East Hill into Dartford by the site of a Roman cemetery.¹⁰  At the foot of the hill the paved surface was uncovered in 1897, 2 ft. 6 in. below the modern road; it was said to resemble the paved road found at Strood, and the paving stones were set in gravel.¹¹ 
   Between Telegraph Hill and Shooter’s Hill, a distance of 10½ miles, the road is in a straight line, lying along the modern road, except at Dartford and Crayford, where it alters its course very slightly to cross the Darent and the Cray, and it is followed for a long way by parish boundaries. Discoveries of Roman relics have been made along its line, chiefly near Dartford and Crayford, where settlements grew at the river crossings, and Hasted says that in 1790 the road was plainly visible on Bexley Heath and at Welling. Before the road has reached Shooter’s Hill it has passed over the boundary of Kent; half a mile beyond Shooter’s Hill it could be seen in 1719, high, and with ditches on either side of it, and five years later, Stukeley remarked that portions of the road still existed.
   Near Kidbrooke, the modern main road bends southward, but the line of the road, followed by parish boundaries, can be traced along a secondary road to within half a mile of Blackheath ; on Blackheath the road was seen by Harris in 1719. From Blackheath to Southwark the line of the road can only be guessed at; in all probability it made a slight detour to the southward to avoid a wide area of marshland near Greenwich, and it is hard to see how a case can be made out for a route across Greenwich Park towards the mouth of Deptford Creek.¹²
   (2) ROCHESTER—MAIDSTONE—WEALD.—The next road for consideration is one which ran from Rochester by Maidstone, Staplehurst, and Hemsted across the Weald nearly to Bodiam in Sussex, a distance of over 25 miles; its course from Rochester to Maidstone has long been recognised, but it is only recently that Mr. O. G. S. Crawford has traced it as far south as Sandhurst,¹³  and thanks are due to him for most of the information contained in this account.
   The line of the road from Eastgate, Rochester, by Star Hill, Delce Road, and Dark Lane to Horsted is almost straight, and Roman remains have been found close to the road near Horsted (Pl. XXIX). At Bridgewood it joins another road from Rochester, but three-quarters of a mile further on it bends slightly south-east; from this point it is not visible on the ground, but air photographs¹⁴  taken in 1928 show clearly that it crosses three fields, falling in the same line as the trackway on the west boundary of Podkin Wood, and becomes a line towards the old chalk pits. If the alignment is now produced, it will be found to pass close to the site of Roman finds on Bluebell Hill and to meet the modern Maidstone road south of Warren Cottage, and thence its line is southward by Tyland to Sandling (by the site of burials near Abbey Court, Boxley) and Maidstone. Springfield Avenue (known sometimes as Radford Lane or Monkton’s Lane or Medway Mill Lane, and for long only a small trackway) leaving the Maidstone road just north of Springfield, probably led to a river crossing at Radford, and thus connected the villas at Buckland and Allington with
   ¹⁰  See page 136.                   ¹¹  Arch. Cant. xxiii, 8—9.
   ¹²  Professor Haverfield considered this suggested detour in Roman Britain in 1914 (B. Acad. Supp. Papers), p. 45. For a discussion on the line of the road in the neighbourhood of London, see Hut. Moni. Comm. Inv. Roman London, p. 50 ff.
   ¹³  But see Journ. R. Studies, xii, 277, and the Ordnance Survey Map of Roman Britain, 2nd edition.
   ¹⁴  Numbers 2,741, Chatham 03,526, Chatham 03,545 in the Ordnance Survey Collection.