The whole indicates no more than half a dozen graves. Farther west, towards Maxton, burials have been found at various places and dates since 1851. In particular, a field about a mile west of Dover market-place was dug for brick earth in 1867 and yielded several graves—a barrel-shaped urn 11 in. high, cordoned somewhat in Late Celtic fashion, which contained burnt bones, a largish urn of red ware with bones and a fibula, several smaller urns of black ware, some with bones, a skull with a bronze circlet, and some few coins of Severus, Postumus, and Constantine.⁷²  (3) On the north of the town near the London Road, which roughly reproduces the Roman road to Canterbury, still more has come to light. The museum possesses urns from Biggin Street, which seem in part at least sepulchral. Farther on, near the end of Bridge Street, Charlton, a builder found some burials in 1864. One large round-bellied ‘dolium,’ 22 in. high by 18 in. in diameter, covered with a tile, contained burnt bones and a graceful long-necked vessel of glass, said to be inscribed but illegible. A similar big urn contained a Samian saucer said to be stamped CNNTOS, burnt bones, and snail shells—presumably not Roman snails, but later intruders. A smaller jar of black ware, 14 in. high, contained only burnt bones. Scratches like XX or X and V were noticed outside it, but are probably only ornament.⁷³   Still farther along the same road, at Buckland, much was found, both when the railway was made in 1859 and subsequently. Scores, if not hundreds, of burials must have been unearthed and destroyed in 1859, most of them (it is said) consisting of a large urn with bones and two or three smaller urns near it. Two potters’ marks on Samian came to light, HABI?ISF, Habilis fecit, now in the museum, and FRONTINI, recorded by Mr. Payne.⁷⁴  As amphora handles and pelves, now in the museum, were also found here, it is possible that there were cottages as well as a graveyard.  (4) Pottery has also been found on the north or Castle side, both in Castle Street and on Castle Hill near the waterworks. But whether these belong to burials is doubtful.⁷⁵
   (iii) One other building of Roman Dover still survives to our day. It is famous, and deservedly, for it is probably unique among the remains of Roman antiquity in the world.⁷⁶ This is the pharos on the Castle Hill (P1. X). It has suffered many changes—in Saxon times probably a church tower, later a fortification, and later still a belfry with five bells, and for a while a Government storehouse, patched and added to by many builders, and again left until recently to fall into ruin. Its original object is, nevertheless, plain, and has not been doubted since Stukeley. It is a tower 62 ft. high. To the height of 43 ft. it is mostly Roman work; the top story and battlements, 19 ft. in all, were put on in the Middle Ages. The Roman portion, which alone concerns us, is a hollow structure. Internally it is 14 ft. square, with sides rising vertically. Externally it is an octagon, now gradually diminishing upwards in one uniform batter, but at one time with vertical walls receding
   ⁷²  Dover Museum; Arch. Journ. xxiv, 280; Arch. Cant. xviii, 204, plate nos. A, B, C. The locality is misstated in the latter passage, and in general it is hard to be sure which urns were found in which Dover cemeteries. But the difference to the result is not great.
   ⁷³  Arch. 7ourn. xxi, 183, hence Gent. Mag. 1864, ii, 18; Dover Museum.
   ⁷⁴  Arch. 7ourn. xvi, 297; Arch. Cant. xviii, 204 plate (part only); Dover Museum; photographs in C. R. Smith’s scrapbooks in Exeter Museum.
   ⁷⁵  Dover Museum; Arch. Cant. xviii, 204.
   ⁷⁶  The tower on Garreg hill, overlooking the Dee near Holywell, has been called a Roman lighthouse (Arch. Journ. lx, 254). It does not seem to be either Roman or a lighthouse.