embankment of the Roman road is very lofty and remarkably perfect.'*

In Scotland the Roman roads appear to have been from 18 to 24 feet wide, with a broad ditch on either side for the purpose of drainage, and to have been composed of various materials, according to the nature of the soil of the country through which they passed. Stuart tells us that where freestone could be used, it was shaped into square blocks, which gave the surface of the road on which they were placed 'something of the appearance of a well-built wall laid on its side.t Roy says:

'Where granite or dry stone of a hard and durable nature was found near at hand, there they seem to have paved their roads, forming them into a sort of rough causeway, not much elevated in the middle. Where the materials consisted of soft freestone or of coarse gravel, they appear to have disposed of them stratum super stratum, in the same manner as the modern turnpike roads were constructed. In other places, where stone and gravel were scarce that is to say, had to be brought from a distance, which is but seldom the case in North Britain-the Romans seem not only to have made their roads broader, but likewise higher, too, in proportion, from the promiscuous materials which the side ditches afforded, cementing them with a thinner coat of the hard stuff at top.'‡

The lesser cross-roads, country roads, and by-roads, though, as might be expected, constructed with much less care than the military ways which we have been *The Celt, the Roman, and the Saxon,' p. 221.

+ Caledonia Romana,' pp. 255, 256.

The Military Antiquities of the Romans in Britain,' p. 108.

considering, were yet made sufficiently durable to have left distinct traces down to the present day. They appear to have been sometimes paved with flag-stones, as in the case of a road traversing the hills near Monmouth, in which the stones are of all shapes and sizes, though carefully fitted together. One of the best specimens of a Roman by-road is the so-called‹ Fishwife's Causey,' which runs from Edinburgh to the seaside town of Portobello. The lesser roads, which have been referred to as having been made for commercial purposes, would also seem to have been constructed much in the same manner as the lesser cross-roads.

The Roman military roads were carried over the rivers which crossed their route by an extensive system of bridges. As early as the time of Cæsar the Roman armies were provided with bridging equipment, consisting of platforms of timber supported by wickerwork vessels, covered with skins of animals, while bridges on piles were constructed in the conquered provinces in connection with the military ways, one of the most remarkable of which was that built by Cæsar himself across the Rhine.* Vegetius, who wrote in the reign of Theodosius (A.D. 386), says that the Roman armies used to carry with them small boats hollowed out of the trunks of trees, together with planks and nails for the purpose of constructing temporary bridges, which were bound together with ropes. In the permanent bridges the width of the passage way was usually narrower than that of modern structures of the kind, and corresponded with the road leading to and from it. It consisted of three parts-a central road for horses * 'Military Bridges,' by Sir Howard Douglas, p. 88.

and carriages, which was called agger or iter, and two raised footpaths on each side (decursoria), protected by parapet walls or balustrades. Mr. Wright is of opinion that a large number of the Roman bridges were still in existence at the time of the Norman Conquest, and it is certain that the remains of many formed the foundations of modern structures. They appear for the most part to have been built of timber upon stone piers without arches. There is, however, a semicircular arched bridge over the river Cock, near its junction with the Wharfe, about half a mile from Tadcaster (Calcaria) on the Roman road leading southward from the town, which Mr. Roach Smith considers to be of Roman workmanship. On some of the stones of this bridge the mason's mark, an R, is still distinctly visible. London Bridge, which, judging from the coins found in a continuous series in the river, was built early in the Roman occupation, probably at first consisted of great beams founded on piles. The piers of the old bridge over the Tyne at Newcastle, which was taken down in 1771, were of Roman masonry, the foundations being laid on piles of fine black oak still in a state of perfect preservation, and decided evidence of the Roman origin of the structure was furnished by the discovery of coins of Hadrian, which must have been buried for sixteen centuries since its erection in A.D. 120, as well as of Trajan, Antoninus Pius, Faustina the Elder, and other succeeding Emperors, which were probably deposited during alterations and repairs. Roman work, equally well preserved, was also discovered on the destruction, in 1815, of the bridge over the Teign in Devonshire, by means of which the Roman road to Totnes and Ply

mouth crossed the river. Dr. Bruce says that the foundations of three Roman bridges still remain in the district of Hadrian's Wall one over the Tyne at Corbridge (Corstopitum), another over the North Tyne at Walwick Chesters (Cilurnum), and the third over the Rede Water at Risingham (Habitancum). None of these in his opinion had arches.

'The piers are of size and strength sufficient to withstand the thrust of the waters without the aid of an arch, and in at least one of these cases the requisite spring of the arch would have raised the road to an inconvenient height. An experienced mason, who examined carefully the ruins of the bridge at Habitancum, told me that he observed that all the stones which encumbered the spot were square, none of them having the shape of stones used in building arches.'*

Another example of a Roman bridge of similar construction was in existence little over a hundred years ago, which connected Caerleon with a hamlet, still known as Ultra Pontem, on the south side of the Usk, and Archdeacon Coxe, writing at the end of the eighteenth century, mentions that he nearly fell into the river owing to the looseness of the planks. It is stated by Pliny in his Natural History' that it was an article of religious faith with the Romans never to nail down the planks of a bridge, but the more probable object of this practice was to facilitate their immediate removal on the approach of an enemy.†

* 'The Roman Wall,' by the Rev. J. C. Bruce, p. 102. + Proceedings of the Cotswold Field Club, vol. xii., pp. 14, 15.

CHAPTER VIII

MILESTONES, POST-STATIONS, AND ROADSIDE INNS

Road measurement-Difference between Roman and English miles-Varieties of Roman milestones (milliaria) found in Britain-Milliaria of twenty-one Emperors, from Hadrian to Constantine Junior, discovered in various counties-The milliarium aureum at Rome-Provincial milliaria aurea— Claims of London Stone' to the title-Governmental poststations-Roadside inns-Posting-stations and inns enumerated in some of the Roman road books-Defects of Itinerary of Antoninus in this respect-Inns rarely used by wealthy travellers--Landowners sometimes built roadside taverns for the sale of their wine and farm produce.

THE distances on the Roman roads were made known to the traveller by milestones, usually called milliaria, but sometimes lapides. The former term was derived from the length of the Roman mile, which consisted of 1,000 paces (mille passuum); the latter was used in a more familiar sense, as may be gathered from the fact that we find the phrase ad tertium lapidem or ad tertium used to express the distance of three miles from Rome, and a station on the coast road, between Bittern (Clausentum), near Southampton, and Richborough (Rutupia), called Ad Decimum, to denote its being ten miles from Chichester (Regnum).

The exact length of the Roman mile, according to