Trifling repairs. Channels worn ; a few small holes. Otherwise in good condition. Slight repairs. Some short holes. Generally in good condition. The first pavement laid in the City. Nine very small repairs made. A few holes. Generally in good condition. ! Narrow way frequently repaired. Channels deeply worn. Well- marked depression at middle by wheels. Many short holes at east end. Much wear. One slight repair. A few small holes. Otherwise in good condi- tion. Slight repairs. Surface worn ; otherwise in good condition. No repairs ; in good condition. TR AVERS (MASTIC). : No repairs. Small depressions on surface , but generally in good ! condition. 254 ASPHALTE PAVEMENTS. TABLE No. 37. CITY OF LONDON : DURATION AND LlMMER LOCALITY. 2 ?! ftO) ft Vehicles in 12 hours in 1872 -73. Time down at March 1, 1873. Total. Per foot of width Lombard Street \ * Feet. 17 Vehi- cles. 2,499 Vehi- cles. 147 Yrs. Mos. 1 9} Moorgate Street * ... 32 23 6,000 3,507 188 1 5J llf | Princes Street Princes Street BARNETT'S Moorgate Street ... 32 6,000 188 1 4J Carter Lane . Lothbury ..... 8 43 317 1,612 40 38 5| 4f Bishopsgate St. Within 28 6,048 216 2J MONTROTIER . | 24 || 5,628 1 235 || 6J | SOCIETE FRANCAISE DBS 21 |] 5,628 1 268 II 7J I ASPHALTE PAVEMENTS. 255 REPAIR OF ASPHALTE PAVEMENTS, 1873 continued. ASPHALTE (MASTIC). Repairs and Condition on March 1, 1873. Extensively repaired in last 3 months. Channels worn. Many short holes at middle of way. Extensively repaired in last 3 months. Considerable indications of surface wear in the middle ; a few bad holes and depressions. Otherwise good. Trifling repairs. Channels slightly worn. Middle of way con- siderably worn; surface rough. ASPHALTE (MASTIC). Extensively repaired. Many small holes, especially at south end, which was laid in very bad weather. Surface in good condition. Large portion badly laid, re-laid. Surface variable, some parts smooth, others worn and rough. No repairs. Numerous holes; and indications of loose structure or of wear. Eastern part rough, indicating inferior material or bad laying. Weather very bad when it was laid. ASPHALTE (COMPRESSED). | No repairs. In good condition. ASPHALTES (COMPRESSED). No repairs. Unusually wavy. Generally in good condition. 256 ASPHALTE PAVEMENTS. In compiling this table, an attempt has been made to gauge the intensity of the traffic by a calculation of the daily number of vehicles by which each street was tra- versed, per foot of the width of carriage-way. These traffics are placed in the following table, No. 38, for comparison with the loss of thickness, taken at per year, in the last column : TABLJE No. 38. CITY or LONDON : WEAR OP ASPHALTE PAYE- MENTS, 1873. VAL DE TRAVERS ASPHALTE (COMPRESSED). LOCALITY. Vehicles in 12 hours per foot of width. Time down at March 1, 1873. Loss of nominal thickness Total. Per year. Vehicles. Yrs. Mos. Inch. Inch. Poultry : next kerb 363 2 2f 44 to -94 20 to -42 Do. general surface . 363 n 375 to -75 17 to -34 Cheapside : next kerb . 298 2 2| 81 to -94 36 to 42 Do. general surface . 298 n f 17 Old and New Broad \ Streets . . . j 103 1 11J 31 16 Queen Street 153 1 10 f 41 Moorgate Street . 188 1 6J 118 08 /I 2J 1-00 (next 87 Threadneedle Street, east end and centre . 142 yi kerb) iV (general 06 \ surface Average for general i surface . . . / 208 19 LIMMER ASPHALTE (MASTIC). Lombard Street . . || 147 || 1 9J || | | -07 BARNETT'S ASPHALTE (MASTIC). Moorgate Street . . || 188 || 1 4J || & | -05 MONTROTIER ASPHALTE (COMPRESSED). Princes Street . . |l 235 II 6J Ij J to f I -35 to 1-40 The data contained in this table, No. 38, are too scanty for the purpose of making a useful comparison of the ASPHALTE PAVEMENTS. 257 wears of the" different asphaltes ; but a general relation may be observed between the traffics per foot of width and the vertical wears per year of the pavements in Val de Tr avers Compressed Asphalt e ; and the averages are as follows : Val de Travers Compressed AspTialte Pavement for Carriage-ways. Vehicles in 12 hours per foot of width .... 208 vehicles. Loss of thickness per year (say 5 inch) . . . -19 inch. It has been argued that the reduction of thickness is the result of simple compression under the traffic, and not sensibly of wear : an argument which is countenanced by Colonel Haywood, who found that asphalte is not so noise- less after it has been down for two or three months, as it is when first laid, and ascribed the difference to compression and solidification under traffic. Be that as it may, de- terioration of the surface is augmented by the action of the traffic; and, ultimately, the abrasion, and the conse- quent loss of thickness, must proceed in an increasing ratio, for it can scarcely be questioned that the material next the upper surface is originally more durable than what is under it. With the experience of nearly a year later than the date of his report from which the foregoing data are derived, but without sufficient experience to determine the actual durability of asphalte in the City, Colonel Haywood, after full consideration of all the circumstances, thought " that, without much repair, none of the asphalt es would last more than from four to six years ; and that, in the course of from six to ten years, the entire surface of all will have been renewed.' ' In explanation, it is to be said that the repair, however slight, of an asphalte carriage-way pave- ment, requires that entirely new material should be ap- plied in replacement of that which is in a state of dis- repair. That is to say, there is no such operation as 258 ASPHALTE PAVEMENTS. the relaying of an asphalte pavement. But, the asphalte that is removed, is utilised by being converted into a liquid asphalte for the formation of footpaths. In the Appendix, a table, extracted from a recent report of Colonel Haywood, gives the condition of the asphalte carriage-way pavements in the City of London, on the 1st of February, 1877. It may be noted that the works of two of the Asphalte Companies, the Montro- tier, and Barnett's, no longer exist. The asphalte pavements in the City of London are all maintained for periods of 16 or 17 years, by contract with the respective companies by whom they were laid. VAL DE TRAVERS ASPHALTE, (Compressed.) Cheapside . . . Poultry. Old Broad Street . Gracecliurcli Street Finsbury Pavement . Moorgate Street (N. end) Queen Street . . LIMMER ASPHALTE (Mastic). Moorgate Street (central) Lombard Street . . CornhiU Mincing Lane . . BARXETT'S ASPHALTE (Mastic). Moorgate Street (S. end) VAL DE TRAVERS (Mastic). George Yard . Concrete, inches. Asphalte. inches. 9 2 i 9 2 i 6 2 9 6 2* 2 6 2 9 2J 9 2 9 2 9 2 6 2 Completed. Dec. 1870 Dec. 1870 March, 1871 July, 1871 Aug. 1871 Aug. 1871 April, 1871 Sept. 1871 May, 1871 March, 1872 Aug. 1873 Oct. 1871 April, 1871 The terms of the contracts for these streets are given in Table No. 39, which follows : ASPHALTE PAVEMENTS. 259 TABLE No. 39. CITY OP LONDON: COST or ASPHALTE PAVEMENTS. Laid and Maintained by Contract. -1 11 First cost per square yard. Contract cost for Total cost per square yard. LOCALITY. maintenance per square : S a I "a Foun- da- tion. As- phalte. Total. yard. Total. Per year. VAL DK TRAVEES (compressed). Yrs. s. d. s. d. s. d. s. d. s. d- s. d. Cheapside . Poultry . 17 17 1 9 1 9 16 3 16 3 18 18 2 yrs. fi ee. 15 at Is. 6d. = 22 6 22 6 40 6 40 6 S3 Old Broad St. . 17 1 9 14 3 16 2 yrs. free. 15 at9d. =11 3 27 3 i n Gracechurch St. 17 1 9 15 3 17 2 free. 15 at Is. =^ 15 32 I 10J Finsbury Pave- ment 17 1 9 14 3 16 2 free. 15 at 9d. 11 3 27 3 1 7 Moorgate St. 17 1 9 14 3 16 11 27 3 1 7| Queen St. . 17 1 9 14 3 16 11 3 27 3 1 7^ LIMMER (mastic). Moorgate St. . 17 2 8 13 4 16 2 yrs. free. 15 at9d. =11 3 27 3 1 7J Lombard St. 17 2 8 13 4 16 11 3 27 3 1 7i Cornhill . 17 2 8 12 4 15 11 3 26 3 1 6 Mincing Lane . 17 1 9 10 3 12 11 3 23 3 1 4| BARNETT'S (mastic). Moorgate St. . 18 2 8 10 6 13 2 3 yrs. free. 15 at Is. 4|d.=20 7 33 9 1 1C VAL DE TRAVERS (mastic). George Yard . 10 1 9 10 3 12 10 free. 12 1 21 ! !i .- NOTES TO TABLE. 1. Excavation is not included. 2. The pavements to be given up as good as new. ASPHALTE PAVEMENT IN MANCHESTER. A specimen of Yal de Travers asphalte, 1,000 square yards, was laid in a street in Manchester, and was down two years. During 17 days in December, 1873, and January, 1874, 61 horses fell, or 3'6 horses per day: comprising 12 carriage horses and cab horses, 39 lurry 260 ASPHALTE PAVEMENTS. horses, and 10 cart horses. The surface was very slip- pery, and considered to be extremely dangerous except in exceptionally dry weather, or wet weather, and not suitable for the humid atmosphere of the Manchester district. The pavement was replaced with granite. CHAPTEE XVIII. OTHER PAVEMENTS. Metropolitan Compound Metallic Paving. This pavement is formed of blocks of a hard cement, having an iron frame bedded on the top enclosing a certain portion of wood. A specimen was laid in Threadneedle Street, in 1853; it did not prove satisfactory, and was replaced by the proprietors by a new and better specimen, in January, 1854. By the end of the year, it became very unsatis- factory, and was extensively repaired. In June, 1855, it became dangerous, and was finally removed. Cast Iron Paving. General Krapp's cast-iron pavement, an American invention, consists of cast-iron frames divided into sections, the divisions being sufficiently close together to prevent the admission of horses' hoofs. The runs and divisions are about 1 inch in width on the top, closely grooved to about an inch in depth. The frames are so designed that, to some extent, there would be a connection and mutual support throughout a paving laid with them. They are laid upon the usual substratum, in the same manner as granite paving, the interstices being filled with gravel, stones, or concrete. The weight of the pavement is 250 pounds per square yard ; it costs, when complete, about 185. per yard. A specimen was laid in Leadenhall Street, early in 1855. By the middle of the year, it got very much out of repair, and it was relaid on concrete, the interstices being filled with concrete ; but its condition was not very satisfactory. In February, 1857, it was in 262 OTHER PAVEMENTS. an extremely bad state of repair ; and in April, it was finally removed, after having been down for a little over 2 years. The Cellular Iron Pavement. This pavement consists of cast-iron blocks, 1 4 inches square, having a flat sole with a downward stiffening rib at each edge, and honeycomb reticulations on the upper side, which are open through the casting. The six-side cells so formed, are oblong, there being six cells side by side, one way, and four the other way of the block. The cells overhang the edges a little, and are the means of interlocking the blocks verti- cally when the blocks are placed in position. Thus, it was designed, the paving would rest solidly and immovably on a common substratum ; and any substance like sand or gravel thrown on the pavement, was to be driven through the cells by the weight of the traffic, and would in fact constitute a continuous packing. This pavement was tried in the Poultry in 1863. Poletti and Dimpfl's Artificial Granite Pavement. This pavement is made into blocks, 3 inches wide, 6 inches deep, 8 inches long, composed of ordinary clay, highly compressed and then burnt. Each block weighs about 12 pounds. The bed for the pavement is levelled and made up with ordinary ballast, and the joints between the blocks, J-inch wide, are filled up with the same material. Compound Wood-and- Stone Pavement. Mr. Newlands, in 1855, tried a system of paving granite sets and wood blocks in alternate courses, for the purpose of lessening the noise of granite pavement. A portion of Great Howard Street, Liverpool, was paved in this manner, and, for a short time, it appeared to answer the purpose. But the wood speedily became depressed below the surface of the granite, either by compression or by wear, and the gaps thus made between the stones had the effect of increasing instead of diminishing the noise. CONCRETE PAVEMENT. 263 Concrete Pavement. This pavement, designed by Mr. Joseph Mitchell, is laid on a bed of Portland-cement concrete, 3 inches deep. The bed may be made with broken stones or with gravel. It is allowed to remain 10 days for consolidation. Paving sets, 2J or 3 inches wide, and 5 inches deep, are built on the bed, with cement mortar, and the joints are filled up with cement grout. The best cement only is used, and the pavement becomes, in fact, a work of masonry 8 inches deep. A piece of this pavement was laid on George IV. Bridge, Edinburgh, in 1866, where the traffic was heavy and continuous; and, after 3J years of work, it was in excellent order, water running freely off the surface, whilst the wear was scarcely perceptible, and the surface was said to be much freer from mud and dust than ordinary pavement. The cost was said not to exceed that of ordinary pavement made with sets 9 inches deep. CHAPTEE XIX. COMPARISON OF CAKRIAGE-WAY PAVEMENTS. Comparative Costs. The experience of wood-pavements of the best construction, and also that of asphalte, for car- riage-ways, in England, is of recent origin, and the terms of a comparative statement of their costs must be approxima- tive. The results of various estimates already given, are here grouped together for easy reference, not as matured estimates, nor even as dd;a for direct comparison : Granite Pavement, in the City of London, excluding Foundation. Dur i tion . Total cost per year. Gracechurch Street (page 185) 25 years 7'92d. per square yard London Bridge (page 184) . 9 Is. 10}d. Five Thoroughfares (page 188) 15^ Is. 7d. Wood Pavements, in the City of London, including Bjllast Foundations. Carey's six thoroughfares . 10 years 2s. 6d. per square yard Improved "Wood four do. . 16 ,, (cont.) 2s. 4d. Asphalte Pavements, in the City of London, including Foundations of Concrete. Val de Travers ) , (Compressed) . ( 7 yrS< ^ '^ 7 * * per Squar yard Limmer (Mastic) . 17 Is. 4d. to Is. 7Jd. Barnett's ( ) . 18 . Is. 10d. An allowance, at the rate of Is. 9^. per square yard for the first cost of a concrete foundation, may be made, in esti- mating the addition to be made to the above-noted costs of granite pavements, to bring them into direct comparison with the other pavements. But, the costs as given for wood COSTS. 265 and asphalte are only nominal. There is reason for believ- ing that the total cost for construction and maintenance of wood-pavement of the best recent construction will be found to be much less than the costs above given for it ; and that the cost for asphalte pavement will be much greater than the costs above given for it. The cost of Yorkshire paving, 3 inches thick, laid in the City of London, amounted, in 1850, to from 6s. to 6s. 6d. per square yard ; and the average cost for repair amounted to about one penny per square yard per year. After remaining down for about 18 years in principal thoroughfares, the paving was removed and relaid in courts or in streets of inferior traffic, where it was esti- mated it would last from 15 to 20 years longer. The total life was taken at 36 years, which was about the same length of time as the life of granite carriage-way pave- ment at the same epoch. But it has recently, in 1875, been estimated that the Yorkshire stone foot-pavements of London, now last only 7 years in the busiest localities, or 12 years in localities of small traffic.* There appears to be an error in the estimate. The cost of supplying and laying 3 -inch Yorkshire flag-stones in Liverpool, before the introduction of Caith- ness flags, was from 3s. 6d. to 3s. 9d. per square yard. The following table is given by Mr. GK J. Crosbie Dawson, for the average costs of foot pavements : Per square yatd s. d. Yorkshire Flags . . . . .99 Asphalte, Val de Travera . , . .60 Blue Brick Paving . . . . 4 6 Gravel Paths ..... . 2 3 Comparative Slipperiness. Colonel Hay wood, in 1873, * The Builder, July 31, 1875, page 679 K 266 COMPARISON OF CARRIAGE-WAY PAVEMENTS. made exhaustive observations on the accidents to horses on carriage-way pavements in the City of London : granite, wood, and asphalte. The wood pavements con- sisted of the " Improved Wood," and the " Ligno-Mineral " pavements. The asphalte pavement was the Yal de Travers (compressed). The average of 50 days' observations showed that granite was found to be the most slippery, asphalte next, and wood the least. That, Granite was most slippery when dry ; safest when wet. Asphalte damp; dry. Wood damp; dry. The numerical data upon which these conclusions are founded, are summarised in the Table No. 40. TABLE No. 40. CITY OF LONDON : COMPARATIVE SLIPPERINESS OF CARRIAGE-WAY PAVEMENTS. PAVEMENT. Distance travelled before a horse fell. Dry. Damp. Wet. Average of 50 days. Granite .... Asphalte .... Wood (two kinds) Miles. 78 223 646 Miles. 168 125 193 Miles. 537 192 432 Miles. 132 191 330 446 58 Improved Wood . Ligno-Mineral Here it is to be observed, that the worst condition of pavement was dry granite, and the best condition was dry wood. The Ligno-mineral wood pavement is obviously far inferior, in point of safety, to the Improved Wood. Improved Wood may be taken as representative of other CONVENIENCE. 267 wood-pavements recently laid, in respect of safety to horses. Comparative Convenience. Wood pavements make the least noise ; and asphalte pavements make less than granite. Of the wood pavements, the quietest are those which are laid on concrete ; and the closer the joints, the quieter and steadier is the motion over the pavement. For, in all wood pavements, the fibre is turned over at the edges into the interspaces ; and a series of hollows is established between the blocks, which, in the more widely set pavements, give rise to unpleasant vibration in vehicles passing over them, like that produced by a stiff spring insufficiently loaded. Henson's pavement is the only one in which such vibration is prevented ; and it is no doubt the dryest, the smoothest, and the most silent of wood pavements, whilst it is likely to be amongst the most durable. Wood absorbs moisture, and is frequently damp when asphalte is dry ; but if it be reasonably clean, the damp- ness does not affect the safety or the comfort of the traffic. Colonel Haywood says that, although some streets in the City have been paved with wood for thirty years, no com- plaints of offensive smell, or of unhealthiness, have been made to the Commission. The Committee of the Society of Arts, on Traction on Roads, express, in their report,* very strong sanitary objections to wood as a material for pavements. Since that report was written the practice of wood paving has been much improved. As Colonel Haywood puts it, ' * in con- fined places, and under some conditions, wood might be objectionable. I have seen it decaying in confined places without traffic." * Printed in the Journal of the Society of Arts ', June 25, 1875. K2 CHAPTEE XX. CLEANSING OF PAVEMENTS. THE dust and mud of the streets of great towns con- sist generally of four things : comminuted stone, horse- dung, shoe-iron, and shoe-leather. House-refuse is not an unknown element ; it counts for something in the mass of articles that is collected in and removed from the streets and roads. Dr. Letheby, in 1867, analysed dry mud from the streets of the City of London, dried by exposure for many hours to a temperature of from 266 to 300 Fahr. At the same time, he analysed, for comparison, well-dried fresh horse-dung, and common farm-yard dung. The results of the analyses, for the mud of stone-pave- ments, are given in table No. 41. TABLE No. 41. COMPOSITION OF MUD FROM STONE-PAVED STREETS, HORSE-DUNG, OR FARMYARD DUNG. Dried at 300 Fahr, p - L Mud from stone -paved streets. CONSTITUENTS. horse- dung. yard dung. Maximum organic (d y weather). Minimum organic (wet weather). Average. Per cent. Per cent. Per cent. Per cent. Per cent. Organic 82-7 69-9 58-2 20-5 47-2 Mineral 17-3 30-1 41-8 79-5 52-8 100-0 100-0 100-0 100-0 100-0 CLEANSING OF PAVEMENTS. 269 The higher proportion of mineral matter in wet weather proves that in such weather the abrasion of stone and iron is greatest. Dr. Letheby estimated that the average proportions of stone, iron, and dung in the muds were : Horse -dung . , . .57 per cent. Abraded stone . . . 30 Abraded iron 13 100 The mud was so finely comminuted that it floated freely away in a stream of water. In the mud of wood pavements, the proportion of organic matter in the dried mud was larger than in the mud of stone pavements. It amounted to about 60 per cent. The amount of moisture in the street mud varied ac- cording to the state of the weather : STOXE PAVEMENT. Moisture. In the driest weather , . rarely less than 35 per cent. In ordinary weather . 48J- In wet weather ... 70 to 90 By far the greatest proportion of the detritus of mac- adamised roads consists of the worn material of the road ; and the important principle was early revealed by experience, " that the oftener that streets are cleansed, the less is the mud which is created and removed, whilst the attendant expenses are by no means increased, and the roads are kept in a better state of preservation." This principle is, besides, logically deducible from the fact that the worn particles, if left on the surface, act as a grinding powder under the wheels and the horses' feet, to reduce to similar powder the surface of the road ; and that the mud which is formed with the detritus, when rain falls upon and is mixed with it, operates 270 CLEAKS1NG CF PAVEMENTS. like a sponge in retaining the moisture upon the sur- face. The upper crust as well as the substratum, under these circumstances, become saturated with moisture, softened, and " rotten ;" just as a gravel footpath, hard and solid in ordinary weather, becomes sodden and pulpy when it is lapped for a time by a covering of half- melted snow. The macadamised carriage-way, thus re- duced, is exposed to rapid deterioration by the traffic, which increases in a highly accelerated ratio with the period during which the road is left uncleansed. The statistics of cleansing unanimously support these conclu- sions. Shortly after Sir Joseph Whitworth introduced his street-scraping machine in Manchester, it was ascer- tained from calculations made by the municipal authorities on the relative advantages of machinery and manual labour, that, by cleansing the macadamised streets with the Whitworth machine, three times a week, the quantity of mud produced on the surface was only one-fifth of what was produced when they were swept by hand twice in three weeks, and only one-thirteenth of what was produced when they were swept once a week. The following are the statistics referred to : MANCHESTER : Area of district, 5,500,000 square yards. Average area swept to Loads produce a removed. load. Square yards Swept by machine three times a week . 1,285 4,388 Swept by manual labour twice in three weeks. Township, 1841-42 . . 6,400 859 Swept by manual labour once a week. Township, 1838-39 .... 17,000 343 There is a want of harmony in this statement, from which it appears that by hand-scraping once a week more mud was collected than by scraping once in 10 days CLEANSING OF PAVEMENTS. 271 There was probably a difference of circumstances, as there was an interval of three years between the two series of observations. Be that as it may, it is obvious that the more frequent cleansing performed by the ma- chine, was instrumental, not only in keeping the streets in better condition, but in reducing the tear and wear of the streets, as represented by the decreased quantity of detritus which was generated and removed.* When a road is very soft, it should not be touched; or, at the most, only the loose liquid mud should be removed, by light sweeping, with, at the same time, a copious use of water. When a road is in a soft con- dition, and is covered with a thick tenacious mud which. adheres to the stones, it is a common opinion that the use of water injures the road by softening the surface, and causing a removal of the material. But the truth is exactly the reverse. A road in this condition can- not be effectually cleansed without water : by the free use of water, the mud is softened, is liquefied, and is easily removed without impairing the integrity of the surface, or removing the useful grit. The difficulty and cost of cleansing streets effectually by hand-labour, are materially diminished by the sub- stitution of the sweeping machine. The Whitworth machine has been much employed for cleansing streets. The apparatus consists of a series of broad brooms, usually about 30 inches wide, attached to a pair of endless chains which turn upon an upper and a lower set of pulleys suspended in a light wrought-iron frame, behind a cart, the body of which is near the ground. As the cart-wheels revolve, they communicate, by gear- ing, a rotary motion to the pulleys carrying the endless * See a paper " On the Present State of the Streets of the Metro- polis," Proceedings of tlte Institution of Civil Engineers, 1843 ; vol. ii page 202. 272 CLEANSING OF PAVEMENTS. chains and the brooms attached to them ; the brooms being sufficiently lowered so as to bear upon the surface of the street, they successively sweep the sur- face, and carry up the soil on an inclined carrier-plate, over the top of which it is dropped into the receptacle. The oblong arrangement of brooms, effected by the em- ployment of the endless chains, contrasts favourably with a simply circular or revolving broom ; for the ends of the brooms move most rapidly in the act of revolving on the lower pulley, whilst they are in contact with the ground, and they move towards the upper end at a less speed, whilst conveying the soil up the incline. In turning over the upper pulley, the velocity of the ends of the brooms is again raised to the maximum, and the mud is at this place stripped from them by a bar called a " doctor." If, for instance, the horse travels at the rate of 3 miles per hour, the speed of the brooms, whilst sweeping the ground, is 9 miles per hour ; whilst passing under the doctor, 6 miles per hour ; whilst they convey the mud up the incline at 3 miles per hour. The brooms are raised and lowered by hand, by means of a coiled spring and chain. The spring relieves the pressure of the brooms upon the ground, according to the inequalities. A strong horse is required to work and to draw the machine. The area of surface swept daily in Manchester, in 1847, by one of Whitworth's machines, with one man and one horse, varied from 14,000 to 20,000 square yards. The quantity removed amounted to from one to three loads per acre, according as the surface was dry or wet.* 4 The cost of working each machine was from 10s. to 1 5s. per day, exclusive of all the expenses attending the disposal of the sweepings, &c. Sweeping continuously, each machine * "Mechanical Street Cleansing," Proceedings of the Institution of Civil Engineers, 1847 ; vol. vi. p. 431. CLEANSING OF PAVEMENTS. 273 would cleanse, in a day, a strip of roadway 30 inches wide, and 20 miles long, making an area of 29,334 square yards. As a fact, more than 30,000 square yards have been swept in a day. But, a third, at least, of the whole time that the machine is at work, is consumed in conveying it to and from the place of deposit, and unloading. During the second half of the year 1843, each machine swept an average of 18,189 square yards per day, in Manchester. Against this performance it appears, from official returns, that in the years 1840 42, each able-bodied man em- ployed swept, on an average, 1,190 square yards per day, and each pauper swept about 600 square yards. From these data, it appears that the effective work of one machine equalled the work of 17 able-bodied sweepers. The comparative costs may be placed as follows : Sweeping by hand : Seventeen sweepers at 2s. per day . 34s. Cart, horse, and driver ... 8s. 42s. per day. Sweeping by machine, say . . 13s. Difference, saving by machine . . 29s. showing that the cost by machine was less than one-third the cost of manual labour. In Salford, the results of comparative trials of sweeping by hand and by machine, showed that the cost of cleansing was Square yards Per square per day. yard. By hand . . 674 8s. 8d. sweeping 15,904 . 10'18d. By one machine . 195 9s. 5d. 14,165 , 3-31 In Birmingham, according to Mr. Pigott Smith, the extent of surface of urban carriage-way, swept by eight machines, was 254,000 square yards. The whole of this area was macadamised, and was kept well cleansed by the eight machines. The area of surface averaged about 32,000 square yards for one machine, which would have been N 3 274 CLEANSING OF PAVEMENTS. traversed three times a week, allowing a traverse of 16,000 square yards per day. "I keep the brooms," says Mr. Smith, " suitable for the weather : if very wet, I use new brooms ; if dry and sticky, I use those that have been worn down ; if very sticky, I send out water-carts before the machines, and the effect is excellent. I endeavour to do all the sweeping between the hours of two and eight in the morning." In 1846, 8,339,760 loads of water were laid on the streets, at a cost of 1,788 : partly to prevent dust, partly to facilitate the cleansing. When the sweep- ing-machines were charged, they were drawn to the gully- grates, into which the fluid portion of the material swept up was allowed to drain off, whilst the grit remaining at the bottom was used as binding. Mr. Pigott Smith, referring to his experience of the roads in Birmingham, communicates some important data on the reduction of waste, and the saving of material for the repair of roads which are efficiently cleansed.* " Long experience has demonstrated that by the use of the sweeping machines, with the proper employment of water, a saving of full one-third of the material required for the repair of the roads has been effected in one particular district in the Borough of Birmingham. The following," he con- tinues, " is the result of an experiment to ascertain whether useful grit was or was not removed by water with machine-sweeping. On the 22nd of March, 1849, the Quadrant, Eegent Street, London, was covered with a thick layer of adhesive mud, which was causing great annoyance, as well as considerable injury to the road, but could not be taken up by scraping, without removing also much of the new stone to which it adhered. It was deter- mined to sweep one half of it whilst dry, and the other half after proper watering. This was done, and the * " Macadamised Roads," in the Proceedings of the Institution of Oivil Engineers, 1853-5-1 ; vol. xiii. page 226. CLEANSING OF PAVEMENTS. 275 sweepings were washed, to separate the refuse from the strong matter mixed with it. It was found that one-thii d of that which was taken up dry, consisted of coarse grit, which would have been useful on the road ; whilst one- twelfth part only of that which was removed in the form of liquid mud was stony matter, and it was so completely pulverised, as to be of scarcely any use. After the two portions of the road had been cleansed, the difference between them was very striking : that which was swept whilst dry was still covered with adhesive mud, which, together with the stones to which it adhered, was lifted by the wheels, the whole road being rough and uneven, whilst the portion which had been swept after watering was smooth and even. On the 24th, both portions were again swept, but only one quarter as much dirt was taken from that which had been water-swept as from that which, being in a sticky condition, had been swept without the addition of water. On the 26th, rain fell, and three times as much slop was taken off the part of the road which had not been swept on the 22nd. The preservative effect of water-sweeping by mechanical means, was demonstrated by the decidedly better condition of that portion of the road so cleansed." The detritus of the material of a granite pavement con- stitutes but a very small proportion of the total quantity of mud-forming dust. Colonel Haywood exemplified this proportion in an interesting manner, taking the instance of the granite pavement of London Bridge, 3 -inch Aberdeen granite sets, which was removed in 1851, after having been down nine years. The average loss of granite over an area of 3,950 square yards, he estimated, was equal to 2 inches of vertical wear. The total ^olume of granite worn away was, therefore, about 219 J cubic yards, assum- ing that the surface was a continuous mass of granite, though there was of course a considerable superficial area 276 CLEANSING OF PAVEMENTS. of joints. Assuming that the granite worn off was re- duced to the state of fine powder, it was increased in bulk probably one half, and its volume had been (219J x 1^- =) 329 cubic yards. Adding 5 per cent, for the loss upon stones removed and replaced from time to time, the total quantity worn off and reduced to powder, and carried away, mixed with the dust of the street, and mud, would only have amounted to 345*7 cubic yards for nine years, equivalent to a wear of *105 cubic yard about a tenth of a cubic yard per day. Whereas, the quantity of dust removed, daily, in dry, calm weather, was from 3 to 3 J cubic yards, over thirty times as much as the granite detritus. So much for horse-droppings and shoe-leather, which must have constituted 29-30ths of the total accu- mulation, independent of the contributions of house- refuse, in the inhabited streets. Seeing that, according to the results of these observa- tions, the granitic element of the refuse constituted little more than 3 per cent, of the whole, the evidence of the exceedingly greater wear of a macadam surface is strik- ingly exhibited in the contrast of the comparative costs for cleansing the two kinds of surface. Mr. Royle states that, in the spring of 1870, there were laid in Chester Street, Manchester, 10,000 square yards of granite pave- ment, and an area of equal extent of macadam. The condition of each surface was maintained uniform through- out the period of trial, by sweeping, for which the respec- tive costs were 9J^. per 1,000 square yards, and 5s. 2^d. per 1,000 square yards, in the ratio of 1 to 7. The average costs throughout the year, ranged in the ratio of 1 to 5, The great bulk of the excess for macadam, com- pared with pavement, must have been due to the vastly augmented wear of granite, even under equal traffics, since the dust and mud from other sources must have been the same as a constant quantity for both. CLEANSING OF PAVEMENTS. 277 In Heading,* which has about 23 miles of macadam roads, 17 miles are watered twice a day ; for which pur- pose 11 watering-carts and barrels, and 4 hand-watering machines, have been employed. One cart waters a length of 5,962 lineal yards, having an area of 23,849 square yards, twice a day, at a cost of 8s. for horse, cart, and man, and Is. 5d. for maintenance of cart, and harness, and shoeing ; total, 9s. 5d. per day. With one hand- machine, 23,740 square yards are watered twice daily at a cost of 2s. Wd. each for two men, and Id. for main- tenance j making a total daily cost of 6s. 3d. Area watered Cost Water delivered twice daily. per day. per square yd. One water-cart . 23,849 square yards 9s. 5d. '51 gallons One hand- machine 23,740 6s. 3d. 1-30 Watering the Streets of the City of London with Jet and Hose. In 1867, and 1872, experiments were made in washing the carriage-ways of Cheapside and the Poultry when laid with granite pavement, and afterwards with asphalte pavement. The asphalte pavement of Old and New Broad Streets was likewise washed experimentally. 'Hydrants were erected in Cheapside, at distances of 33 feet apart, and in Broad Streets at 140 feet apart. Six men were employed at each washing of Cheapside in the first trial, and 10 men in the second trial ; for which two jets of water were used during the time of the first trial. The work of the 10 men in the second trial was thus distributed: four in playing the jets, two in moving the hose from place to place, and four with brooms in sweeping the surface of the asphalte, and keeping the channels free from straw and larger refuse. These men also used "squeegees" (india-rubber sweeps) to dry the surface of the asphalte. Of the six * Mr. Ellice- Clark's Report, pages 25, 26. 278 CLEANSING OF PAVEMENTS. men employed on the granite pavement, four played the jets, and two moved the hose, and plied the brooms. TABLE No. 42. WASHING CARRIAGE-WAY PAVEMENTS WITH JET AND HOSE. KeLiilts of Experiments in the City of London. PARTICULARS. 1867. Granite pavement. Cheapside, &c. 1872. Asphal^e pavement. Cheapside, &c. Broad Streets. Length .... feet 2,000 2,000 1,436 Area of pavement washed square yards 9,792 10,353 3,671 Time washing 2h. 19m. 2h. 4m. Ih. 3m. Area washed per hour square yards 4,220 5,000 3,500 Water consumed . gallons 19,500 19,726 9,786 Do. do. per square yard . . . gallons 1-99 1-90 2-66 Cost for labour 9s. 10|d. 9s. 4s. lOfd- Cost for water, at 6d. per 1,000 gallons .... 9s. 9d. 9s. lOjd. 4s. 9d. Cost for labour and water, one day's washing 19s. 7|d. 18s. lOjd. 9s. 7|d. Cost for labour and water, per square yard; . pence 0241 0218 0315 Cost for labour and water, per square yard, per year ; pence 7-510 6-823 9859 Add for supervision, preparing apparatus, wear and tear, &c., 20 per cent. . 1-500 1 377 1-972 Total cost . 9-010 8-200 11-831 The granite pavement was more thoroughly cleansed by hose than in the ordinary way, by scavengers. The cost of washing asphalte was only 3 or 4 per cent, less than that for granite. Mr. T. Lovictts Experiments in Washing the Streets of London. In evidence before the General Board of Health, in 1850, Mr. Lovick stated that in surface-cleansing by a jet and hose, illustrated in Figs. 57 and 58, the quantity CLUANSING OF PAVEMENTS. 279 of water consumed was at the rate of rather less than 1 gallon per squaie yard of surface of carriage-way, working at " extremely low pressures." He mentions experiments made by Mr. Lee, at Sheffield, with "very high pressures," in which Mr. Lee effected the cleans- ing with a consumption of less than a third of a gallon per square yard. Mr. Lovick estimated that the foot- paths could be cleansed with a consumption of half a Fig. 57. Cleansing the Surfaces of Streets. gallon of water per square yard. He adds that the daily cleansing could be effected at a cost of 3^. per house per week, exclusive of the cost for water ; for which he allowed 28 square yards of carriage-way and 16 square yards of footway, per house. At this rate, the first item of cost per square yard of surface would amount to 3%d. per square yard per year. For water, the cost at Qd. per 1,000 gallons, would be \^d. per 280 CLEANSING OF PAVEMENTS. square yard per year. The total cost would be 5d. per square yard per year. Mr. Lovick, at the same time, estimated that the cost of cleansing by ordinary scaveugii g was nearly double Fig. 58. Cleansing close Courts and Alleys. the cost by jet and hose, whilst the jet had been shown to be much more efficacious in removing evaporating matter and filth.* * He was able " not only to cleanse the pavements by this means, but also to cleanse the walls from urine stains and other filth." CLEANSING OF PAVEMENTS. 281 It has been found, as the result of special observa- tions, that, For 4 loads of mud or slop removed from a macadam surface, 1 load is removed from a granite pavement ; and load is removed from a wood pavement. Cleansing in Paris. In Paris, since 1855, many attempts have been made to substitute sweeping machines for hand labour. But the inconvenience attending their use the need for a depot to hold them when not in use, the need for horses to work them, and the diffi- culties of continuously employing them in frequented thoroughfares, have hitherto prevented their being gene- rally employed. One of these machines, constructed by M. Tailfer, which worked very well, performed the work of 8 or 10 sweepers. It is fitted with a revolving brush, 5 feet 8 inches long, placed obliquely, and sweep- ing a breadth of 5 feet at one passage. In some comparative trials which were made, one man, with a good brush, swept 720 square yards of macadam pet hour, dry, lightly watered, without heaping the mud ; 480 square yards covered with liquid mud ; 360 square yards covered with thick mud. On a paved street, he swept from 600 to 820 square yards per hour. M. Tailfer' s machine performed the work of 7 men in the first case just recorded, 10 men in the second case, and 12 men in the third case ; but, in the third case, the work was not perfect. Taking all charges into the account, it appeared that the work done by the machine cost about the same as the work done by hand, but it was done much more expeditiously. CHAPTEB XXI. MOUNTAIN-ROADS. THE first thing to be observed by the engineer in lay- ing out a mountain-road, is that he should personally examine every possible route before deciding upon the line to be followed. An old-established line of traffic should not be hastily abandoned for one that presents greater engineering facilities, and apparently equal com- mercial advantages. Major James Browne* mentions an instance where, in one district of the Punjab, a new road was made from the hills to the plains, which, it was thought, would at once draw away all traffic from the old, circuitous, and greatly more difficult line. Never- theless, the old road is used, and the new one is not used, for the simple reason that, whereas both lines are thickly wooded, the trees on the new line are acacias and thorn bushes, whilst those on the old line provide excellent forage for the beasts of burden, for five days' march. Major Browne adds, that the natives of India have generally had good economical and local reasons for the selection of their old lines of traffic. The broad principle most generally applicable in the selection of mountain-roads, is, according to Major Browne, "The best line for a mountain-road is that on which the total sum of the ascents and of the descents between * " On the Tracing and Construction of Roads in Mountainous Tropical Countries." Proceedings of the Institution of Civil Engineers 1873-4; vol. xxxviii. page 67. MOUNTAIN-ROADS. 283 the extreme points, is the least." At elevations exceed- ing 8,000 feet, attention is to be given to the action of snow in winter ; for want of such consideration, roads, otherwise well laid out, are rendered impassable, for many weeks, by avalanches. In the Himalayas, the northern slopes are thickly wooded, where the southern slopes are bare. The wooded side should be selected ; for the trees break the force of the rain as it falls, and the road is better protected than it would be on the bare hillside. Where the road-trace crosses a deep chasm formed by a river, the approaches should be laid down stream, to have the benefit of the slope of the Fig. 59. Road-Trace approaching a deep Chasm. river bed. Major Browne illustrates this proposition, by drawing the horizontal line, n H, Fig. 59, through the point E, where the road is to cross the sloping bed of the river, A B, parallel to the bank, c D, of the vertical heighb E M. Draw E F and E G through E, making equal angles with the horizontal line. The down-stream trace clears the bank in the distance E F, which is shorter than the up-stream trace E G. The distinction, though obvious, is frequently overlooked. Heavy expenses and difficult drainage are often avoided by the use of zig-zags in the route. In marking out the formation-level, cuttings which would exceed 10 or 15 feet in depth should be avoided, if possible. In the stony 284 MOUNTAIN- ROADS. soil of a hilly country, no trustworthy information is to "be had by boring, for rock of the toughest and hardest de- scription may crop up where least expected. In tracing a Himalayan cart-road, length is the main thing sought for, to surmount the immense heights met with. Every possible foot of rise should be gained, and never lost. The steepest admissible gradient for an unmetalled mountain cart-road in India, is usually considered to be 1 in 18, or 5 -55 in 100. Where the trace rises steadily, the gradient is broken at every 500 or 600 yards by 100 feet or so, of slight counter-slope, not merely to ease the cattle, but also to break the drainage. Mr. Dobson, writing from his experience of road-making in New Zealand,* says that, whilst, in level districts, it is sometimes worth while to incur considerable cost for the sake of avoiding inclines steeper than 1 in 30, in hilly country, a gradient of 1 in 10 may often be considered a fair working gradient ; that a gradient of 1 in 9 is a limit which should not be exceeded for a cart-road, and that it is seldom worth while to increase constructive costs for the sake of attaining gradients flatter than 1 in 22. In India, the usual width of a mountain cart-road varies from 18 feet in open ground, to 12 feet along cliffs or in very difficult places ; the maximum gradients vary from 1 in 18 to 1 in 25. The metalled road-surface in cutting along a mountain- side, is usually sloped inwards from the outside, at the rate of 1 in 18. The metalling consists of a 9-inch layer of broken granite, kunkur rock (a concre- tion of carbonate of lime), or coarse slate shingle. The cost of the construction of a road in India, ex- clusive of metalling, according to Major Browne, may vary from 1,500, where the bridging is light and labour is abundant, to 2,700 per mile, where the excavation is heavy and the labour is scarce. An addition of 300 per * " Pioneer Engineering," page 61. MOUNTAIN-ROADS. 285 mile is to be made for metalling, which would be at the rate of about Id. per square yard. A portion of the Hindustan and Thibet road, anunmetalled mule-road, 7 feet in width, with gradients of from 1 in 6 to 1 in 4, cost about 500 per mile. ] I y The Grand Trunk road, in India, 837 miles in length, constructed to connect J i ^ distant parts of the Bengal Presidency, s] gti has been in use for upward of 25 years. It j j | is raised at every part, about 18 inches above the greatest known height of in- undations. The standard width of the surface was, in the first instance, 30 feet ; this was afterwards increased to 40 feet, with side slopes of 4 to 1. The central portion, 1 6 feet wide, is metalled with broken granite, or kunkur, which is laid 8 inches thick, and is rolled or beaten down < to 6 inches, by which means a smooth road is formed. When reduced by wear to a thickness of 4 inches, which is generally done in three years, the surface is picked over to the depth of half an inch, and a thickness of 4 or 5 inches of new metal is laid upon it ; and it is again rolled or rammed, and at the same time well watered, to bind and consolidate the new material. It is calculated that the whole depth of metalling is renewed every six years. The cost of the road is about 500 per mile, exclusive of large bridges. The cost of laying the metalling is stated as 162 2s. per mile, or 6 '22^. per square yard ; the cost for repair and main- tenance of metalling is 23 12s. per mile, or 1*38^. per square yard per year. 286 MOUNTAIN-ROADS. The usual section for a first-class road in India, like what has just been described, is shown in embankment, Fig-. 61. Roads in India : First-class Road in Heavy Cutting. in Fig. 60 ; and in heavy cutting in Fig. 61. Hill roads are shown in section in Figs. 62, 63, and 64. Fig. 62. Roads In India : Hill Roact Fig. 63. Roads in India : Hill Road. The vertical lines of cliffs present the most formidable obstacles to the formation of a road along their faces. In MOUNTAIN-ROADS. 287 Fig. 65, if A B c be the section of the cliff, and the rock be sofncienfcly hard and stiff, a half-tunnel like DBF may be lig. 64. Roads in India : Hill Ro. d. Fig. 66. Formation of a Road in the face of a Cliff. . 66. Formation of a Road in the face of a Cliff. blasted out ; but if it be too soft and rotten to admit of this 288 MOUNTAIN- ROADS. being done, the best plan, if the cliff be of any great height, B F, above the formation level, is to blow out the whole piece G E F by a large mine at E. Mining should not, as a rule, be employed where there is a chance of the strata being blown out downwards, according to the dip ; for a piece may be blown out, like the shaded portion, Fig. 66, when Fig. 67. Mode of formation off a Road in the face of a Cliff. much time and expense are entailed in rectifying the level. The general mode of attacking a vertical cliff, and of forming a half-tunnel, is shown in Fig. 67. The large blasts, 0, a, a, a, driven 8 feet in depth, at an angle of 45, are 7 feet 3 inches apart horizontally, and 5 feet vertically. The small holes #, #, &c., 3 feet apart and 3 feet deep, which are not fired, serve to determine and facilitate rupture at the proper level. These blasts, when fired, MOUNTAIN-ROADS. 289 generally blow out or loosen a piece like A B c D. The remaining space, B E F, is blown out in the same manner.* "It sometimes happens," says Mr. Dobson, "that ad- vantage can be taken of the natural stratification to econo- mise work in a long side-cutting. This was done by the Author in the case of a road over Evan's Pass, at Port Lyttelton, New Zealand. The descent of the Pass was on the side of a long volcanic spur, formed by a succession of lava streams, dipping at an angle of 1 in 12, the lower part of each lava stream being hard volcanic rock, whilst the upper portion was soft and easily worked. " The line was originally set out with a gradient of 1 in 17, which would have entailed a series of cuttings through the hard rock, and retaining-walls in front of the softer portions. By altering the gradient, however, to that of the lava-streams, a solid floor was obtained throughout, the retaining-walls were dispensed with, and the excava- tion was made chiefly in soft material. The alteration effected considerable saving in time and first cost, as well as in the cost of maintenance."! * These particulars are derived from Major Browne's paper, already referred to. f See Mr. Dobson' s work on Pioneer Engineering for valuable infor- mation on the tracing of roads in mountainous districts. OHAPTEE XXH RESISTANCE TO TRACTION ON COMMON ROADS. IN investigations into the resistance to wheels on com- mon roads, it is usual to construct a diagram showing a wheel on a horizontal plane in contact with a stone, over which it is to be pulled ; and the force required to pull it over the obstacle is calculated. But this is not the kind of re- sistance worth investigating, and it certainly has no relation to the kind of resistance which is usually opposed to the wheels of vehicles on inferior roads. The resistance is that of a medium distributed over the submerged por- tion of the circum- ference of the wheel, in advance of the perpendic- ular line drawn from the centre of the wheel to the plane of the road. Let a o I be such a drawn over Fig. 68. Rolling Resistance to Traction. the horizontal SUr- face c d e of the road, in the direction of; and let the road be of such a consistency that the wheel penetrates to the depth d I below the surface, leaving a track b g behind it. The arc I c is the submerged portion of the circumference, and it may be assumed to be identical with the chord of the arc, b c?. Now. the resistance is dis- ROLLING RESISTANCE. 291 tributed over the surface I c, and it may be taken as acting on this surface perpendicularly to the plane of the road, or vertically and directly opposed to the gross weight con- sisting of the weight of the wheel and the load upon it. To simplify the investigation, let it be supposed that the upper portion of the road is homogeneous, as clay or sand ; then, the resistance to penetration is nothing at the surface, and it increases as the depth ; and the upward resistance along the line of submersion b c, is a maximum at I and it vanishes at c, and the varying intensity of the graduated pressure may be represented by an isosceles triangle, of which the centre of gravity h, situated at one-third of its length, b h, from the base #, is also the centre of resistance, and therefore also the centre of pressure under the load ; and the radial line o h is the resultant of the pressure of the load, measured in force and direction by the vertical o i, and the traction force, measured by the horizontal line h i, or o k. But the vertical o i may be taken as equal to the radius o b, and the horizontal h i may be taken as one- third of the semichord of submersion c d\ whence the simple proportion, Load : tractive force : o b : : : radius of wheel : J semichord ; and the resistance to traction is equal to the product of the load by the third of the semichord, divided by the radius of the wheel. But the length of the semichord c d may be more easily determined by calculation from the measured depth of submersion d b. It is equal to the square root of the prc - duct of the segments into which the diameter a b is divided by the plane of the road c d e, or to ^/ a d X d b ; and the /rhole of the calculation is embraced by the equation, Tractive force o k = } X - *** (1.) o 2 292 RESISTANCE TO TRACTION. The equation is, no doubt, applicable, with a sufficient degree of accuracy, for any real needs, for calculating the resistance of gravel, loose stones, soft earth, or clay. The work done in compressing the material of the road is easily indicated diagrammatically, by supposing the wheel to advance through a, space equal to the semi- chord c d, or the length of submer- sion. Thus, in Fig. 69, the wheel a b is supposed to roll forward and to occupy the po- sition d It '. The Fig. 69.-Rolling Resistance to Traction. WQrk done j n CQm _ pressing the road is proportioned to the four-sided area b c c' I', comprised between the circumferential segments b c and I' c ; and this area is, by the properties of the circle, equal to the original rectangular area Id cV. Now. suppose a wheel ABA, Fig. 70, of larger diameter with the same grossweight, to travel over the same surface. It Fig. 70. Rolling Resistance to Traction. is obvious that, if it could sink to the snme depth, d b, as that for the smaller wheel, the length of immersion, d c, ROLLING RESISTANCE. 293 would be increased, and the rectangle d b X d c, representing work, would be greater than that performed by the smaller wheel in the first example. Such a supposition cannot be admitted : the depth of immersion d B, for the larger wheel, must be less than that, d b, for the smaller wheel, though the length of immersion d c, must be greater than that, d c, for the smaller wheel ; but not so much greater as if the wheel were sunk to the first depth d b. In fine, larger wheels sink less, but spread more, into the surface, than smaller wheels, in such proportion that the area of the rectangle representing work of submersion is constant for all sizes of wheels. In this instance, accord- ingly, the rectangle d b X d c = the rectangle d B X d c. It might be thought that, on this principle of the con- stancy of the work of submersion, in a soft road, the re- sistance to traction must be the same for all diameters of wheels. But, as the rectangle of work is spread over a longer space, d c, for the larger wheel, than the space, d c, for the smaller wheel, it follows, on the contrary, that the resistance or force of traction varies in some proportion inversely as the diameter, being less as the diameter is greater. This conclusion accords with experience ; but, though the actual law of variation may not be strictly deducible in the line of reasoning here traced, it is never- theless useful to carry the reasoning to its logical conclu- sion. Let a and A be the diameters respectively of the smaller and the larger wheels ; b and B the depths of im- mersion ; and c and c the lengths of immersion, or d c and d c respectively. As already stated, the areas of immer- sion are equal to each other ; or, b c = B C. (2.) Also, the values of c and c are, by the properties of the circle, expressible by the products V a b. and V A B, fo] 294 RESISTANCE TO TRACTION. all cases that need occur in practice ; and, by substitution in the equation (2), b\/al = BV/AB; (3.) and, squaring both sides, a b 3 = A B 3 . (4.) Finally, extracting the cube root of each side of this equation (4), the equation (5) is obtained, which may be developed into the proportion, b : B : : v/A : vV; (6.) showing that the depth of immersion varies as the cube root of the diameter. But, as e = B c, and I : B : : c : 0, then, C : c : : ^A~ : v^ (70 showing that the length of immersion is as the cube root of the diameter. It has already been seen that the force of traction is as the length of immersion ; therefore, finally, The circumferential or rolling resistance of wheels to traction on a level road, is inversely proportional to the cube root of the diameter. On this principle of resistance, it follows that, to reduce the rolling resistance of a wheel to one-half, for instance, the diameter must be enlarged to eight times the primary diameter. The deduction of M. Morin, mentioned at page 51, that the resistance varies simply in the inverse ratio of the diameter of the wheel, so that, for example, a wheel of twice the diameter would only incur half the resistance, has been generally accepted. But this deduction is not supported by the foregoing analysis of forces, and there is good reason for renouncing it, in the more recent experi- M. DEBATJVE S DATA. 295 ments of M. Dupuit. He placed model wheels or rollers of various diameters at the summit of an inclined plane, succeeded by a horizontal plane, on which they rolled down by the force of gravity and arrived at a state of rest after having expended the energy acquired in falling through the height of the plane. From these and other experi- ments he drew the following deductions : On macadamised roads in good condition, and on uni- form surfaces generally, 1. The resistance to traction is directly proportional to the pressure. 2. It is independent of the width of the tyre. 3. It is inversely as the square root of the diameter. 4. It is independent of the speed. M. Dupuit admits that on paved roads, which give rise to constant concussion, the resistance increases with the speed, whilst it is diminished by an enlargement of the tyre up to a certain limit. M. Debauve* submits the following data for resistance of vehicles on common roads, in which it may be remarked that the resistances decrease in chronological order, - owing, no doubt, as he says, to the progressive improve- ments in vehicles and in roads. The data are here given in English measures : 1st. On Metalled Roads : Count Rumford Gordon Coste and Perdonnet Navier Per cent. Per ton. . 7-7 or -13th of the wt. ; or 172-5 Ibs. . 6-2 or -16th 139 Ibs. b . .2-9 or -35th 65 Ibs. . 2-2 or -46th 49-3 Ibs. Roads : . 5-5 or -18th of the wt.; or 123-2 Ibs. e) . . 4-0 or -25th 89-6 Ibs. . 7-1 or -14th 159-0 Ibs. ; . .3-3 or l-30th 73-9 Ibs. Rumford Navier (walking pace) . Do. (at a trot) Coste and Perdonnet * Manuel de Vlngenieur des Fonts tt Chaussces, 9 me Fascicule ; page 31. 296 RESISTANCE TO TRACTION. TABLE No. 43. KESISTANCE TO TRACTION ON COMMON ROADS. (M. Debauve.) Draft, in pounds per ton of the weight. Vehicles. oj the wheels. oi the tires. Uniform metaJroad, Paved road. at a walk, 01 at a trot. at a walk. at a trot Cart feet, inches. 6 inches. 2 pounds. 72 pounds. 47 pounds. 63 6 1 3 69 46 jmuiUiU. v 6 2J 4-4 67 39 _ ( i 6 3 5-6 67 37 Waggon 6 5 6 65 40 Cabriolet 4 10J 2 81 54 76 Jaunting-car j 4 11 2 10 2 2 81 81 67 67 83 83 Stage-coach j 4 11 3 1J 5.2 5-2 65 65 36 36 45 45 M. Debauve makes the general deductions from this table, that the advantage of the pavement over the metalled road is considerable for waggons, is less for stage-coaches, and is nearly nothing for voitures de luxe, or the cabriolet and the jaunting-car. The efi'ect of the tabulated values is, in summary, as follows : VEHICLES. Waggon Stage-coach Cabriolet RESISTANCE TO TRACTION On metalled roads. On pavements. . 67 lb. per ton 38 Ib. per ton. 67 lb. 45 lb. . 81 lb. 76 to 83 lb. M. Tresca tested the resistance of an omnibus on Loubat's system, adapted with wheels for running on a common road. The experiments were made on an inclined street in Paris, in good condition, having ascending gradients of 1 in 55, one part of which was paved, and another part was macadamised. The frictional resistance, after the gravitation on the incline was eliminated, \vas as follows : EXPERIMENTS AT BEDFORD. 297 SURFACE. Macadam Pavement Frictional Gross weight. Speed, resistance. Tons. Miles per hour. Lbs. per ton. . 5-67 10-7 83 . 5-67 10-1 66 The resistance to traction, of agricultural carts and waggons, was tested at Bedford in July, 1874, by means of a new horse-dynamometer designed by Messrs. Eastons and Anderson.* The first course was a piece of hard road rising 1 in 430 ; it was dry and in fair condition, largely made of gravel. The surface was, in many places, some- what loose. The second course was along an arable field, growing oats, on a rising gradient of 1 in 1,000; it was very dry, and was harder than in average condition. The fore wheels of the waggons averaged 3 feet 3 inches, and the hind wheels 4 feet 9 inches in diameter ; the width of the tyres was from 2 to 4 inches. The weight, empty, averaged about a ton, and it was nearly equally divided between the fore and hind wheels. The cart wheels were, say, 4 feet 6 inches high, with tyres 3j- and 4 inches wide. The weight of the empty carts averaged 10 cwt. The loads were from 2 to 4 tons in a waggon, and 1 ton in a cart. The following results are deduced from the given data ; the speeds averaging 2J miles per hour : f Draft on road. Draft on field. VEHICLE. Total. Eeduced to a level. Total. Eeduced to a level. Per ton, Per ton, gross. gross. Pounds. Pounds. Pounds. Pounds. Pair-horse waggon without springs .... 159 43-5 700 210 2 pair-horse waggons with- out springs 251 44.5 997 194 Pair-horse waggon with springs .... 133 34-7 710 210 1- horse cart, without springs 49-4 28 212 140 * See a report of the trials in Engineering, July 10, 1874, page 23. f See Manual of Rules, Tables, and Data for Mechanical Engineers 1877, page 962. 3 298 RESISTANCE TO TRACTION. The results of Sir John Macneil's experiments on the tractive force required to draw a waggon on various kinds of road has already been given, page 52. He made other experiments on the traction of a stage-coach on a section of the Holyhead Eoad. The weight of the coach empty was 18 cwt. ; and the weight of seven passengers in addition, allowing 1J cwt. for each passenger,, was 10J- cwt. ; total weight, 28 j- cwt. The first part of the following table has been arranged for various gradients and various speeds, from the statement calculated by the experimenter. The second part of the table is added to show the net frictional resistance after the elimination of the resistance of gravity : TABLE No. 44. TRACTIVE FORCE REQUIRED TO DRAW A STAGE- COACH : GROSS WEIGHT 1*42 TONS. GRADIENT. Total tractive force required. Net frictional resistance per ton. Speed in miles per hour. Speed in miles per hour. 6 I 8 10 6 8 10 1 in 20 1 in 26 1 in 30 1 in 40 1 in 600 Ibs. 268 213 165 160 111 Ibs. 296 219 196 166 120 Ibs. 318 225 200 172 128 Ibs. 76 63 41 56 72 Ibs. 96 68 63 61 78 Ibs. 112 72 66 65 81 Averages . 62 73 79 The net frictional resistances at equal speeds, vary very much for different gradients, by some unexplained cause. They are a maximum for the steepest gradient, and a minimum for gradients of 1 in 30 and 1 in 40 ; for these they are less than for 1 in 600, which is nearly a level. The mode of action of the horses upon the carriage may have been an influential element. The averages show: FORMULA FOR RESISTANCE. 299 FOR A STAGE COACH: At 6 miles per hour, 62 Ibs. per ton frictional resistance. At 8 73 Ibs. At 10 79 Ibs. With these may be associated, for the purpose of deducing a formula, the resistance of a waggon on a good road, given at page 52, at, say, 2^ miles per hour, 44 Ibs. per ton, frictional resistance. Plotting the resistances per ton for the above four speeds, the following formula is deduced : Frictional Resistance to Traction of a Stage-coach. K = 30 4- 40 - R = the frictional resistance to traction per ton. V = the speed in miles per hour. NOTE. The formula is applicable for waggons at low speed. This formula is simpler than the formulas deduced by Sir John Macneil, page 53, and it is equally approximate and comprehensive. M. Charie-Marsaines made observations of a general character on the performance of Flemish horses drawing loads upon the paved and the macadamised roads in the north of France, where the country is flat, and the loads are considerable. The results of his observations, pub- lished by M. Debauve, are here given in English measures, m Table No. 45. TABLE No. 45. PERFORMANCE OF HORSES ON ROADS IN FRANCE. (M. Charie-Marsaines.) Work Season of the year. Description of road. Weight per horse. Speed in miles per hour. done per Ratio of paved hour, in road to tons macadamised drawn one road. mile. Tors. Miles. Ton-miles. Winter | Pavement . Macadam . 1-306 851 2-05 1-91 2-677 1-625 | 1-644 to! Summer \ Pavement . Macadam . 1-395 1-141 2-15 2*16 3-027 2-464 } 1-229 to 1 300 RESISTANCE TO TRACTION. The table shows that there is a clear inferiority of per- formance in summer ; and that the pavement is superior, even in winter, to the macadam in summer Ratio. Pavement, summer , . 3-027 tons drawn one mile, 1-86 winter . . 2-677 1-65 Macadam, summer . .2-464 1-52 winter , , 1-625 1-00 The average daily work of a Flemish horse in the north of France is, on the same authority, equivalent to 21-82 tons drawn one mile, in winter ; and to 27*28 tons drawn one mile, in summer ; giving a mean for the year, of, say, 25 tons drawn one mile. The horses are powerful, and the roads are easy. M. Charie-Marsaines observed that the harness lasted six years, over pavement ; and only five years over ma- cadam. Also, that the waggons lasted 7 years on the pavement, and 9 years or the macadam : whilst the horses lasted much less time on the macadam than on the pave- ment. The greater mortality on the macadam is ascribed to the inhaling of the fine silicious dust which rises from it. It is stated by Mr. D. K. Clark,* that a good horse can draw a load of 1 ton at 2^ miles per hour, for from 10 to 12 hours a day; equivalent to (1 X 2 X 10 =) 25 ton-miles per day. This is the same performance as is above given by M. Charie-Marsaines. * Manual of Eules, Tables^ and Data for Mechanical Engineers^ 1877, page 720. APPENDICES. APPENDIX I. ON POLLING NEW-MADE POADS. BY GENERAL SIR JOHN F. BURGOYNE, BART. [This paper, written in 1843, is valuable now, and is here reproduced from tha fourth Edition of the present work.] THE importance of rolling roads, either newly constructed or when subjected to extensive repairs, seems never to have been duly appre- ciated. Lines of any length of new-laid broken stone may be deemed nearly impracticable to ordinary traffic ; the worst and most hilly old roads are always taken in preference to the new roads while in that state, although the latter may be much shorter, and with very improved levels. At length the old road is shut up, carriages are forced to take the new, occasioning the greatest inconvenience and drawback to the in- tercourse for perhaps a year or more, a great wear and waste of the material, and a considerable expenditure in watching and maintenance, until the material, or what remains of it, shall be finally consolidated, and even then in a very imperfect form, unless great pains are taken with it. The rolling is, in fact, effected, but in the most distressing and expensive manner, and by carriages and horses very ill adapted to it. These evils may be entirely prevented, the road put at once into good working condition, and, certainly, a considerable expense even- tually saved, by thorough systematic rolling ; nor ought any road to be considered as made until that operation shall be completely effected. Three reasons have probably operated to prevent this principle having been acknowledged and acted upon. 302 APPENDICES. 1. Because the traffic on the road will, sooner or later, do the work, thereby apparently reducing, in a small degree, the cost of the original construction or repair. 2. Because a roller is not usually at hand, and, from its weight and unmanageable character, it is most inconvenient and expen- sive to be removed from one place to another, so that in most jases one would have to be constructed for the purpose, and then be useless. 3. Much uncertainty, as yet, as to the best manner of operating, its efficacy, and the expense. The first reason is founded completely on error ; it is manifest that this manner of completing the road by the traffic is most inconvenient, and occasions enormous sacrifices by the parties using the road, and consequently a great loss to the public in general ; nor can there be a doubt but that the actual expenditure on the subsequent early maintenance oj the road itself is greater than would be incurred by at once operating thoroughly with the roller. With regard to the second reason, there are many ways in which the objection can be greatly alleviated. Although there is some justice in the third, and that the most per- fect mode of proceeding is not yet perhaps understood, there is so much useful effect to be produced by any, that it is surprising that it has not been reduced to just principles by experiment, and generally adopted. The practice of rolling has been rare, and almost entirely confined to gentlemen's demesnes, and occasionally to the macadamised road- ways in some cities ; but in the latter, it is believed, without the appli- cation of sufficient means for the purpose. There are certain considerations which may serve as guides to arrive at just conclusions with regard to this proceeding. 1. A roller should not be too heavy in proportion to its bearing surface, or, instead of binding the material in the position and form laid down and desired, it will press it more or less into the substratum ; much of the material will thus become useless, and it will be very troublesome to obtain the necessary resistance for the consolidation. 2. It must not be too light, or the effect will be too small ever to gain the object fully ; or at any rate without an extent of operation that would be very costly or inconvenient. It is believed that the ordinary rollers are too light, which may have thrown the practice into disrepute. For the Dublin streets they have a roller of two contiguous cylinders, each of 4 feet diameter, and 1 foot 6 inches in width, APPENDICES. 303 making in all a "bearing of 3 feet ; it weighs 2 tons 3 cwt. ; only two horses are attached to it, but the work is exceedingly heavy. It is applied to successive layers of material, in new formations, and about an inch of gravel is worked into the upper layer or surface. It is said to consolidate the roadway very effectually, but might probably be improved by adding to its weight.* From other recorded trials, however, there is reason to believe, that a road roller should not be lighter than 28 cwt. for every 12 inches lineal of bearing on the road; that is, if 4 feet wide, that it should weigh 5 tons 12 cwt. ; if 3 feet, 4 tons 4 cwt., &c. ; and that it should only be applied to the upper surface of all. A roller somewhat heavier than 28 cwt. per foot would be more effective, but it is better after that limit to gain the object rather by adding to the number of times passing over the surface than incur the inconveniences of the heavier machinery. This is one very interesting point to prove, namely, the relative effects of light and heavy rollers, taking into account the number of turns required by each. 3. For effect, the wider a roller can be, the better, because the operation will be more quickly performed, and because, in pro- portion as it is narrow, will there be a tendency to force the broken stone laterally from under its action ; but, as the weight must be in proportion to its bearing surface, the width must be limited to a degree that will prevent that weight being too un- wieldy ; a very narrow roller might also have a tendency to over- turn. On the other hand, one that is very wide may take up toe- much room, if the road is open to traffic during the time of its use. 4. Horses should not be obliged to use very great exertions in drawing a roller, or the action of their feet will discompose the loose stones very inconveniently ; therefore, as the draught is very heavy at first, and never very light up to the last of the operation, they should not have more than from ten to twelve cwt. each to draw at first, nor so much as a ton each at last. 5. It would be desirable not to put more than four horses to such a machine, because, as the number of horses is multiplied, it becomes more difficult to obtain a perfectly united effort from them ; but on the above data a roller of four tons maximum weight might be too small for the best service, and as six horses may per- haps be applied without much inconvenience, it is proposed to give that number as a limit, and to allow 5 tons 12 cwt. as the maximum weight for the roller ; this, at 28 cwt. per foot of bearing, would give it a width of four feet. A short street (Herbert Street), made in 183fi, and +hen weJl rolled, has never required repair or new material since, up to 1843; it is a good street, but not entirely built on, nor a great thoroughfare. 304 APPENDICES. From the Continent we have records of several trials that have been made of late years of the effect of rolling new-laid material on roads ; although there are discrepancies in some of the particulars, there are many in which all agree ; and in all the practice has been strongly recommended. The one that seems to be the most practical is a roller described as first used in the Prussian provinces on the Rhine, and from thence introduced with some modification into a neighbouring district in France. It consists of a cylinder of cast iron of about 4 feet 3J inches wide and 4 feet 3 inches diameter.* On the axle by means of iron stanchions is fixed a large wooden case of 6 feet 4J inches long, 5 feet 8J inches wid.e, and 1 foot 8 inches high, open at top. This roller has a pole before and behind, in order to be able to draw it in either direction without turning ; the hind pole is sometimes used to assist in guiding it. It has also a drag, by the pressure of a board on its face, in the manner used for French waggons. The cylinder and other iron work weighs nearly 2 tons ; the case and woodwork 'about 19 cwt., making the whole 2 tons 19 cwt. The case will contain a weight of stone of 2 tons 19 cwt. when com- pletely loaded; therefore the entire weight can be brought up to ft tons 18 cwt. Six strong horses worked it well. It is passed over the entire surface of the road once or twice without any loading, weighing consequently nearly three tons, to obtain a first settlement of the loose material ; then one or two turns with about 1 J tons loading, making ^ tons ; and then the last turns, making ten in all, with the full loading, when it becomes 5 tons 18 cwt. Traversing 12 miles, it will thus completely roll about 3,000 square yards f in one day, or about a quarter of a mile of a road of 21 feet width. All accounts agree as to the absolute necessity, and the best manner of applying some gravel, or other sharp, gritty, very fine stuff on the surface, during the operation, without which it will not be thoroughly bound. The consolidation commences with the lower part, which is the first to get fixed and arranged ; and when, after about six turns over the whole, the upper layers have become tolerably firm and well bedded, some sand or stone-dust, or, what is best of all, sharp gravel, is very lightly sprinkled over it by degrees at every successive rolling, solely * These and other dimensions are necessarily in odd numbers, owing to reducing them from French measures and weights. t Some of the calculations are not strictly in accordance with the data, because the data themselves are not given in minute fractional parts, and consequently the reduction of the results will show a difference ; but it is very small and of no consequence in a general consideration of the matter. APPENDICES. 305 for the purpose of filling up the interstices of the broken stone, and not to cover it ; about 3 cubic yards in the whole per 100 square yards (equal to about an inch in thickness if spread over the whole surface) will be required. It is essential that this small stuff be not applied earlier, or it will get to the lower strata, and not only be wasted but injurious ; the object is that it should penetrate for two or three inches only, to help to bind the surface. Provided the upper interstices are filled, the less gravel used the better ; therefore it is applied by little and little after each of the three or four last successive passages of the roller, and then only over the places where there are open joints. After the work, if well done, is completed, it is stated that such is the effect, that the upper crust may be raised in cakes of six or seven square feet at a time, which could never be without the gravel. The effect may be improved also by having the upper inch or two of stone finer than the rest, say to pass a ring of 1 \ inch or \\ inch. This work should be done m wet weather, or the material will require to be proiusely watered artificially. It will be better that it should not absolutely rain, unless very lightly, when the gravel is applied (although the stoning should be wet), as it will cause it to adhere to the roller, and even at times to bring up the broken stone with it. In frost it is of no use attempting to roll. The state of the material, as regards its being wet or dry, will have great influence in the success of the operation. The form of the road will be best preserved by rolling from the two sides towards the middle, and not commencing along the latter. The calculated expense of the work in France was For six horses and two drivers per day .140 .For six labourers attending on the road, assisting at the roller, levelling inequali- ties, spreading gravel, &c. . . .070 Total for 3,000 square yards , * 1 11 Being about one penny for eight square yards, or one penny per running yard of road, twenty-four feet wide, and will amount to about 7 5s. per mile. For Ireland, these prices would have to be increased, thus For six horses, with drivers, per day . .170 For six labourers, at Is. 4d, . . .080 1 15 It is considered that a modification would be desirable in this foreign roller, by making it only four feet wide ; its weight might then be, with its box for the additional loading, &c., about 2 tons, which with 306 APPENDICES. an extra loading of 2 tons 2 cwt. would bring it to 5 tons 12 cwt. for its extreme weight, at 28 cwt. per foot. Such a roller passing ten times over every part, and working twelve miles per day, would require five days, and the operation costs about 8 15s. per mile of road, of twenty-four feet wide completed. The gravel ought to be considered as material, but in this case it is an addition to what would otherwise be applied; the cost therefore must be added. Suppose it to amount to one shilling per cube yard, the expense, at thirty-six square yftrds per cube yard of gravel will be about 19 5s. per mile of road of twenty-four feet wide. This would bring the whole to an amount of 28 per mile. However perfect the rolling may be, there will be at the end a slight elasticity and yielding of the surface, which will only become quite firm and hard after some days' traffic, say from six to ten when tole- rably frequented, during which its form and smoothness must be care- fully attended to ; add, therefore, 2 per mile for that extra work, and the cost will be 30 per mile. The expense of the operation of the roller (independent of the gravel) is so small, that if the weight is under- estimated, so that the width of the roller should require to be reduced to three feet, thus adding one-fourth to that part of the outlay, or that it would require to be passed a greater number of times more than calculated, that increase would not be of essential importance on the gross amount. If artificial watering should be necessary, that expense also must be added, but it would be small. The subsequent wear of material, under proper care, will be most trifling ; one French engineer states, that where the rolling in this manner has been successfully performed, there has never been a necessity for applying above one cubic yard of broken stone per 300 square yards of road in the next year ; that in one instance only one cubic yard per 1,500 square yards was used, although on a road subject to the passage of 400 horses in draft per day ; and on another road no fresh stone was laid for three year? To make a more direct comparison, however, of expense, it may be assumed that a much greater diminution of thickness will take place in the consolidation by the traffic than if effected at once by the rolling, because the narrow wheels of ordinary carriages penetrate into the loose matter, and force the lower part of it partially into the subsoil. The displacement and grinding and crushing is also very great; whereas, in rolling, the entire is preserved and in its proper place ; it may therefore not be too much to estimate, that if it require ten inches of loose material to bind into six inches by the ordinary process, as it probably would, eight inches well rolled would give the same ; if so, the saving at once would be very great ; thus, suppose the covering of APPENDICES. 307 one inch of stone to cost as much as two inches of gravel, that is, if the gravel is valued as above, at Is. per cubic yard, that the stone be valued at 2s., then we have four times the cost of the gravel, which was stated to be 20 per mile, or 80, to set against the 30, estimated expense of rolling. If the rolling effected a saving of only one inch of the broken stone, still the cost of that one inch would exceed that of the rolling, including the gravel. This last calculation is given only as a proof of the saving, and not as recommending the reduction of the mass of material laid down to a minimum; on the contrary, as the rolling of the surface is a final measure, and requires no renewal until the road is worn to a minimum thickness, the most economical plan probably would be to apply a con- siderable degree of substance at once, enough to last some years, so as to reduce the number of periods when rolling would be necessary. A roller, on what is here assumed to be the best construction, namely, of 4 feet bearing in one cylinder, and weighing 2 J tons, would not be very convenient to move from place to place for any consider- able distances ; an idea was therefore suggested of constructing one as a cart, on two wheels, the tires of which might be of 9, or 12, or 18 inches in width. Such a machine would weigh from one to two tons. The axle might be bent, and under the body, or might be straight, and pass through it, so that the loading of two or three tons might lie very low. The distance between the wheels should be a certain number of times the width of the tires ; namely, if the latter are 9 inches, the distance asunder might be 3 feet, or 3 feet 9 inches ; if 12 inches, they might be 3 or 4 feet ; and if 18 inches, 3 feet. Such a roller might be very convenient in many respects, but woulJ be subject to two objections : 1st, The tendency to force the material laterally from under its pressure, by the little width of each roller. 2nd, The impossibility by successive passages of giving every part of the road precisely the same amount of rolling; some parts must have more turns than others. In some few situations the very formation of the road may be made subservient to its rolling. In the construction of a new road over the Carey mountain, in the County of Antrim, material for stoning the road was quarried in several parts of the mountain up to its summit. Some carts were made with wheels of four-inch tires, and the laying of the broken stone being commenced close to the quarry, the work was carried on from each quarry down hill, the loaded carts being taken over the new-laid material, working systematically over 308 APPENDICES. the entire width of the road, and discharged below, returning up the hill light. By the time the work was completed the road had acquired in this way a considerable degree of consolidation without extra labour. A roller of the weight of five or six tons may be worked up inclines of one in twenty by increasing the number of horses, but not steeper ; if at all exceeding one in thirty it would probably be better to apply the roller in its lightest state, and increase the number of passages. It is very desirable to complete rapidly what is once begun, but it is attended with the disadvantage of taking up short lengths at a -time, which leads to the occasion for turning the roller very frequently, a manoeuvre that is particularly inconvenient. Although certain dimensions and weights are suggested to be likely to prove the most efficient, any other kinds that happen to be in pos- session might be tried and adapted to the above principles, which will usually require weight to be added with the successive rollings ; this may be done in various ways according to circumstances and situa- tions ; the most simple, but most cumbrous and troublesome of appli- cation, will be a large case on the roller for loading with stone. In or near towns, iron weights might be used instead of stone, partly on or suspended to the axle, within the cylinder, or in a case outside, which might be then much smaller, and the weight be more compact and more easily shifted ; or for use IN a town, when the most efficient dimensions and weights were ascertained, rollers might be prepared of two or three qualities, that is, all of the same extent of bearing, but of cylinders of different weights, from the lightest to the heaviest, and "brought in succession on to the work. APPENDICES. 309 APPENDIX II. EXTRACTS FROM "REPORT ON THE ECONOMY OF Eo AD- MAINTENANCE AND HORSE-DRAFT THROUGH STEAM ROAD -ROLLING, WITH SPECIAL REFERENCE TO THE METROPOLIS." (Printed by Order of the Metropolitan Board of Works.) BY FREDERICK A. PAGET, C.E. 1870. THE SPECIAL SUITABLENESS OF ROAD-ROLLING TO A MOIST CLIMATE. There is a point much in favour of an extended use of road-rolling in London and England generally ; making it probable that there is no other country in the world from which such good effects are to be expected from steam road-rolling. The climate of England, from its humidity and the best judges, Macadam amongst others, consider the presence of moisture as extremely favourable to road-making and its freedom from either long frosts or long droughts, is peculiarly favourable to the employment of rolling. In France and Prussia the rollers are used by preference in rainy weather, in order to save water- ing, as wet is essential for consolidating the metal. From want of experience in road-rolling some English surveyors object to the appli- cation of sand, which it is necessary to use of course in not too large quantities with rolling, in order to prevent the crushing of the metal. Upon the Holyhead roads, consolidated in the ordinary way, Telford always applied clean gravel to the new metal ; and this is now done in France, Prussia, and on all roads where road-rolling is understood and practised. The binding of rolled roads in Prussia by means of the sand, is so strong that united blocks 18 inches wide can often be removed. According to the official Hanoverian " special instructions " on road- rolling, when rolling the covering of the road, the binding or sand is only to be applied by the time that about two-thirds of the operation is completed. The amount of sand must be only about one-sixth of that of the metalling. The watering should be very carefully carried out, so that the bottoming be softened as little as possible. THE APPLICATION OF THE STEAM ROLLER TO THE FORMATION OF THE FOUNDATIONS OF NEW ROADS. France. Except with very heavy traffic, or with a very soft sub- stratum of the road, the Inyenieurs des Ponts ^ Chaussees do not use 310 APPENDICES. paved, and very seldom concrete, foundations or bottoming. They compress the bottom by rolling ; and then roll down upon it successive layers of road material about four inches thick. Germany. When rolling road bottomings, the German engineers do not take the roller over the sides, but rather lightly compress the crown of the road foundation. According to the official Hanoverian " special instructions " about rolling the bottomings of roads, this must be done with a light roller, and must not be carried out till the bottom is quite hard. No water must be used, and the operation should be carried out in dry weather, especially with a clayey soil. Liverpool. The thirty-ton steam road-roller, by Messrs. Aveling and Porter, was set to work in Liverpool in October, 1867. As we have already remarked, Mr. James Newlands, the borough engineer of Liverpool, has not yet been able, in so short a time for such a pur- pose, to determine the actual saving in maintenance effected by it. He writes us, however, that "besides its advantage in making a newly-coated macadam road perfectly smooth in a single night, it is of no less utility in forming the foundations of new roads. Formerly the traffic had to be turned on these foundations to consolidate ana render them fit to receive the protective coatings of paving or macadam respectively a work which took from three to six months, according to the locality. Now, when the foundation is laid, it can be rendered fit for paving and macadam in a day or two." There are upwards of forty miles of macadamised roads in the borough of Liverpool. SUBSIDIARY APPLICATIONS OF THE STEAM ROAD- ROLLER. Gravel Roads. As regards the application of the steam-roller to gravel roads, of which areas still greater than those of granite and flint roads are in existence, but little has yet been done except in parks in this direction, and certainly nothing has yet been published as to the economical results. Macadam wrote, some fifty years ago, " In the neighbourhood of London, the roads are formed of gravel," "the component parts" of which "are round, and want the angular points of contact, by which broken stone unites, and forms a solid body;" and large areas are still laid down on the outskirts of the metropolis. The only surveyor who has spoken from experience as to the economical value of rolling gravel roads is Mr. Howell, of St. James's. But, as has been well observed to us by Messrs. Amies and Barford, "If a garden-roller is useful in a degree, surely a much heavier one must be still more efficient for light vehicle traffic." Paved Streets. In Paris, successful, but partial, experiments have been made with the steam road-roller as a substitute for the paviour's rammer. APPENDICES. 311 Heavy rollers are very extensively applied in Hanover to pave- ments; and the official Hanoverian instructions for keeping up the roads contain several directions as to the time and way in which the roller is to be applied. Experience has shown that the rollers should be as heavy as possible ; that they should be applied to any uneven- ness of considerable area; that pavements are best rolled after the breaking up of frost ; and that the surfaces should first be swept, and then the rolling be continued until the stones no more give way. Herr von Kaven, in his " Lectures on the Engineering Sciences," gives elabo- rate figures, based on some experiments conducted in Hanover in 1863, showing the economy of this method of laying down and keeping pavement in repair. It is clear that rolling must also be peculiarly applicable to the construction of foundations for paving setts. Picking up Roads by Steam. The steam road-roller, though still a comparatively new implement, has thus already made itself useful in a variety of ways, and has adapted itself to several important functions in road-making and repairing, besides that of rolling only. "With respect to picking up, or " lifting " the macadam of old roads, before laying down a fresh coating of metal, there can be no doubt that a great economy of manual labour can be effected in this direction. Mr. Aveling's apparatus for "lifting" roads simply consists of a series of spikes bolted on plates let into holes in the periphery of the roller. A machine on this principle, the invention of Mr. Browse, the general surveyor of the Metropolis Roads Office, has long been used by Messrs. Mowlem, Burt, and Freeman, for picking up roads in the metropolis. It simply consists of a heavily-laden cart, having its wheels armed with spikes. At Manchester, one of Aveling and Porter's engines picked up an area of 2,048 square yards in three hours forty minutes ; and the cross-picking was afterwards done by hand labour, only equal to that of one man working sixty hours. This application of steam will no doubt lead to very great economies on many country and other roads that have been long wastefully managed. What Macadam wrote still applies in many cases. " Gene- rally the roads of the kingdom contain a supply of material sufficient for their use for several years, if they were properly lifted and applied" with the difference in favour of the present time that much of this metal has been previously broken. On any soft, ill-drained founda- tion, successive supplies are often driven in by the superincumbent traffic, the intermediate mud working its way up. Some road sur- veyors consider that many old roads, on which there is often as much as two feet of metal, might be mended by the steam-roller without using any new metalling. They could be picked up and then steam- rolled, with an excellent new surface as a result. Breaking and Conveying Metalling. It is obvious that the steam 312 APPENDICES. road -roller could be easily applied, not merely to driving stone- breaking machines, but also conveying the metalling to the spots where required. SUBSIDIARY ADVANTAGES OF THE STEAM ROAD- ROLLER. Saving in Sweeping and Scavengering. In all these instances, it is quite clear that any saving in road metal must be accompanied, as in the case of the substitution of pavement for macadam, though, of course, to a slighter extent, by a proportionate saving in the expenses of scavengering and removal of detritus. Diminution in Paved Areas. It is also clear that its well-considered and properly carried-out adoption in our large towns must tend, as in Paris, to narrow the encroachments of the uncomfortable, if cheap, paving stones. The main advantage, in certain situations, of pave- ments is in the diminution of cost of maintenance. At speeds higher than at a walk, General Morin's experiments prove that the resistance to draught on pavements increases, which is not the case with a well- metalled road in a dry state. RESUMfi OF THE MAIN ADVANTAGES OF STEAM ROAD-ROLLING. Briefly, the three principal advantages obtained by steam road- rolling over horse-rolling are, as observed in the French Official Report on the Exhibits at the late Paris Exhibition, when alluding to the Ballaison steam-roller, worked in Paris by Gellerat and Co., " saving of expense, saving in time from rapidity of execution, and a better quality of work." All that has been said with respect to the great diminution in draught by horse -rolling applies with much greater force to rolling by steam. But even if steam-rolled roads cost more instead of at least fifty per cent, less in maintenance than traffic-rolled roads, there would be good reasons for adopting steam-rolling, as the public is well served under the first circumstance, and very badly served under the second ; so that, even if the expenses were greater, there would be a balance of advantage and expense. As a corre- spondent, "H. R.," in 1868, observed in a letter to the Times : "When one considers the injury done to carriages and horses, and the delays caused by the practice of leaving the stones to be ground in by the carriage-wheels, one feels that even if rolling in the stones would be more costly to the parishes, the pecuniary gain to the public would be very considerable." In the words of Mr. Robert Mallet, the well-known engineer and author, writing in 1866, " there is still room for very great improvement in the laying and repairs of the macadamised streets and APPENDICES. 313 roads of the metropolis, although the practice has of late years in some respects improved. The admirable practice of Paris, where, with far worse material than is at our command, the results obtained are so much better, should be in part our school and example. " We have tardily adopted from thence the steam road-roller ; let its use be extended, and let us also adopt the admirable methods of cleansing, watering, and repairing of the French." The question is, as Mr. Tomkins, the surveyor of the roads in the district of St. George's Hanover Square, very neatly defines it, whether the roads shall be consolidated " by the traffic or for the traffic." ECONOMICAL RESULTS OF THE GENERAL USE OF STEAM ROAD-ROLLING IN THE METROPOLIS. Remarks on tlie Table. The annexed table* is almost sufficiently explanatory of itself. It has been made as complete as has been possible ; and when we state that, without counting the Parliamentary Returns, not fewer than forty-four different authorities have been applied to for returns, and that probably each different authority keeps its books on a different plan, some notion of the labour undergone in compiling the table will have been formed. It will be seen that the thirty-eight Metropolitan Vestries and District Boards of Works appear in the same order as usual in such returns ; the Vestries being separated from the Districts. If it had been possible, we should have preferred a different arrangement, based on other considerations. The great roads of London, and their minor feeders, radiate from the City as a centre, something like the threads of a completed spider's web. As we leave the centre, the main roads necessarily diverge farther apart, until lost in the surrounding country. If the metropolitan area were divided into small regular squares, or hexagons, of equal areas, it would be possible, starting from the City as a centre, to pretty closely classify portions of the divisions, now coming under the headings of parishes and districts, into four principal zones. Within a circle of about two miles radius from St. Paul's, enclosing an area of some thirteen square miles, we should find that all the streets are paved. Another circle, or perhaps rather an ellipse, with the Thames from the Tower to Somerset House as its major axis, would enclose the principal Guernsey granite areas ; farther out we should find, still of course deal- ing with the generality of minor roads, a zone mainly composed of flint roads ; until, at .the farthest distances from the City, and of course in regions of lesser traffic, we reached the outermost zone of gravel or pebble roads. It is of course at once seen that, as the traffic increases towards the centre of the web, the roads have to be made stronger and * The table here referred to is given in abstract at page 152, ante. D.K.C. P APPENDICES. stronger, until a centre is reached entirely paved with the hardest granite. We must note that the term "macadamised" is by some London surveyors strictly confined to a road laid with granite ; by others it is extended to flint roads, which again are, by some surveyors, put under the category of gravel roads. We have tried in the table to separate as much as possible the granite from the flint and gravel roads. It will be observed how the average width of the streets and roads increases in the wealthier parishes, and gradually diminishes in the poorer and older ones. In St. James's, Westminster, and Battersea, the roads are, on the average, 50 feet wide ; in Paddington we have an average width of 45 feet ; in St. George's, Hanover Square, 40 feet; and in St. George's in the East only 18 feet. The Present Extent of the Roads in the Metropolis. Mr. George Vul- liamy, architect of the Metropolitan Board of Works, in his " Memo- randum as to the Means of providing Fire-plugs in the Metropolis," assumes the length of streets in London at from 2,000 to 2,500 miles. This rough estimate includes the paved roads. Our table, based on thirty-six different returns, from as many parish and district surveyors, gives more than 1,100 miles of macadamised roads in the metropolitan limits, with an estimated total area of nearly twenty-three millions of square yards. These totals are really far below, rather than over, the mark, as they do not take into account the areas of private roads, of which there exist a great number, especially at the outskirts, such as towards Battersea, Lewisham, Islington, Hackney, and other districts. Besides, in the few cases, such as Marylebone and others, in which incomplete returns of lengths or areas could be obtained, the figures given have been estimated low. These totals also exclude the gravel footpaths, the areas of which are very considerable in many parishes. The public and private parks are also not included. In the former, Hyde and St. James's Parks alone contain eight miles of roadway under Government supervision. With Regent's Park, Battersea, Victoria and the other Parks, at the very least thirty miles of roads could be added. It will be observed on looking at any map, that the metropolitan area of parishes has grown into the shape of a huge bunch of grapes from the Thames as the stalk of the whole bunch. Between the fringes of its irregular contour are great areas of inhabited country, often as much covered with roads as the adjacent metropolitan parishes proper. There are thus immense areas of roads on the outskirts. For instance, Mr. Browse, surveyor of the Metropolis Eoads Office, has not less than seventy miles of macadamised turnpike roads, beyond the metropolitan area, but still in Middlesex. These seventy miles havo iui average width of 40 feet, without- reckoning the footpaths. APPENDICES. 315 Thus, taking the districts of Plumstead and Lewisham, the farthest from the centre of London, as giving a measure, in the ratio of the area of their roads to acreage, of the proportion of roads on the out- skirts, \ve should find that, within a radius of only twelve miles from Charing Cross, there are at least forty millions of square miles of macadamised roads. The Future Extent of the Roads of the Metropolis. As might naturally be expected, considerable areas of roads are being laid down every year in the metropolis, and especially in the outlying metropolitan districts. In many districts, such as, for instance, Wandsworth, at least two or three miles are laid down annually. The total population of 1861 was three times that of 1801, having trebled itself in sixty years. From the figures giving the population of London at the end of every decade since 1801, it can be seen that, since the beginning of the present century, the metropolitan popula- tion has doubled itself in about forty years. In fact, the Registrar- General estimates that the metropolitan population, which in 1866 was more than 3,000,000, will, at the present rate of increase, in forty years be doubled, or will rise to 6,000,000. Should the population double itself in the next forty years, the mean annual rate of increase, calcu- lating from 1865, during that period, will be about 75,000. After making every allowance for the "tendency to build houses much higher than was formerly the custom," and without taking into account the opposing and corrective tendency to lay out wider streets than formerly, Mr. Haywood estimates that " at the expiration of thirty- nine years hence, perhaps forty or fifty square miles of open country will be covered, more or less closely, with houses, for the additional three millions of inhabitants which will then exist. But," he goes on to say, " there are other causes which have arisen of late years tending largely to disperse and radiate that part of the existing population which is above the operative classes, the principal agency being the facilities for transit offered by railways. The tendency of that class, undoubtedly, is to seek cheaper residences and a purer atmosphere, and consequently to encroach still further upon the open area surrounding the metropolis, so that probably sixty square miles of open country, if not a considerably larger area, will be covered and occupied by the time the population reaches six millions." This would give an area one-half more than at present. The minimum Annual Saving to London Ratepayers derivable from the nse of Steam Road-rolling. After taking the most minute precautions against error, we feel some confidence that our figures of economy through road-rolling are much below the mark. "We see by the table that there are at least 1,126 miles of macadamised road in the thirty- nine metropolitan parishes and districts, and that their total area is at least 21,562,605 square yards. Of course, the true measure of the 316 APPENDICES. work to "be done would be a figure giving the exact cubic contents of exactly the same kind of metalling consumed in a given period of time over the given area. It is impossible to obtain this, though we know that the surfaces of many London roads have to be renewed several times a year. We find, however, as can be seen in the table, that there were expended on the roads and streets in the London parishes and districts during the /ear 1866-7 714,662; and during 1867-8, 781,003. From a number of returns forwarded to us by different London surveyors, we estimate that the average annual cost of main- tenance, exclusive of cleansing and watering, of the London maca- damised roads, is 280,750, or at the rate of. in round figures, 250 per mile. We have seen, from not less than seven estimates, that the saving in maintenance through horse road-rolling can be estimated at 40 per cent., and that the French, after nearly ten years' experience in Paris, consider that they save 50 per cent, in maintenance over and above horse-rolling. Therefore, even if we assume that only 50 per cent, is saved by steam -rolling, Tre have a gain of more than 140,000 per annum, without reckoning the diminution of scavengering and water- ing, and many other sources of economy. It will be observed that this sum is less than one-fifth of the grand total of that spent in 1867-8 on all the charges for the London roads ; and that, capitalised at three per cent., it represents more than four and a half millions. The Classes directly interested in the Application of Steam Eoad-r oiling. As every one uses roads, and is directly or indirectly a ratepayer, every one may be said to be more or less interested. This of course applies more directly to every owner of a horse, for the reasons we have already given at length, under the heading of saving in draught ; but still to every user whatsoever of the road. As a London local paper, the " Parochial Critic," observes, " an uncovered part is seized upon by every humane driver, and quite a contest takes place for possession of the smooth portion. When all is covered by loose sharp stones the drivers and horses yield themselves to their fate, and drag through the punishment as best they can. A return of the number and character of accidents occasioned by the present system of road-making for the last seven years would surprise the ratepayers." As usual in such cases, although the complaints of the wealthy are heard more loudly, the poor probably suffer most. It is said that, in some parishes, the habit is to lay down the loose stones in October, and to leave them to be consolidated by the passing traffic before the rich inhabitants return to town in February. The work of rolling is thus thrown on draught -carts, the vehicles of the poorer classes, and especially the omnibus the poor man's own conveyance. The dimi- APPENDICES. 317 nation in the work of traction produced by road-rolling certainly ought to direct the attention of the omnibus companies, now serving a passenger traffic in which they have to compete with the metropolitan railways, and soon with the tramways the proprietors of railway vans, such as those of Messrs. Pickford and Chaplin and Home, and of the railway coal-carts to this mode of saving wear and tear of horseflesh and vehicles. Any road whatever is neither more nor less than an apparatus for saving draught ; and the only advantage of a railroad is that it has so much less draught on its rails that it econo- mises the work of traction. The roads of a city are in fact the per- manent way and fixed plant of its means of locomotion. The Amount of Capital invested in Horses and Vehicles in the Metropolis. In order to obtain some farther measure of the extent of the classes directly interested in the application of steam road- rolling, we will form an estimate of what an engineer would term the rolling-stock of the London roads. According to recent fiscal arrangements, all horses whatsoever are taxed, and also all vehicles, with the sole exception of trade carts for burthens, on which are painted the name and address of the proprietor. This gives an easy means of getting at the number of horses and, approximately, the number of vehicles. The Honour- able Commissioners of Inland Eevenue have obligingly furnished us with figures, giving the number of horses, and of taxed vehicles, within a circle of a radius of 4^- miles from Charing Cross, and hence covering the greater part of the metropolitan area. There are thus 71,903 horses in all, and 24,095 taxed carriages, within this space. On account of the untaxed trade carts, we may safely raise the number of vehicles to 30,000. In the absence of any exact data, there may probably be different opinions as to the relative numbers of the different kinds ; but the following rough estimate will not be very far from exactitude, especially as we have verified the numbers we have estimated for the cabs and omnibuses by inquiry at Scotland Yard. Of light carts, vans, gigs, phaetons, 15,000, which, at the price of 30 each, would give a sum of 450,000 ; cabs, 7,000 at 35 each, give 245,000 ; omnibuses, 2,000 at 60, 160,000 ; heavy carts and wag- gons, 3,500 at 50 each, 175,000 ; carriages, broughams, &c., 2,500 at 120 each, 300,000. This would afford a total of 1,290,000 invested in vehicles ; and harness may be estimated at 40,000 sets, at the rate of 4 each, or 160,000. The depreciation of this capital is very heavy, as experts estimate that vehicles, the first prices of which are 20 to 150, cost from 4 to 15 per annum in repairs; and the wear and tear of harness is always very considerable. Each of the 15,000 light cart, gig, phaeton, or van horses we will take to require one horse, of the value of 20, or a total value of 300,000. Each of the 7,000 harder worked cabs would require two horses, as a horse must be in stable at least three days a week ; but 318 APPENDICES. their average value can only be taken as 15, or a total of 225,000, As few omnibus horses do more than one journey a day, and even then often rest one or two days a week, we probably have 30,000 horses to the 2,000 omnibuses ; and being of a somewhat superior kind, they must be estimated at the rate of 30 each, or a total of 900,000. Heavy cart or waggon horses, 7,000 at 60 each, and two per cart, will give 420,000 ; and carriage and brougham horses, two per vehicle, 5,000 at 100 each, wiU employ the sum of 500,000. On account of their inconsiderable number, especially in the centre of London; for we are only considering the area within a radius of 4^ miles from Charing Cross we have left out the riding-horses. The total capital thus invested in horse-flesh we may estimate at 2,345,000 ; and, adding to this 1,290,000 for vehicles, and 160,000 for harness, we have a total of 3,735,000, or of nearly four millions of pounds sterling, invested in the rolling-stock of the metropolitan highways. THE HISTORY OF HORSE ROAD-ROLLING. The first recorded allusion to road-rolling seems to have been made in the letters patent granted in 1619 to a certain John Shotbolte, who, without enrolling any specification, speaks of using "land stearnes, scowrers, trundlers, and other strong and massy engines," "in the making and repairing of highways and roads." In Partington's " British Cyclopaedia of the Arts and Sciences" (1835) we find it stated that it was soon after the year 1700 that a part of the charge of repairing roads was taken off the parishes and levied on the traveller by means of turnpike-gates. It was, nevertheless, a com- plaint that the roads were little, if at all, improved by the expenditure of the money so raised. This complaint is energetically advanced in "a dissertation concerning the present state of the high-roads of England, especially of that near London, wherein is proposed a new method of repairing and maintaining them," read before the Royal Society in the winter of 1736-7 by Robert Phillips, and printed in a small separate volume. " The author's great object is to recommend washing the roads by a constant stream, if possible;" at any rate, washing the materials in which he partly anticipates Macadam of which they are composed. Phillips strongly remonstrates against " the practice of laying down large heaps of unprepared gravel to be gradually consolidated into a harder mass, at the expense of the intolerable labour of the poor animals that are obliged to grind it down." His complaint has been re-echoed in London for more than one hundred and thirty years, but with little practical results. The first proposal of a road-roller, on a sound and scientific basis, was made by a French Royal Engineer of Roads ; and already, in 1787, M. de Cessart, then Inspecteur General des fonts et Chaussees, APPENDICES. 319 recommended a oast -iron roller for rolling down newly metalled roads, lie chose it 8 feet long. 3 feet in diameter, and of a weight of 7,000 old French pounds. Very soundly he compared its action on broken stone roads to that of a pavior's hammer on paving stones, and altogether his description is what might he expected from a scientific engineer. The AssembUe des Ponts et Chaussees strongly advised a careful trial ; but the disturbed political state of France, then in the throes of her first Revolution, evidently had the effect of preventing the continuation of the experiment. De Cessart's description is republished in the volume for 1844 of the jAnn ales des Punts et Chaussees. Philip Hutchinson Clay patented and specified in England, in 1817, " a large roller which, with the carriage over it, empty or filled with gravel or other material, may be made to press upon the road any weight from 6 tons to 20 tons, and by being used frequently, will press the moisture to the surface, and admit the sun and rain to act upon it, at the same time pressing down the loose material and making the road more even." That Clay under- stood the subject is evident from his using " a gravel cart," " fixed upon a roller, to carry gravel or other material used in the repairs of roads, and deposit it where necessary, the great advantage being that it improves the roads it passes over, and will be particularly useful in districts where material is scarce, whereas the present mode of carrying it in carts with narrow wheels destroys the road it passes over." This plan was afterwards carried out in Ireland by Sir John Burgoyne John Biddle patented, in 1825, "a combination of machinery for making, repairing, and cleansing roads, consisting of scrapers com- bined with rollers." The road-roller thus seems to have been used by several English inventors ; and it is acknowledged by the French that a sight of some at work in London revived the attention of Continental engineers to their use. A road-roller was tried as far back as 1826 on the road near Gottingen from Hanover to Cassel ; but the experiments did not succeed at the time, and it was only between 1837 and 1840 that Herr E. Bokelberg, now Wegbaurath (literally " Road-making Councillor"), practically introduced the process into Hanover. The first roller adopted in actual practice in France appears to have been one by M. Polonceau, the celebrated builder of the Carrousel Bridge, in 1829. This roller is said to have been built up of wooden lagging tied together with iron hoops, and weighted with stones, being, as to construction, long behind that of De Cessart. Wooden rollers, filled with stones, sand, or clay, were in use in France, about thirty years ago, in the department of the Marne et Loire ; and one of this kind is illustrated in a volume for 1840 of the Annales des Pouts et Chaussees. Rollers made of solid stone and granite were also used at the beginning. Some French writers state that the horse-roller was first re introduced from England into France, in 1833, by M. Fortin. Dr. Riihlmann, of Hanover, unknown to ourselves, arrived some years ago at the same 320 APrE.NDlCF.S. conclusion that road- rollers first began to appear in actual practice in 1830, or thereabouts; and that they were introduced into Hanover in 1837, or after their appearance in France. As we know, the Hano- verian administration of the roads has the reputation of being the very best in all Germany. The year 1830 is about the date given by Mr. YTgnoles as that of the general and official adoption by the French of Macadam's system of roads, of which they at once perceived that rolling is the indispensable concomitant. They hold that an unrolled macadamised road is only half finished ; and the centralization of the French, Hanoverian, and Prussian administrations of the roads has i'acilitated the universal combination of rolling with the system of macadamisation. In 1837, Mr. (now Sir John) Macneil patented the use of a kind of road consisting of pieces of cast or wrought iron mixed with broken stone and gravel. It was to be consolidated by horse-rolling in preference to wheel traffic. That, however, road- rolling was not much known in England by 1838, appears from an examination of Sir Henry Parnell's celebrated work on roads, the second edition of which was published in that year. The Eight Honourable Sir Henry Parnell, Bart., was constituted the Government Commissioner of the Holyhead roads, laid out by Telford; and the plans, specifications, and contracts drawn by Telford are printed in the body of the work. No mention is made of road-rolling as little as by Macadam himself, in the sixth edition of his work, published in 1822. It is well known that Sir John Loudon Macadam rather directed his attention to the construction of new roads and to restoring bad ones than to the maintenance proper of good roads. Some of his principles are no doubt applicable to maintenance as well as to making and repairing, but there is a good distance from that to a complete method. The first recommendation in the English language of horse road- rolling, as a measure of economy, was published by Colonel (now Field-Marshal) Sir John F. Burgoyne, E.E., in 1843, when Chairman of the Board of Works in Ireland. He was one of the first engineers who used it in their own work, and certainly the first in England to scientifically recommend it, not as a refinement, but as a necessity. Before 1840, as appears from the Verhandlungen des Vereins zur Befor- derung des Gewerbfleisses in Preussen, road-rolling was officially and universally introduced upon the Prussian roads. There can be little doubt that sooner or later all the rest of Europe will imitate Prussia in this, the only logical and truly economical mode of road mainte- nance, just as the Prussian authorities are being tardily copied in their first use of breechloaders, in their army administration, com- pulsory education, preliminary examination of patents for inventions, the stringent management of their civil services, and other matters. A great number of different constructions of horse-rollers are in use, designed with regard to facility in turning at the ends, facility for APPENDICES. 221 increasing or diminishing the load on the roller, and ease of draught. The most elegant of these contrivances is the roller ballasted with water ; obviously affording a very easy means for lightening or increasing the load. It is generally believed in England to be a homo invention, though really first applied and described in 1853, by a Government Engineer, Gr. Nell, employed in Prussian Silesia, THE HISTORY OF STEAM ROAD-ROLLING. In the presence of the settled employment in France of horse-rollers, it was natural to expect that French inventive talent should first seek for the usual substitution, in an implement in much requisition, of steam for horses. The limitation of the areas paved with " setts " in Paris and other large cities, said to be ordered by Napoleon III. as a timely precaution against street barricades, has also greatly favoured the employment of the steam-roller by raising the demand for a very high and durable class of macadam, especially for tho centres of Parisian traffic. The main impulse towards the use of the steam- roller in France was, however, the extraordinary economy in main- tenance found to be derived from horse, and a fortiori to be expected from steam, road-rolling. The saving to the public in wear and tear cf horse-flesh, vehicles, and harness, however great, could scarcely have so much stimulated the road authorities. The publication more than a quarter of a century ago of these economical results of horse-rolling has had little or no effect on our road management ; the yet greater advantages obtained in France from steam road-rolling may, however, possibly do more. The first patent for a steam road-roller was taken out in France, at the beginning of 1859, by M. Louis Lemoine, of Bordeaux. His roller was not patented in England, nor ever fully described in any publica- tion. At the meeting on the 13th of February, 1861, of the Societe d" En- couragement, the well-known engineer, M. Combes, is recorded to have described though his description is not given in the Bulletin of the Society Lemoine' s roller, which he had seen at work on the Bordeaux roads. M. Lemoine was then an employe of the Bordeaux municipality. This steam-roller appears to have consisted of one main roller for doing the work ; the engine, weighing in all only about ten tons, being steered by a pair of side wheels, the axles of which, by means of gear- ing, could be set at varying angles to the main roller. In 1860 M. Lemoine' s roller was tried on the road of the Bo is de Boulogne. Ballai- son's steam-roller, patented in France in August, 1860, and now in use by Gellerat and Co., the Paris Steam Road-rolling Company, was also there tried in August of the following year. In 1862 experiments wore carried out with both these steam-rollers by the French engineers of the State, M.M. Darcel and Labry, who gave an account of them in r 3 322 APPENDICES. an unpublished report of May of that year. They expressed strong opinions in favour of steam-rolling in general, giving the preference to the Ballaison roller. An extract from this report was printed in M. Homberg's Notice sur ks voles em^icrrees et asphaltce* de Puris (1865). They observe that the only possible apparent objection to steam-rollei a is their frightening the horses. We understand from M. Lemoine that since these trials he has made several different forms of rollers, now at work in Bordeaux and its environs, By this time the system has firmly established itself in Paris and elsewhere in France by a success extending over more than seven years. The results obtained in Paris during the couple of years after 1861 attracted little or no attention in England. The main reason for this is to be found in the slight importance attached by most of our road authorities to rolling. It is not regarded as a necessity but as a luxury. Quite independently of the French inventors, Mr. W. Clark, chief engineer to the municipality of Calcutta, in 1863 conceived the idea of a self-propelling steam road-rolling engine, with its weight uniformly distributed over the whole width of the rollers The different English traction engine-makers to whom Mr. Clark applied were unwilling to make designs and patterns for a single implement ; and at last Mr. Clark having consulted Mr. W. F. Batho, of Birming- ham, the engineering manager of Mr. Josiah Mason's vast under- takings, and well known in the profession for his original constructive talent, an engine on Mr. Batho' s designs was made by Mr. Worsdell of that town and sent out to Calcutta, where it has done good service. This design, patented in 1865 ly Messrs. Clark and Batho, was there- fore that of the first steam road-roller ever tried or patented in Great Britain. The chief features of Clark and Batho' s patent are the use of three sets of rollers, two in front acting as drivers, while the third is set up in a turn-table, being adjustable so as to steer the engine, at the same time overlapping the space between the two outside drivers. At about this time Mr. Batho went to much trouble in trying to pre- vail upon the road authorities of many of the larger towns to adopt steam-rolling, but without the least success. In a history of the sub- ject he must, however, figure as the first Englishman who fully com- prehended the national importance of steam road-rolling ; and we understand that, although, as we have stated, the first experiments were made in France, yet his steam road-roller was the first thoroughly successful implement of the kind. About a year afterwards, Ballai- son's steam-roller appeared on t*ie English patent lists as a communica- tion from Messrs. E. Gellerat and Co., of Paris ; and in May, 1865, that Company concluded a six years' contract with the Administration of Paris for rolling the roads of that capital. In the meantime the roller made b} r Mr. Batho for Calcutta had been at work there sinca 1861, more than repaying its entire first cost. The Bombay autho- APPENDICES. 323 rities accordingly instructed their agent to order one of Clark and Batho's engines, through Captain Trever, R.E., who, however, pre- ferred to employ Messrs. Moreland and Son, of London. These gentlemen accordingly made two engines which were sent out to Bombay. In 1866, Messrs. Eastons, Amos, and Anderson undertook some contracts for rolling roads. Mr. Anderson informs us that the firm "happened to have one of Aveling and Porter's 12-horse traction- engines, and it occurred to them to harness it to a 10-ton roller," 6 feet in diameter and length. They " commenced operations in 1866 by rolling the new roads of the Belvedere Estate," Belvedere, near Erith. The firm "subsequently made a contract with the First Commissioner of Works to roll the roads in Hyde Park," and, " during the autumn of 1866 and the early part of 1867" the engine and roller were kept in that park, though the Government officials were anything but favourable to its employment. It again illustrates the little atten- tion paid in England to what is being done abroad, that, though the city of Paris had, nearly a year before, concluded a six years' contract with the Paris Steam Road-rolling Company, yet this eminent firm employed a certainly ingenious, but comparatively ineffi- cient, combination for the purpose. It was soon found that the wheels of the traction-engine left deep ruts that were not made good by the roller, and that much time was lost in turning ; but that the dead load of the boiler and engine useless in the traction-engine could bo made to do good work in rolling. What with these observations, and the reports probably more or less current about the success of the Paris and the Calcutta rollers, Messrs. Aveling and Porter determined to adapt their form of traction-engines to this purpose. The result is a combination of their simple and efficient form of traction-engine with the arrangement of rollers and turn-table patented in 1863 by Messrs. Clark and Batho. Towards the end of September, 1867, Messrs. Aveling and Porter supplied a 30-ton steam road-roller to the borough authorities of Liverpool, where it has been ever since at work to the satisfaction of the engineer. Subsequently to this, Messrs. Moreland and Son hired to the Government the steam roller now [1870] in use in Hyde Park. In 1867 steam road-rollers began to be introduced into Hanover. The manufacture of the rollers on the Ballaison construction, was taken up towards the end of 1868 by Messrs. Manning, Wardle, and Co., of Leeds ; and, at. the beginning of last year, a paper giving a description ox the steam-roller used in Paris, was brought before the Institution of Mechanical Engineers, Birmingham, by M. Gellerat, of Paris y eliciting an interesting discussion on the whole subject. 324 AITEJJ DICES. w tf d _o if i; Jl i So "fl'i g'h, | g i^ p "o ci) i |J3 o'i si fi |I1 1 1 || OJ *s I- 1 ci5 o ^ | ^ ^i ill li '^2 PH^ g K j i >> 1 s S* FI 1| rt rt o o II ,g s <^ ri ^ $ lit Brunswick Rock Asphalte Paving Co. Val de Travers Asphalte Pav- ing Co. 6 P 6 P Limmer As- phalte Pav. Co. ; o J ^^ o - - . ."i g *5| o 2^1 H - H -4-3 ff\ ^ 7^ J5 WOQ 8 w o L C W HEAPS1DE ORNHIIL . ENCHURCH STRE Gracechurch Railway Pla RVCHURCH STRE Railway Place 1 FTTER LANE (N Bartlett'sPass born FTTER LANE (M IJartlett's Pa Meur-de-Lis RIDAY STREET . INSBURY PAVEM 1 M a is Q rH P-< -H R r^ ^ r^ 326 APPENDICES. ii s i 1 " CD b f- ^ 1 g * Q-J pS O> i 'i J -i o*"8,"| 1 d d d j l^gjl ^ o 4 '1 R o O 8 o I" 1 O iii ; l : t I PH E rCD o ^. M 2 s S 9 M ofl o " 2 * 1 *a gl| CD 1 o CD 1 R S 1 M PH O 1 d o d o d o 1 1 1 || 05 <0 00 CO M * S Po GO 1C o CN o (M CO 1 >5 V frj j| O s TJ p^ ^ "^ \ 02 "73 d ^ . E 11 JS s .2 i i j jT ^ 'eg I^H i i J2 r? 5 PH ^ P" CQ P- o ^^ CD EH S CD B H o S pq ^ rj ^h-l M v < A r^ J 1 cc Q O ECHURCH STRI T BELL ALLEY H m 's ARMS YARD e Street) STREET (Chea WlLLTAM ST tween Nichola: irid Clement's '. o e 33 "A x -q M P4 3i pj wJ B O c O X w 1^1 APPENDICES. 327 1 a I < ** K 02 I i|| * -i p o flJli d O o '1 1 O O dn o o >^ 02 2 . g* 1 -gii . 1 . Oj '1 1 S *Pn6 w o 8 fe o 1 1 A B 1 g 1 ID d d r^ O 03 ^H d d rv. O > ~^ ^ 3 > s a > BlFsif ' S^B pa d - H -7 J "S H ^ is f-t 2 ^2 Sg o 02 ^,^-4-3^ r ^fl W g 50 11! I OMDARD STRE ONDOX WALL OTHKURY 'AXSELL STREI 'AXSION Hous St. Mildred' Cornhill .ILK STREET 328 APPENDICES. CD -g >-, i ^ +3 f -M . ^ J J ^^ ^ S-| 1 |^| || -So o ,! fl If^ 1 w ^l^ll *g- 1 llll i*'l "cS 1 |.af| O ^ t> P^ O QQ |il PJ^ 02 Pn d .3 S i^ ^.S ^ iSi T3T3 O .S 3 & CD g ^ 1 \ 1 O 1*1 a M JQ W P ^> d) r^ "^ Pn g PH 1 1 rC ' o 5 T 72 U5 1>- CO J^ 05 1 _ ) - w ^'i 05 3 CO CM CO CO S ^>H r-T r-T r-T TJH" i CO ^-1 ^ i 1 ^ CD w c CO CO CO CO CO CO p^ 'fe'd r 1 T 1 1 T-H H 1 1 9 1 I O (T) 1 1 1 Pavement laid by Limmer As- ! phalte Paving Co. 5^t 1^ m Jt5 PP H Val de Travers Asphalte Pav- ing* Co. 6 P fi . .,-j . ^^ -* % o ll II J-3 Situation. 3NVJ OXIOXI OORGATE STREET Lothbury to Great Alley OORGATE STREET From Great Bell to near Coleman t Buildings OORGATE STREET From Coleman S B uildings to Rope: fetreet EW BROAD STREET . Vv r ormwood Stree Liverpool Street EWGATE STREET H ^ ^ SEi >q ^ APPENDICES. 329 - CM CO O O i O5 CO O rH CN rH CM ^ CM t~*- CO 5 CO i I -<*i i t CO GO CO OO CO CO CO CM CO CO oo 6 r-^F ( r-^r-^, | 11 "I 1 1 r-H ft jiiii i 1 i m ft 'S T3 > GQ ^ OQ 1 g 1 1 Q 1 9t" 3^.3 K ,i || | D r-y DQ fcD r i 5 >** g* If! !i ; J I '3 - A - 5 H 'oT fe, g 1 !^ d 1 H 22 g^'S H M H K r3 5S II H ^ a a 111 I ^1^ S SH LD JEWRY ^3 cc rt 1 CO ft ft o o PH rH 330 APPENDICES. a 4-3 id T^ JO 1 ^ O O S3 $ I t || R ' .S^tf ^ 1 ' ' 8 ! Jjffi 05 .& i 1 '3~ | > g '1 ' ^^ o ,d 3 o ^ 1 * 1 TJ CD CD t* 1 g j i-S i* 03 g "^ o ^~ .9 1 ^| 1 Jfo rt "g PI > } 1 !ll 6 ^ '> I rd 1^> 'i rd J~f CD S r^ *& d O 1^ in i o O % o +3 43 1 o o O ill ; ; ; ; TJ ; ; ; lr S CD 1|| cF 1 1 CD CD CD O i j-j ^ OO CO CO co Q "^^ t-T r-T cT to CM _ CD ^ to to ^ 1 I'd 00 CO CO CO OO CO 1 1 co oo CD^S V ~ oT r * ^ r -g i o | o o ^* ft H o p Ha ^ jS ^ CD _ i '3 d lid Pavement ! 6 P 1 I P 3 o^S SJ1 a^'fl M 6 ft o p I ^ ;* e ^j - '3^ rt . e rrt CD CD 39 C b Situation, BARIIICAN (Middle) . BARTHOLOMEW LANE . (North BARTHOLOMEW LANK. (South BIRCHIN LANE (South BlSHOFSGATE ST. \ViT BLOMFIELD STREET . BuiDEWiiLL PLACE CANNON STREET St. Paul's Churchya South-west come Abchuroh Lane APPENDICES. B33 M 1 M ^3 a o o |8 O O o o o 'o "5 o o *-* b o So o 1 O CO O H O O o 1 | S 8 1 o o o o O ^ 0) . ^ O CO CO CO oT <& d d ao ti ) CO o t-~ CO 4 1-3 & t? 1 ? 5 . ^ sphalticWood Pavement Co. i 1 ) 6 _, d i R I mpr o v e d Wood Pave- d d o d II f *"Nl d ^ **""* |s.sj id O f-i p. 1 1 o a f d P <1 a ii PH i*\ M ^ ^rd cj ;-( 'I D ' H z 11 d M o ^ J o > G Q O d o 1 i 5^ 05 H 25 . 1 KTTER LANE (> Fleet Street t Lis Court ORE STREET RACECHURCH S RF.AT TOWFR SEETHING LA OLBORN . OUNDSDITCH half) >f ^ L;WRT STREET l*i PH O W H w Hj APPENDICES. 335 i o o d o O d o o O 2 O o O Needs relay of the wh< surface. o O . . . - v si I s CO . . * 1 . . S O '"tn .y 5 o d ^ IE OB | o Pi o pi pi 'I s 1 ! o to S O^ o to i to i & & & * CS 4 CO o y> C7S CSI o ft s ^ s M 1 I If s^ a i i 03 'S* 1 KING EDWARD STREET KING WILLIAM STREEI KING WILLIAM STREEI ADELAIDE PLACE LEADENHALL STREET (Western half) LEADENHALL STREET (Eastern half) LITTLE BRITAIN LONDON HOUSE YARD LUDGATE HILL . NEW BRIDGE STREET . Pi %\ 2 i . , . O> f ao & o ||| 1 1 |-| o a o o "S o I 1 i O M H ^ ^ ^ ^ ,. 11 ji g* ^ o o 5s H go ~ (M * i .2^6 "S^d 2 r ~ l 6 >^3 . Pavemenl III |l << HH ** g"*2 ^^ ?H ^ o ^ t 1 * 1 O "3- ' og. -g^ ^ a 6 ^ fi 's X |! tj < 5rJ S < ^ ^ g g S 2 2 Situation. 02 05 * r1 S * H S^ tss ^ H> g ^ 03 ^ ^ S 0) &t 4 CO iO CO 10 ( 8 CO 00 oo ! 1 00 00 00 'o CD d 1* 1 I i CQ CD" 4 J* T4 > & CD rrt > 6 CD a l| Improve < Wood Pa ment Co. Mowlem & Improve Wood Pa ment Co. Mowlem & ^ O CH Mowlem & CD .rH > ;+3 00 ^ c3 . . . . , , "o bD . QD.g i EH CD O h3 CO H * S H H H M & "J|rf "g s EH 1 W 1 ^ & O P O ? 02 & t S (j Ej 5 H CD SI > S P o oJ pb w ^ {^ t^ WPk INDEX. A SPHALTE for pavement, 17, A 130 ; artificial asphaite, 131 ; wheel-tracks, 134, 214, 242 Asphaite, JBarnett's liquid iron, pavement, 245, 249, 254, 256, 258, 259 ; cost, 259 ; wear, 254, 256 Asphaite, Bennett's foothold me- tallic, pavement, 246 Asphaite, Lillie's composite, pave- ment, 247 Asphaite, Limmer mastic, pave- ment, laid in the City of Lon- don, 244, 249, 254, 256, 258, 259 ; cost, 237, 256, 259 ; wear, 254, 256 Asphaite, Maestu compound, pave- ment, 246 Asphaite, Montrotier com- pressed, pavement, 246, 249, 254, 256 ; cost, 246 ; wear, 254, 256 Asphaite, patent British, pave- ment, 245 Asphaltes, Societe Francaise des, pavement, by, 246, 249, 254 Asphaite, Stone's slipless, pave- ment, 246 Asphaite, Trinidad, pavement, 245 Asphaite, Val de Travers, pave- ment, laid in Paris, 242 ; in the City of London, 243, 249, 2o2, 256, 258, 259 ; cost, 237, 242, 244, 259 ; wear, 250, 252, 256 ; average wear, 257 Asphaite joints, granite pave- ment with, in Manchester, 199 ; in the City of London, 247 Asphaite pavement, wear of, 250, 252, 254, 256 Asphaite pavements, 242 : Yal de Travers asphaite laid in Paris, 242 ; asphaite pavements in the City of London, 243 ; cost, 259 ; in Manchester, 259 Asphalte-gramte pavement, in Manchester, 199, in the City of London, 247 Asphaltic wood pavement, 229, 235, 236, 237, 238 BALFOUK, ME. D., on the cost of maintenance of macadam roads at Sunderland, 161 Barnett's liquid iron asphaite pavement, 245. (See Asphaite, Barnutt'*, &c.) Bennett's foothold metallic as- phaite pavement, 246 Birmingham : Wear and cost of macadamised roads, by Mr. J. P. Smith, 155 ; comparative cost of macadam and paved roads, by Mr. W. Taylor, 158 ; cleansing the streets of, 273 Blackfriars Bridge, stone pave- ment, 173, 187 Bode, Baron de, on wood pave- ment in Russia, 13 Bokeberg, Herr E., on the inter- spaces in broken metal, 145 Boning rods, 134 Boulder pavement, 11 ; in the City of London, 173 ; in Liver- pool, 195; in Manchester, 198 INDEX. 339 British, patent, asphalte pave- ment, 245 Browne. Major James, on moun- tnin roads, 2 S3 Burgovne, Sir John, on road- rolling, 11; appendix, 301 CAREY'S wood pavement, 17, 217, 234, 236, 237, 238, 239 Cast-iron paving, 261 Cell alar iron pavement, 262 Chulk as a building material for roads, 90 Chaiie-Marsaines on performance of Flemish hors- s, 299 Clark, D. K., on the performance of horses, 300 Cleansing of roads and streets, 268. (See Roads and Streets, Cleansing of) Commercial Road, stone tram- ways in, 208 Concrete as a foundation for roads, by Mr. Hughes, 81 ; and by Mr. Pen fold, 81 Concrete pavement, McDonnell's adamantean, 247 ; Mitchell's, 263 Concrete roads, by Mr. Joseph Mitchell, 162 Cons'ruction of roads, 40, 79 ; earthwork, 40 ; working plan, 40 ; slopes, 43, 46 ; drainage, 46, 70 ; embankments, 47 ; catch-w;iter drains and cioss drains, 48 ; roads on the side of a hill, 48 ; side-cuttings, 49 ; spoil-bank, 50 ; maximum gra- dients, 63, 65 ; minimum gra- dients, 64 ; foundation and su- perstructure, 79 ; three kinds of foundation, 80; con Crete founda- tion, by Mr. Hughes, 81 ; and by Mr. Penfold, 81 ; foundation of pavement, by Mr. Tel ford, 83 ; broken stone first recom- mended by Macadam, 86 ; Mr. Hughes on broken stone cover- ing, 88 ; chalk as a binding material, 90 ; Mr. Walker on iron scraps, as a binding ma- terial, 91 ; foundation for paved streets, 109 ; materials employed in the construction of roadi and streets, 123 ; modern macadamised roads, 134 Contour lines and maps, 24 Cost of asphalte pavements, 259, 264, 265 Cost of macadamised roads, 151, 156, 158, 161, 162, 163 Cost of stone pavements : First cost and repair of streets in the City of London, 171, 183, 184, 18S; Euston pavement, 176; experimental paving in Moor- gate Street, 177 ; London Bridge, 186; Blackfriars Bridi^e, 187; Liverpool, 194, 195; com- parative costs, 265 Cost of Yorkshire paving, 265 Co.-tof Wood Pavements, 217, 236 Crossings, 112 Cunningham, Mr. J. H., on the cost of construction of macadam roads, near Edinburgh, Glas- gow, and Carlisle, 162 Curbs, 110 DERBY, cost of macadamised streets in, by Mr. E. B. Ellice-Clarke, 158, 161 Debauve, resistance to traction on common roads, 295 Dobson, Mr., on mountain roads, 284 Dockray, Mr., his opinion of the Kuston pavement, 176 Drainage, 46, 48, 74 Dumas on French roads, 19 ; on gradients of roads, 63 ; on mo- dern macadamised roads in France, 165 Dupuit, M., on the gradients of roads, 64 ; on the width of sur- face of tyres in contact, 144 ; experiments on resistance to traction, 295 TMRTHWORK, 40 JU Eastons and Anderson on resist- ance of carts and waggons, 297 Edgeworth, on the construction of roads, 9 340 INDEX. Edinburgh, Glasgow, and Car- lisle, cost of construction of macadam roads near, by Mr. J. H. Cunningham, 162 Ellice-Clark, Mr. E. B., on the cost of macadamised streets in Derby, 158, 161 ; estimate cost for paving and maintaining streets in Derby, 160 Embankments, 47 Estimates, taking out quantities for, 114 Exploration for roads, 21 ; laying out, 21, 36; contour-lines Hud maps, 24 ; working sections, 37 , MR. JOHN, on the Jj work of horses and the wear of roads, 140 Fences, 104. (See Hedges and Fences] Footpaths, 77, 112; stones used for, 128 Foster, P. Le Neve, Jun., on tram- ways in Milan, 208 Foundations for pavements, 108, 109 Foundation of roads, 79. (See Construction of Roads) France, roads in, 18, 165. (See Roads in France) pABRlEL'S wood pavement, VJ 233, 235 Gradients, maximum, 63, 65 ; minimum, 64 Granite, 122 ; comparative wear, 126 Granite pavement, artificial, Po- letti and Dimpfl's, 262 Granite tramways, 208. (See Stone Tramways) Greywacke, 127 TTARRISON'S wood pavement, n 231 Haywood, Colonel, Reports of, to the Commissioners of Hewers of the City of London, 170; earliest carriage - way pave- ments in the City of London, 171 ; pavements in 1848, length of carriage-way in 1851 and 1S66, three-inch granite sets m;nie the best pavement, 174; experimental paving in Moor- gate Street, 177; traffic in the City of London in Ls50, 1857, 1865, and 1871, 181 ; duration of three-inch set pavements, 182; cost, 184; on the wear and cost of paving for London Bridge, 185; estimate of dura- bility and cost of pavements in principal streets of the City, 187 ; typical sections of a fifty- feet street, 188; table, showing the condition of wood and asphalte carriage -way pave- ments in the City of London, on 1st February, 1877, 324 Hedges and fences, 104 ; stone fences, 104 ; post and rail fVnce, 10-5 ; quickset hedge, 105 ; Professor Mahan on fences, 106 ; Sir John Macneil on the evil of close fences, 106 ; Mr. Walker on the same, 106 Heivson's wood pavement, 231, 237 Hope, D. T., his experiments on wear of wood in pavements, 132 Horse tracks, 1 Hughes, Mr., on the form of the bed of a roa